Thesis Nordmark

Structural and Functional Changes in the Brain after Surgically Repaired Median Nerve Injury

Per Nordmark

Department of Integrative Medical Biology, Physiology Section Department of Surgical and Perioperative Sciences, Section for Hand and Plastic Surgery Umeå University, Umeå 2019

Responsible publisher under Swedish law: the Dean of the Medical Faculty This work is protected by the Swedish Copyright Legislation (Act 1960:729) Dissertation for PhD ISBN: 978-91-7855-072-2 ISSN: 0346-6612 New Series No 2035 Electronic version available at: http://umu.diva-portal.org/ Printed by: Cityprint i Norra AB, Umeå SE-901 87 Umeå, Sweden

Figures 1-8: Illustrations by Per Nordmark

Image on p. 1: Zoom in on Adam and Gods hands from ‘The Creation of Adam’ by Michelangelo Buonarroti, fresco in the ceiling in the Sistine Chapel – (private photograph).

N.B. A peculiar interpretation of this fresco’s depiction of God is that suggested by Frank Lynn Meshberger; that the outlines of God and the persons surrounding him make up the contour of the midsagittal plane of a brain and that this was a tribute by Michelangelo to the brain and the intellect (Meshberger, 1990).

Image on p. 39: Studies of arms and hands, 1513-1520, Michelangelo (Public domain) Image on p. 43: Studies of folded hands, 1511-1512, Michelangelo (Public domain) Image on p. 55: Scheme for the Sistine chapel ceiling; studies of arms, 1508, Michelangelo (Public domain) Image on p. 61: Anatomical studies of arms and shoulders, 1513-1520, Michelangelo (Public domain) Image on p. 77: Studies of an outstretched arm for the fresco ‘The Drunkenness of Noah’ in the Sistine Chapel, 1508-9, Michelangelo (Public domain)

To my family

Table of Contents

Abstract ...... iii Abbreviations ...... iv Original Papers ...... v Introduction ...... 1 General introduction ...... 2 Short historical review ...... 4 History of peripheral nerve injuries and their treatment ...... 4 Historical aspects on understanding the function of the brain ...... 8 Normal hand function ...... 11 Peripheral sensory mechanisms supporting manual dexterity ...... 12 Cortical brain areas associated with somatosensory and multisensory processing ...... 16 Cortical motor areas ...... 23 Prefrontal cortex ...... 26 Hand function after peripheral nerve injury ...... 29 Classification of nerve injuries ...... 29 Reinnervation after surgical repair ...... 31 Median nerve injury...... 32 CNS changes after peripheral nerve injury ...... 34 Findings in monkeys ...... 34 Findings in humans ...... 36 Aims of projects within this Thesis ...... 39 Specific Aims ...... 40 Materials and Methods ...... 43 Brief description of methodology and rationale for each paper ...... 44 Paper I ...... 44 Paper II ...... 45 Paper III ...... 45 Material ...... 47 Participants, inclusion and exclusion criteria ...... 47 Methods ...... 48 Magnetic Resonance Imaging – a brief description ...... 48 Functional Magnetic Resonance Imaging– a brief description ...... 49 Voxel-Based Morphometry – a brief description ...... 50 Novel apparatus for tactile stimulations in the MRI-scanner ...... 50 Data acquisition and analysis ...... 50 Ethical considerations ...... 53 Results ...... 55 Summary of Results presented in Paper I ...... 56 Summary of Results presented in Paper II ...... 57 Summary of Results presented in Paper III ...... 58

Discussion ...... 61 Paper I ...... 63 Paper II ...... 65 Paper III ...... 66 Synthesis ...... 69 Reliability ...... 70 Statistical choices and issues ...... 72 Future directions ...... 74 Conclusions ...... 77 Acknowledgements ...... 79 References ...... 80

Abstract

Despite the best available surgical repair, traumatic median nerve injury within the forearm typically causes lifelong impairment in hand function. This stems from an inadequate reinnervation of the nerves supporting sensory functions of the thumb, index and long finger, and of nerves supplying intrinsic hand muscles. This thesis examines whether median nerve injuries can cause structural and functional changes in the brain. Understanding such changes can help the development of new treatments for improved recovery of hand function. The first study introduces a novel apparatus and paradigm for examining tactile neural processing with fMRI under well-controlled behavioral conditions. The scientific issue challenged was whether, in healthy adults, different cortical areas could be involved in processing tactile stimuli depending on their temporal frequency content. In a threshold-tracking paradigm, the participants’ task was to detect oscillatory mechanical stimulations of various frequencies delivered to the tip of either left or right middle finger. Regardless of stimulated hand, tactile detection of audible frequencies (20 and 100 Hz) engaged the left auditory cortex while detection of slow object displacements (3 Hz) engaged visual cortex. These results corroborate and advance the metamodal theory of brain function, which posits that brain areas can contribute to sensory processing by performing specific computations – those for which they are specialized – regardless of input modality. The second and third studies concern structural and functional changes in the brain of adults with one reinnervated hand after an injury transecting the median nerve in the forearm. Healthy individuals matched for sex, handedness and age served as controls. Irrespective of side of injury (left or right), voxel-based morphometry applied on T1 MR-images revealed reductions of gray matter in the left ventral and right dorsal premotor cortex, and reductions of white matter in related commissural pathways. We interpreted these as activity-dependent structural adaptations to reduced neural processing linked to restrictions in the diversity of the natural manual dexterity repertoire caused by a disturbed innervation of the hand. Conversely, increases in gray matter were observed bilaterally in a motion-processing visual cortical area. We interpreted this as a structural manifestation of increased neural processing linked to greater dependence on vision for controlling manual dexterity due to impaired tactile innervation of the affected hand. To reveal functional changes in tactile cortical processing after median nerve reinnervation, we recorded brain activity using fMRI when study participants performed perceptually demanding tactile threshold-tracking and oddball detection tasks with our novel apparatus. The hand representation of the contralesional primary somatosensory cortex (S1) showed greater activity compared to the controls when the reinnervated index finger was engaged in the tasks, but strikingly also when fingers of both hands innervated by uninjured nerves were engaged, i.e., little finger of the reinnervated hand and index and little finger of the other hand. The generally increased activity indicates a general disinhibition of contralateral S1, suggesting that increased functional reorganization is an ongoing process of chronic nerve injury. In addition, prefrontal areas implicated in processes that support decision-making and response selection showed increased activity, suggesting that such processes were more computationally demanding after nerve injury. Together, these results indicate that brain areas can undergo significant changes after peripheral nerve injury, even when followed by best available surgical repair and reinnervation conditions. These changes can include activity-dependent structural adaptations consisting of either regional decreases or increases in gray matter concentration, which likely depend on an area's functional specialization and on changes in its processing load due to behavioral constraints imposed by the injury. Moreover, the results also suggest that the affected hand's primary cortical projection area is still in a state of ongoing functional reorganization despite the fact that peripheral reinnervation of the hand should have been completed long ago, which should inspire the development of new therapeutic regimens for what today is considered a chronic impairment. Key words: Humans, Hand, Touch, Peripheral nerve injury, Magnetic resonance imaging, Somatosensory Cortex, Cortical plasticity

iii

Abbreviations

A1 Primary auditory cortex ACC Anterior cingulate cortex ANOVA ANalysis-Of-VAriance BA Brodmann Area BOLD Blood oxygen level dependent CCZ Caudal cingulate zone CM Corticomotoneuronal CMA-r/-d/-v Rostral-/dorsal-/ventral- cingulate motor area CNS Central nervous system DL-/DM- PFC Dorsolateral/dorsomedial prefrontal cortex EEG Electroencephalography FA-1 Fast adapting type 1 afferent FA-2 Fast adapting type 2 afferent fMRI functional Magnetic Resonance Imaging FWHM Full Width at Half Maximum LMN Lower motor neurons LTMR Low-threshold mechanoreceptor M1 Primary motor cortex MEG Magnetoencephalography MNI Montreal Neurological Institute MRI Magnetic Resonance Imaging MST Medial superior temporal area OP Parietal operculum PMd Dorsal premotor cortex PMv Ventral premotor cortex PNS Peripheral nervous system PPC Posterior parietal cortex Pre-SMA Pre-supplementary motor area RCZ-a/-p Anterior-/posterior- rostral cingulate zone S1 Primary somatosensory cortex S2 Secondary somatosensory cortex SA-1 Slowly adapting type 1 afferent SA-2 Slowly adapting type 2 afferent SMA Supplementary motor area SMC Sensorimotor cortex SPM Statistical Parametric Mapping SVC analysis Small volume correction analysis TMS Transcranial magnetic stimulation V1 Primary visual cortex VBM Voxel-based morphometry VL-/VM- PFC Ventrolateral/ventromedial prefrontal cortex V5/hMT+ Visual area 5/human middle temporal visual complex

Original Papers

The thesis is based on the following papers that will be referred to by their Roman numerals

I BOLD responses to tactile stimuli in visual and auditory cortex depend on the frequency content of stimulation PF Nordmark, JA Pruszynski, RS Johansson, Journal of cognitive neuroscience 24 (10), 2120-2134 (2012)

II Structural changes in hand related cortical areas after median nerve injury and repair PF Nordmark, C Ljungberg, RS Johansson, Scientific Reports volume 8, Article number: 4485 (2018)

III Disinhibition of human primary somatosensory cortex after median nerve transection and reinnervation PF Nordmark & RS Johansson, Manuscript (2019)

All papers are reproduced with permission from the copyright holders.

Introduction

A man paints with his brains and not with his hands. Michelangelo (1475-1564)

General introduction The brain’s skilled use of our hands is behind many of the cultural and technological achievements that mark us as humans. The brain and the hands are close associates in two important and closely interconnected functions: the exploration and reshaping of the physical world. This functional collaboration requires that the brain, via peripheral (afferent) nerves, obtains rich information of mechanical events that take place when the hand interacts with objects and a highly developed central neural control system that, via peripheral (efferent) nerves, can orchestrate the musculature of the hand and arm. Accordingly, injuries to peripheral nerves that supply the hand cause impaired hand function due to immediate and detrimental loss of tactile sensibility from the normally innervated skin areas and paralysis of muscles innervated by injured nerves. Such injuries to peripheral nerves can be repaired with microsurgical techniques that re-adapts the interrupted nerve sheet. Successful surgical intervention enables nerve fibers (axons) to re-grow out over the injury site and further from the injury site at a speed of about 1 mm a day for later reinnervation of the affected skin and muscles. The reinnervation is necessary to regain sensory and motor functions of the affected nerves. However, not all axons survive to regenerate, which results in fewer axons reaching reinnervated areas. Furthermore, during regeneration, axons are misguided to varying degree, which results in mismatches concerning the location and type of end-organs of reinnervated axons compared to before the injury. These and other factors affect both sensory and motor functions and generally lead to lasting impaired sensorimotor control of the affected hand even after state-of-the-art treatment (Lundborg and Rosen, 2007). Age, severity and site of injury together with the delay in surgical repair significantly influence the functional outcome. A follow-up time of 2–3 years is considered necessary to evaluate the outcome of treatment of the injury (Bowlby, 1890; Nicholson and Seddon, 1957; Sunderland, 1968; Ruijs et al., 2005). The lasting impairments consist of various degrees of altered sensation: affected persons may show increased touch thresholds, have trouble localizing tactile stimuli in the affected area of the hand, lack the ability to appreciate the three-dimensional form of objects by touch (stereognosis), and to various degrees also experience cold intolerance and pain. Mostly related to the sensory impairment, nerve injury leads to clumsiness of the affected hand and especially problems with everyday tasks that require fine motor skills, e.g., buttoning a shirt, tying shoes or performing hand actions without seeing their affected hand.

Nerve injuries in the upper limb typically result from accidental blunt or sharp trauma, e.g., from traffic accidents or work-related injuries, but also from complications at childbirth, from poorly performed surgery (iatrogenic injuries) and self-inflicted injuries (Ciaramitaro et al., 2010). Based on Swedish statistics, it is estimated that trauma to peripheral nerves innervating the hand affects more

than 80,000 persons every year in Europe (incidence 11 per 100 000 person- years; Asplund et al., 2009). Nerve injuries often occur in rather young individuals (average age 28 years; Asplund et al., 2009) and the resultant lifelong impairment in hand function affects daily activities and may prevent the injured person from returning to work (Bruyns et al., 2003; Dahlin and Wiberg, 2017).

The principal goal of this thesis was to further the understanding of how the brain might change its neural processing in response to peripheral nerve injury in the upper limb. For this purpose, I have investigated persons who suffered a traumatic total transecting injury to their median nerve in the distal forearm followed by state-of-the-art surgical repair. More than 2 years after the injury and surgical repair, they were assessed with respect to structural changes in the brain by using voxel-based morphometry (VBM) and functional changes in the brain by using functional magnetic resonance imaging (fMRI). The overall objective of the thesis work was to contribute to a better understanding of how changes in the brain relate to how nerve-injured persons recover hand function and to inspire further studies of new aspects of structural and functional plasticity that can help the development of new, more efficient, treatments.

The main theme of the thesis thus concerns neuroplasticity, and more specifically, brain plasticity at adulthood associated with a defined peripheral nerve injury (Paper II and III). However, the first study, presented in Paper I, deals with healthy adults and introduces a novel apparatus and paradigm for examining tactile neural processing under well-controlled behavioral conditions using fMRI. This development also laid the foundation for the study of functional changes in the brain following median nerve injury (Paper III). The scientific question addressed in Paper I relates to the specificity with which different brain regions process information from different sensory modalities, in this case, different sub-modalities within the sense of touch.

In the Introduction sections that follows, I first briefly present historical aspects of the treatment of peripheral nerve injuries and of how our understanding of brain function has developed, with emphasis on the human cerebral cortex. Then, I present current views on the organization of the nervous system related to hand function under normal healthy conditions, with a focus on peripheral sensory mechanisms and cortical organization of brain areas involved in goal-oriented hand actions. Finally, I address clinical aspects of injuries to peripheral nerves innervating the hand, with emphasis on the median nerve, and current views of how the central nervous system might be affected by such injuries.

Short historical review

History of peripheral nerve injuries and their treatment One of the oldest known recommendations regarding treatment of peripheral nerve injury is that of the Greek physician and philosopher Galen of Pergamon (AD 129-216, Gross, 1987): he discouraged suturing of severed nerve ends since it would lead to unbearable pain and “epilepsy”. Paul of Aegina (AD 625-690) was the first to describe approximation of nerve ends with wound closure. He lived and worked during the Byzantine era and in his epical work Medical Compendium in Seven Books (Aegina, 1914) he emphasized the importance of not further injuring the nerve when taking care of the wound and advocated avoidance in piercing the nerve with the needle when adapting skin and subcutaneous tissue. Thus, instead of direct repair by suturing the nerve, he recommended bringing the nerve ends closer through careful adaptation of the surrounding tissues. He also seemed to have understood the close relationship between peripheral nerves and the brain: “But nothing may touch the wound, for the nerve is very sensitive and cold in its temperament, and forms union of the principal parts with the brain.” “When the nerve is completely cut, there is no danger, but the limb becomes power- less” Over more than a thousand years it was debated whether to suture nerve injuries or not. Although successful outcomes after nerve suture were described in the early 19th century, the results were probably exaggerated and reflected poor understanding of how to assess surgical outcomes after nerve injury and repair: “Many experiments have been made by physiologists, to prove that when a nerve is divided all sensation and motion are lost in the parts to which it was distributed, and that after the reunion of the divided parts it performs its functions as well as before.” Joseph Swan, 1820 (Swan, 1820) “…very frequently in fact the phenomena of sensation and motion which had been believed to be permanently destroyed, are ultimately more or less perfectly reestablished.” Alfred LM Velpau, 1845 (Velpeau, 1846) The problems with severe postoperative pain, fear of painful chronic neuralgias and the lack of functional restitution kept many renowned surgeons from attempting nerve repair and instead they recommended cutting out the nerve (neurotomy) or amputating the limb/finger(-s) affected by an injury (Guthrie, 1820).

The at the time prevailing idea that the nerve fibers (axons) in the two stumps of the nerve would simply re-unite immediately with the suturing of the nerve, through so called prima intentio nervorum, might have obscured the assessment and realization of the actual post-operative results. The exaggerated

interpretations of successful results within 2-3 weeks after the suturing of injured peripheral nerves are ground for hesitation as to whether there were any successful nerve repairs in the 19th century (Trendelenburg, 1923). Furthermore, some of the techniques of suturing peripheral nerves at that time also indicate the understanding of the physiology of nerve signals and the concept of nerve regeneration after peripheral nerve injury was very limited (see Fig. 1).

Figure 1. Old discarded, and currently used techniques for nerve suture

Historical exposé of suturing techniques used for adaptation of severed nerve endings. It is easy to stand in hindsight today and say that some of the techniques that were suggested would never work. A: The severed nerve, B: Ferara´s side-to-side nerve suture from the 17th century; C: Wedge suture, among those “shown only to be condemned” in (Wolsey 1907); D: End-to-end with sutures passing through nerve fascicles; E: Rawa´s co-apted suture, with nerve stumps sutured together but without any possibility of regeneration of axons; F: epineural microsutures, the current state-of-the-art-technique. For injuries with significant gap between nerve ends, that cannot be closed without tension in the nerve; G: The intermediate suture or ‘Suture á la distance’; H: Neuroplasty for bridging the gap presented by Woolsey after grafting-failures; I: Grafting with autogenous nerve material (from same individual), the currently used technique when nerve injury has significant gap. References; B: (Browne, 1951) C: (Woolsey, 1907), E: (Rawa, 1885), D, G and H described in (Woolsey, 1907) and (Babcock, 1927).

However, with expanding clinical experience from patients coming from the wars of the late 19th and early 20th century, peripheral nerve surgery became established and popularized.

August Waller refuted the ideas of axon reuniting through prima intentio nervorum by showing that axons in frogs distal to the injury would degenerate after the nerve was divided (so called Wallerian degeneration: Waller, 1850). That axons actually had the opportunity to grow out from the site of injury was made clear by the demonstration of the regenerating axonal growth cone in chicken in 1890 (Ramon y Cajal, 1890). With that, it became evident that time was needed for nerves to regain their function after injury.

During the American Civil War, the Federal Armies established a medical unit for nerve injuries staffed by the surgeons Weir Mitchell, Morehouse and Keen, and based on their experiences from the treatments and observations of 120 cases they wrote Gunshot Wounds and other injuries of nerves in 1864. This is probably the first major summary of cases of nerve injuries, important for an overall understanding of consequences of nerve injury although they did not in particular describe surgical nerve injury interventions. Based on symptoms of some of their patients they described causalgia for the first time, today known as complex regional pain syndrome type II, CRPS II (Weir Mitchell, 1864).

In 1889 Anthony Bowlby (Bowlby, 1890), a surgical pathologist active in London, followed up 20 cases treated with primary nerve repair and found that only two of those had what could be considered a sufficient return of hand function. He also established that one often had to wait 2-3 years before evaluating the final results, a follow up-time that is more or less in agreement with current knowledge (Bowlby, 1890; Nicholson and Seddon, 1957; Sunderland, 1968; Ruijs et al., 2005).

The idea that nerves injuries with large gaps could be treated with nerve grafts, i.e., harvesting another nerve and suture it to bridge the gap, won some popularity between the world wars after Sterling Bunnell showed success in severed nerves in the upper limb (Bunnell, 1927; Bunnell, 1939). Before this, nerve grafting had been considered more or less useless due to poor outcome, probably due to lack of knowledge of graft-versus-host reactions associated with use of nerve grafts from animals (Woolsey, 1907; Trendelenburg, 1923). The lasting hesitation to graft nerves, even using nerve grafts from the same person (autografts), is illustrated by the fact that still in the late 1970s, surgeons instead shortened bones to enable tension free nerve adaptation (Mackinnon, 2015).

Sir Herbert Seddon took interest in the fact that there exists… “among clinicians…considerable confusion when we come to examine those numerous cases in which function is lost although anatomical continuity of the nerve is more or less preserved.” From this interest and the study of 460 cases at the Oxford Centre, he 1941 classified nerve injuries in three groups: (i) complete division, (ii) lesion in continuity and (iii) transient block. The next year, Sir Seddon published a paper in which he applied the advice from Henry Cohen, a general internist in Liverpool and the President of the Royal Society of Medicine, on the nomenclature. As a result, the different types of injury were termed neurotmesis, axonotmesis and neurapraxia, respectively (Seddon, 1942).

Sir Sidney Sunderland further advanced modern peripheral nerve surgery by meticulously assessing injuries and the post-operative results. He added two gra- des on Seddon’s grading of peripheral nerve injuries, between axonotmesis and neurotmesis: ‘loss of continuity of nerve fibres’ and loss of perineurium and fascicles. He also emphasized the importance of gaining knowledge and experience from case histories and, accordingly, initiated follow-up studies over more than a ten-year period of soldiers from the Australian Forces who had sustained peripheral nerve injuries and returned to their homeland for treatment (Sunderland, 1968).

Erik Moberg took the assessment of hand function even further than just considering sensory and motor function: It has been born home to me with the passage of the years how little the results of customary tests of sensibility in an injured hand correspond with the actual ability of the patient to use his hand (Moberg, 1958) By declaring that the return of function of the hand is the therapeutic goal after nerves injuries, and by implementing multiple tests that assessed compound sensorimotor function of the hand, he set new standards for the assessment of hand injuries, for instance, in measuring of outcomes after treated upper limb peripheral nerve injury.

Today, the surgical treatment of early diagnosed sharp total peripheral nerve injury is immediate primary repair with epineural suture (Mackinnon, 2015; Dahlin and Wiberg, 2017), to adapt the nerve stumps such that the proximal axons can grow out (regenerate) and potentially regain contact with sensory receptors or muscles and thereby restore their former function. After surgical treatment, the patient is followed by experienced physiotherapists and occupational therapists through the period of hand rehabilitation.

The expansion of microsurgical treatment of peripheral nerve injuries would not have been possible without parallel advancements in medicine, such as, antibio- tics, anesthetics, sterile techniques and various technical advances (the microscope/magnifying loupes, microsurgical tools, etc.). And surely, without the contribution of either successes or failures by the pioneers of nerve injury treatment, of which many are not mentioned in this brief historical exposé, as well as the hardships endured by patients with peripheral nerve injuries who provided … …lessons which could be learnt in the ward where each [nerve] lesion was regarded as a providential experiment from which every possible advantage should be taken” (Sunderland, 1968)

… peripheral nerve surgery would perhaps be as discouraging today as it was back in the ancient Greece.

Historical aspects on understanding the function of the brain The historic origins of comprehension of human cerebral functions largely traces it roots in case studies after trauma attained on the battlefield. The Edvin Smith Surgical Papyrus from the 1,6oo BC, describes two cases of head injury with subsequent loss of the ability to speak (aphasia). However, the idea of the brain as an organ of the mind took time to gain acceptance is apparent from the writings of the eminent Greek philosopher and scientist Aristotle (384-322 BC) who 13 centuries later proclaimed that intelligence and thoughts resided in the heart1, whereas the brain was merely considered to be a cooling system for blood. After Aristotle’s death, Herophilos of Chalcedon (335-280 BC) became the first ancient Greek physician to perform systematic dissections of human cadavers. He performed about 600 dissections of presumably executed prisoners. He is supposed to have described the nervous system in some detail although none of his original works have remained. However, human dissections became forbidden after his death and his ideas was outcompeted by older thoughts or simply forgotten (Ghosh, 2015).

Galen of Pergamon (AD 129-216) instead dissected animals (Gross, 1987). Based on the macroscopic appearances of each part of the central nervous system (CNS), he believed that the cerebellum controlled muscles whereas the cerebrum processed senses. Although he was in many respects quite far from correct, he considered the brain to be responsible for thoughts and the nerves to be responsible for relaying sensations.

1 Consider how this idea lives on in with sentences such “I love you with all my heart”, whereas “…with all my brain” would scientifically be more accurate, although perhaps considered as less romantic.

It was not until the 19th century that ideas of functional specialization of specific brain areas began to be accepted (Boling et al., 2002). Franz Joseph Gall forwarded in 1796 the theory that (i) humans possess innate moral and intellectual abilities that depend on the brain, (ii) the brain comprises a set of “organs” each related to its specific faculty, but also (iii) that the skull reflects the shape of the brain and therefore invites assessment of a person’s capabilities by means of so- called cranioscopy. This scientific proposal was quickly transformed into lucra- tive quackery and pseudo-science by Johann Spurzheim (1776-1832), who spread the teaching of phrenology and very imaginative phrenological maps across Europe and the USA (Zola-Morgan, 1995). Gall will probably forever be associated with phrenology and yet his proposal that humans possess innate abilities that depend on the brain and that the brain consist of functionally distinct areas are now recognized as scientific facts.

With Paul Broca’s presentation in 1861 of his observations of two patients who had lost their ability to speak after suffering injury to the inferior frontal gyrus in the left hemisphere of their brains (Broca, 1861), the idea of functional organization of the brain gained further ground. In the decade after Broca’s discovery, Fritsch and Hitzig furthered this idea when they discovered the excitability of the cortex and localized contralateral motor function in the anterior part of the cortex of the two hemispheres of dogs, i.e., electrical current applied in the left hemisphere was accompanied with motion in the right half of the body (Fritsch, 1870). Albert Grünbaum and Sir Charles Sherrington further mapped the cortical functions of motor and sensory cortices of monkeys using electrical stimulation and ablation of specific cortical areas (Grünbaum, 1902). In parallel with the electrophysiological studies of the cortex, histological assessment with cytoarchitectural mapping techniques of the brain confirmed the physiological boundaries of, for instance, the motor areas (Campbell, 1905; Brodmann, 1909); the areas defined by these boundaries are still in use. Campbell may, in fact, have been the first to report on cortical neuroplasticity related to peripheral nerve injury when he described cellular changes in the contralateral primary motor cortex after limb amputation.

With the emergence of neurosurgery of epileptic foci in humans and the use of galvanic (electrical) cortical stimulation, the similar cortical maps in humans could be confirmed (Krause, 1908; Cushing, 1909; Snyder and Whitaker, 2013). With additional sophistication of cortical stimulation techniques in the mid-20th century it was possible to map the functional organization of the cortex with high precision in awake humans during surgery. Through this, Wilder Penfield furthered the concept of somatotopy of the primary sensory and motor cortices – the idea of representations of body parts spread in a mapped order in the cortex of the brain. Penfield´s biggest impact came perhaps from the introduction of the cortical homunculus (Penfield, 1937), the “little man” with body parts in sizes

relative to their sensory or motor representation sizes. One of Penfield’s students, Donald Hebb, postulated in 1949 the basic idea behind neuroplasticity, that two neurons that are repeatedly active together become structurally more associated, so that activity in one facilitates activity in the other (Hebbian learning: Hebb, 1949). Penfield later also presented the first evidence of a human secondary somatosensory cortex (Penfield, 1954).

With the introduction of functional magnetic resonance imaging (fMRI) in 1990, it became possible to evaluate brain activity non-invasively whilst performing various tasks in the MRI-scanner (Ogawa et al., 1990). Accordingly, the possibility to investigate human brain functions expanded vastly. Up to now, more than 50,000 scientific articles with the search word ‘fMRI’ have been catalogued in PubMed, with more than 5,000 articles added just in 20182. Continued improvement in fMRI sampling and analyzing techniques together with other old and new techniques used for studying brain function (humans and animal: intracortical neuron recordings, electroencephalography; EEG, magneto-encephalography; MEG, Optical imaging techniques etc.) warrant that the progress of our understanding of the brain is likely to continue.

2 Statistics from www.pubmed.gov, search "functional magnetic resonance imaging"[All Fields] OR "fmri"[All Fields] performed 2019/05/13. For data of publications in 2018 the search terms: AND "2018/01/01"[PDAT] : "2018/12/31"[PDAT] were added.

Normal hand function “What can the eyes see? What can the ear hear? What can the hand do?” With these words Erik Moberg beautifully summed up that hand function assessment demands testing of the composite function of the hand, and not only ‘sensation’ or ‘motion ability’ (Moberg, 1966). Put in other words, dexterous manual actions critically depend on both motor function for movement of the hand and arm, and sensation for feedback control and their complex sensorimotor interactions.

To exemplify: if the goal is to contact a specific object, the hand needs to properly reach the object, for this programming of muscle commands depends on integration of visual information and somatosensory information from the arm (proprioception). Vision can provide a sensory framework for positioning, orientation and pre-shaping of the hand during reaching and gripping of objects, but offers limited information useful for controlling the action of the digits in dexterous tasks once an object is contacted. After contacting the object, sensory information from the tactile receptors in the glabrous skin on the inside of the hand becomes critical.

Figure 2 shows, in a very simplified and schematic form, the role of sensory inputs in motor control. The illustrated principle enables correction with both visual and proprioceptive information during reaching, while tactile signals would dominate during manipulation. Tactile sensory information is forwarded to the brain, providing information on mechanical events in object interfaces, including forces applied on the object, and information about properties of the object necessary for planning and predictive control of hand actions, e.g. object weight and friction in relation to the skin (Johansson and Flanagan, 2009). The action goals, formulated by the brain, are then transformed to desired motor commands, which specifications depends on multimodal sensory state information.

Figure 2. Sensorimotor feedback loop for the execution of a motor action

Sensory information is not only used for planning motor commands, but also for automatic (reflex) correction of evolving commands transmitted to the effectuating muscles. Of course, the actions governed by the muscles result in changes in the sensory input/state which close the sensorimotor loop. Without sensory feedback there is no perception of how much power is applied to the object, and consequently it will either be held too tightly and possibly crushed or be held too loosely and possibly dropped. As such, in dexterous manipulation proper fine motor actions critically depend on tactile feedback from the fingertips contact with the object.

In the sections that follow, I will briefly describe critical components of the sensorimotor mechanisms supporting manual dexterity, with emphasis on the peripheral sensory mechanisms and cortical organization of brain areas involved in goal-oriented hand actions.

Peripheral sensory mechanisms supporting manual dexterity The number of axons within the brachial plexus and the nerve roots that supply each upper limb has been estimated to about 350,000. Notably, about 90% of these are afferent axons providing the CNS with sensory information, whereas only about 10% serve motor functions (Gesslbauer et al., 2017), which illustrates the fundamental importance of sensory inputs for the sensorimotor control of the hand and arm. In this brief review, I will focus on the low-threshold mechanoreceptors (LTMRs) situated in the glabrous skin of the palmar surface of the human hand because they are particularly pertinent to my thesis work: critical information for the control of manual dexterity consists of information on mechanical interactions between the hands and objects that originate in these receptors (Johansson and Flanagan, 2009). Vision only provides indirect information about such events and proprioceptive receptors, which deliver information about joint configurations and muscle states, poorly reflect details of mechanical events at the fingertips (Hager-Ross and Johansson, 1996; Macefield et al., 1996; Macefield and Johansson, 1996; Dimitriou and Edin, 2008).

LTMRs are responsible for the tactile sensibility of the hand (Vallbo and Johansson, 1984; Johnson, 2001) and signals from these mechanoreceptors are conveyed to the brain by fast conducting thick myelinated Aβ-afferent axons belonging to first-order tactile neurons with cell bodies in the dorsal root ganglia (DRG). Other sensory signals from the hand and arm important for normal hand functions are signals from sensors in muscles conducted in thick myelinated Aα- afferent axons that indicate muscles’ states (muscle spindles and Golgi tendon organ; Phillips et al., 1971) and from nociceptors (pain) and thermoreceptors conducted in thin myelinated Aδ- and unmyelinated C-afferent axons.

The physiology of the tactile afferent neurons conveying signals from LTMRs in the glabrous skin of the human hand has been investigated thoroughly in the last ~50 years (Hagbarth and Vallbo, 1967; Knibestol and Vallbo, 1970; Johansson and Vallbo, 1979; Vallbo and Johansson, 1984; Johansson and Flanagan, 2009; Abraira and Ginty, 2013; Vallbo, 2018). Each fingertip is innervated by about 2,000 of tactile neurons, whereas some 10,000 innervate the palm (Johansson and Vallbo, 1979).

Four types of tactile neurons innervate different types of LTMRs and signal complementary aspects of skin deformation (Fig. 3). These are considered as different sub-modalities in the cutaneous sensibility of touch. Two of these types show fast-adapting (FA) responses, i.e., they respond with action potentials only during skin deformation changes. These afferent neurons innervate Meissner and Pacinian LTMRs. The two other types, innervating Merkel and Ruffini LTMRs, are slowly-adapting (SA), i.e., in addition to responsiveness to skin deformation changes they also respond to maintained tissue deformation. In addition, the tactile afferent neurons can be subdivided with respect to the characteristics of their cutaneous receptive fields as defined by gentle pointed touch stimuli: Type I afferents (FA-1, SA-1) originate in the superficial layer of the skin and have small and well-defined receptive fields (typically ~10 mm2) composed of several “hotspots” due to branching of the distal axon in the skin (Johansson, 1978; Nolano et al., 2003; see Fig. 3), while Type II (FA-2, SA-2) afferents originate in deeper layers and have larger and more diffusive receptive fields.

Figure 3. The location of LTMRs in skin and their relationships to tactile afferents

The innervation density in the skin of the Type I tactile neurons shows a marked increase in the proximo-distal direction of the hand, i.e., the density is particularly high in the fingertips. Accordingly, these neurons enable the fingertips to provide high spatial tactile acuity. In contrast, Type II afferents show a more uniform innervation density. The FA-2s (Pacinians) resemble seismographs as they are exceptionally responsive to mechanical transients that propagate through the tissues even from remote sites. Likewise, SA-2s are responsive to stimuli applied remotely from the receptor itself by gauging strain in the dermal and subdermal fibrous tissues.

Because each of the FA-1, FA-2 and SA-1 afferents are most easily excited by oscillatory skin displacements of different frequency (Johansson et al., 1982), each type can be stimulated in isolation using threshold near amplitudes of sinusoidal mechanical stimulations of the skin. Each type also typically evokes a characteristic feeling when stimulated in isolation (Ochoa and Torebjork, 1983; Lofvenberg and Johansson, 1984; Vallbo et al., 1984). Stimulation of FA-1 afferents, which are most sensitive to skin deformation changes of frequencies between ~5 and ~50 Hz, give rise to a sensation described as “flutter”, while stimulation of FA-2 afferents, which are exceptionally sensitive to high-frequency vibrations (~40 – 400 Hz), bring about a sensation of diffuse “vibration”. Finally, stimulating SA-1 afferent neurons in isolation, most sensitive to skin deformation changes of low-frequency (<~ 5 Hz) and to static skin displacements, cause “pressure” sensations.

Afferents from adjacent peripheral receptors and axons innervating adjacent muscle fibers merge to form small nerve fascicles which are further joined to larger fascicles that together form peripheral nerves. Near the spinal cord afferent axons and efferent (motor) axons are separated such that they form dorsal sensory and ventral motor roots, respectively.

The cell bodies of the first-order sensory neurons that innervate receptors in the various body tissues, including tactile LTMRs, are located in dorsal root ganglia adjacent to the spinal cord (Fig. 4). These neurons are classified as pseudo- unipolar because they have a single axon with two branches, one reaching distally to the end-organs, and the other projecting proximally into the spinal cord. A collateral of the proximal axon of tactile afferent neurons ascends in the ipsilateral dorsal column of the spinal cord and reaches the cuneate nucleus located in the medulla oblongata, where the afferent signals are processed (Witham and Baker, 2011; Bengtsson et al., 2013). From the 2nd order tactile afferent neurons in the cuneate nuclei, the tactile information is further propagated centrally in axons that cross the body midline in the medial lemniscus

Figure 4. Schematic illustration of sensory input and motor output

Sensory afferent signals from LTMRs in the skin are relayed to CNS through Aβ- fibers of 1st-order sensory neuron (red line in cross sections) with cell bodies located in the dorsal root ganglia (DRG – cross section labelled 4). Axonal collaterals ascend through dorsal column (DC – cross section labelled 3) of the spinal cord to the cuneate nuclei (NC – cross section labelled 2). 2nd-order sensory neurons cross the midline of the body and relay signals (blue line) to the ventroposterior nucleus (VPN) of thalamus (T). From VPN, 3rd-order neurons relay signals (yellow line) to the 4th layer of the primary somatosensory cortex (S1 – cross section labelled 1). Motor efferent signals from the upper motor neuron in the primary motor cortex (M1) travel in corticospinal fibers (light green line) that crosses the midline in the medulla and carry information down to activate interneurons and lower motor neurons in the ventral horn of the spinal cord. Axons of lower motor neuron (orange line) innervate muscles and signal muscles to contract. and terminate in the ventro-posterior nucleus of thalamus (VPN). Both the cuneate nucleus and VPN has a somatotopic organization (Kaas et al., 1984; Davis et al., 1996), and cortical top-down signals can tune the processing in the somatosensory ascending pathways in a task and context dependent manner (Harris et al., 1965; Adkins et al., 1966; Canedo, 1997; Ergenzinger et al., 1998; Palmeri et al., 1999). For example, the relationship between afferents from thalamus to cortex and corticothalamic efferents is 1:10 (Liu et al., 1995) suggesting that corticothalamic information has major influences on the processing of the afferent signals before they are further relayed to the cortex. From the VPN, axons of 3rd-order afferent neurons convey processed information

to the primary somatosensory area (S1) in the anterior parietal cortex, but to some degree also directly to the secondary somatosensory cortex (S2) in the parietal operculum on the ceiling of the lateral sulcus (Pons et al., 1992).

Cortical brain areas associated with somatosensory and multisensory processing

Primary somatosensory cortex The primary somatosensory cortex (S1) plays a critical role in processing afferent somatosensory input and contributes to the integration of sensory and motor signals necessary for skilled movement. S1 is located in the anterior parietal cortex, in the posterior bank of the central sulcus and on the postcentral gyrus, and is the major cortical receiver of somatosensory information which arise from receptors positioned throughout the body. The S1 of each hemisphere primarily contains a tactile representation of the opposite (contralateral) side of the body and the representations of the body parts are orderly arranged: from toes at the medial aspect of the hemisphere to the mouth at the lateral inferior aspect. The somatosensory representation of the hand is located at the postcentral gyrus posterior and adjacent to the primary motor cortex of the hand in the precentral gyrus.

S1 consists of four different Brodmann areas (BA), area 3a, 3b, 1 and 2, organized in an anterior to posterior order (see Fig. 5). Each area contains its own somatotopic map of the body, including separate representations of each digit (Kaas, 1993; Gelnar et al., 1998; Maldjian et al., 1999; Kurth et al., 2000; Sato et al., 2002; Chen et al., 2003; Ruben et al., 2006). In area 3a the thalamic input originates primarily in muscle spindles (Phillips et al., 1971; Kaas, 2004) and this area is thought to provide an important source of proprioceptive information for the adjacent primary motor cortex. BA3b is by some referred to as S1 proper because its main inputs originate from cutaneous receptors (Wu and Kaas, 2003; Liao et al., 2013). This area has small, intermingled domains that are activated by different types of tactile stimulation, such as vibration, pressure, or flutter, reflecting distinct pathways from the various types of peripheral receptors in the skin (Sur et al., 1981). Neurons in BA1 receive strong activating inputs from BA3b and thus this area is thought to be involved in a secondary cortical stage of tactile processing (Papadelis et al., 2011), but they also receive direct input from VPN. BA1 neurons show larger and more complex cutaneous receptive fields than those of BA3b and encode more complex stimuli, such as direction of object moving over the skin surface. Evidence from non-human primates shows that neurons in BA1 can encode various features of various kinds contact events at the fingertips (Callier et al., 2019).

Figure 5. Schematic illustration of finger representations in S1, and their input and output

Signals from LTMRs reach the primary somatrosensory cortex through neurons from the ventroposterior nucleus (VPN) of thalamus. Input is processed and relayed to neighbouring and more distant areas of the brain involved in perception and control of actions and decisions. BA4 is the equivalent of the primary motor cortex. BA: Brodmann Area; PPC, Posterior Parietal Cortex; VPN, ventroposterior nucleus of thalamus; VPS, ventroposterior superior nucleus of thalamus; Cx, Cortex; S2, secondary somatosensory cortex. Roman figures I-V denote schematically mediolateral organization of finger representations. Colors of arrows denote major routes of feedforward connections, notably all of these show reciprocal elements.

BA 2 receives inputs from BA3b and BA1 and therefore constitutes a third level of cortical processing of tactile and proprioceptive information (Pons and Kaas, 1986). This cortical input is combined with proprioceptive inputs directly from thalamus. Neurons in BA2 are especially responsive when objects are actively explored or manipulated with the hands so that tactile and proprioceptive afferent information is combined for more complex representation of objects necessary for dexterous object manipulation and haptics in general, as exemplified by stereognosis. Essentially, all connections between the Brodmann areas of S1 involve reciprocal elements (Kaas, 2004; Keysers et al., 2010).

The processing of tactile information in S1 involves intra-areal lateral connections allowing for processes where the various finger representations dynamically interact, mainly through complex lateral inhibitory mechanisms often reminiscent of surround suppression (Ruben et al., 2006; Lipton et al., 2010; Reed et al., 2010; Thakur et al., 2012; Martuzzi et al., 2014). Such interactions likely exists to enhance spatial processing of tactile stimuli often with regard to

contrast, orientation, and direction of motion. Of relevance to the study in the current thesis on how unilateral median nerve injury can affect brain function (Paper III) is that there are dynamic interactions also between hand S1 cortices of the two hemispheres. This implies that stimulation of one hand not only give rise to contralateral cortical activity, but also to ipsilateral activity changes (Hlushchuk and Hari, 2006; Klingner et al., 2015; Tal et al., 2017). However, there is little evidence of direct connections between the BA3b’s of the two hemispheres. Communication between the two BA3b is thought to be through sensory areas showing interhemispheric connections with which BA3b directly communicates, partially BA1 and BA2 but especially "higher-level" sensory areas, including S2 (Killackey et al., 1983; Iwamura et al., 2002; Reed et al., 2011).

Lesions affecting the hand representation of postcentral gyrus cause deficits in two-point discrimination, position sense and point localization, object size, shape, and texture discrimination (Corkin et al., 1970; Roland and Mortensen, 1987; Kaas, 2004).

Sensory association cortices and multisensory processing Processing of somatosensory input beyond S1 occurs in several cortical areas (Kaas, 2004; Dijkerman and de Haan, 2007). The characteristics of the neuronal signals become increasingly more complex the further away the neurons are from the thalamic input to the cortex. The cortical association areas, i.e., the brain areas outside the primary regions, typically not exclusively process one type of sensory input, but are involved in multisensory/multimodal integration. Association areas with a close relationship to S1 are the secondary somatosensory cortex (S2), the posterior parietal cortex (PPC) and the (posterior) insular cortex (Culham and Kanwisher, 2001).

Traditionally, the primary sensory areas process signals belonging to only one sensory modality (vision, hearing, somatosensation, etc.), but more recently there is evidence that virtually all cortical regions can receive and be influenced by inputs in more than one sensory modality. That is, while processing in the primary sensory areas is dominated by modality specific information, these areas can still process signals originating in other modalities. This can happen "when there is a need" to perform specific calculations that the primary area are specialized in regardless of input modality (Pascual-Leone and Hamilton, 2001). An example on such metamodal processing is that congenital or early blind persons activate visual cortex when reading Braille (Sadato et al., 1996; Sadato et al., 1998). In the case of healthy individuals, the primary auditory cortex (A1), located in the Heschl’s gyrus of the temporal lobe, has been reported to be involved in processing of tactile stimuli (Foxe et al., 2000; Schroeder et al., 2001; Fu et al., 2003; Caetano and Jousmaki, 2006; Schurmann et al., 2006). This metamodal mechanism was further explored in Paper I of the current thesis, in

which we investigated brain activity in healthy adults to examine whether different cortical areas could be involved in processing tactile stimuli depending on the temporal frequency content of the stimuli. That is, whether cortical brain activity would differ in response to tactile oscillatory stimulations at 3, 20 and 100 Hz delivered to the fingertips.

Considering the secondary somatosensory cortex (S2), single neuron recordings in trained monkeys during tactile perceptual tasks engaging their fingertips show that attention modulates neuronal discharge at a higher rate in S2 (~80% of neurons) than in S1 (~ 50%), and that S2 neurons code more complex, integrated somatosensory information (Hsiao et al., 1993). For tactile processing of object shape for action and perception, BA2 of S1 and S2 are considered responsible for three-dimensional shape based on integration of tactile and proprioceptive inputs while two-dimensional tactile shape (and movement) are processed in areas 3b and 1 of S1 by neurons with receptive fields consisting of excitatory and inhibitory subregions (Hsiao, 2008).

The human S2, located in the parietal operculum on the ceiling of the lateral sulcus (Sylvian fissure) holds multiple bilateral representations of the body surface (Sanchez Panchuelo et al., 2018). S2 receives inputs from S1 through reciprocal cortico-cortical connections and to some extent also from thalamus. S2 exchanges information with other associative areas involved in higher-level processing such as PPC and insula. Human S2 is considered to consist of four areas that are called parietal operculum 1-4 (OP1-4) based on cytoarchitectonic distinctions (Eickhoff et al., 2006a). The four areas are believed to be involved in different functions (Eickhoff et al., 2006b). Specifically, areas OP1, 3 and 4 are considered to be directly related to somatosensory processing with independent representations of the body, whereas OP2 seems to be involved in multimodal processing by responding to visual, auditory and vestibular stimuli and during whole body movements (Eickhoff et al., 2006c; Shinder and Newlands, 2014; Frank and Greenlee, 2018). Functional and structural analysis of the connections between brain areas in humans show that OP1 is more closely integrated with the anterior inferior parietal cortex and VPN of thalamus, while OP4 shows a higher level of integration with premotor areas, inferior frontal cortex, postcentral gyrus and primary motor areas (Eickhoff et al., 2010). This suggests that OP1 is more closely connected to the parietal networks for higher-level somatosensory processing and OP4 more closely connected to higher-level motor control, which is in agreement with cortical connection of analogue brain areas in monkeys (Disbrow et al., 2003), although both areas have connections with premotor areas (Gharbawie et al., 2011). The function and connectivity pattern of OP3 have not yet been as well characterized as the other areas. Overall, the detailed functions of the S2 subareas are largely unknown, but in general, they seem to possess associative functions comparable to those described for areas of the posterior

parietal cortex. As such, the name “secondary somatosensory cortex” perhaps more describes that the area was the second to be found showing somatosensory responses (Penfield, 1954) rather than being the sole ‘second area’ in a hierarchy of somatosensory processing.

The posterior parietal cortex (PPC), located between the S1 and the visual cortex in the occipital lobe, is a section of the cortex that links and processes somatosensory and visual information (Creem-Regehr, 2009). Processed information is then fed to the prefrontal and premotor areas for sensory control of actions (Gentile et al., 2011; Gallivan and Culham, 2015). The integration of sensorimotor information in the PPC are exemplified by, for instance, regional increases in brain activity during observation, planning and execution of reaching and grasping actions (Prado et al., 2005; Beurze et al., 2007; Kroliczak et al., 2008), and tool perception and tool use (Hermsdorfer et al., 2007; Higuchi et al., 2007; Imazu et al., 2007; Vaesen, 2012). The generality and associativeness of the PPC in the construction of multisensory/multimodal based maps (visual/somatosensory) of the peripersonal space supporting control of actions are further exemplified by the fact that PPC lesions can give rise to symptoms such as simultagnosia, oculomotor apraxia, and optic ataxia3, a triad first described by Rezso Balint in 1909 (Balint, 1909; Hausser et al., 1980). Also, damage to PPC is associated with a profound lack of awareness of bodily posture or the position of limbs. Furthermore, “temporary lesions” produced with transcranial magnetic stimulation (TMS) suggests that PPC also uses tactile and proprioceptive information (position and force related signals) for our percept of stiffness of a contacted object, for instance when feeling the ripeness of a fruit in the food store (Reichenbach et al., 2011; Ferrari-Toniolo et al., 2015).

Dorsal and ventral visual and somatosensory processing streams Visual information is important for positioning one’s body, limb and hands to enable proper object-oriented manual actions. The cortical area primarily receiving visual input is the primary visual cortex (V1) located in the most posterior part of the occipital lobe. In analogy with the somatotopic organization of S1, in V1 there is a spatially ordered map of the visual field seen by the eyes (retinotopy). V1 feeds information to other vision processing areas, termed the extrastriatal cortex, which in turn relay processed visual information further across the brain through different processing streams.

Ungerleider and Mishkin introduced the notion that visual information is processed in a ventral visual stream for recognition and identification of objects

3 Simultagnosia: inability to perceive more than one item in the visual field; Oculomotor apraxia: difficulty in making targeted eye movements; Optic ataxia: the inability to make visually guided arm and hand movements

and in a dorsal visual stream for visually guided actions. They showed that lesions of the inferio-temporal cortical regions in monkeys affected visual discrimination and recognition of an object whereas posterior parietal lesions resulted in impairment in visuospatial functions (Mishkin et al., 1982; Mishkin and Ungerleider, 1982; Mishkin, 1983). Evidence of a corresponding distinction of visual processing pathways in human was then generated by Goodale and Milner who launched the concept of the now well-known "what" and "how" streams (Goodale and Milner, 1992). The latter (“how”) was based on the observation that damage to the posterior parietal cortex not only impaired the control of the hand’s position during reaching but also the adaptation of the shape and orientation of the hand to the orientation and shape of the target object (Perenin and Vighetto, 1988). The dorsal “how” stream has subsequently been divided in two: (i) A dorsomedial stream where visual information is processed in a pathway that includes the superior parieto-occipital cortex, and the dorsal premotor cortex (PMd) in the frontal lobe (Galletti et al., 2001; Gallivan et al., 2009). (ii) A dorsolateral pathway pass through the inferior parietal lobe of the PPC and the ventral premotor cortex (PMv) in the frontal lobe (Rizzolatti and Matelli, 2003; Gallivan and Culham, 2015). In this distinction, the dorsomedial stream has been implicated in planning and online corrections of skilled reaching and grasping whereas the dorsolateral stream in higher-level manual dexterity such as tool use or hand gestures that rely on visual information to recognize, in particular, movement of objects during self-movements (Johnson-Frey, 2004; Galletti and Fattori, 2018).

The dorsal stream features early integration of tactile and visual stimuli relevant for control of manual actions. The extrastriate “middle temporal” area (MT/V5) of the primate brain (in humans: human middle temporal visual complex; hMT+) is considered located in the dorsal visual stream and is specialized in the processing of visual motion information. It belongs to a region in the posterior temporal cortex near the so-called occipito-temporal junction that holds association areas that respond to multisensory information. Of these areas, the lateral occipital cortex and the hMT+ respond to tactile stimuli in addition to visual stimuli. The lateral occipital cortex is located just posterior to hMT+ and is claimed to be involved in processing visual and tactile shape information of objects (as opposed to texture, Amedi et al., 2001; Beauchamp, 2005). The hMT+ responds to both visual and tactile motion (Hagen et al., 2002) and its direction (van Kemenade et al., 2014), and when watching images of moveable objects learned through touch (Chan et al., 2010) and during 3D object haptic perception (Blake et al., 2004).

The ventral stream is considered to go from the early visual cortices (striate cortex) through areas of the inferotemporal cortex and further in the temporal cortex to the anterior inferotemporal area, and from there to prefrontal areas: the ventrolateral prefrontal cortex (VLPFC) and the orbitofrontal cortex (Kravitz et al., 2013). Analysis of connections between areas using resting state functional MRI data of 470 individuals also finds support for three cortical pathways (Haak and Beckmann, 2018; see Fig. 6).

Figure 6. Lateral schematic view of the cortical visual streams

Areas described in this section schematically illustrated. S1: Primary somatosensory cortex, M1: Primary motor cortex, A1: Primary auditory cortex, V1: Primary visual cortex, PPC: Posterior Parietal Cortex, Cx: Cortex. Extrastriate cortex refers to non- primary visual processing areas.

Whether there are two (or three) visual streams has been debated, partly because there are quite a few overlapping functions and well-developed connections between the proposed streams. Nevertheless, many researchers in the field consider this model as relevant for investigating integration of visual (and somatosensory; see further below) information for planning and control motor actions (e.g., the journal Cortex devoted an entire volume to this subject in 2018; de Haan et al., 2018). However, the initial ideas from Mishkin and Ungerleider to consider the cortical streams as fixed series of interconnected cortical areas have been replaced over time by the view that they rather represent interconnected neuronal

networks involved in a number of functional processes and whose activation changes dynamically according to the context (Galletti and Fattori, 2018).

Mishkin also proposed a similar dual organization of the somatosensory pathways for touch and proprioception (Mishkin, 1979), a notion that has gained further support (Dijkerman and de Haan, 2007; Delhaye et al., 2018). In this model, the somatosensory ventral stream originates in the projections from S1 to S2 cortex, including the neighboring zones along the upper bank of the lateral (Sylvian) fissure and the parietal operculum. The ventral network has strong anatomical links through the insular cortex to the medial temporal lobe and hippocampus, providing a cortico-limbic circuit for object recognition and memory acquired through touch. The insular cortex is located deep in the fold between the temporal, parietal and frontal lobes, and is covered by the temporal, frontal and parietal operculum. Insula is thought to play a role in diverse functions usually linked to emotion or the regulation of the body's homeostasis. These functions include compassion and empathy, perception, motor control, self-awareness, cognitive functioning, and interpersonal experience. The posterior insula connects reciprocally with S2 and receives input from thalamic nuclei, including from the ventromedial nucleus (posterior part) that are highly specialized to convey homeostatic-related information such as pain, temperature, itch, local oxygen status, and sensual touch. The somatosensory dorsal stream is conveyed through the PPC, particularly the region bordering the intraparietal sulcus in humans and monkeys. The dorsal stream represents information gathered by touch and proprioception during interactions with objects and the environment. The PPC projects primarily to the premotor areas of the frontal lobe, providing a pathway for initiating and controlling motor behavior. These frontoparietal networks play a key role in voluntary movement, particularly the skilled behaviors of the hand and arm: reaching, grasping and tool use.

Cortical motor areas The phrase “motor area” refers commonly to brain areas located anterior to the central sulcus that project to the ventral horn of the spinal segments harboring interneurons as well as α motor neurons responsible for activating skeletal muscles.

The primary motor cortex, or M1, was given its name because very small electrical stimuli can evoke limb movements, whereas the term premotor nowadays refers to areas in the frontal lobe with direct projections to M1. Studies in non- human primates show six definable premotor areas; on the lateral surface of the hemispheres the ventral and dorsal premotor areas (PMv and PMd) and in the medial surface the supplementary motor area (SMA), and the rostral, dorsal and ventral cingulate motor areas (CMAr, CMAd and CMAv; Riehle, 2005). Neuronal tracing studies show that the premotor cortex and M1 hold similar amounts of

corticospinal neurons (Riehle, 2005), although a significant fraction of M1 neurons directly innervate the motor neuron whereas premotor neurons essentially innervate only interneurons (Lemon et al., 2002). Electrical stimulation in each of the premotor areas evoke limb movements, but in a smaller proportion of neurons than in M1, and at intracortical stimulation intensities typically higher than those that evoke movements when applied to M1. Furthermore, there is evidence that motor effects evoked from the premotor areas at least in part depend on interactions with M1 (Riehle, 2005; Schmidlin et al., 2008). As such, each premotor area has the potential to influence the generation and control of movement at the level of the spinal cord, as well as at the level of M1.

Furthermore, premotor areas are typically activated bilaterally even during strictly unilateral upper limb movements (Ehrsson et al., 2000), suggesting that they are involved in higher-level representations and processing of information and action goals. In contrast, M1 is considered to be essentially unilaterally active in unimanual actions, again suggesting a more direct coupling with the motor neurons of the contralateral limb (but see for instance: Ganguly et al., 2009; Heming, 2019, for examples of ipsilateral M1 activity). The exact functional division of labor between the premotor areas are yet to be understood, although vast research has been devoted to this issue.

The motor areas also each have different degrees of reciprocal connections with parietal areas such as Brodmann areas 3a, 1, 2, and the areas within the PPC. The complexity of these parietofrontal networks have been explored using neuronal tracing techniques in non-human primates, lesion studies, electrical stimulations, etc. According to these studies, each premotor area receives a unique pattern of input from the various subdivisions of the parietal lobe and provides a functional-connectivity basis for characterizing the various premotor areas. Understanding how the premotor areas interact and function in humans is challenging, because most of what is known stem from studies in non-human primates with which humans do not necessarily share detailed neuroanatomical traits. It is, for instance, largely unknown if connections between non-primary motor areas to the lower motor neuron in the spinal cord exist in human (Riehle, 2005).

Primary motor cortex Primary motor cortex controls body movements through corticospinal projections to contralateral spinal interneurons and motor neurons (lower motor neurons, LMNs) that activate muscles, creating limb and body movements. Rather than a somatotopic organization or a representation of isolated muscle (i.e., myotopic representation), the primary motor cortex consists of several overlapping distributed body-part-modules that each processes and controls

composite movements involving several muscles often actuating an operating across several joints (Ejaz et al., 2015). That stimulation of individual M1 corticomotoneuronal (CM) neurons can activate two or more muscles that operate over both distal and proximal joints of the arm indicates that individual CM neurons can contribute to complex multi-joint movements (McKiernan et al., 1998). Conversely, evidence that LMN’s that innervate intrinsic hand muscles can be driven by neuronal CM pools targeting single muscles harmonizes with the fact that finger movements can be individualized and more finely tuned.

Input to M1 from other cortical areas arises from regions in the frontal and parietal lobes that, like M1, are the origin of projections to the spinal cord (Dum and Strick, 2005; Riehle, 2005). These corticospinal-tract projecting areas include all the premotor areas in the frontal lobe (defined above) and portions of the anterior superior parietal lobe. M1 has no substantial connections with the prefrontal and limbic cortices.

Premotor cortex The ventral premotor area (PMv), which in the left hemisphere contains the Broca’s area attributed to speech production, has been shown to also be involved in comprehension, execution and observation of sequential actions (Fadiga et al., 2009), i.e., functions not exclusively related to language processing. For example, TMS over the left PMv has shown that it is involved in higher-order processing of perceptual sequences also when the task is neither motor nor linguistic (Alamia et al., 2016). PMv’s role in higher-level control and processing of hand actions in object manipulation is evident from studies showing its engagement in selection of specific grip postures (Raos et al., 2006), but not to a particular object (Kroliczak et al., 2008; for a review about PMv, see: Grafton, 2010).

The role of the dorsal premotor area (PMd) in manual dexterity is even less well understood than that of PMv. PMd seems, however, primarily to be involved in the planning of manual actions in the reachable space and to be especially involved in the control of bimanual actions (Sadato et al., 1997). Virtual lesions in PMd, produced by TMS in humans, disrupt both the coupling of grasp and lift forces in precision object lifting (Davare et al., 2006) and the ability to associate object mass with symbolic information (Nowak et al., 2009). The dorsomedial visual stream that codes information for reaching projects to the PMd, whereas the dorsolateral visual stream that codes information for dexterous object handling and tool use project to the PMv.

The supplementary motor area’s (SMA) hierarchical intermediate position between higher-level processing areas and the executive motor areas is reflected by its balanced connections with prefrontal areas without direct corticospinal connections and its projections to PMd, PMv and M1. The SMA’s projections to PMd and PMv are more substantial than its projections to M1, suggesting that its

influence one motor activity is exerted through connections with other premotor areas rather than directly to M1 (Dum and Strick, 2005; Nachev et al., 2008).

For the premotor areas of cingulate cortex in the medial wall of the hemispheres, electrophysiological and lesion studies in monkeys, along with virtual lesion studies using TMS and neuroimaging data in humans suggest that they are responsible for higher-level planning and control of goal directed manual actions. To this end, they integrate information about (internal) action-goals from prefrontal cortex and multimodal sensory state information (external and internal) from parietal cortex, including somatosensory information from the hands (Binkofski et al., 1999; Dum and Strick, 2005; Frey, 2008; Grafton, 2010; Davare et al., 2011; Brozzoli et al., 2014; Rizzolatti et al., 2014).

In non-human primates, the cingulate premotor areas are subdivided into rostral, ventral and dorsal parts. The human counterparts (Picard and Strick, 1996) are considered to be the anterior and posterior rostral cingulate zones (RCZa and RCZp) and the caudal cingulate zone (CCZ). In humans, as for non-human primates, these areas are involved in multiple aspects of motor control and are considered holding separate upper limb representations (Picard and Strick, 1996). Intracortical electrical stimulation in humans assessed for epileptic- attenuating surgery showed that an area corresponding to the RCZa was most reliable in promoting reaching and grasping as well as body-directed arm or hand actions with a contralateral preference while more anterior areas did not evoke any limb movements (Caruana et al., 2018).

Prefrontal cortex The prefrontal cortex (PFC) is located rostral to the premotor cortices (see Fig. 6). The subregions of the prefrontal cortex have been defined in detail using cytoarchitectonic mapping techniques (Donahue et al., 2018) and based on connectivity analysis of fMRI-data (Fan et al., 2016). However, the functional definitions of these areas are not characterized to a similar degree of detail. The functions of the prefrontal cortex is considered to be “knowledge about behavior” (Wise, 2008), perhaps more often described as executive functions and cognitive control. This includes planning, selection and moderation of complex cognitive behavioral by weighting internal and external goals against the predicted consequences of optional actions, considering factors such as long-versus-short- term rewards and punishments. The PFC is thought to have a division of labor according to a multidimensional spatial organization, which in one dimension involve a gradual shift from the posterior regions implicated in optimizing the performance of tasks at hand to the frontal regions more implicated in global abstract functions such as exploring the environment in relation to internal goals for potentially changing to action strategies considered more important or beneficial for the individual (O'Reilly, 2010; O'Reilly et al., 2010).

The prefrontal cortex is highly interconnected with many parts of the brain, including extensive connections with other cortical, subcortical and brain stem sites. In contrast to the premotor areas in the frontal cortex, the prefrontal cortex (PFC) shows no projections to M1 and no corticospinal projections. Accordingly, electrical stimulations do not evoke any body movements. Hence, PFC has more general functions in goal-directed behaviors that relate to choices in the forthcoming and ongoing behavior, and choices about changing ongoing behavior (Miller and Cohen, 2001). Signals from PFC feed into premotor areas (but not to M1), and, as such, can be considered representing higher levels of decision- making. Pathology disturbing these areas causes shifts in personality and affects, for instance, inhibition control. A famous example is that of Phineas Gage, who in 1848 suffered an injury to his PFC, resulting in profound changes in his personality traits including lack of impulse control, whereas he showed no apparent changes in other functions such as memory, communication or sensorimotor skills (Harlow, 1869).

The function of PFC may seem largely unrelated to manual actions and how upper limb peripheral nerve injury affects hand functions. However, given that peripheral nerve injury imposes behavioral changes that must be dealt with in terms of planning and selection of actions as well as implementation of compensatory strategies, it seems reasonable to assume that prefrontal areas can show injury-related functional and/or structural changes.

Based on anatomy and proposed functional specializations different ways of subdividing the PFC can be found in the literature, but differentially labeled subareas often show overlapping complementary and related functions. The PFC is often divided into a dorsolateral area (DLPFC) and a ventrolateral area (VLPFC) located the lateral aspect of the frontal lobe, whereas the medial aspect of the frontal lobe is divided into the pre-supplementary motor area (pre-SMA) and the anterior cingulate cortex (ACC), and more anteriorly a dorsomedial PFC area (DMPFC) and a ventromedial PFC area (VMPFC).

The pre-SMA, located just anterior to SMA, shows extensive connections with prefrontal and premotor areas (Bates and Goldman-Rakic, 1993), but little or no connections with M1 and no corticospinal projections (Riehle, 2005). The pre- SMA shows more abstract couplings of activity to motor output than its neighboring area SMA and has been, for example, implicated in situations of response conflicts (Picard and Strick, 1996; Nachev et al., 2007; Nachev et al., 2008).

The anterior cingulate cortex (ACC) is located rostral to the RCZa and is thought to play a critical role in forming associations between rewards (/punishments) and actions (Behrens et al., 2007; Jocham et al., 2009). Its dorsal part is mainly involved in cognitive aspects of decision making, which is reflected in its dense

connections with parietal cortex and other prefrontal areas, while its ventral parts are more involved in emotional and motivational aspects as reflected by denser connections with amygdala and anterior insula. ACC is especially involved in effortful tasks, such as for instance learning a new task or performing problem- solving (Kolling et al., 2016; Shenhav et al., 2016). Together with the anterior insula it belongs to a network that responds to salient information about bodily states (involving visceral and somatosensory information available through the anterior insula) that through ACC signal the need for behavioral change (Dosenbach et al., 2007; Seeley et al., 2007; Menon and Uddin, 2010).

DLPFC has been proposed to be the “endpoint” of the dorsal visual pathway and mainly involved in processing of how to interact with stimuli, whereas VLPFC has been suggested to be the “endpoint” of the ventral visual pathway and primarily involved in characterizing stimuli (Bunge et al., 2005; Dosenbach et al., 2007; Gilbert and Sigman, 2007; Dosenbach et al., 2008; O'Reilly, 2010). However, both these regions seem to be involved in task-switching, inhibition of actions, and action orientation. The DMPFC, which to some respects correspond to the frontotemporal pole or the anterior prefrontal cortex, is considered involved in assessing whether to exploit the current task or to explore other potentially more beneficial tasks. This include balancing one’s behavior by processing benefits and risks while attending and pursuing the currently performed task, or flexible change between tasks (Donoso et al., 2014; Koechlin, 2014, 2016; Mansouri et al., 2017). VMPFC corresponds in part to the orbitofrontal cortex, which is one of the least understood regions of the human brain. It is proposed to be involved in higher levels of sensory integrations, such as in reinforcements of rewards as well as punishments and implementing stimulus-reinforcement associations (Kringelbach and Rolls, 2004; Kringelbach, 2005).

Hand function after peripheral nerve injury

Classification of nerve injuries Trauma to a peripheral nerve can cause damage of various degrees to the nerve fibers (axons) and may also be associated with injuries to tendons, bone, connective tissue, and blood vessels, all in potential need of surgical treatment. For adequate treatment, it is important to assess correctly the degree of nerve injury and to determine if surgery is necessary or not. Today, Sunderland’s grading of nerve injury is in clinical use. It consists of a five-degree scale that is based on the nature of axonal injury and the extent of damage to the connective tissues in the nerve that protect and guide the axons to their distal targets during reinnervation (Pfister et al., 2011; for details, see Fig. 7). In addition, Sunderland described patients in the potential need of surgical intervention who suffered variable degrees of mixed or partial injury to a nerve (Sunderland, 1968). This was later introduced in as a sixth category of nerve injury (Ray and Mackinnon, 2010).

It can be difficult to clinically diagnose correctly the degree of nerve damage. For example, in a person with a wound at the wrist who suffers impaired sensation within a skin territory that conforms to the innervation territory of a nerve at the site of the injury (with or without motor symptoms) any degree of nerve injury may be present. A thorough clinical assessment in combination with a description of the injury course will, however, help decide whether the injury will need surgical exploration and perhaps surgical intervention or if it is appropriate to await spontaneous recovery. Currently, when patients presented early and when there is even a suspicion of 5th degree injury, such as in the patients investigated in Paper II and III, immediate primary epineural suture is the preferable surgical approach (Pederson, 2014; Mackinnon, 2015; Dahlin and Wiberg, 2017).

Figure 7. Classification of nerve injuries

1st degree “Neurapraxia” Temporary conduction block with axons and supportive structures not disrupted. Surgical intervention not needed, will recover in days to weeks. Most commonly caused by compression, such as the radial nerve compression seen in cases of “Saturday night palsy”. 2nd degree “Axonotmesis” Purely axonal damage with preservation of the connective tissue framework of the nerve. Disintegration of the axon distal to the injury site accompanied by breakdown of the myelin sheet (Wallerian degeneration). Proximal axons attempt to regenerate guided by the distal nerve to reinnervate their targets. Surgical intervention not needed. Outcome can vary depending on the location of injury, a more proximal injury has worse prognosis. Most commonly caused by crush injury. 3rd degree Axonal damage and disruption of the Schwann cell basal lamina and concomitant injury to the connective tissue closest to the axon, the endoneurium. Partly disorganizes the fascicular axonal structure, can give rise to distal mismatches in reinnervation. Surgery typically not needed. Most commonly caused by crush injury. 4th degree As 3rd degree, but more severe injury involving disrupted perineurium. Poor tissue architecture and possible scar formation that may hinder regeneration of axons. Some spontaneous recovery may occur, but excision of the involved nerve segment and surgical repair of the nerve is typically required. Most commonly caused by crush injury. 5th degree “Neurotmesis” Complete transection of nerve, in need of surgical repair. Options are primary repair, nerve grafting if injury caused distance between nerve endings such that they cannot be easily re-adapted without tension, or possibly, if motor function is affected and the injury is proximal, nerve transfer. Most commonly caused by nerve transection or laceration.

Reinnervation after surgical repair

As mentioned in GENERAL INTRODUCTION, suturing of the nerve stumps using microsurgical techniques that re-adapts the interrupted nerve sheet allows axons of the injured nerve the ability to regenerate over the injury site. The regeneration rate is typically about 1 mm a day for later reinnervation of denervated peripheral tissues. However, following transection of a peripheral nerve, about one-third of the sensory neurons in the DRG die due to stress-induced apoptosis (programmed cell death; Liss et al., 1996). For surviving neurons, the regeneration and their path finding at outgrowth depend on various types of complex molecular mechanisms that stimulate or inhibit the axonal outgrowth (Lundborg and Rosen, 2007; Allodi et al., 2012; Grinsell and Keating, 2014; Dahlin and Wiberg, 2017; Krarup et al., 2017).

After axotomy, neuronal phenotype switches from a transmitter to a regenerative state, inducing the down- and up-regulation of numerous cellular components as well as the synthesis de novo of some molecules normally not expressed in adult neurons. Regardless of how well the nerve fascicles are re-adapted during surgical repair, misdirection of axonal outgrowth appears inevitable given that tens of thousands of axons may be involved. For example, cutaneous afferent axons may be misdirected to motor endplates or sensory end-organs of inappropriate modality or locations (Witzel et al., 2005). Meissner and Merkel end-organs appear to be reinnervated to a higher degree than Pacinian corpuscles (Dellon, 1981; Hallin et al., 1981; Mackel, 1985). Furthermore, during early stages of reinnervation (7–23 months postoperatively), as compared to late stages (after ≥3 years), there are at least three abnormalities: single afferent axons can show multiple cutaneous receptive fields (FA-1, SA-1), can have unusually small or large receptive fields (FA-1, SA-1), and show pronounced fatigue to repeated stimulations (SA-1: Mackel et al., 1983a; Mackel, 1985). This suggests that the transitional properties of regenerating afferents reflect unstable connections between axons and end-organ as well as immature axonal properties, including changes in axonal branching patterns in the skin.

Reinnervated sensory neurons may not only show receptor dysfunction but regularly show abnormal conduction properties characterized by abnormally high dispersion of conduction velocities, giving rise to desynchronized inputs to the CNS compared to normal conditions (Buchthal and Kuhl, 1979; Hallin et al., 1981; Krarup et al., 2017). Together, these abnormalities contribute to incomplete recovery of tactile sensory functions after reinnervation and also make prognosis for tactile recovery unpredictable. That is, the spatial misalignment of regenerating axons, which was first recognized by John Kearsley Mitchell in 1895 (Mitchell, 1895), and the abnormal axonal conduction properties may interfere with the spatial and temporal structuring of the afferent neural activity necessary for well-functioning central processing of sensory signals.

A nerve injury affecting axons of motor neurons will leave the previously innervated muscle fibers paralyzed until reinnervated. Although α motor neurons can also enter apoptosis, they tend to do so to a lesser extent than sensory neurons (Li et al., 1994; Novikov et al., 1995). Likewise, with adequate surgical treatment, including good matching of motor fascicles in the proximal and distal stump of the injured nerve, the efferent axons can, to a high degree, regenerate and reinnervate muscles. As with reinnervation by sensory axons, reinnervation by motor axons typically results in some misdirection of axons during regeneration with reference to the muscle fibers that were originally innervated. Little is known about reinnervation of muscle sensory organs (muscle spindles and Golgi tendon organs etc.).

With developments of new neuroprotection techniques, the problems of neuronal death of α motor neurons and sensory neurons in the DRG after peripheral nerve lesions may be reduced in the future (Hart et al., 2004; Hart et al., 2008). However, mismatches of axonal reinnervation concerning location and type of targeted end-organs will almost certainly remain even with an increased number of surviving neurons. Consequently, the brain's ability to learn to utilize spatiotemporal aberrant signal patterns in the populations of affected afferent neurons remains to be crucial for restoring hand function after peripheral nerve injury. By better understanding how functional plasticity of CNS might help or hinder the regaining of hand function, new approaches to treatment of nerve injuries may be identified, with the long-term goal of improving patient outcomes.

Median nerve injury Injury to the median nerve constitutes the most common form of peripheral nerve injury following injury to digital nerves (Asplund et al., 2009), and is especially detrimental for hand function because the median nerve supports the brain with sensory information from the digits that are most important for manual dexterity, i.e., the thumb, index- and middle finger. The inclusion criteria for the nerve- injured persons studied in this thesis (Paper II and III) were that they more than 2 years before the study had experienced a 5th degree unilateral median nerve injury (complete nerve transection) which was treated within 24 hours after the injury and that the injury was located at the forearm level, distal to the sites where the axons that innervate the extrinsic hand muscles leave the nerve.

These criteria defined a group of median nerve injured and reinnervated study participants with an affected sensory innervation in a well-defined hand area: the volar aspect of the thumb, index and long finger, the radial half of the ring finger, and the radial aspect of the palm, whereas the effects on muscle function where limited. The muscles that could be affected are intrinsic hand muscles that contribute to fine thumb movements (opponens pollicis muscle, abductor pollicis

brevis muscle and the superficial part of flexor pollicis brevis muscle) and to fine control of index and long finger movements (1st and 2nd lumbar muscles).

After microsurgical nerve repair, for a patient suffering a 5th degree nerve injury in the distal forearm, it may take about 200 days before sensory reinnervation of the fingertips is initiated given the typical axonal regeneration rate of 1 mm/day. About 2 years after the injury, the signals relayed by reinnervated afferent axons seem to have matured and the peripheral situation is stabilized (Mackel et al., 1983a; Mackel et al., 1983b), which corresponds to the time after which little further improvement is seen in hand function (Dellon, 1981; Rosen and Lundborg, 2001; Ruijs et al., 2005). However, even with the best available clinical treatment, a median nerve transection at the distal forearm level results in lasting impairment of hand function (Dellon, 1981; Jaquet et al., 2001; Ruijs et al., 2005; Lundborg and Rosen, 2007; Vordemvenne et al., 2007; Mavrogenis et al., 2009; Pederson, 2014).

The typical clinical outcome at late stage after an median injury at the distal forearm level is near normalized hand mobility, including for the thumb, while the tactile sensibility remains affected as evidenced by, for example, increased touch thresholds and increased moving and static two-point-discrimination thresholds (Chemnitz et al., 2013b). However, the patients rarely express their problems in sensory terms, but rather comment on shortcomings in activities of daily life and especially in tasks that require fine finger dexterity. Some activities are often described as impossible to perform, e.g., buttoning a shirt without sight, picking up small objects from a tabletop. Patients also commonly report on accidentally burning or cutting their injured hand because of impaired sensation. They also commonly comment on increased dependence on vision when performing fine manual tasks such as tying shoelaces and handling cutlery. Symptoms connected to cold intolerance are also described, including freezing more quickly, color changes, stiffness, and reduced dexterity (Chemnitz et al., 2013a).

CNS changes after peripheral nerve injury Injury to a peripheral nerve results in CNS partially or totally losing control of the affected body parts concerning sensory, motor and autonomic functions. Functional deficits caused by a peripheral nerve injury can be compensated by essentially two neural mechanisms: the reinnervation of denervated targets by regeneration of injured axons and the remodeling of nervous system circuitry related to the lost functions. Plasticity of central connections may to some degree compensate functionally for the lack of specificity in target reinnervation and distorted temporal structuring of the peripheral afferent signals; plasticity in human has, however, limited effects on disturbed sensory localization, discrimination or fine motor control after injuries, and may even result in maladaptive changes.

From the brain’s perspective, sensory information relayed through the affected nerves goes missing immediately with injury, and during the period of axonal regeneration there are no sensory signals from the previously innervated segments of the body and after reinnervation there is a spatiotemporal aberrant signal pattern in the populations of affected afferent neurons. Basic knowledge of how the brain responds to such changes come from neuronal recordings using intracortical electrodes in brains of monkeys that have been experimentally exposed to injuries to peripheral nerves (Paul et al., 1972; Kaas et al., 1983; Merzenich et al., 1983b; Wall et al., 1986; Sanes et al., 1988; Florence et al., 1994; Jones, 2000).

Findings in monkeys Studies in non-human primates have shown that functional reorganization of sensory and motor systems following peripheral damage is an activity-dependent multiple-step phenomenon affecting wide neural networks including spinal cord, brainstem, thalamus and also cortical regions directly or indirectly involved in the processing of the impacted nerve (for reviews see: Jain et al., 1998; Chen et al., 2002; Wall et al., 2002; Kaas and Collins, 2003). However, plasticity following peripheral nerve injury and deprivation has been mainly monitored in primary somatosensory cortex which will be focused on here.

Peripheral nerve transection implies an acute deafferentation which can trigger immediate and evolving long-standing plastic changes on the corresponding cortical representation areas as well as in adjacent brain regions (Merzenich et al., 1983b; Merzenich et al., 1983a; Wall et al., 1986; Garraghty and Kaas, 1991; Silva et al., 1996). Peripheral nerve injuries result in an obvious loss of evoked activity in the corresponding cortical map in response to stimulation of the skin covered by the injured nerve. Shortly thereafter, the same brain region becomes responsive to inputs originated in other body parts with adjacent

cortical representations. The result is shrinkage in size of the cortical field deprived from its original inputs, and its takeover by adjacent representations for the processing of information conveyed by intact sources. The rapid time-course of these changes suggests that there is an extended network of reciprocal connectivity across cortical representations of different sensory areas, so that sensory inputs from one finger normally inhibit existing inputs from adjacent fingers. The fast changes, occurring within minutes after deafferentation, are probably based on the unmasking of previously present, but functionally inactive, synaptic connections. Although, unmasking of connections probably involves various mechanisms influencing the balance of cortical excitation-inhibition (Isaacson and Scanziani, 2011), removal of inhibition of excitatory synapses due to reduction of GABAergic inhibition has been suggested as the most important to cause plastic changes (Sammons and Keck, 2015). The consolidation of longer lasting plastic reorganization might require structural processes consisting in the generation of new projections through sprouting of axon collaterals, dendritic elongation and branching and formation of new synaptic connections. The extent and relevance of these mechanisms, however, remain still poorly understood.

If peripheral nerves are surgically repaired and axonal regeneration occurs, the cortical remodeling may be totally or partially restored depending on the severity of the injury and the capability to accurately reinnervate their former targets.

However, misdirection in axonal paths during regeneration results in motor and sensory axons that can aberrantly reinnervate targets that do not match their original organ, function or territory. This result in an abnormal mapping between cortical representations of the affected body segments regarding both input and output functions. In response to this, within limits the primary somatosensory cortex (S1) reorganizes by showing a new and distorted mapping of the skin areas originally innervated by injured nerves (Wall et al., 1986; for reviews see: Kaas et al., 1983; Jain et al., 1998). Accordingly, the originally well-organized somatotopic digit representation can change to a mosaic-like pattern where individual digits can be represented across multiple discontinuous areas and individual cortical neurons can respond to stimuli in multiple areas of the skin rendering fractionated cutaneous receptive fields. In addition to a confused topography of inputs from the hand, the relative timing of impulses in ensembles of afferents will be distorted, which could hinder the normally occurring rapid recognition/classification of tactile stimuli (Johansson and Birznieks, 2004; Johansson and Flanagan, 2009; Pruszynski et al., 2018). In fact, a distorted timing of neural signals as such might cause remodeling of hand representation in adult S1 as a result of Hebbian-like functional plasticity (Wang et al., 1995; Pleger et al., 2001).

Findings in humans Brain plasticity in contexts of sensory and sensorimotor processing has been intensively studied. Structural changes of varying complexity have been associated with adaptations to peripheral conditions affecting perceptual and movement abilities as well as skill learning (Sluming et al., 2002; Draganski et al., 2004; Bengtsson et al., 2005; Draganski et al., 2006; Draganski and May, 2008; Scholz et al., 2009; Langer et al., 2012; Makin et al., 2013b; Makin et al., 2013a; Preissler et al., 2013). For example, persons suffering limb amputation show signs of gray matter reduction in thalamus (Draganski et al., 2006) and in cortical motor areas (Makin et al., 2013a) related to the lost limb, but also gray matter increases which might reflect increased processing associated with compensatory behavioral changes (Preissler et al., 2013). Studies have also showed that skill learning can increase regional gray matter volume in certain areas of the brain, as for example in the left PMv in professional musicians (Sluming et al., 2002) and in association areas of the parietal cortex after intensive training of a visuomotor task (juggling; Draganski et al., 2004). Furthermore, visual deprivation in adults can induce gray and white matter decreases in the primary visual pathways (Mendola et al., 2005; Xie et al., 2007; Hernowo et al., 2014) and hearing loss can result in an analogous degeneration in auditory areas (Husain et al., 2011). Thus, it is well-established that both sensory deprivation-related plasticity and use-dependent plasticity can cause structural changes in the human brain.

When it comes to median nerve injury, I only know one previous study that examined structural brain changes (Taylor et al., 2009). This study reported decreases in gray matter after injury and repair of the median and/or ulnar nerve and in brain areas not specifically associated with hand control, including areas of postcentral gyrus seemingly distant from sensorimotor hand representations. In Paper II, we asked whether such areas can be seen showing structural changes by using an alternative technique, voxel-based morphometry (VBM), which is additionally sensitive to local surface area and cortical folding compared with the voxel-based cortical thickness analysis applied in the previous study (Ashburner et al., 2003; Hutton et al., 2009; Winkler et al., 2010).

Concerning previous studies on effect of median nerve injury on brain activity in humans, fMRI studies have shown that tactile stimulation of digits can activate the contralesional S1 to a similar or even greater extent after unilateral hand nerve injury and reinnervation compared with healthy individuals (Hansson and Brismar, 2003; Taylor et al., 2009; Fornander et al., 2010; Chemnitz et al., 2015; Bjorkman and Weibull, 2017). However, these previous studies have mainly focused on stimulations within the reinnervated skin area. In Paper III, we asked to which degree median nerve injury might cause changes in cortical responses to tactile input from non-median innervated skin areas within the affected hand

since the activity within S1's finger representations dynamically interact under normal conditions and are subject to complex lateral, mainly inhibitory, intra- areal interactions (Ruben et al., 2006; Lipton et al., 2010; Reed et al., 2010; Thakur et al., 2012; Martuzzi et al., 2014). We also asked whether nerve injury might affect cortical responses to inputs from the hand of the non-injured arm since stimulation of one hand in healthy individuals also affect the brain activity within the ipsilateral S1 (Hlushchuk and Hari, 2006; Klingner et al., 2015; Tal et al., 2017). One conceivable pathway for such interactions between the S1 cortices of the two hemispheres is that signals are conveyed via callosal connections between higher-level somatosensory areas (e.g., S2) and that the effects are transmitted to S1 via feedback connections from higher-level somatosensory areas (Tommerdahl et al., 2006; Reed et al., 2011; Klingner et al., 2015). To address these issues, in Paper III, we recorded brain activity using fMRI during tactile stimulation targeting the tip of the index finger innervated by the reinnervated median nerve, but also the tip of the little finger of the same hand innervated by the uninjured ulnar nerve and the tips of the index and little fingers of the other unaffected hand. Furthermore, although it is well established that the neural processing of tactile stimuli even in early somatosensory areas varies with task requirements (Staines et al., 2002; Nelson et al., 2004), previous imaging studies on changes in brain processing after median nerve reinnervation rarely controlled adequately for the task-/attentional-set. Therefore, it is unclear whether median nerve injury can affect the functioning of top-down mechanisms that normally control the brain's processing of sensory information from the hand based on task-set.

To address this issue, in the present thesis recordings of brain activity with fMRI were compared under conditions where the participants were required to continually attend to and process peripheral afferent signals that emerged from tactile stimulations at their fingertips (i.e., during a tactile task), but also under conditions with identical tactile stimulations when of no importance for their current task (i.e., during a visual task).

Aims of projects within this Thesis

You miss 100% of the shots you don't take. Wayne Gretzky (1961 - )

The overall aim of the empirical studies within this thesis was to investigate changes in the brain after median nerve injury, i.e., after surgical repair, regeneration and matured reinnervation. To reveal changes in the primary somatosensory and motor areas as well as areas beyond these primary areas, VBM (Voxel-Based Morphometry) was applied to determine structural changes in the structure of gray and white matter while fMRI (functional Magnetic Resonance Imaging) was applied to determine changes in cortical activation during functional tasks.

Specific Aims Paper I: The traditional view is that sensory perception relies on brain areas specialized to process specific sensory modalities has been challenged by the metamodal theory of brain function, which posits that brain areas contribute to sensory processing by performing specific computations – those for which they are specialized – regardless of input modality (Pascual-Leone and Hamilton, 2001). We set out to:

• investigate differential brain activity during tactile stimulation targeting the different sub-modalities of touch by delivering tactile stimulations of oscillatory frequencies (3/20/100 Hz). These are for each frequency principally based on inputs from a specific type of cutaneous mechanoreceptor in the glabrous skin of the hand (3Hz: SA-1, 20Hz: FA- 1 and 100 Hz: FA-2; Talbot et al., 1968; Freeman and Johnson, 1982; Johansson et al., 1982; Lofvenberg and Johansson, 1984). • investigate whether detection of tactile stimulation by the fingertips at hearable frequencies (> 12 Hz) would engage auditory areas because of their known computational specialization for detecting temporal features encoded in periodic ∼mechanoreceptive afferent signals. We also considered that visual areas might be engaged for infrasonic frequencies at which humans visually perceive object motion (< 12 Hz). • test if such potential effects could depend on whether the tactile ∼ stimulations were critically used or not (i.e., task-set dependent effects).

However, to test these hypotheses it was also necessary to develop and test a novel apparatus and paradigms for implementation of the required well-controlled mechanical stimulation protocols in the electromagnetically harsh MRI- environment. The development of this apparatus and tactile paradigms to test effects of task-set was also important for later investigation of persons suffering median nerve injury presented in Paper III.

Paper II: Inspired by recent findings on activity-dependent structural plasticity in the adult brain, the aim of the study presented in Paper II was to test whether

the imperfect reinnervation and the resulting deterioration in fine dexterity after repaired median nerve injuries can cause structural changes in the brain.

• Specifically, we expected that activity-dependent structural changes might occur in the hand area of the S1 contralateral to the injured side since experimental peripheral nerve lesions in animals can cause substantial functional reorganizations both in contralesional S1 and M1. We also considered that brain areas functionally associated with higher- level planning and control of hand actions might be affected because of altered neural processing linked to more general changes in behavior related to hand use after median nerve injury.

Paper III: Previous fMRI studies in humans have reported that tactile stimulation an affected digit can activate the contralateral S1 to a greater extent after median nerve damage and reinnervation than in healthy individuals. However, there are no systematic studies of how median nerve injury might affect the responsiveness of the contralesional S1 during tactile stimulation of non-median innervated areas within the affected hand and stimulation of the hand of the non-injured arm.

• A specific aim of the study of Paper III was to investigate with fMRI whether and how median nerve injury can affect the complex lateral interactions that normally occur between the finger representations within S1 and its responsiveness to stimulation of the ipsilateral hand. • Another aim of this study was to test whether median nerve injury might affect modulation of brain processing of sensory information from the hand based on task-set, primarily focusing on the contralesional S1. • Since median nerve injury leads to behavioral changes related to various aspects of hand function, we also sought for injury effects on brain activity in cortical areas outside the primary sensorimotor cortices when the study participants performed demanding tactile tasks engaging their fingertips.

Materials and Methods

I have not failed. I´ve just found 10 000 ways that won’t work. - Thomas A. Edison (1847 – 1931)

In this section, I give an overview of, and rationales for the techniques and methods used in the three papers on which this thesis is based.

Brief description of methodology and rationale for each paper

Paper I In Paper I, we used fMRI to identify brain areas differentially involved in the processing of tactile information depending on the temporal frequency of oscillatory skin displacements. To this end, an apparatus was designed to provide precise tactile stimulation to fingertips in the MRI-scanner. Sixteen healthy individuals performed a tactile detection task during which brain activity was measured. The participants performed the task receiving oscillatory skin displacements at 3-, 20-, and 100-Hz delivered to the middle finger of the left and right hand in separate consecutive scanning sessions. The detection task was based on a variant of the von Bekesy’s threshold-tracking method (von Bekesy, 1947). In the threshold-tracking tasks the participants responded to detecting the stimuli at the fingertip by pressing a button with the big toe contralateral to the hand receiving the tactile stimuli.

The frequencies (3, 20, 100 Hz) were chosen because tactile detection at these frequencies is for each frequency principally based on inputs from a specific type of cutaneous mechanoreceptor in the glabrous skin of the hand (3Hz: SA-1, 20Hz: FA-1 and 100 Hz: FA-2; Talbot et al., 1968; Freeman and Johnson, 1982; Johansson et al., 1982; Lofvenberg and Johansson, 1984). In addition, they represent both infrasonic (3 Hz) and hearable (20 and 100 Hz) frequencies. As presented in the introduction, these frequencies evoke different percepts. Stimulation at 3 Hz is often described in visual terms, such as a gradual position change of an object moving the skin (Lofvenberg, 1984), and stimulation at higher frequencies are described in more auditory related terms (20 Hz: wobble or flutter, 100 Hz: diffuse vibration; Talbot et al., 1968).

The rationale for employing the threshold-tracking task was threefold: First, threshold-tracking kept the subjects actively engaged in the task because they had to use the tactile information continuously. Second, threshold-tracking ensured that the effective stimulus intensity was normalized in perceptual terms across the different frequencies. Since the sensitivity to tactile stimuli varies with frequency (Talbot et al., 1968; Lofvenberg and Johansson, 1984) and the effective stimulus intensity can influence brain activation levels (Backes et al., 2000; Arthurs and Boniface, 2002; Nelson et al., 2004; Siedentopf et al., 2008), normalizing the stimulus amplitude to perceptual threshold ensured that any observed effects were related to differences in stimulus frequency rather than

intensity. And third, by keeping the signals near perceptual threshold, we minimized both central and peripheral adaption to maintained tactile stimulation (Lundstrom and Johansson, 1986; Leung et al., 2005; Tommerdahl et al., 2005).

Important for the interpretation of potential differences in brain activity between trial types, we wanted to have similar amounts of button presses performed by each participant. For this we had programmed an adaptive algorithm that also prevented the participants from simply controlling the stimulus amplitude by rhythmically pressing the button (see methods section in Paper I for further details).

In addition, to analyze whether task set would affect brain activity in relevant brain areas, the participants also performed a visual threshold-tracking task during which they received the same tactile stimulations of the fingers as delivered during the tactile trials. That is, during each tactile trial, the oscillatory skin stimulation was recorded and then replayed to the fingertip in the subsequent visual trial. During the visual trials the participants’ task was to detect a circular image displayed in the center of the computer screen against a white background.

Paper II For Paper II, we used VBM, an MRI technique that enables investigations of regional structural changes in gray and white matter of the brain (Ashburner and Friston, 2000, 2001), on sixteen right-handed nerve-injured adults. Sixteen healthy individuals matched by age, sex and handedness served as control group. The nerve-injured had more than two years earlier (range: 2.4 – 17.1 years) suffered sharp complete transection (Sunderland 5th degree nerve injury) of the median nerve in their left (n = 11) or right (n = 5) distal forearm followed by microsurgical epineural end-to-end repair within 24 hours. We chose to exclude patients with injuries that had happened less than two years after such an injury, since most of the attainable recovery of hand function is believed to have occurred at about two years after such injuries.

Paper III In Paper III, we used fMRI to identify brain areas with changed activity after median nerve injury as compared to age, sex and handedness matched controls. The same apparatus and similar threshold-tracking tasks as used in Paper I was applied together with an additional tactile task that consisted of detecting deviant oddball stimuli delivered among trains of standard suprathreshold stimuli. The same participants as in Paper II performed tactile tasks in the MRI-scanner. In this paper, for presentational reasons, we focused on the eleven participants that

has suffered left hand injury and their controls, although also right-side injured and their controls was analyzed but not presented.

During the tactile threshold-tracking task, in an interleaved manner, the index or the little finger of either the left of the right hand received oscillatory tactile stimulations at 20 Hz, and the left and right hand was addressed in separate consecutive scanning sessions. As in Paper I, we included visual threshold- tracking trials interleaved with tactile threshold-tracking trials. The visual trials were included to investigate whether the nerve-injured participants maintained the ability of healthy individuals to suppress tactile afferent-induced activity in somatosensory brain areas while task-irrelevant (Nordmark et al., 2012; Hsiao et al., 1993; Johansen-Berg et al., 2000). Also here did we apply an adaptive algorithm to normalize the amount of button presses (and decisions) between trial types and participants, important for the interpretation of potential differences in brain activity between these (see Paper I and III for details).

In the tactile oddball-detection task, the participant’s assignment was to detect rarely occurring epochs of 40 Hz sinusoidal stimulations among regular epochs of 20 Hz stimulations. That is, the participants were requested to respond whenever they perceived a deviant stimuli. As in the threshold-tracking task they performed the oddball task with their index and little finger of their right and left hand and responded by pressing the pushbutton with the big toe of the contralateral side. The principal rationale for including the oddball task was to complement the threshold-tracking task providing threshold near stimulation amplitudes with a task with suprathreshold stimulations (4 x perception threshold) that also involved frequency discrimination. It has been reported in animal studies that various brain areas, including S1, may be more critically involved in tasks requiring tactile frequency discrimination than in tasks requiring detection of stimuli at threshold (LaMotte and Mountcastle, 1979; Romo et al., 2012). Although, the participants were much less active in button- presses in the oddball task (0.03 s-1) than in the threshold-tracking task (0.2 s-1), they still needed to continuously focus on the tactile stimulation. Importantly, the study was not intended to analyze brain activity in connection with detection of oddballs; such an analysis would have required a significantly higher frequency of oddball stimuli.

We chose to use 20 Hz stimulations since they primarily evoke responses in the FA-1/Meissner corpuscles, which seem most reestablished after reinnervation following peripheral nerve injury (Dellon, 1981). This notion was supported by a pilot study in which I exposed persons with reinnervated unilateral median and ulnar nerve injury to vibrotactile stimuli of different frequencies (3, 20 and 100 Hz, data not published). The findings indicated that the ratio of tactile thresholds between hands was lowest at 20 Hz stimulation.

The study presented in this paper had several critical features that we considered important for revealing possible functional changes in the brain after median nerve injury: . It had passed at least two years between the participants´ nerve injury and repair and their participation in the study (range: 2.5 – 17.1 years), i.e., a time after which these patients show little further functional improvement. . The paradigms forced the participants to actively attend the stimuli. This implied, and experiments verified, that all participants used the tactile afferent signals evoked by the stimuli. Indeed, it is well established that the neural processing of peripheral tactile input vary with task requirements even in early somatosensory areas as also shown in Paper I (Johansen-Berg et al., 2000; Brooks et al., 2002; Johansen-Berg and Matthews, 2002; Puckett et al., 2017; Nordmark et al., 2012). . The stimulus paradigms used assured that the effective stimulus intensity was normalized to perceptual thresholds across the different stimulation sites and participants, including the site in the reinnervated skin area of the nerve-injured participants. We consider the individually normalized stimulus intensities important to allow comparisons of brain activation both within a person and between persons given that the intensity can affect the level of brain activity (Arthurs and Boniface, 2002; Nelson et al., 2004; Siedentopf et al., 2008; Backes et al., 2000). . The experimental design made it possible to investigate the effects of injury not only when a reinnervated finger was stimulated in the tactile task but also injury effects on the responses in especially the contralesional S1 to stimulation of fingers supplied by non-injured nerves, both belonging to affected hand and the other hand.

Material

Participants, inclusion and exclusion criteria Paper I: The participants included 16 right-handed healthy individuals (eight women and eight men, aged 19–28 years) without any neurological disorder. They were recruited to attend after they had reported interest in participating in the study after reading about the study on posters at Umeå University’s campus area.

Paper II and III: The participants included 16 adult right-handed persons that had suffered median nerve injury and 16 healthy controls, matched to age, sex and handedness. The nerve-injured participants were recruited from patients

(age 18-65 years) treated at the Hand and Plastic Surgery Clinic, University Hospital of Umeå, Sweden, for accidental sharp nerve injury isolated to their median nerve. To obtain a homogeneous group of study participants with well- defined and well-characterized type of nerve injury, we chose to investigate persons that had suffered unilateral median nerve injury at the level of the distal forearm. In addition they were required to have had undergone early surgery with primary nerve repair and had reached the time after surgery considered needed for a steady state recovery of the function of the affected hand. After excluding potential participants because of a history of other neurological deficits or other hand injuries, diabetes, or other disabilities, 36 participants were contacted and out of these, 18 met the inclusion criteria and agreed to participate in the study. Two of them suffered claustrophobic reactions when entering the MRI-scanner hindering their participation, leaving us with a total of 16 nerve-injured participants.

Methods

Magnetic Resonance Imaging – a brief description MRI (Magnetic Resonance Imaging) is used clinically in the diagnosis of pathology in for instance brains, organs, limbs, etc. The basic technique was initially used to investigate resonance within bulk matter (Nobel Prize in Physics – 1952, Bloch and Purcell, 1952), but was later further developed to also investigate biological tissues (Nobel Prize in Medicine or Physiology – 2003, Lauterbu and Mansfield, 2003). In short, an atomic nucleus—most commonly hydrogen—is perturbed by a weak oscillating magnetic field (radio-frequency pulse or RF-pulse) whilst located in a strong static magnetic field. The strong static field positions the axis of all hydrogen atoms along the magnetic field such that they have a low-energy or high-energy spin while the oscillating magnetic field will cause resonance in the intrinsic frequency spin of the atoms at a given frequency (the Larmor frequency) depending on the static magnetic field. The oscillating field thus provides the means for exciting the nuclei with RF-pulses, lifting the nuclei to their high-energy state. When the pulse vanishes, the excited nuclei will relax to their low-energy state and at the same time produce an electromagnetic signal with a frequency characteristic of the magnetic field at the nucleus. By recording the electromagnetic signal the amount of hydrogen atoms can be measured and by adding slight differences in the magnetic field of three planes of the area measured (X, Y, Z), using a so called gradient field, differences in the electromagnetic signals recorded can be used to define the location of the signals. Eventually, by detecting the signals and analyzing them, sets of in-plane slices that build up a 3D volume of the assessed body part are produced (Huettel et al., 2009).

Functional Magnetic Resonance Imaging– a brief description BOLD-fMRI (Blood Oxygenated Level-Dependent functional Magnetic Reso- nance Imaging) was developed by Ogawa and colleagues and first published in 1990 (Ogawa et al., 1990). fMRI is currently a vastly used non-invasive neuroimaging technique for measuring brain activity by detecting changes associated with regional cerebral blood flow. The technique takes advantage of the magnetic properties of hemoglobin molecules that have released oxygen in tissues, i.e., deoxyhemoglobin, to indirect measure the amount of blood flow in small brain volumes, so called voxels, which usually comprise about 3×3×3 mm. Changes in blood flow are detected as changes in the BOLD signal such that the increased flow of oxygenated blood, considered to reflect local increased glucose consumption, is delivered in excess of the metabolic requirement of oxygen which leads to wash out of the magnetic deoxyhemoglobin (Arthurs and Boniface, 2002; Glover, 2011).

The BOLD-signal was earlier considered to indirectly measure regional neuronal activity, however, recent studies have questioned this coupling, and it is more or less agreed today that the signal not only reflects neuronal activity but to some degree also consumption of glucose by supportive cell types (Hillman, 2014; Logothetis, 2008).

The physiological blood-flow response, the hemodynamic response, lags the cellular events that trigger it by a couple of seconds, since the vascular system respond with some delay to the brain’s regional metabolic demands. Therefore, there is a delay in both the increase and the decrease of the BOLD signal after local changes in brain activity: the peak in blood-flow typically lags a stimulus by about 5 seconds and the increase in the signal takes several seconds to vanish after the stimulus has ended.

To sample brain activity of the whole brain, and to produce three-dimensional fMRI brain volumes of data, sampling of fMRI-data requires sampling of multiple adjacent thin brain slices. Typically these are sampled with a rate of one complete brain volume every 1.5 or 2 s, which is called the repetition time. By repeating the stimulus that is to cause the brain activity, and sample data containing hemodynamic responses during all stimulations, each voxel’s BOLD-signal response can be correlated to a temporal model of the normal hemodynamic response. By correlating the measured and the temporal hemodynamic response, a β-value is calculated for each sampled voxel. It is thus possible to analyze statistically if a specific voxel shows a significantly positively or negatively correlated BOLD-signal with the stimulations provided. Given this, fMRI-data usually contain a spatial resolution of the sampled voxels of about mm3 size and a temporal resolution that is in the order of seconds.

These spatial and temporal limitations are common to all fMRI experiments and have led to two main experimental paradigms: either repetitive blocks (“trials”) of stimulations or single stimulations (“events”). The trials or events are interleaved with interstimulus intervals of defined rest during which the recorded BOLD-signal is considered to reach its baseline value – corresponding to the measured voxels baseline energy consumption.

Voxel-Based Morphometry – a brief description Voxel-based morphometry (VBM) is a neuroimaging technique for measuring gray or white matter volumes to detect local differences in brain anatomy (Ashburner and Friston, 2000, 2001). It uses the signal intensity from T1-MR images to compare regional gray or white matter concentration between groups of participants or longitudinally within specific participants (Good et al., 2001; Draganski et al., 2004; Draganski and May, 2008). Sampled MR images are segmented (separated) into gray matter, white matter and cerebrospinal fluid, then spatially normalized to an average-brain template and spatially “smoothed” to account for possible anatomical differences between individuals that are not dealt with by the spatial normalization. As for the fMRI-data, the voxels sampled and analyzed in the structural analysis contain a vast number of neurons, and the analysis method does not measure the actual number of neurons but rather the signal intensity of the gray matter which is translated to gray matter concentration in gray matter areas and white matter concentration in white matter tracts.

Novel apparatus for tactile stimulations in the MRI-scanner Roland S. Johansson, Göran Westling and Anders Bäckström designed, developed and produced the apparatus used in Paper I and III for tactile stimulations of the fingertips in the MR environment. It consists of a rectangular box containing tactile stimulators that can act on the left or the right hand. The fingertips of the stimulated hand contacted the vertical surface of one side of the box and the thumb contacts the opposing surface. For each stimulated fingertip the oscillatory skin displacements were created by the movements of a protruding ridge extending in the proximal-distal direction of the fingertip (see methods in Paper I for details). To instruct the participants about the procedure and how to perform the tasks in the MRI scanner, before the MR scanning session they practiced short versions of the different tasks on a replica of the apparatus outside the MRI-scanner.

Data acquisition and analysis For all data acquisitions, processing and statistical analysis of the MRI-data, we used standardized methods. All functional and structural brain scans were

performed at the University hospital of Northern Sweden, Umeå. For Paper I we used a 3-Tesla MRI-scanner (Achieva 3.0, Phillips Medical Systems, Eindhoven, The Netherlands) using an eight-channel SENSE head coil. For Papers II and III we instead used a scanner not available during the data collection for Paper I, i.e., Discovery MR750 3-Tesla scanner equipped with a 32-channel head coil (GE Medical Systems, Milwaukee, Wisconsin, USA).

For Paper II and III we had a priori hypotheses about structural (II) and functional (III) changes in the primary somatosensory cortices based on previous research. However, considering the novelty of the tasks performed and the paradigms applied, we were also open for effects in other parts of the brain. Therefore, in these papers, we performed both analyses focusing on the hand areas of the primary somatosensory cortices with statistical thresholds corrected for multiple comparisons in the search area, so called small volume correction (SVC) analysis, and whole-brain analysis with correction for multiple comparisons taking into account all registered voxels. fMRI-data Selection of MRI parameters during fMRI data acquisition depend on the analyses to be pursued because there is a trade-off between sampling the whole brain which requires comparatively thicker slices and focusing on smaller details that requires a thinner slice thickness. For these reasons and given the scanners available for our studies, in Paper I we used a slice thickness of 4.65 mm with an in- plane resolution of 3.44 × 3.44 mm2 (64 × 64 matrix), whereas in Paper III we used a slice thickness of 4.5 mm and an in-plane resolution of 1.95 × 1.95 mm2 (128×128 matrix).

As detailed in the papers, fMRI-data were analyzed with the SPM software4. These analyses included standard procedures such as correcting for head movements, spatial and temporal smoothing and normalization to a standard brain template (the standard Montreal Neurological Institute, MNI, EPI template) to allow group analysis. We applied a rather limited spatial smoothing with an isotropic Gaussian kernel of 6 mm full width at half maximum (FWHM) due to the fact that we did not tolerate larger spatial shifts of potential effects given our interest in location of the effects (Mikl et al., 2008). High-pass filtering and grand means scaling were applied to reduce participant-specific drifts in the BOLD-signal and slow global changes in BOLD activity. Batching of analyses, visualization, and extraction of parameter estimates across clusters were performed with software developed in-house (DataZ).

4 http://www.fil.ion.ucl.ac.uk/spm

Our main rationale for choosing Analysis of variance (ANOVA) as the statistical model for analyzing our fMRI-data was to optimize our ability to find reliable effects with our novel experimental paradigms. ANOVAs, as compared to paired t-tests have the advantage of detecting effects between given factors with fewer statistical tests. For instance, using Student’s t-test for analyzing the 2 × 3 design in Paper I (2 hands × 3 frequencies), would have required 15 t-tests which, if not handled properly with correction for multiple tests, would result in p-value inflation. In contrast, given that the underlying assumptions behind the analysis are fulfilled, the corresponding ANOVA allows proper conclusions of both the main effects and any interactions.

We applied group level analysis using SPM’s flexible factorial design (Paper I) and full factorial design (Paper III) for statistical inferences; the former does not allow repeated measures analysis of group effects, i.e., will produce false positives if comparing groups of participants (McLaren, 2014). SPM enables analyses of at most 2 × 3 matrices. Accordingly, it does not allow analyses of designs that otherwise would have suited us, e.g., a multifactorial design, such as, group × hand × finger × task.

In addition, to explore and evaluate brain activations common to different experimental conditions and participant groups with reference to the baseline activity recorded during the intertrial periods, we performed conjunction analyses (Nichols et al., 2005).

In addition to whole-brain analyses performed for Paper I and III, for Paper III, given the strong a priori assumption of effects in the hand primary sensorimotor cortex after peripheral nerve injury, we performed constrained small volume correction (SVC) analyses to the pre- and postcentral gyrus. In these analyses we applied a cluster extent threshold of 20 continuous voxels combined with a voxel threshold of P <0.005.

VBM data The T1-MR images for VBM analyses were sampled with 1 mm thick slices and an in-plane resolution of 0.49 × 0.49 mm (512 × 512 matrix). Here we chose a FWHM of 8 mm as recommended in the SPM manual5. We then entered the preprocessed single-participant gray and white matter images into group analyses using the VBM-software provided in SPM8 (Ashburner and Friston, 2000). A general linear model analysis was performed at each voxel where group of participants constituted the predictor of prime interest. We used total intracranial gray and white matter volume as a “global nuisance effect” both in the gray and white matter analysis (global normalization, ANCOVA in SPM8).

5 https://www.fil.ion.ucl.ac.uk/spm/doc/#Manual, p. 474

Furthermore, to statistically control for influences of age and sex on regional gray and white matter volumes (Good et al., 2001), age constituted a continuous predictor and sex a categorical predictor in all analyses. To protect against false positives while at the same time retain the power to detect meaningful effects in the face of multiple comparisons across image voxels (Roiser et al., 2016), we subjected the preprocessed morphometric images to a two-step threshold approach in all statistical analyses of effects of nerve injury (Forman et al., 1995). We combined a voxel-based threshold of P < 0.005 with a cluster size threshold of P < 0.05 corrected for so-called family-wise error rate.

Given the a priori assumption of effects in the hand primary sensorimotor cortex after peripheral nerve injury, we performed small volume correction (SVC) analyses for these areas, focusing target spheres with a diameter of 20 mm centered on the MNI-coordinates of the hand SMC established in a meta-analysis based on fMRI (Mayka et al., 2006).

For the whole-brain analysis a cluster size threshold of one-half of that in the standard double-threshold approach was used to search for areas showing symmetrically localized bilateral effects on the gray and white matter volume. We motivate this approach with the argument that bilateral homologous effects with high probability indicate meaningful effects given the strong structural and functional connectivity between bilaterally connected areas (Salvador et al., 2005; Hofer and Frahm, 2006; Johnston et al., 2008; Chao et al., 2009).

We chose to report the standardized effect sizes using Cohen’s d, which describes the effect size as the difference between the means of the compared groups divided by the pooled standard deviation.

Ethical considerations The studies on human subjects included in this thesis was approved by the Regional Ethical Review Board in Umeå, Sweden (DNR 07-087M). In all the studies within this thesis, the principles stated in the Declaration of Helsinki (2008, sixth version) were followed in study implementation and design.

Results

People love chopping wood. In this activity one immediately sees results. Albert Einstein (1879-1955)

Summary of Results presented in Paper I The main findings of Paper I were:

• that the novel apparatus and paradigms for assessment of tactile sensibility of the fingertips based on well-controlled mechanical parameters were found to operate flawlessly in the MRI-environment • the demonstration of an effect of stimulus frequency during the tactile threshold-tracking in two cortical regions: the auditory cortex and the visual cortex o the BOLD-response was stronger in the left auditory cortex during detection of tactile stimuli at hearable frequencies (20 and 100 Hz), compared to during detection of tactile stimuli at infrasonic frequencies (3 Hz). o the BOLD-response was suppressed in the right lingual gyrus of visual cortex during detection of tactile stimuli of infrasonic frequencies (3 Hz), whereas there was no difference compared to baseline during detection of tactile stimuli at hearable frequencies (2o and 100 Hz). • the finding that the frequency-dependent effects in both these areas disappeared when the participants performed a visual threshold-tracking task despite receiving identical tactile stimuli as in the tactile threshold- tracking tasks.

Analysis of behavioral data showed the new methodology met the prescribed goals. For example, the tactile detection thresholds measured when the tactile threshold-tracking tasks were performed in the scanner were similar and varied with stimulus frequency in the same manner as previously reported for oscillatory displacements of the skin of the fingertips in traditional laboratory settings. Furthermore, in accordance with the methodological design, the amount of response button actions (performed with the foot contralateral to the stimulated hand) were similar for all tested conditions both under the tactile and the visual threshold-tracking trials. This minimized the risk that comparisons of brain activity between different test conditions were affected by the neural processes related to frequency of decision-makings and response selections.

Brain activity, measured as BOLD-signals, differed depending on hand involved during the tactile threshold-tracking task: the primary somatosensory cortex contralateral to the hand receiving the tactile stimuli showed higher activity, as did the primary sensorimotor cortex of the foot contralateral to the foot responding with button presses and the ipsilateral anterior foot areas of the cerebellum.

A comparison of BOLD-signals during the tactile and the visual task, independent of hand receiving tactile stimuli (and foot controlling push button), showed widespread effects. Higher activity during the tactile task was observed bilaterally

in the supplementary motor areas, in the parietal cortex extending from the hand area of the postcentral gyrus and anterior PPC along a belt that included the secondary somatosensory cortex and bilaterally the inferior frontal gyrus corresponding to PMv. Higher brain activity during the visual task was seen in vast areas of the visual cortices of the occipital lobe.

Analysis of common brain activity during the tactile threshold-tracking task independent of stimulated hand and stimulation frequency showed bilateral activation in the perisylvian cortex including the secondary somatosensory cortex, the superior temporal cortex, inferior parietal lobule, supramarginal gyrus and insula. Bilateral activation was also present in pre-SMA, anterior cingulate cortex and in the putamen and palladium of basal ganglia. Bilaterally, but to a larger extent for the right hemisphere, a cluster with increased activity stretched along the anterior of precentral gyrus from the posterior part of the ventral premotor cortex and insula medial to the dorsal premotor cortex. A separate prominent cluster engaged the right dorsolateral prefrontal cortex.

Summary of Results presented in Paper II The main findings of Paper II were that the median nerve-injured study participants, compared with controls, showed:

• reduction of gray matter in two premotor areas independent of side of injury (left and right forearm). • reductions of white matter pathways that connect premotor cortices of the two hemispheres independent of injured side. • increase, independent of injured side, of gray matter bilaterally in a motion-processing visual area located in the middle temporal gyrus, also independent of side of injury. • reduction of gray matter in the right M1 of left-sided injured participants.

Analysis of changes in gray matter concentration restricted to the primary sensorimotor cortex showed decrease in an area of the contralateral primary motor cortex of the left side median nerve-injured participants.

Whole-brain analysis revealed gray matter decreases for the nerve-injured, independent of injured side, in the left ventral premotor cortex (PMv) and the right dorsal premotor cortex (PMd). White matter decreases were found in the middle part of corpus callosum and in a sector of white matter subcortically radiating from corpus callosum to the right PMd. The decrease in gray matter in the right PMd and the adjacent decrease in white matter in the right radiation of corpus callosum showed a positive correlation for the nerve-injured participants.

Whole-brain analysis also revealed bilateral areas in extrastriate visual cortex with increased gray matter concentration for the nerve-injured participants, namely bilaterally in the posterior part of middle temporal gyrus near the occipito-temporal junction. The location of this effect was attributed to the medial superior temporal area (MST) in the anterior-superior zone of the human homolog of the macaque motion complex (V5/hMT+) of the visual cortex.

Summary of Results presented in Paper III The main findings of Paper III were that the left-sided median nerve injured study participants, compared with controls, showed:

• higher brain activity in the hand representation of the contralesional right primary somatosensory cortex (S1) when performing tactile tasks (tactile threshold-tracking and tactile oddball-detection tasks) o when using the left reinnervated index finger, but also when using the little finger of the left-hand and the fingers of the other hand innervated, all innervated by uninjured nerves. • increased brain activity in prefrontal cortical areas implicated in higher- level behavioral processing. • that there were no effects of injury when the participants performed a visual task while they received identical tactile stimuli as in the tactile threshold-tracking task.

Behavioral data showed that the detection thresholds measured during the tactile threshold-tracking task based on 20 Hz oscillatory displacements of the skin of the fingertips were similar for the nerve-injured and the control participants, with the exception of the reinnervated index finger of the nerve-injured that showed an increased threshold. The amount of button presses per trial were similar for both groups of participants during matching test conditions.

During both the tactile threshold-tracking and the oddball-detection trials, the nerve-injured participants showed increased brain activity in contralesional S1 in a zone stretching from the index to the little finger representation regardless of stimulated finger, i.e., left and right index and little finger. There was no effect of nerve injury in the corresponding areas of the left, ipsilesional, S1. Analysis of brain activity within these areas during visual threshold-tracking showed no effect of injury, suggesting that the injury effect depended on task-set.

Analysis of brain activity during both the tactile threshold-tracking task and the oddball-detection task showed individual representations of the index and little finger in contralateral S1, for both hands and for both the controls and the nerve- injured participants. For both identified finger representations in contralesional S1, the increase in brain activity of the nerve-injured compared to the control participants was similar regardless of which finger was stimulated. In fact, for

both finger representations, the control participants showed negative BOLD- signals when the other tested finger of the same hand and when the fingers of the ipsilateral right hand received tactile stimuli, whereas the injured participants basically lacked such negative BOLD responses within the identified finger representations.

The whole-brain analysis showed a main effect of injury in three areas of the prefrontal cortex when the reinnervated left index finger was engaged in the tactile tasks: the dorsal anterior cingulate cortex, the left ventrolateral prefrontal cortex (VLPFC) on the orbital part of inferior frontal gyrus and the right dorsolateral prefrontal cortex (DLPFC) of the middle frontal gyrus.

Analysis of brain areas that were jointly activated independent of stimulated hand, task and injury during the tactile tasks showed results that were practically indistinguishable from those of the conjunction analysis of the BOLD-signals recorded during the tactile threshold-tacking task performed by healthy adults in Paper I in which common activity independent of stimulated hand and stimulation frequency was analyzed.

Discussion

It is the mark of an educated mind to be able to entertain a thought without accepting it. Aristotle (384-322 BC)

The overall aim of this thesis has been to expand the knowledge of structural and functional changes in the brain occurring as a consequence of median nerve injury.

In a first study, presented in Paper I, we addressed 16 healthy individuals to validate the use of a novel apparatus for tactile stimulation of fingertips with high- precision in the MRI-scanner and developed a task paradigm that would later be used to study persons with median nerve injury (Paper III). In Paper I we provided evidence for metamodal processing in healthy humans, specifically that tactile stimuli with frequency contents in the audible range are processed not only in the classical primary somatosensory cortical areas but also in the primary auditory cortex.

Papers II and III addressed persons that had suffered median nerve injury and had regained a stable degree of sensory function after reinnervation of the hand. In Paper II we addressed structural changes in the brains of the nerve-injured using the MR-technique voxel-based morphometry. The VBM-analysis showed that median nerve injury caused gray matter reductions in two premotor areas (left PMv and right PMd), two white matter regions that correspond to the interhemispheric connections between the premotor areas of the two hemispheres, and bilateral increases of gray matter volume in an extrastriate brain area in the middle temporal gyri involved in visuomotor guidance of the hands. Also, the left side injured (n =11) showed decreased gray matter volume in the contralesional primary motor area of the hand. These structural findings are interpreted as expressions of activity-dependent plasticity secondary to the consequences of the nerve injury, given that the neurons in the areas showing structural changes are multiple synaptic connections away from the injured neurons of the peripheral nerve.

In Paper III we focused on the eleven left-side nerve-injured participants and their controls and showed that the entire hand area of the contralesional S1 – also including the representation of the little finger innervated by the uninjured ulnar nerve – had increased activity during tactile tasks independent of finger or hand being stimulated. We interpret these general increases in activity in the contralesional S1 hand area as changes in GABAergic inhibition potentially promoting enhanced functional reorganization capacity. We also found increased brain activity in three areas in the prefrontal cortex that have been implicated in higher-level goal-oriented behavioral processing such as decision-making and response selection, suggesting that such processing is computationally more demanding during tactile tasks after median nerve injury.

In the sections that follow, I will summarize discussions for each paper, provide a synthesis of the findings, discuss the reliability of the findings, and lastly

comment on open ends that should be further studied regarding peripheral nerve injury affecting hand function.

Paper I The main novel finding in Paper I was that vibratory fingertip stimulation with frequencies in the hearable range affected brain activity in the primary auditory cortex, whereas stimulations at inaudible (infrasonic) frequencies did not. That is, the effect was driven by the 20 and 100 Hz stimulations but not by the 3 Hz stimulation. Although we did not map out A1 with auditory stimuli, Heschl’s gyrus, to which the activity was localized, is considered as one of the more reliable anatomical structures in terms of functional and structural coupling in the brain. Previous studies have shown responses in the auditory cortex to touch, both in man (Foxe et al., 2000; Caetano and Jousmaki, 2006; Schurmann et al., 2006) and monkey (Schroeder et al., 2001; Fu et al., 2003) and with different types of techniques such as intracortical recordings (monkeys) and fMRI and MEG (humans). However, in none of these studies were stimuli with a frequency content outside that of the hearable range presented. By also presenting effective tactile stimuli at inaudible (infrasonic) frequencies, we could show that the responses to touch stimuli in the auditory cortex were exclusive to stimuli with frequencies within the hearable range. Accordingly, our conclusion is that— whether afferent signals originate in the mechanoreceptors in the cochlea or the skin—the auditory cortex can be recruited to take advantage of its specialization in detecting temporal features encoded in periodic afferent signals.

On a speculative note, it is intriguing to consider the similarities between hearing and touch, given that both percepts are caused by mechanical stimulations. Indeed, similarities between hearing and touch was noted already in the 19th century by Ernst Heinrich Weber (Weber, 1851) and was quite extensively investigated by George von Bekesy (von Bekesy, 1947, 1959). The increase in vibrotactile perceptual thresholds caused by simultaneous auditory stimulation with frequency-matched pulsed tones (Verrillo and Capraro, 1974) also matches our results. That is, the elevation in tactile threshold would depend on the auditory cortex being caught up by the auditory stimuli (i.e., the primary modality), reducing its utility as a coprocessor available to the somatosensory system.

From an evolutionary point of view, it is worth to comment on the possibility that there is a common ancestor of the low-threshold mechanoreceptors located in the skin and the mechanoreceptors in the organ of Corti in the cochlea of the inner ear. As an example, fish have sensitive pressure receptors in the lateral line system, which can detect vibrations relayed through the water, enabling, for instance, a shark to detect stimuli in the range 25-50 Hz (Bleckmann and Zelick, 2009). Perhaps did these pressure organs evolve in two separate ways through

evolution of land living creatures, one ending up as receptors in the cochlea of the ear (Popper and Fay, 1997; Fay and Popper, 2000) and one as pressure receptors in the skin. Along the same lines, a common “pressure sensory cortex” would have been divided into what is currently known as primary somatosensory cortex (S1) and primary auditory cortex (A1). Indeed, these cortical areas are in close vicinity in humans, and are even closer in “lower mammals” (see: Kaas, 2004, for the relationship of S1-A1 in the evolution steps represented by opossum, rat, owl monkey and more).

Stimulations of the fingertip at 3 Hz also affected the activity in the lingual gyrus. We speculate that this BOLD-signal suppression represents decreased processing in that area. This suppression may reflect a mechanism that contributes to task performance by restricting the crossmodal responsiveness of this area to object motion. In the absence of any concurrent visual information about moving objects, such suppression may minimize interference between the tactile and visual modality regarding perception of object motion. Such suppression may have been especially important in our experimental paradigm, where the participants performed a tactile task that required maximum focus on detecting tactile events of amplitudes hovering around the perception threshold.

Moreover, the finding that frequency effects were present only during the tactile task and not during the visual task emphasizes the task dependence of the neural processing of sensory information (Hyvarinen et al., 1980; Drevets et al., 1995; Carlsson et al., 2000; Johansen-Berg and Lloyd, 2000; Staines et al., 2002; Nelson et al., 2004; Porro et al., 2004; Ghazanfar and Schroeder, 2006). Furthermore, the widespread effect of task on brain activity (see Figure 5B in Paper I) suggests that the responsiveness of traditionally somatosensory areas to tactile stimulation is determined by interactions between the tactile input and complex task-related cognitive processes.

That we did not see any effects of stimulation frequency in S1 does not exclude that tactile afferent information from different types of cutaneous mechanoreceptors (SA-1s, FA-1s, FA-2s) can be processed in different cortical modules of S1 as indicated in animal studies (Sur et al., 1984; Friedman et al., 2004). Because these modules consist of small, intermingled, clusters called cortical columns or modules ( 500 μm in diameter) within S1, even with the highest today available spatial resolution provided by ultra-high field human MRI (≥7 Tesla) one would not be able∼ to detect the small scale of this organization (Huber et al., 2017; Kashyap et al., 2018).

Paper II The main findings in Paper II were that median nerve-injured show gray matter decreases in two cortical areas involved in higher-lever processing of hand function (PMv, PMd) with two related white matter changes (in corpus callosum and right radiation of corpus callosum), as well as gray matter increase bilaterally in an area involved in visual spatial processing of information supporting hand actions. The left side injured participants also showed reduction in gray matter in the contralesional primary motor cortex. That the detected effects were synaptically remote from the lesioned neurons and involved reductions as well as increases of cortical gray matter volume suggests that they represent complex activity-dependent, injury driven, adaptations in specific processing units in the brain (Jones, 2000; Jaquet et al., 2001; Navarro et al., 2007; Fields, 2015; Zatorre et al., 2012).

The lack of detectable effects in contralesional S1 and limited effects in contralesional M1 stand in contrast to the considerable gray matter effects in these areas reported in persons with sensorimotor deprivation caused by amputation or immobilization of an upper extremity (Langer et al., 2012; Makin et al., 2013a; Preissler et al., 2013). A possible explanation for the overall spared primary sensorimotor areas in the nerve-injured participants in the present study is that they regularly used their affected hand in everyday uni- and bimanual activities. With the increased activity during tactile tasks in the contralesional S1 as shown in Paper III, perhaps even an increased gray matter volume in S1 could have been reasonable. Another explanation, is the possibility that reinnervation may prevent transneuronal atrophic effects propagating through the ascending somatosensory pathways to S1 (Jones, 2000), given that as many as 70–80% of the DRG neurons survive section of a peripheral nerve (Risling et al., 1983; Devor et al., 1985; Arvidsson et al., 1986; Himes and Tessler, 1989; Lekan et al., 1997). Other possible explanations of no evident decrease in gray matter volume in S1 may be that there is a compensatory increase after an initial decrease during the first period after injury, or that immediate mechanisms of plasticity related to “overtaking” by adjacent somatosensory areas prevents an initial decrease.

That the reduction in gray matter was present in the left PMv, i.e., the opercular part of Broca’s area, irrespective of affected hand is consistent with the left hemisphere´s dominance in manual praxis in right-handed humans (Haaland et al., 2000; Grezes and Decety, 2001; Johnson-Frey et al., 2005; Lewis, 2006). Moreover, it is consistent with the idea that this cortical area constitutes a repository for a "vocabulary" of manual actions containing “words” representing motor acts of different degrees of abstraction, complexity and hierarchical relationships (Rizzolatti et al., 1988; Jeannerod et al., 1995; Binkofski and Buccino, 2006; Fadiga et al., 2009; Rizzolatti et al., 2014). As such, the loss of gray matter in the left PMv would depend on reduced use of neuronal populations that

represent certain action phases (i.e., “words” in the analogy of speech) that are essential parts of the repertoire of actions of daily life. The failure of nerve-injured individuals to express certain action phases would then originate from a lack of reliable afferent signals that would, in turn, prevent successful execution of one or several of the sequentially linked action phases that are necessary to achieve reliably goals of motor acts (Johansson and Flanagan, 2009).

The gray matter increase observed bilaterally in the posterior middle temporal gyrus matches the increased dependence on vision in activities requiring skillful object handling in nerve-injured persons with impaired tactile innervation of a hand (Brink and Mackel, 1987; Jerosch-Herold, 1993; Jenmalm and Johansson, 1997; Jenmalm et al., 2000; Chemnitz et al., 2013a). The localization of the effect to the V5/hMT+ complex agrees with previous observations that this area constitutes a node in the dorsal visual stream that supports visually guided actions (Goodale and Milner, 1992; Grefkes et al., 2004; Schenk et al., 2005; Ilg and Schumann, 2007; Whitney et al., 2007; Kravitz et al., 2013). Likewise, manual visuomotor skill acquisition is associated with increased gray matter density especially in this extrastriate visual region (Draganski et al., 2004; Boyke et al., 2008). Moreover, the V5/hMT+ complex is considered as the earliest site in the parieto–premotor (dorsomedial) pathway where visual motion signals and tactile information are integrated (Hagen et al., 2002; Blake et al., 2004; Beauchamp et al., 2007; Ricciardi et al., 2007; van Kemenade et al., 2014). This suggests that the increase in gray matter in this area reflects an intensified crossmodal processing for functional compensation of the somatosensory deficiencies of the affected hand.

Paper III The main finding in Paper III was that the hand representation of the contralesional (right) primary somatosensory cortex (S1) of the nerve-injured showed greater brain activity compared to the controls not only when the left reinnervated index finger was engaged in the tactile tasks, but also when the left little finger and the fingers of the ipsilateral (right) hand innervated by uninjured nerves were engaged. We also found that the nerve-injured participants showed elevated brain activity in prefrontal cortical areas during execution of the tactile tasks. We interpret this as representing that top-down processes supporting decision-making and response selection are computationally more demanding for the injured participants due to their compromised tactile sensibility compared to the control participants. The trend towards increased brain activity within these prefrontal areas also during stimulation of non-affected digits may reflect the fact that these higher-level processing areas are more generally involved in hand function, i.e., these areas increase their activity independently of whether the task involves affected or non-affected skin areas.

Even after state-of-the-art surgical repair and reinnervation of injured hand nerves, functional reorganizational changes are thought to occur in hand representations of the human S1 (Lundborg and Rosen, 2007). Yet, it is not obvious that such changes per se explain the increased brain activity in the nerve- injured participants of the present study. Moreover, the similarity of the localization of the identified index finger representation of the nerve-injured and the controls indicates that such reorganizational changes were limited in our study material. As such, our findings are consistent with those found in non- human primates in which that the median nerve after transection and surgical repair largely resumes its original projection area in S1, but with some distortion of its internal topography (Wall et al., 1986; Merzenich and Jenkins, 1993). Furthermore, there is growing evidence that cortical representation of a limb remains remarkably stable despite loss of its major peripheral input (Kikkert et al., 2016; Makin and Bensmaia, 2017).

The increased brain activity in the contralesional S1 during tactile tasks after reinnervation most likely reflected a reduction of stimulation-induced inhibition that changes the overall balance of excitation-inhibition towards excitation. Since cortical inhibition is generated by neurons releasing GABA, it seems reasonable to assume that the elevated brain activity in the contralateral S1 is caused by a reduction in intra-areal stimulation-induced GABAergic inhibition (the absence of a detectable effect during the cue period suggests that this inhibition required the presence of tactile stimuli rather than just focused attention on sensing them). This, in turn, suggests that processes promoting increased plasticity were still ongoing in the nerve-injured participants. The main reason for this interpretation is that a reduction in stimulation-induced GABAergic inhibition seems crucial for inducing (and maintaining) excitatory cell reorganizations in the deprived projection zone of a lesioned peripheral nerve in adulthood (Garraghty et al., 1991; Arckens et al., 2000; Levy et al., 2002; Garraghty et al., 2006; Marik et al., 2010; Chen et al., 2011). Whether this interpretation of increased activity in the right S1 really depend on GABAergic disinhibition could be addressed in future studies by using Magnetic Resonance Spectroscopy which can measure local cortical GABA concentrations (Stagg et al., 2011; Puts and Edden, 2012).

A sign that the reorganization of newly formed circuits are stabilizing is that GABAergic inhibition (and cortical excitability) gradually approach baseline levels (Sammons and Keck, 2015) and thus re-establishes a more normal inhibition/excitation-balance, to a more inhibited state. Accordingly, although the injury occurred several years ago (on average 8.1 years) and the attainable peripheral reinnervation of the hand were completed, the overall increase in BOLD-signal in the contralesional hand S1 representation in the nerve-injured participants may reflect an ongoing reorganizational plasticity. Such ongoing reorganizational plasticity may relate to limitations in the Hebbian learning rules

considered to account for the formation of cortical topographic maps and cell assemblies in S1 that represent natural stimuli (Buonomano and Merzenich, 1998). Due to the spatiotemporal abnormal noisy activity in ensembles of 1st- order tactile neurons after reinnervation, these rules may partially fail to achieve a complete stabilization of S1 since they are based on the detection of spatiotemporally correlated inputs with precise time criteria and from a limited cutaneous area (Pons et al., 1991; Song and Abbott, 2001; Feldman, 2012; Chang and Merzenich, 2003; Zhou et al., 2011).

During the visual task, the BOLD-signals were suppressed to the same levels in the injured and the control participants. This suggests another type of inhibitory mechanism than that behind the difference between the groups in the tactile task. Hence, the modulation dependent on task modality was likely caused by “top- down” attentional-related global modulation of somatosensory processing pathways driven by higher-level processing distributed within frontal and parietal cortex, cerebellum and thalamus (Gilbert and Sigman, 2007; Dosenbach et al., 2008).

The three prefrontal areas showing higher activity for the nerve-injured participants during the tactile tasks all show complex interactions in cognitive control of goal-oriented behavior. One possible interpretation of these areas increased activity is, according to previous human fMRI studies, that they represent components of two separate networks supporting different types of top-down control (Dosenbach et al., 2008). The dorsal ACC and VLPFC together with anterior insula and thalamus have been implicated in processes supporting stable maintenance of task sets in goal directed behaviors. In contrast, DLPFC together with the inferior parietal lobule, intraparietal sulcus, precuneus and middle cingulate cortex are considered engaged on a trial-by-trial basis in error-related activity and initiation and adaptations of control. These complex prefrontal effects are to interpret, however, we consider that they reflect that tactile tasks performed with the hands after median nerve injury are more cognitively demanding, causing additional neuronal processing compared to healthy controls. Perhaps are these effects related to behavioral changes that must be dealt with regarding choices in the ongoing or forthcoming behavior. Or in the case of performing tactile tasks, conflicts about staying in the current task-set and continuing the performance of the task although it involves their impaired hand.

The tendency to elevated brain activity regardless of finger used in the tactile tasks likely reflects that the control of everyday dexterous actions practically always engage multiple digits and also engage both hands more often than infrequently (Kilbreath and Heard, 2005; Ingram and Wolpert, 2011). Hence, the compromised tactile inputs from one hand's median nerve innervation area could affect the decoding and higher-level processing of tactile information normally based on spatiotemporally organized inputs from digits of both hands. Moreover,

the prefrontal effects, especially that in the left VLPFC, might also reflect processes involved in coping with pain, numbness, stiffness, weakness and swelling related to cold intolerance often experienced during use of the affected hand (Irwin et al., 1997; Chemnitz et al., 2013b; Kringelbach and Rolls, 2004).

Synthesis Changes in central nervous sensorimotor processing after median nerve injury, repair and reinnervation, are amply exemplified in this thesis work: structural and functional changes not only occur in primary somatosensory and motor areas of the hands, but also in premotor, prefrontal and extrastriatal visual brain areas (Fig. 8).

Figure 8 – Schematic illustration of brain areas affected by median nerve injury after reinnervation.

Findings of brain areas affected by median nerve injury detected in Paper II and III schematically illustrated on a lateral view of the right hemisphere for visualization. Note that some of the illustrated changes were bilateral and some occurred in only one of the hemispheres regardless of side of injury. Green circle denotes increase in gray matter, red circles denote decrease in gray matter and blue circles denote increased brain activity during tactile tasks.

The implementation of most daily hand actions involves several brain areas, including those areas which in this thesis work have been shown affected by median nerve injury. The areas identified are implicated in seeing and localizing an item to be manipulated, deciding what goal-oriented behavior to (or not to) perform, generate motor commands to move the hands, adapt the contact conditions between the hand and seized objects to their mechanical properties for successful object manipulation outcomes etc. In planning and control of goal-

oriented manual behaviors one can consider multiple sensorimotor control loops (Fig. 2) that may be affected after nerve injury due to the impaired tactile sensation. For the brain, a spatiotemporally distorted afferent input from the hand after nerve injury leads to unreliable interpretations of the mechanical events occurring in the affected skin areas. This may cause insecurities and “clumsiness” in nerve-injured when performing some manual actions, which may lead them to search for, and choose, alternative behavioral strategies or perhaps strictly avoiding certain manual tasks. Such behavioral changes likely require increased cortical processing resources for decision making and action selection, i.e., processing that primarily rely on operations of the prefrontal cortex. The seemingly most ‘serious’ structural changes—the gray matter shrinkages—are observed in areas associated with the repository of processes that represent various manual skills. This suggests that the gray matter reduction reflects a decrease in neuronal activity associated with a lasting limitation of the manual action repertoire caused by the nerve injury. Conversely, the bilateral increase of gray matter volume in the extrastriatal cortex likely reflects effects of use- dependent plasticity related to compensatory visuomotor processing linked to the impaired tactile sensibility.

However, the indications on increased ongoing reorganization plasticity in contralesional S1 in combination with improvements in neuroprotection that may reduce the loss of sensory DRG neurons and prolonged rehabilitation schemes, create hope for improved functional return after median nerve injury in the future. Ideas for further studies that might contribute to such a development are presented in the section below entitled ‘Future directions’.

Reliability In this section, I intend to bring up a few points that argue for the reliability of the results achieved. First, for the fMRI analyzes presented in Paper I and III, we consider that the study paradigms used successfully promoted adequate comparisons of brain activity (BOLD-signals) between different stimulus conditions within the participants, but also for comparisons performed at group level. The behavioral/psychophysical results provide support that the participants made use of the tactile inputs that they were exposed to in the tactile tasks and that they performed the tasks in agreement with the instructions given. The measured tactile thresholds along with the observed responses to oddball stimulus were consistent with this. Likewise, the algorithm used for balancing the amount of button presses in each threshold-tracking trials worked as expected, i.e., it steered the stimuli such that similar number of responses were made in all trials by all participants. Furthermore, the intensity of the tactile stimulations in the tactile tasks were perceptually equalized across all comparable trials by all participants.

The scientifically most interesting effects on brain activity presented in these papers (Paper I: effects of stimulus frequency, and Paper III: effect of injury and of finger) can therefore not be easily explained by differences in perceptual intensities of the stimuli, differences in behavior or task-set. Notably, studies investigating stimulation-evoked brain activity that lack control of such potentially confounding factors may show differences in brain activity that depend on differences not controlled for in the experimental conditions. Also, despite partly different stimulus parameters in Paper I and III, we observed a striking overlap of effects in conjunction analyses looking for increased brain activity common for the tactile trial irrespective of stimulation condition (stimulated hand/finger, tactile task) and group of study participants (younger healthy adults in Paper I, partly older healthy adults and nerve-injured adults in Paper III). The brain areas commonly engaged in our tactile tasks included not only the traditional somatosensory areas but also cerebellum, posterior parietal, premotor and prefrontal areas, which are areas shown to be active also in other types of demanding tactile or sensorimotor tasks (Burton et al., 2008; Kroliczak et al., 2008; Gallivan and Culham, 2015). Another factor that I consider supports the reliability considering the interpretation of the present findings is that the included nerve-injured participants were quite homogenous with regard to type of nerve injury, surgery and state of reinnervation. Although the nerve-injured participants showed differences in concomitant injuries (e.g., injuries to tendons and arteries, see: Nordmark et al., 2018), they showed no apparent lasting disabilities related to this, such as reduced range of motion. Nor did the amount of concomitant injuries correlate with their self-reported disabilities in everyday actions or cold intolerance, indicating that the nerve injury was itself the most important factor.

However, apart from potential statistical issues commented on further below, there are limitations regarding the generalizability of the results in the present studies, especially those involving nerve-injured participants. One limitation of the studies involving nerve-injured is that no drop-out analysis was performed. Therefore, we cannot say whether there were differences in the studied group compared to the nerve-injured persons that did not want to participate in the study or could not endure the MRI session. It is, for instance, possible that the included participants were those with the best clinical outcomes. Since we did not obtain consent from those that rejected participation, it would, in our case, have been unethical to perform drop-out analyses that might have revealed differences in outcome potentially available in clinical data. Another limitation concerns gender aspects. Injuries to peripheral nerves are more common in men than women (Asplund et al., 2009), and therefore perhaps not surprisingly, more nerve-injured men than women were available for participation for the studies presented in Paper II and III. To the best of my knowledge, sex differences in anatomy of the peripheral nerves, treatment of nerve injuries or injury type and

outcome of injuries have never been reported, and this thesis did not aim to investigate potential differences between the sexes. However, to avoid potential effects on group level we factored out potential effects of gender in Paper II and III. Notably, SPM analyses of the factor gender did not produce any statistically significant effects, but the statistical power of this analysis was low given that only four in each of the groups of the injured and controls were women.

Statistical choices and issues The choices of statistical thresholds in neuroimaging techniques using fMRI is a topic that is often debated and should be debated. Due to the immense number of statistical tests that can be performed given some 200,000 voxels sampled multiple times for whole-brain analyses, the risk of false positive voxels with wrongly performed statistical tests may be high. Today, especially given signs of low reproducibility of fMRI studies (Open Science, 2015) and recent statistical concerns questioning the validity of many scientific articles based on fMRI-data (Eklund et al., 2016), it is important to consider the validity of statistical analyses when performing brain imaging studies. In the studies presented in this thesis we have applied a double-step statistical threshold with the first threshold at voxel- level and the second at cluster-size level, to reduce the risk of false positive clusters. However, the double-step thresholding approach used in this thesis was likely to classify smaller clusters as non-significant since it implied a quite conservative cut-off cluster-size for significant clusters in its second step. For the SVC analyses in Paper III, given the strong a priori assumption of effects of factor finger and potentially group, we decided to use a cluster extent threshold of 20 continuous voxels combined with the voxel threshold of p < 0.005 which can be considered as rather liberal.

In this respect, I derive satisfaction from seeing the reproducibility in the conjunction analyses performed in Paper I and III mentioned above. That is, despite partly different stimulus parameters in Paper I and III, we observed a striking overlap of effects in conjunction analyses looking for increased brain activity common for the tactile trial irrespective of stimulation condition (stimulated hand/finger, tactile task) and group of study participants (younger healthy adults in Paper I and relatively older healthy and nerve-injured adults in Paper III). The brain areas commonly engaged in our tactile tasks included not only the traditional somatosensory areas but also cerebellum, posterior parietal, premotor and prefrontal areas, which are areas shown to be active also in other types of demanding tactile or sensorimotor tasks (Burton et al., 2008; Kroliczak et al., 2008; Gallivan and Culham, 2015). I also especially derive satisfaction from seeing the similarity of the main effects of finger in the right S1 between the separate analyses of data from the tactile threshold-tracking and oddball- detection tasks, and between separate analyses of data from each group of

participants (see Figures 3b and 3c in Paper III). Finally, another important point in this respect is that the significant effects presented in the fMRI studies in this thesis – with the exception of the conjunction tests - are derived from comparisons of brain activity when the participants were engaged in tasks, as opposed to comparing activity between data recorded during trials with that sampled during intertrial periods.

Thus, given the statistical criteria applied and the overall sensible outcome of the analyses with reference to established knowledge about brain areas implicated in hand functions, I consider all effects of the addressed experimental factors presented in this thesis as robust. However, with more participants additional areas with significant effects might have been detected, but with regard to peripheral nerve injury, as for studies of other clinical conditions, there is a potential trade-off between including many participants and striving for homogeneous groups. Perhaps, the most obvious analysis where more nerve- injured participants would have been desired was for the right-side nerve-injured (N = 5) for the studies presented in Papers II and III.

Although previous studies have strived to increase the statistical power by pooling data from left- and right-side nerve-injured participants after “flipping” of hemispheres (left-right) for achieving symmetry regarding injury side (Makin et al., 2013a; Chemnitz et al., 2015), we chose not do so because of known asymmetries between the hemispheres. These include structural asymmetries (Watkins et al., 2001; Luders et al., 2006) as well as functional asymmetries that certainly applies to brain areas implicated in higher-level action planning and manual praxis (Haaland and Harrington, 1996; Schluter et al., 2001; Johnson-Frey et al., 2005; Lewis, 2006). It even applies to primary sensorimotor cortex (Kim et al., 1993; Ziemann and Hallett, 2001; Driver et al., 2015). Indeed, that many of the significant effects detected in the thesis studies were lateralized to either the left or the right hemisphere regardless of side of injury (Paper II) or hand stimulated (Paper I and III) confirmed the potential interpretational danger of applying such procedure of “flipping” of hemispheres.

Future directions From the work leading to this thesis and the findings presented herein, I have identified three “research areas” that I consider interesting and important, as well as feasible to pursue: (i) further brain-imaging studies in persons with peripheral nerve injury, (ii) studies of injury effects on the peripheral nerve, (iii) studies that investigate composite hand function.

The most obvious area to proceed with is perhaps the first of those indicated above, i.e., further fMRI studies in patients with peripheral nerve injury. This may, for example, involve taking the stimulus paradigms used in Paper I and III further or by applying other perceptual paradigms during sampling of fMRI-data in nerve-injured persons. A feasible idea within reach would be to dive deeper into the complex interactions in the hand representation of S1 both under healthy conditions and after peripheral nerve injury. For this, similar paradigms as in Papers I and III could be used to address stimuli to the thumb/index-/long-/ring- /little-finger of each hand. In addition, stimulation to more than one finger at a time would allow for investigation of potentially more complex interactions, including lateral inhibitory mechanisms. With a field-of-view focused on S1 and with the possibility of using fMRI sampling at higher magnetic field (7 Tesla), the sampled data would offer higher spatial resolution than those of the present thesis. Other paradigms focusing on sensorimotor control could investigate brain activity during uni- and/or bimanual object manipulation tasks to search for changes in brain activity that could be related to the structural changes seen in Paper II, i.e., exploring eye-hand coordination, bimanual coordination or the narrowing in manual repertoire.

For improved understanding of structural changes after peripheral nerve injury, a longitudinal study approach would have been preferred. However, given the time for a PhD project, the cross-sectional approach involving comparisons of nerve-injured and matched controls was realistic. It would have taken about 19 years to complete a longitudinal study with the nerve-injured patient material available for this study. That is, the time from the first MR scan just after the earliest incoming patient was operated until the last received patient's clinical condition would have been considered stable. Longitudinal studies with adequate power performed within a more reasonable time frame should, however, be feasible as multicenter enterprises given that MRI scanners are commonly available even at regional hospitals. Of special interest would be to combine longitudinally followed structural changes with parallel observations of functional development over time, and how various rehabilitation regimes or treatments could affect these results.

(ii) The second research area proposed; further studies of injury effects on the peripheral nerve, would focus on how the nerve’s signaling and stimulus encoding

properties are affected after reinnervation. A better understanding of changes in the peripheral nerve in this respect would be of great importance for the understanding the pathophysiology behind the lasting hand impairment. Such knowledge could help better understand the challenges for the CNS to interpret and adapt to the input coming from the affected nerve.

Questions such as to which degree LTMRs of the various functional types are represented in reinnervated peripheral nerves could, for example, be investigated using psychophysical tests combined with fMRI in paradigms similar to the tactile threshold-tracking paradigm used in Paper I. By performing the threshold- tracking task with frequencies adjusted to primarily evoke responses in a specific type of tactile afferent neurons (3Hz: SA-1, 20Hz: FA-1 and 100 Hz: FA-2), and comparing the detection thresholds to those at normally innervated glabrous skin areas, the degree of functional reinnervation could be estimated. As mentioned in the methods section, a pilot study that I did suggest that patients with unilateral median and ulnar nerve injuries have “more normal” tactile thresholds for FA-1 neurons after reinnervation as compared to SA-1s and FA-2s. Furthermore, reinnervated afferents could be investigated using microneurography addressing individual afferents that have reinnervated the skin of the hand. Such investigation could answer whether or to which degree reinnervated afferents reclaim similar receptive field properties (Pruszynski and Johansson, 2014; Johansson, 1978) as before the injury. Microneurography was used by Mackel and colleagues (Mackel et al., 1983a; Mackel et al., 1983b; Brink and Mackel, 1987) to investigate reinnervated humans in the 1980s, and they already then suggested (Mackel et al., 1983b) to use better receptive field mapping techniques similar to those used by (Johansson, 1978) to answer such questions. However, to the best of my knowledge, no one has pursued such investigations. Furthermore, since the receptive fields depend on the terminal branching of the axons in the skin, histological techniques investigating biopsies from the affected skin areas of the hand, such as that used by Nolano and colleagues (Nolano et al., 2003), could help to shed light on this matter.

(iii) The third identified research area is perhaps a bit paradoxical to come out from the studying of structural and functional changes in the brains of persons that suffered median nerve injury: the need for more knowledge about behavioral changes after nerve injury. Although important and vast efforts have been made to characterize hand function after peripheral nerve injuries (see for instance: Moberg, 1958; Sollerman and Ejeskar, 1995; Rosen and Lundborg, 2001; Jerosch-Herold, 2005; Geere et al., 2007; Lundborg and Rosen, 2007; Chemnitz et al., 2013a; Ashwood et al., 2017; Krarup et al., 2017), there are still, what I consider, critical questions remaining. In my mind, a burning question revolves around statistics on how nerve-injured persons use their hands in everyday life, including more complex aspects of the dexterous hand function. Measurable

behavioral changes after injury would be interesting to correlate with structural and functional changes identified in the brain, but most importantly the behavior represented by the actual natural use of the hands is the outcome that the clinical treatment aim to reach.

Data that could be of importance for evaluation and development of treatments are those that also identify compensatory mechanisms that the patients use in everyday manual actions, in addition to objective measures of their “clumsiness” and restrictions in manual action repertoire. For this, based on the structural changes seen in Paper II and on notions from clinical experience, I propose studies on both eye-hand-coordination and bimanual coordination, which may answer and describe how nerve-injured individuals compensate their somatosensory deficit. In addition, fundamental knowledge about how much they actually use their hands is important considering that the amount of use likely correlate with the functional recovery. Although the measurement of daily use of the hands can be difficult to achieve, given the fast development of many smart electronical devices that can be worn without hassle, such data collection should soon, if not already, be within reach. Daily-use measurements of hand use could help in the timing of introduction of clinical rehabilitation interventions and also help determine when to increase the complexity of manual training exercises. Another application of data sampled that way could be to provide the patient with daily feedback to encourage “normality” in the use of the hands.

In summary, for those who are curious and want to develop the area scientifically, the possibilities for future investigations in the work of improving hand function after peripheral nerve injury seem rich.

Conclusions

I am just a child who has never grown up. I still keep asking these 'how' and 'why' questions. Occasionally, I find an answer. Stephen Hawking (1942-2018)

Based on the findings presented in this thesis, it is concluded that:

In healthy people, the temporal frequency content of tactile stimuli influences brain activity during tactile detection tasks (Paper I). . fMRI-analysis of brain activity showed that during tactile detection of 3, 20, and 100 Hz stimulations, the left auditory cortex shows increased activity, during stimulation with hearable frequencies (20 and 100 Hz), and an area within visual cortex exhibits decreased activity at infrasound frequencies (3 Hz). Thus, the frequency content of the tactile input influences the processing in two traditionally unimodal brain areas, neither of which are primarily associated with processing tactile information. This finding advances the metamodal theory of brain function.

Median nerve injury can cause long-term structural changes in the brain (Paper II). . VBM-analysis showed that the left ventral and right dorsal premotor cortex show gray matter reductions irrespective of side of median nerve injury, and reductions of white matter are seen in related commissural pathways interconnecting the premotor areas of the two hemispheres. Left-side injured show gray matter reduction in the contralesional primary motor cortex. All these structural brain changes likely represent manifestations of reduced neural processing linked to restrictions in the diversity of the natural manual dexterity repertoire caused by the peripheral nerve injury. . An area involved in visual movement processing show bilateral increase in gray matter, which likely reflects compensatory visuomotor processing linked to the impaired tactile sensibility.

Median nerve injury can cause long-term functional changes in the brain (Paper III). . fMRI-analysis of brain activity showed that the hand representation of the contralesional primary somatosensory cortex (S1) shows increased activity compared to controls when nerve-injured use a reinnervated finger in tactile tasks, but also when using fingers innervated by uninjured nerves of either hand. This indicates that the injury causes a general disinhibition of the contralesional S1, suggesting that increased reorganization plasticity is a continuous feature of chronic recovery from peripheral nerve injury. . Nerve-injured also show increased activity in prefrontal cortical areas, suggesting that top-down processes supporting decision-making and response selection are computationally more demanding for the injured participants due to their compromised tactile sensibility compared to the control participants.

Acknowledgements

I want to express my appreciation to all those that have contributed to my graduate studies by catalyzing and helping in the scientific work, or by easing the everyday life as a PhD-student. First, I would like to acknowledge the studied patients. Taking time away from work, families and everyday activities for participation, travelling far, unselfishly for the potential future gain of other nerve injured persons. I would also like to express my deepest gratitude to my employer, ‘Region Västerbotten’, for providing the financial means for me to be a part-time graduate student in combination with clinical work for more than nine years. Especially, I would like to acknowledge Roland Johansson, for excellent supervision, for leading by example, for always having an open door and mind for discussions about research in specific and everything in general, for pushing me those extra miles beyond what I imagined a PhD could ever look like. You are the most curious person I have ever met. Göran Westling, for your excellence in materializing thoughts into gadgets, devices and thingies such as the apparatus used in Paper I and III, but even more for your everyday encouragements that over these years have meant more than you can imagine. Anders Bäckström, for your brilliance in programming and your tremendous skill in always having a sharpest mind in trouble-shooting and problem-solving. Thank you for all of your patience in every last week-/day-/hour-/minute-changes in the protocols used for data- collection for Paper I and III. Micael Andersson, for your Matlab-wiz skills that have saved me so much time over the years, and for your patience in letting me take a big part of that saved time for discussions and endless questions that you have answered. Also, thanks for not letting me miss out on the lunches! Andrew Pruszynski, for a fruitful collaboration in Paper I and for being a young and humble scientific role model. Christina Ljungberg, for all the encouragement and valuable discussions. Mikael Wiberg, for capturing me in my last lecture as a student-tutor in anatomy more than ten years ago, proposing that I should go into the research that led me to this thesis and into hand surgery, thank you. Ben Edin & KG Westberg, my first teachers in physiology, for valuable discussions regarding physiology and for input on my scientific work over the years. Lars Nyberg & Johan Eriksson for your always valued comments and discussions over my years as a PhD-student, and in leading the weekly/monthly journal clubs. To previous and current PhD- students/post-docs/researchers/staff of IMB & UFBI, for everything from work related feedback to chit-chats over a coffee: Anita, Anna, Anna-Lena, Alireza, Anders, Anders, Andreas, Amar, Bror, Carola, CJ, Clémence, Daniel, Erland, Ewa, Fredrik, Filip, Greger, Gregoria, Herbert, Ingalill, Jarkko, Janove, Kamil, Karolina, Kristina, Kurt, Lenita, Lars-Erik, Lars, Lars, Linnea, Michael, Michael, Mikael, Michael, Nina, PA, Per, Peter, Roger, Staffan, Sara, Sara, Toshiki, Urban. For those outside the direct research, but still perhaps having to hear my frequent updates, for listening, and for at times leading the discussion on to something else. To all my colleagues and the staff at my other work at the Hand and Plastic Surgery Clinic. To Anders, friend, colleague, room-/team-/ beer-brewer-mate, hunting-leader, for always being able to fill ‘spare-time’ with exiting side-projects. Pamela, Johan, Erika, Ida, Pär, Martin, Mari-Therese, Daniel, Marjan for making weekday- and weekend-life good. Mattias, Thomas, Björn, Jonas, friends since toddlers, for keeping me firmly on the ground since before we could walk, and for not letting me be ‘the BOSS’ all the time: I believe our friendship defines who we are. Sebastian, Amar, Gustav, for talks about high and low, for all the entertainment and ideas, and for putting a smile on the workday. Lina, Tove, Kajsa, Gunilla, Magnus, Karl for making med school pass in a blink, yet feeling like it happened yesterday. Johan & Elof, for biking in line through NZ; loads of memories for a rainy day! To my parents, Christina & Roger, for your unconditional love and for your endless (endless!) support, I couldn’t have asked for more. To Claes, for paving the way for your little brother by, early on and thereafter, showing that limits can be stretched by will, work and ambition. Lastly, to my family, Ingela, Elsa & Agnes, for accepting my absence in participation, and for welcoming me with open arms on my return, for your support, for your comfort, for your love. There are not words enough. You are my true inspiration.

References

Abraira VE, Ginty DD (2013) The sensory neurons of touch. Neuron 79:618-639. Adkins RJ, Morse RW, Towe AL (1966) Control of somatosensory input by cerebral cortex. Science 153:1020-1022. Aegina Po (1914) Paulos von Aegina - Des Besten Arztes Sieben Büscher (German translation): Vormals E.J. Brill. Alamia A, Solopchuk O, D'Ausilio A, Van Bever V, Fadiga L, Olivier E, Zenon A (2016) Disruption of Broca's Area Alters Higher-order Chunking Processing during Perceptual Sequence Learning. J Cogn Neurosci 28:402-417. Allodi I, Udina E, Navarro X (2012) Specificity of peripheral nerve regeneration: interactions at the axon level. Prog Neurobiol 98:16-37. Amedi A, Malach R, Hendler T, Peled S, Zohary E (2001) Visuo-haptic object-related activation in the ventral visual pathway. Nat Neurosci 4:324-330. Arckens L, Schweigart G, Qu Y, Wouters G, Pow DV, Vandesande F, Eysel UT, Orban GA (2000) Cooperative changes in GABA, glutamate and activity levels: the missing link in cortical plasticity. Eur J Neurosci 12:4222-4232. Arthurs OJ, Boniface S (2002) How well do we understand the neural origins of the fMRI BOLD signal? Trends Neurosci 25:27-31. Arvidsson J, Ygge J, Grant G (1986) Cell loss in lumbar dorsal root ganglia and transganglionic degeneration after sciatic nerve resection in the rat. Brain Res 373:15-21. Ashburner J, Friston KJ (2000) Voxel-based morphometry--the methods. Neuroimage 11:805-821. Ashburner J, Friston KJ (2001) Why voxel-based morphometry should be used. Neuroimage 14:1238-1243. Ashburner J, Csernansky JG, Davatzikos C, Fox NC, Frisoni GB, Thompson PM (2003) Computer-assisted imaging to assess brain structure in healthy and diseased brains. Lancet Neurol 2:79-88. Ashwood M, Jerosch-Herold C, Shepstone L (2017) Learning to live with a hand nerve disorder: A constructed grounded theory. J Hand Ther. Asplund M, Nilsson M, Jacobsson A, von Holst H (2009) Incidence of traumatic peripheral nerve injuries and amputations in Sweden between 1998 and 2006. Neuroepidemiology 32:217-228. Babcock W (1927) A standard technique for operations on peripheral nerves: With especial reference to the closure of large gaps. Surgery, Gynecology & Obstetrics 46:364-378. Backes WH, Mess WH, van Kranen-Mastenbroek V, Reulen JP (2000) Somatosensory cortex responses to median nerve stimulation: fMRI effects of current amplitude and selective attention. Clin Neurophysiol 111:1738-1744. Balint R (1909) Seelenlähmung des "Schauens" optische Ataxia, räumliche Störung des Aufmerksamkeit. Monatsschrift für Psychiatrie und Neurologie 25:51-81. Bates JF, Goldman-Rakic PS (1993) Prefrontal connections of medial motor areas in the rhesus monkey. J Comp Neurol 336:211-228. Beauchamp MS (2005) See me, hear me, touch me: multisensory integration in lateral occipital-temporal cortex. Curr Opin Neurobiol 15:145-153. Beauchamp MS, Yasar NE, Kishan N, Ro T (2007) Human MST but not MT responds to tactile stimulation. J Neurosci 27:8261-8267. Behrens TE, Woolrich MW, Walton ME, Rushworth MF (2007) Learning the value of information in an uncertain world. Nat Neurosci 10:1214-1221. Bengtsson F, Brasselet R, Johansson RS, Arleo A, Jorntell H (2013) Integration of sensory quanta in cuneate nucleus neurons in vivo. PLoS One 8:e56630.

Bengtsson SL, Nagy Z, Skare S, Forsman L, Forssberg H, Ullen F (2005) Extensive piano practicing has regionally specific effects on white matter development. Nat Neurosci 8:1148-1150. Beurze SM, de Lange FP, Toni I, Medendorp WP (2007) Integration of target and effector information in the human brain during reach planning. J Neurophysiol 97:188-199. Binkofski F, Buccino G (2006) The role of ventral premotor cortex in action execution and action understanding. J Physiol Paris 99:396-405. Binkofski F, Buccino G, Posse S, Seitz RJ, Rizzolatti G, Freund H (1999) A fronto-parietal circuit for object manipulation in man: evidence from an fMRI-study. Eur J Neurosci 11:3276-3286. Bjorkman A, Weibull A (2017) Loss of inhibition in ipsilateral somatosensory areas following altered afferent nerve signaling from the hand. Neurosci Res. Blake R, Sobel KV, James TW (2004) Neural synergy between kinetic vision and touch. Psychological Science 15:397-402. Bleckmann H, Zelick R (2009) Lateral line system of fish. Integr Zool 4:13-25. Bloch F, Purcell EM (1952) The Nobel Prize in Physics 1952 - "for their development of new methods for nuclear magnetic precision measurements and discoveries in connection therewith.". In: NobelPrize.org. Nobel Media AB 2019. Boling W, Olivier A, Fabinyi G (2002) Historical contributions to the modern understanding of function in the central area. Neurosurgery 50:1296-1309, discussion 1309-1210. Bowlby AA (1890) Injuries and Diseases of Nerves. J Anat Physiol 24:472-474. Boyke J, Driemeyer J, Gaser C, Buchel C, May A (2008) Training-induced brain structure changes in the elderly. J Neurosci 28:7031-7035. Brink EE, Mackel R (1987) Sensorimotor performance of the hand during peripheral nerve regeneration. J Neurol Sci 77:249-266. Broca P (1861) Remarks on the Seat of the Faculty of Articulated Language, Following an Observation of Aphemia (Loss of Speech). Bulletin de la Société Anatomique 6:330-357. Brodmann K (1909) Vergleichende Lokalisationslehre der Grosshirnrinde. Leipzig: Verlag von Johann Ambrosius Barth. Brooks JC, Nurmikko TJ, Bimson WE, Singh KD, Roberts N (2002) fMRI of thermal pain: effects of stimulus laterality and attention. Neuroimage 15:293-301. Browne K (1951) A history of neurological surgery: Surgery of the peripheral nerves. Hafner, New York. Brozzoli C, Ehrsson HH, Farne A (2014) Multisensory representation of the space near the hand: from perception to action and interindividual interactions. Neuroscientist 20:122-135. Bruyns CN, Jaquet JB, Schreuders TA, Kalmijn S, Kuypers PD, Hovius SE (2003) Predictors for return to work in patients with median and ulnar nerve injuries. J Hand Surg Am 28:28-34. Buchthal F, Kuhl V (1979) Nerve conduction, tactile sensibility, and the electromyogram after suture or compression of peripheral nerve: a longitudinal study in man. J Neurol Neurosurg Psychiatry 42:436-451. Bunge SA, Wallis JD, Parker A, Brass M, Crone EA, Hoshi E, Sakai K (2005) Neural circuitry underlying rule use in humans and nonhuman primates. J Neurosci 25:10347-10350. Bunnell S (1927) Surgery of the nerves of the hand. Surgery, Gynecology and Obstetrics 44:145-152. Bunnell SB, J.H. (1939) Nerve grafts. The American Journal of Surgery 44:64-75. Buonomano DV, Merzenich MM (1998) Cortical plasticity: from synapses to maps. Annu Rev Neurosci 21:149-186.

Burton H, Sinclair RJ, McLaren DG (2008) Cortical network for vibrotactile attention: a fMRI study. Hum Brain Mapp 29:207-221. Caetano G, Jousmaki V (2006) Evidence of vibrotactile input to human auditory cortex. Neuroimage 29:15-28. Callier T, Suresh AK, Bensmaia SJ (2019) Neural Coding of Contact Events in Somatosensory Cortex. Cereb Cortex. Campbell AW (1905) Histological studies of the localisation of cerebral brain function. London: Cambridge University Press Warehouse. Canedo A (1997) Primary motor cortex influences on the descending and ascending systems. Prog Neurobiol 51:287-335. Carlsson K, Petrovic P, Skare S, Petersson KM, Ingvar M (2000) Tickling expectations: neural processing in anticipation of a sensory stimulus. J Cogn Neurosci 12:691- 703. Caruana F, Gerbella M, Avanzini P, Gozzo F, Pelliccia V, Mai R, Abdollahi RO, Cardinale F, Sartori I, Lo Russo G, Rizzolatti G (2018) Motor and emotional behaviours elicited by electrical stimulation of the human cingulate cortex. Brain 141:3035- 3051. Chan JS, Simoes-Franklin C, Garavan H, Newell FN (2010) Static images of novel, moveable objects learned through touch activate visual area hMT+. Neuroimage 49:1708-1716. Chang EF, Merzenich MM (2003) Environmental noise retards auditory cortical development. Science 300:498-502. Chao YP, Cho KH, Yeh CH, Chou KH, Chen JH, Lin CP (2009) Probabilistic topography of human corpus callosum using cytoarchitectural parcellation and high angular resolution diffusion imaging tractography. Hum Brain Mapp 30:3172-3187. Chemnitz A, Dahlin LB, Carlsson IK (2013a) Consequences and adaptation in daily life - patients' experiences three decades after a nerve injury sustained in adolescence. BMC Musculoskelet Disord 14:252. Chemnitz A, Bjorkman A, Dahlin LB, Rosen B (2013b) Functional outcome thirty years after median and ulnar nerve repair in childhood and adolescence. J Bone Joint Surg Am 95:329-337. Chemnitz A, Weibull A, Rosen B, Andersson G, Dahlin LB, Bjorkman A (2015) Normalized activation in the somatosensory cortex 30 years following nerve repair in children: an fMRI study. Eur J Neurosci 42:2022-2027. Chen JL, Lin WC, Cha JW, So PT, Kubota Y, Nedivi E (2011) Structural basis for the role of inhibition in facilitating adult brain plasticity. Nat Neurosci 14:587-594. Chen LM, Friedman RM, Roe AW (2003) Optical imaging of a tactile illusion in area 3b of the primary somatosensory cortex. Science 302:881-885. Chen R, Cohen LG, Hallett M (2002) Nervous system reorganization following injury. Neuroscience 111:761-773. Corkin S, Milner B, Rasmussen T (1970) Somatosensory thresholds--contrasting effects of postcentral-gyrus and posterior parietal-lobe excisions. Arch Neurol 23:41- 58. Creem-Regehr SH (2009) Sensory-motor and cognitive functions of the human posterior parietal cortex involved in manual actions. Neurobiol Learn Mem 91:166-171. Culham JC, Kanwisher NG (2001) Neuroimaging of cognitive functions in human parietal cortex. Curr Opin Neurobiol 11:157-163. Cushing H (1909) A note upon the Faradic stimulation of the postcentral gyrus in conscious patients. Brain 32:44-53. Dahlin LB, Wiberg M (2017) Nerve injuries of the upper extremity and hand. EFORT Open Rev 2:158-170. Davare M, Kraskov A, Rothwell JC, Lemon RN (2011) Interactions between areas of the cortical grasping network. Curr Opin Neurobiol 21:565-570.

Davare M, Andres M, Cosnard G, Thonnard JL, Olivier E (2006) Dissociating the role of ventral and dorsal premotor cortex in precision grasping. J Neurosci 26:2260- 2268. Davis KD, Kiss ZH, Tasker RR, Dostrovsky JO (1996) Thalamic stimulation-evoked sensations in chronic pain patients and in nonpain (movement disorder) patients. J Neurophysiol 75:1026-1037. de Haan EHF, Jackson SR, Schenk T (2018) Where are we now with 'What' and 'How'? Cortex 98:1-7. Delhaye BP, Long KH, Bensmaia SJ (2018) Neural Basis of Touch and Proprioception in Primate Cortex. Compr Physiol 8:1575-1602. Dellon AL (1981) Evalutation of sensibility and re-education of the hand Baltimore, MD Williams & Wilkins Devor M, Govrin-Lippmann R, Frank I, Raber P (1985) Proliferation of primary sensory neurons in adult rat dorsal root ganglion and the kinetics of retrograde cell loss after sciatic nerve section. Somatosens Res 3:139-167. Dijkerman HC, de Haan EH (2007) Somatosensory processes subserving perception and action. Behav Brain Sci 30:189-201; discussion 201-139. Disbrow E, Litinas E, Recanzone GH, Padberg J, Krubitzer L (2003) Cortical connections of the second somatosensory area and the parietal ventral area in macaque monkeys. J Comp Neurol 462:382-399. Donahue CJ, Glasser MF, Preuss TM, Rilling JK, Van Essen DC (2018) Quantitative assessment of prefrontal cortex in humans relative to nonhuman primates. Proc Natl Acad Sci U S A 115:E5183-E5192. Donoso M, Collins AG, Koechlin E (2014) Human cognition. Foundations of human reasoning in the prefrontal cortex. Science 344:1481-1486. Dosenbach NU, Fair DA, Cohen AL, Schlaggar BL, Petersen SE (2008) A dual-networks architecture of top-down control. Trends Cogn Sci 12:99-105. Dosenbach NU, Fair DA, Miezin FM, Cohen AL, Wenger KK, Dosenbach RA, Fox MD, Snyder AZ, Vincent JL, Raichle ME, Schlaggar BL, Petersen SE (2007) Distinct brain networks for adaptive and stable task control in humans. Proc Natl Acad Sci U S A 104:11073-11078. Draganski B, May A (2008) Training-induced structural changes in the adult human brain. Behav Brain Res 192:137-142. Draganski B, Gaser C, Busch V, Schuierer G, Bogdahn U, May A (2004) Neuroplasticity: changes in grey matter induced by training. Nature 427:311-312. Draganski B, Moser T, Lummel N, Ganssbauer S, Bogdahn U, Haas F, May A (2006) Decrease of thalamic gray matter following limb amputation. Neuroimage 31:951-957. Drevets WC, Burton H, Videen TO, Snyder AZ, Simpson JR, Jr., Raichle ME (1995) Blood flow changes in human somatosensory cortex during anticipated stimulation. Nature 373:249-252. Driver ID, Andoh J, Blockley NP, Francis ST, Gowland PA, Paus T (2015) Hemispheric asymmetry in cerebrovascular reactivity of the human primary motor cortex: an in vivo study at 7 T. NMR Biomed 28:538-545. Dum RP, Strick PL (2005) Frontal lobe inputs to the digit representations of the motor areas on the lateral surface of the hemisphere. J Neurosci 25:1375-1386. Ehrsson HH, Fagergren A, Jonsson T, Westling G, Johansson RS, Forssberg H (2000) Cortical activity in precision- versus power-grip tasks: an fMRI study. J Neurophysiol 83:528-536. Eickhoff SB, Schleicher A, Zilles K, Amunts K (2006a) The human parietal operculum. I. Cytoarchitectonic mapping of subdivisions. Cereb Cortex 16:254-267. Eickhoff SB, Amunts K, Mohlberg H, Zilles K (2006b) The human parietal operculum. II. Stereotaxic maps and correlation with functional imaging results. Cereb Cortex 16:268-279.

Eickhoff SB, Weiss PH, Amunts K, Fink GR, Zilles K (2006c) Identifying human parieto- insular vestibular cortex using fMRI and cytoarchitectonic mapping. Hum Brain Mapp 27:611-621. Eickhoff SB, Jbabdi S, Caspers S, Laird AR, Fox PT, Zilles K, Behrens TE (2010) Anatomical and functional connectivity of cytoarchitectonic areas within the human parietal operculum. J Neurosci 30:6409-6421. Ejaz N, Hamada M, Diedrichsen J (2015) Hand use predicts the structure of representations in sensorimotor cortex. Nat Neurosci 18:1034-1040. Eklund A, Nichols TE, Knutsson H (2016) Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates. Proc Natl Acad Sci U S A 113:7900-7905. Ergenzinger ER, Glasier MM, Hahm JO, Pons TP (1998) Cortically induced thalamic plasticity in the primate somatosensory system. Nat Neurosci 1:226-229. Fadiga L, Craighero L, D'Ausilio A (2009) Broca's area in language, action, and music. Ann N Y Acad Sci 1169:448-458. Fan L, Li H, Zhuo J, Zhang Y, Wang J, Chen L, Yang Z, Chu C, Xie S, Laird AR, Fox PT, Eickhoff SB, Yu C, Jiang T (2016) The Human Brainnetome Atlas: A New Brain Atlas Based on Connectional Architecture. Cereb Cortex 26:3508-3526. Fay RR, Popper AN (2000) Evolution of hearing in vertebrates: the inner ears and processing. Hear Res 149:1-10. Feldman DE (2012) The spike-timing dependence of plasticity. Neuron 75:556-571. Ferrari-Toniolo S, Visco-Comandini F, Papazachariadis O, Caminiti R, Battaglia-Mayer A (2015) Posterior Parietal Cortex Encoding of Dynamic Hand Force Underlying Hand-Object Interaction. J Neurosci 35:10899-10910. Fields RD (2015) A new mechanism of nervous system plasticity: activity-dependent myelination. Nat Rev Neurosci 16:756-767. Florence SL, Garraghty PE, Wall JT, Kaas JH (1994) Sensory afferent projections and area 3b somatotopy following median nerve cut and repair in macaque monkeys. Cereb Cortex 4:391-407. Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC (1995) Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold. Magn Reson Med 33:636-647. Fornander L, Nyman T, Hansson T, Ragnehed M, Brismar T (2010) Age- and time- dependent effects on functional outcome and cortical activation pattern in patients with median nerve injury: a functional magnetic resonance imaging study. J Neurosurg 113:122-128. Foxe JJ, Morocz IA, Murray MM, Higgins BA, Javitt DC, Schroeder CE (2000) Multisensory auditory-somatosensory interactions in early cortical processing revealed by high-density electrical mapping. Brain Res Cogn Brain Res 10:77- 83. Frank SM, Greenlee MW (2018) The parieto-insular vestibular cortex in humans: more than a single area? J Neurophysiol 120:1438-1450. Freeman AW, Johnson KO (1982) A model accounting for effects of vibratory amplitude on responses of cutaneous mechanoreceptors in macaque monkey. J Physiol 323:43-64. Frey SH (2008) Tool use, communicative gesture and cerebral asymmetries in the modern human brain. Philos Trans R Soc Lond B Biol Sci 363:1951-1957. Friedman RM, Chen LM, Roe AW (2004) Modality maps within primate somatosensory cortex. Proc Natl Acad Sci U S A 101:12724-12729. Fritsch GH, E. (1870) Über die elektrische Erregbarkeit des Grosshirns (On the Electrical Excitability of the Cerebrum). Archiv für Anatomie, Physiologie und wissenschaftliche Medicin:300-332.

Fu KM, Johnston TA, Shah AS, Arnold L, Smiley J, Hackett TA, Garraghty PE, Schroeder CE (2003) Auditory cortical neurons respond to somatosensory stimulation. J Neurosci 23:7510-7515. Galletti C, Fattori P (2018) The dorsal visual stream revisited: Stable circuits or dynamic pathways? Cortex 98:203-217. Galletti C, Gamberini M, Kutz DF, Fattori P, Luppino G, Matelli M (2001) The cortical connections of area V6: an occipito-parietal network processing visual information. Eur J Neurosci 13:1572-1588. Gallivan JP, Culham JC (2015) Neural coding within human brain areas involved in actions. Curr Opin Neurobiol 33:141-149. Gallivan JP, Cavina-Pratesi C, Culham JC (2009) Is that within reach? fMRI reveals that the human superior parieto-occipital cortex encodes objects reachable by the hand. J Neurosci 29:4381-4391. Ganguly K, Secundo L, Ranade G, Orsborn A, Chang EF, Dimitrov DF, Wallis JD, Barbaro NM, Knight RT, Carmena JM (2009) Cortical representation of ipsilateral arm movements in monkey and man. J Neurosci 29:12948-12956. Garraghty PE, Kaas JH (1991) Large-scale functional reorganization in adult monkey cortex after peripheral nerve injury. Proc Natl Acad Sci U S A 88:6976-6980. Garraghty PE, LaChica EA, Kaas JH (1991) Injury-induced reorganization of somatosensory cortex is accompanied by reductions in GABA staining. Somatosens Mot Res 8:347-354. Garraghty PE, Arnold LL, Wellman CL, Mowery TM (2006) Receptor autoradiographic correlates of deafferentation-induced reorganization in adult primate somatosensory cortex. J Comp Neurol 497:636-645. Geere J, Chester R, Kale S, Jerosch-Herold C (2007) Power grip, pinch grip, manual muscle testing or thenar atrophy - which should be assessed as a motor outcome after carpal tunnel decompression? A systematic review. BMC Musculoskelet Disord 8:114. Gelnar PA, Krauss BR, Szeverenyi NM, Apkarian AV (1998) Fingertip representation in the human somatosensory cortex: an fMRI study. Neuroimage 7:261-283. Gentile G, Petkova VI, Ehrsson HH (2011) Integration of visual and tactile signals from the hand in the human brain: an FMRI study. J Neurophysiol 105:910-922. Gesslbauer B, Hruby LA, Roche AD, Farina D, Blumer R, Aszmann OC (2017) Axonal components of nerves innervating the human arm. Ann Neurol 82:396-408. Gharbawie OA, Stepniewska I, Qi H, Kaas JH (2011) Multiple parietal-frontal pathways mediate grasping in macaque monkeys. J Neurosci 31:11660-11677. Ghazanfar AA, Schroeder CE (2006) Is neocortex essentially multisensory? Trends Cogn Sci 10:278-285. Ghosh SK (2015) Human cadaveric dissection: a historical account from ancient Greece to the modern era. Anat Cell Biol 48:153-169. Gilbert CD, Sigman M (2007) Brain states: top-down influences in sensory processing. Neuron 54:677-696. Glover GH (2011) Overview of functional magnetic resonance imaging. Neurosurg Clin N Am 22:133-139, vii. Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, Frackowiak RS (2001) A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 14:21-36. Goodale MA, Milner AD (1992) Separate visual pathways for perception and action. Trends Neurosci 15:20-25. Grafton ST (2010) The cognitive neuroscience of prehension: recent developments. Exp Brain Res 204:475-491. Grefkes C, Ritzl A, Zilles K, Fink GR (2004) Human medial intraparietal cortex subserves visuomotor coordinate transformation. NeuroImage 23:1494-1506.

Grezes J, Decety J (2001) Functional anatomy of execution, mental simulation, observation, and verb generation of actions: a meta-analysis. Hum Brain Mapp 12:1-19. Grinsell D, Keating CP (2014) Peripheral nerve reconstruction after injury: a review of clinical and experimental therapies. Biomed Res Int 2014:698256. Gross CG (1987) Early history of Neuroscience. In: Encyclopedia of Neuroscience (Adelman G, ed), pp 843-847: Birkauser. Grünbaum ASFS, C.S. (1902) Observations on the physiology of the cerebral cortex of some of the higher apes. Proceedings of the Royal Society of London 69:206- 209. Guthrie GJ (1820) A treatise on gun-shot wounds, on injuries of nerves and on wounds of the extremities, 2 Edition. London: S. Gosnell. Haak KV, Beckmann CF (2018) Objective analysis of the topological organization of the human cortical visual connectome suggests three visual pathways. Cortex 98:73-83. Haaland KY, Harrington DL (1996) Hemispheric asymmetry of movement. Curr Opin Neurobiol 6:796-800. Haaland KY, Harrington DL, Knight RT (2000) Neural representations of skilled movement. Brain 123 ( Pt 11):2306-2313. Hagbarth KE, Vallbo AB (1967) Afferent response to mechanical stimulation of muscle receptors in man. Acta Soc Med Ups 72:102-104. Hagen MC, Franzén O, McGlone F, Essick G, Dancer C, Pardo JV (2002) Tactile motion activates the human middle temporal/V5 (MT/V5) complex. European Journal of Neuroscience 16:957-964. Hallin RG, Wiesenfeld Z, Lindblom U (1981) Neurophysiological studies on patients with sutured median nerves: faulty sensory localization after nerve regeneration and its physiological correlates. Exp Neurol 73:90-106. Hansson T, Brismar T (2003) Loss of sensory discrimination after median nerve injury and activation in the primary somatosensory cortex on functional magnetic resonance imaging. J Neurosurg 99:100-105. Harlow JM (1869) Recovery from the passage of an iron bar through the head. Boston, Massachusetts, USA: Clap. Harris F, Jabbur SJ, Morse RW, Towe AL (1965) Influence of the cerebral cortex on the cuneate nucleus of the monkey. Nature 208:1215-1216. Hart AM, Terenghi G, Wiberg M (2008) Neuronal death after peripheral nerve injury and experimental strategies for neuroprotection. Neurol Res 30:999-1011. Hart AM, Terenghi G, Kellerth JO, Wiberg M (2004) Sensory neuroprotection, mitochondrial preservation, and therapeutic potential of N-acetyl-cysteine after nerve injury. Neuroscience 125:91-101. Hausser CO, Robert F, Giard N (1980) Balint's syndrome. Can J Neurol Sci 7:157-161. Hebb DO (1949) The Organization of Behavior - A Neuropsychological Theory. London: John Wiley & Sons, Inc. Heming EAC, K.P.; Takei, T.; Cook, D.J.; Scott, S.H. (2019) Keep your hands apart: independent representations of ipsilateral and contralateral forelimbs in primary motor cortex. bioRxiv. Hermsdorfer J, Terlinden G, Muhlau M, Goldenberg G, Wohlschlager AM (2007) Neural representations of pantomimed and actual tool use: evidence from an event- related fMRI study. Neuroimage 36 Suppl 2:T109-118. Hernowo AT, Prins D, Baseler HA, Plank T, Gouws AD, Hooymans JM, Morland AB, Greenlee MW, Cornelissen FW (2014) Morphometric analyses of the visual pathways in macular degeneration. Cortex 56:99-110. Higuchi S, Imamizu H, Kawato M (2007) Cerebellar activity evoked by common tool-use execution and imagery tasks: an fMRI study. Cortex 43:350-358.

Hillman EM (2014) Coupling mechanism and significance of the BOLD signal: a status report. Annu Rev Neurosci 37:161-181. Himes BT, Tessler A (1989) Death of some dorsal root ganglion neurons and plasticity of others following sciatic nerve section in adult and neonatal rats. J Comp Neurol 284:215-230. Hlushchuk Y, Hari R (2006) Transient suppression of ipsilateral primary somatosensory cortex during tactile finger stimulation. J Neurosci 26:5819-5824. Hofer S, Frahm J (2006) Topography of the human corpus callosum revisited-- comprehensive fiber tractography using diffusion tensor magnetic resonance imaging. Neuroimage 32:989-994. Hsiao S (2008) Central mechanisms of tactile shape perception. Curr Opin Neurobiol 18:418-424. Hsiao SS, O'Shaughnessy DM, Johnson KO (1993) Effects of selective attention on spatial form processing in monkey primary and secondary somatosensory cortex. J Neurophysiol 70:444-447. Huber L, Handwerker DA, Jangraw DC, Chen G, Hall A, Stuber C, Gonzalez-Castillo J, Ivanov D, Marrett S, Guidi M, Goense J, Poser BA, Bandettini PA (2017) High- Resolution CBV-fMRI Allows Mapping of Laminar Activity and Connectivity of Cortical Input and Output in Human M1. Neuron 96:1253-1263 e1257. Huettel SA, Song WS, McCarthy G (2009) Functional Magnetic Resonance Imaging, 2nd Edition. Sunderland, Massachusetts, USA: Sinauer Associates, Inc. Husain FT, Medina RE, Davis CW, Szymko-Bennett Y, Simonyan K, Pajor NM, Horwitz B (2011) Neuroanatomical changes due to hearing loss and chronic tinnitus: a combined VBM and DTI study. Brain Res 1369:74-88. Hutton C, Draganski B, Ashburner J, Weiskopf N (2009) A comparison between voxel- based cortical thickness and voxel-based morphometry in normal aging. Neuroimage 48:371-380. Hyvarinen J, Poranen A, Jokinen Y (1980) Influence of attentive behavior on neuronal responses to vibration in primary somatosensory cortex of the monkey. J Neurophysiol 43:870-882. Ilg UJ, Schumann S (2007) Primate area MST-l is involved in the generation of goal- directed eye and hand movements. J Neurophysiol 97:761-771. Imazu S, Sugio T, Tanaka S, Inui T (2007) Differences between actual and imagined usage of chopsticks: an fMRI study. Cortex 43:301-307. Ingram JN, Wolpert DM (2011) Naturalistic approaches to sensorimotor control. Prog Brain Res 191:3-29. Irwin MS, Gilbert SE, Terenghi G, Smith RW, Green CJ (1997) Cold intolerance following peripheral nerve injury. Natural history and factors predicting severity of symptoms. J Hand Surg Br 22:308-316. Isaacson JS, Scanziani M (2011) How inhibition shapes cortical activity. Neuron 72:231- 243. Iwamura Y, Tanaka M, Iriki A, Taoka M, Toda T (2002) Processing of tactile and kinesthetic signals from bilateral sides of the body in the postcentral gyrus of awake monkeys. Behav Brain Res 135:185-190. Jain N, Florence SL, Kaas JH (1998) Reorganization of Somatosensory Cortex After Nerve and Spinal Cord Injury. News Physiol Sci 13:143-149. Jaquet JB, Luijsterburg AJ, Kalmijn S, Kuypers PD, Hofman A, Hovius SE (2001) Median, ulnar, and combined median-ulnar nerve injuries: functional outcome and return to productivity. J Trauma 51:687-692. Jeannerod M, Arbib MA, Rizzolatti G, Sakata H (1995) Grasping objects: the cortical mechanisms of visuomotor transformation. Trends Neurosci 18:314-320. Jenmalm P, Johansson RS (1997) Visual and somatosensory information about object shape control manipulative fingertip forces. J Neurosci 17:4486-4499.

Jenmalm P, Dahlstedt S, Johansson RS (2000) Visual and tactile information about object-curvature control fingertip forces and grasp kinematics in human dexterous manipulation. J Neurophysiol 84:2984-2997. Jerosch-Herold C (1993) Measuring outcome in median nerve injuries. J Hand Surg Br 18:624-628. Jerosch-Herold C (2005) Assessment of sensibility after nerve injury and repair: a systematic review of evidence for validity, reliability and responsiveness of tests. J Hand Surg Br 30:252-264. Jocham G, Neumann J, Klein TA, Danielmeier C, Ullsperger M (2009) Adaptive coding of action values in the human rostral cingulate zone. J Neurosci 29:7489-7496. Johansen-Berg H, Lloyd DM (2000) The physiology and psychology of selective attention to touch. Front Biosci 5:D894-904. Johansen-Berg H, Matthews PM (2002) Attention to movement modulates activity in sensori-motor areas, including primary motor cortex. Exp Brain Res 142:13-24. Johansen-Berg H, Christensen V, Woolrich M, Matthews PM (2000) Attention to touch modulates activity in both primary and secondary somatosensory areas. Neuroreport 11:1237-1241. Johansson RS (1978) Tactile sensibility in the human hand: receptive field characteristics of mechanoreceptive units in the glabrous skin area. J Physiol 281:101-125. Johansson RS, Vallbo AB (1979) Tactile sensibility in the human hand: relative and absolute densities of four types of mechanoreceptive units in glabrous skin. J Physiol 286:283-300. Johansson RS, Birznieks I (2004) First spikes in ensembles of human tactile afferents code complex spatial fingertip events. Nat Neurosci 7:170-177. Johansson RS, Flanagan JR (2009) Coding and use of tactile signals from the fingertips in object manipulation tasks. Nat Rev Neurosci 10:345-359. Johansson RS, Landstrom U, Lundstrom R (1982) Responses of mechanoreceptive afferent units in the glabrous skin of the human hand to sinusoidal skin displacements. Brain Res 244:17-25. Johnson-Frey SH (2004) The neural bases of complex tool use in humans. Trends Cogn Sci 8:71-78. Johnson-Frey SH, Newman-Norlund R, Grafton ST (2005) A distributed left hemisphere network active during planning of everyday tool use skills. Cereb Cortex 15:681- 695. Johnson KO (2001) The roles and functions of cutaneous mechanoreceptors. Curr Opin Neurobiol 11:455-461. Johnston JM, Vaishnavi SN, Smyth MD, Zhang D, He BJ, Zempel JM, Shimony JS, Snyder AZ, Raichle ME (2008) Loss of resting interhemispheric functional connectivity after complete section of the corpus callosum. J Neurosci 28:6453- 6458. Jones EG (2000) Cortical and subcortical contributions to activity-dependent plasticity in primate somatosensory cortex. Annu Rev Neurosci 23:1-37. Kaas JH (1993) The functional organization of somatosensory cortex in primates. Ann Anat 175:509-518. Kaas JH (2004) Evolution of somatosensory and motor cortex in primates. Anat Rec A Discov Mol Cell Evol Biol 281:1148-1156. Kaas JH, Collins CE (2003) Anatomic and functional reorganization of somatosensory cortex in mature primates after peripheral nerve and spinal cord injury. Adv Neurol 93:87-95. Kaas JH, Merzenich MM, Killackey HP (1983) The reorganization of somatosensory cortex following peripheral nerve damage in adult and developing mammals. Annu Rev Neurosci 6:325-356.

Kaas JH, Nelson RJ, Sur M, Dykes RW, Merzenich MM (1984) The somatotopic organization of the ventroposterior thalamus of the squirrel monkey, Saimiri sciureus. J Comp Neurol 226:111-140. Kashyap S, Ivanov D, Havlicek M, Sengupta S, Poser BA, Uludag K (2018) Resolving laminar activation in human V1 using ultra-high spatial resolution fMRI at 7T. Sci Rep 8:17063. Keysers C, Kaas JH, Gazzola V (2010) Somatosensation in social perception. Nat Rev Neurosci 11:417-428. Kikkert S, Kolasinski J, Jbabdi S, Tracey I, Beckmann CF, Johansen-Berg H, Makin TR (2016) Revealing the neural fingerprints of a missing hand. Elife 5. Kilbreath SL, Heard RC (2005) Frequency of hand use in healthy older persons. Aust J Physiother 51:119-122. Killackey HP, Gould HJ, 3rd, Cusick CG, Pons TP, Kaas JH (1983) The relation of corpus callosum connections to architectonic fields and body surface maps in sensorimotor cortex of new and old world monkeys. J Comp Neurol 219:384- 419. Kim SG, Ashe J, Georgopoulos AP, Merkle H, Ellermann JM, Menon RS, Ogawa S, Ugurbil K (1993) Functional imaging of human motor cortex at high magnetic field. J Neurophysiol 69:297-302. Klingner CM, Brodoehl S, Witte OW (2015) The importance of the negative blood- oxygenation-level-dependent (BOLD) response in the somatosensory cortex. Rev Neurosci 26:647-653. Knibestol M, Vallbo AB (1970) Single unit analysis of mechanoreceptor activity from the human glabrous skin. Acta Physiol Scand 80:178-195. Koechlin E (2014) An evolutionary computational theory of prefrontal executive function in decision-making. Philos Trans R Soc Lond B Biol Sci 369. Koechlin E (2016) Prefrontal executive function and adaptive behavior in complex environments. Curr Opin Neurobiol 37:1-6. Kolling N, Behrens T, Wittmann MK, Rushworth M (2016) Multiple signals in anterior cingulate cortex. Curr Opin Neurobiol 37:36-43. Krarup C, Rosen B, Boeckstyns M, Ibsen Sorensen A, Lundborg G, Moldovan M, Archibald SJ (2017) Sensation, mechanoreceptor, and nerve fiber function after nerve regeneration. Ann Neurol 82:940-950. Krause F (1908) Chirurgie des Gehirns und Rückenmarks nach eigenen Erfahrungen. Berlin, Germany: Urban & Schwarzenberg. Kravitz DJ, Saleem KS, Baker CI, Ungerleider LG, Mishkin M (2013) The ventral visual pathway: an expanded neural framework for the processing of object quality. Trends Cogn Sci 17:26-49. Kringelbach ML (2005) The human orbitofrontal cortex: linking reward to hedonic experience. Nat Rev Neurosci 6:691-702. Kringelbach ML, Rolls ET (2004) The functional neuroanatomy of the human orbitofrontal cortex: evidence from neuroimaging and neuropsychology. Prog Neurobiol 72:341-372. Kroliczak G, McAdam TD, Quinlan DJ, Culham JC (2008) The human dorsal stream adapts to real actions and 3D shape processing: a functional magnetic resonance imaging study. J Neurophysiol 100:2627-2639. Kurth R, Villringer K, Curio G, Wolf KJ, Krause T, Repenthin J, Schwiemann J, Deuchert M, Villringer A (2000) fMRI shows multiple somatotopic digit representations in human primary somatosensory cortex. Neuroreport 11:1487-1491. LaMotte RH, Mountcastle VB (1979) Disorders in somesthesis following lesions of parietal lobe. J Neurophysiol 42:400-419. Langer N, Hanggi J, Muller NA, Simmen HP, Jancke L (2012) Effects of limb immobilization on brain plasticity. Neurology 78:182-188.

Lauterbu CL, Mansfield P (2003) The Nobel Prize in Physiology or Medicine 2003 - "for their discoveries concerning magnetic resonance imaging.". In: NobelPrize.org. Nobel Media AB 2019. Lekan HA, Chung K, Yoon YW, Chung JM, Coggeshall RE (1997) Loss of dorsal root ganglion cells concomitant with dorsal root axon sprouting following segmental nerve lesions. Neuroscience 81:527-534. Lemon RN, Maier MA, Armand J, Kirkwood PA, Yang HW (2002) Functional differences in corticospinal projections from macaque primary motor cortex and supplementary motor area. Adv Exp Med Biol 508:425-434. Leung YY, Bensmaia SJ, Hsiao SS, Johnson KO (2005) Time-course of vibratory adaptation and recovery in cutaneous mechanoreceptive afferents. J Neurophysiol 94:3037-3045. Lewis JW (2006) Cortical networks related to human use of tools. Neuroscientist 12:211- 231. Levy LM, Ziemann U, Chen R, Cohen LG (2002) Rapid modulation of GABA in sensorimotor cortex induced by acute deafferentation. Ann Neurol 52:755-761. Li L, Oppenheim RW, Lei M, Houenou LJ (1994) Neurotrophic agents prevent motoneuron death following sciatic nerve section in the neonatal mouse. J Neurobiol 25:759-766. Liao CC, Gharbawie OA, Qi H, Kaas JH (2013) Cortical connections to single digit representations in area 3b of somatosensory cortex in squirrel monkeys and prosimian galagos. J Comp Neurol 521:3768-3790. Lipton ML, Liszewski MC, O'Connell MN, Mills A, Smiley JF, Branch CA, Isler JR, Schroeder CE (2010) Interactions within the hand representation in primary somatosensory cortex of primates. J Neurosci 30:15895-15903. Liss AG, af Ekenstam FW, Wiberg M (1996) Loss of neurons in the dorsal root ganglia after transection of a peripheral sensory nerve. An anatomical study in monkeys. Scand J Plast Reconstr Surg Hand Surg 30:1-6. Liu XB, Honda CN, Jones EG (1995) Distribution of four types of synapse on physiologically identified relay neurons in the ventral posterior thalamic nucleus of the cat. J Comp Neurol 352:69-91. Lofvenberg J, Johansson RS (1984) Regional differences and interindividual variability in sensitivity to vibration in the glabrous skin of the human hand. Brain Res 301:65-72. Logothetis NK (2008) What we can do and what we cannot do with fMRI. Nature 453:869-878. Luders E, Narr KL, Thompson PM, Rex DE, Jancke L, Toga AW (2006) Hemispheric asymmetries in cortical thickness. Cereb Cortex 16:1232-1238. Lundborg G, Rosen B (2007) Hand function after nerve repair. Acta Physiol (Oxf) 189:207-217. Lundstrom R, Johansson RS (1986) Acute impairment of the sensitivity of skin mechanoreceptive units caused by vibration exposure of the hand. Ergonomics 29:687-698. Mackel R (1985) Human cutaneous mechanoreceptors during regeneration: physiology and interpretation. Ann Neurol 18:165-172. Mackel R, Brink EE, Wittkowsky G (1983a) Transitional properties of afferents reinnervating mechanoreceptors in the human glabrous skin. Brain Res 276:339-343. Mackel R, Kunesch E, Waldhor F, Struppler A (1983b) Reinnervation of mechanoreceptors in the human glabrous skin following peripheral nerve repair. Brain Res 268:49-65. Mackinnon SE (2015) Nerve Surgery. New York: Thieme Medical Publishers, Inc. Makin TR, Bensmaia SJ (2017) Stability of Sensory Topographies in Adult Cortex. Trends Cogn Sci 21:195-204.

Makin TR, Scholz J, Filippini N, Henderson Slater D, Tracey I, Johansen-Berg H (2013a) Phantom pain is associated with preserved structure and function in the former hand area. Nat Commun 4:1570. Makin TR, Cramer AO, Scholz J, Hahamy A, Henderson Slater D, Tracey I, Johansen- Berg H (2013b) Deprivation-related and use-dependent plasticity go hand in hand. Elife 2:e01273. Maldjian JA, Gottschalk A, Patel RS, Detre JA, Alsop DC (1999) The sensory somatotopic map of the human hand demonstrated at 4 Tesla. Neuroimage 10:55-62. Mansouri FA, Koechlin E, Rosa MGP, Buckley MJ (2017) Managing competing goals - a key role for the frontopolar cortex. Nat Rev Neurosci 18:645-657. Marik SA, Yamahachi H, McManus JN, Szabo G, Gilbert CD (2010) Axonal dynamics of excitatory and inhibitory neurons in somatosensory cortex. PLoS Biol 8:e1000395. Martuzzi R, van der Zwaag W, Farthouat J, Gruetter R, Blanke O (2014) Human finger somatotopy in areas 3b, 1, and 2: a 7T fMRI study using a natural stimulus. Hum Brain Mapp 35:213-226. Mavrogenis AF, Spyridonos SG, Antonopoulos D, Soucacos PN, Papagelopoulos PJ (2009) Effect of sensory re-education after low median nerve complete transection and repair. J Hand Surg Am 34:1210-1215. Mayka MA, Corcos DM, Leurgans SE, Vaillancourt DE (2006) Three-dimensional locations and boundaries of motor and premotor cortices as defined by functional brain imaging: a meta-analysis. Neuroimage 31:1453-1474. McKiernan BJ, Marcario JK, Karrer JH, Cheney PD (1998) Corticomotoneuronal postspike effects in shoulder, elbow, wrist, digit, and intrinsic hand muscles during a reach and prehension task. J Neurophysiol 80:1961-1980. McLaren DG (2014) Jiscmail [SPM]: Choosing which 2nd level analysis approach to use. In. Mendola JD, Conner IP, Roy A, Chan ST, Schwartz TL, Odom JV, Kwong KK (2005) Voxel-based analysis of MRI detects abnormal visual cortex in children and adults with amblyopia. Hum Brain Mapp 25:222-236. Menon V, Uddin LQ (2010) Saliency, switching, attention and control: a network model of insula function. Brain Struct Funct 214:655-667. Merzenich MM, Jenkins WM (1993) Reorganization of cortical representations of the hand following alterations of skin inputs induced by nerve injury, skin island transfers, and experience. J Hand Ther 6:89-104. Merzenich MM, Kaas JH, Wall JT, Sur M, Nelson RJ, Felleman DJ (1983a) Progression of change following median nerve section in the cortical representation of the hand in areas 3b and 1 in adult owl and squirrel monkeys. Neuroscience 10:639- 665. Merzenich MM, Kaas JH, Wall J, Nelson RJ, Sur M, Felleman D (1983b) Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted deafferentation. Neuroscience 8:33-55. Meshberger FL (1990) An interpretation of Michelangelo's Creation of Adam based on neuroanatomy. JAMA 264:1837-1841. Mikl M, Marecek R, Hlustik P, Pavlicova M, Drastich A, Chlebus P, Brazdil M, Krupa P (2008) Effects of spatial smoothing on fMRI group inferences. Magn Reson Imaging 26:490-503. Miller EK, Cohen JD (2001) An integrative theory of prefrontal cortex function. Annu Rev Neurosci 24:167-202. Mishkin M (1979) Analogous neural models for tactual and visual learning. Neuropsychologia 17:139-151. Mishkin M, Ungerleider LG (1982) Contribution of striate inputs to the visuospatial functions of parieto-preoccipital cortex in monkeys. Behav Brain Res 6:57-77.

Mishkin M, Lewis ME, Ungerleider LG (1982) Equivalence of parieto-preoccipital subareas for visuospatial ability in monkeys. Behav Brain Res 6:41-55. Mishkin MU, L.G. Macko, K.A. (1983) Object vision and spatial vision: two cortical pathways. Trends in Neurosciences 6:414-417. Moberg E (1958) Objective methods for determining the functional value of sensibility in the hand. J Bone Joint Surg Br 40-B:454-476. Moberg E (1966) Methods for examining sensibility in the hand. In: Hand Surgery (Flynn JE, ed), pp 435-439. Baltimore: Williams & Wilkins. Nachev P, Kennard C, Husain M (2008) Functional role of the supplementary and pre- supplementary motor areas. Nat Rev Neurosci 9:856-869. Nachev P, Wydell H, O'Neill K, Husain M, Kennard C (2007) The role of the pre- supplementary motor area in the control of action. Neuroimage 36 Suppl 2:T155-163. Navarro X, Vivo M, Valero-Cabre A (2007) Neural plasticity after peripheral nerve injury and regeneration. Prog Neurobiol 82:163-201. Nelson AJ, Staines WR, Graham SJ, McIlroy WE (2004) Activation in SI and SII: the influence of vibrotactile amplitude during passive and task-relevant stimulation. Brain Res Cogn Brain Res 19:174-184. Nichols T, Brett M, Andersson J, Wager T, Poline JB (2005) Valid conjunction inference with the minimum statistic. Neuroimage 25:653-660. Nicholson OR, Seddon HJ (1957) Nerve repair in civil practice; results of treatment of median and ulnar nerve lesions. Br Med J 2:1065-1071. Nolano M, Provitera V, Crisci C, Stancanelli A, Wendelschafer-Crabb G, Kennedy WR, Santoro L (2003) Quantification of myelinated endings and mechanoreceptors in human digital skin. Ann Neurol 54:197-205. Nordmark PF, Pruszynski JA, Johansson RS (2012) BOLD responses to tactile stimuli in visual and auditory cortex depend on the frequency content of stimulation. J Cogn Neurosci 24:2120-2134. Nordmark PF, Ljungberg C, Johansson RS (2018) Structural changes in hand related cortical areas after median nerve injury and repair. Sci Rep 8:4485. Nowak DA, Berner J, Herrnberger B, Kammer T, Gron G, Schonfeldt-Lecuona C (2009) Continuous theta-burst stimulation over the dorsal premotor cortex interferes with associative learning during object lifting. Cortex 45:473-482. Novikov L, Novikova L, Kellerth JO (1995) Brain-derived neurotrophic factor promotes survival and blocks nitric oxide synthase expression in adult rat spinal motoneurons after ventral root avulsion. Neurosci Lett 200:45-48. O'Reilly RC (2010) The What and How of prefrontal cortical organization. Trends Neurosci 33:355-361. O'Reilly RC, Herd SA, Pauli WM (2010) Computational models of cognitive control. Curr Opin Neurobiol 20:257-261. Ochoa J, Torebjork E (1983) Sensations evoked by intraneural microstimulation of single mechanoreceptor units innervating the human hand. J Physiol 342:633-654. Ogawa S, Lee TM, Kay AR, Tank DW (1990) Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci U S A 87:9868- 9872. Open Science C (2015) PSYCHOLOGY. Estimating the reproducibility of psychological science. Science 349:aac4716. Palmeri A, Bellomo M, Giuffrida R, Sapienza S (1999) Motor cortex modulation of exteroceptive information at bulbar and thalamic lemniscal relays in the cat. Neuroscience 88:135-150. Papadelis C, Eickhoff SB, Zilles K, Ioannides AA (2011) BA3b and BA1 activate in a serial fashion after median nerve stimulation: direct evidence from combining source analysis of evoked fields and cytoarchitectonic probabilistic maps. Neuroimage 54:60-73.

Pascual-Leone A, Hamilton R (2001) The metamodal organization of the brain. Prog Brain Res 134:427-445. Paul RL, Goodman H, Merzenich M (1972) Alterations in mechanoreceptor input to Brodmann's areas 1 and 3 of the postcentral hand area of Macaca mulatta after nerve section and regeneration. Brain Res 39:1-19. Pederson WC (2014) Median nerve injury and repair. J Hand Surg Am 39:1216-1222. Penfield WB, E. (1937) Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 60:389-443. Penfield WJ, H. (1954) Epilepsy and the Functional Anatomy of the Human Brain. Boston: Little, Brown & Co. Perenin MT, Vighetto A (1988) Optic ataxia: a specific disruption in visuomotor mechanisms. I. Different aspects of the deficit in reaching for objects. Brain 111 ( Pt 3):643-674. Pfister BJ, Gordon T, Loverde JR, Kochar AS, Mackinnon SE, Cullen DK (2011) Biomedical engineering strategies for peripheral nerve repair: surgical applications, state of the art, and future challenges. Crit Rev Biomed Eng 39:81- 124. Phillips CG, Powell TP, Wiesendanger M (1971) Projection from low-threshold muscle afferents of hand and forearm to area 3a of baboon's cortex. J Physiol 217:419- 446. Picard N, Strick PL (1996) Motor areas of the medial wall: a review of their location and functional activation. Cereb Cortex 6:342-353. Pleger B, Dinse HR, Ragert P, Schwenkreis P, Malin JP, Tegenthoff M (2001) Shifts in cortical representations predict human discrimination improvement. Proc Natl Acad Sci U S A 98:12255-12260. Pons TP, Kaas JH (1986) Corticocortical connections of area 2 of somatosensory cortex in macaque monkeys: a correlative anatomical and electrophysiological study. J Comp Neurol 248:313-335. Pons TP, Garraghty PE, Mishkin M (1992) Serial and parallel processing of tactual information in somatosensory cortex of rhesus monkeys. J Neurophysiol 68:518-527. Pons TP, Garraghty PE, Ommaya AK, Kaas JH, Taub E, Mishkin M (1991) Massive cortical reorganization after sensory deafferentation in adult macaques. Science 252:1857-1860. Popper AN, Fay RR (1997) Evolution of the ear and hearing: issues and questions. Brain Behav Evol 50:213-221. Porro CA, Lui F, Facchin P, Maieron M, Baraldi P (2004) Percept-related activity in the human somatosensory system: functional magnetic resonance imaging studies. Magn Reson Imaging 22:1539-1548. Prado J, Clavagnier S, Otzenberger H, Scheiber C, Kennedy H, Perenin MT (2005) Two cortical systems for reaching in central and peripheral vision. Neuron 48:849- 858. Preissler S, Feiler J, Dietrich C, Hofmann GO, Miltner WH, Weiss T (2013) Gray matter changes following limb amputation with high and low intensities of phantom limb pain. Cereb Cortex 23:1038-1048. Pruszynski JA, Johansson RS (2014) Edge-orientation processing in first-order tactile neurons. Nat Neurosci 17:1404-1409. Pruszynski JA, Flanagan JR, Johansson RS (2018) Fast and accurate edge orientation processing during object manipulation. Elife 7. Puckett AM, Bollmann S, Barth M, Cunnington R (2017) Measuring the effects of attention to individual fingertips in somatosensory cortex using ultra-high field (7T) fMRI. Neuroimage 161:179-187. Puts NA, Edden RA (2012) In vivo magnetic resonance spectroscopy of GABA: a methodological review. Prog Nucl Magn Reson Spectrosc 60:29-41.

Ramon y Cajal S (1890) A quelle epoque apparaissent les expansions des cellule nerveuses de la moelle epinere du poulet. Anat Anzerger 5:609-613. Raos V, Umilta MA, Murata A, Fogassi L, Gallese V (2006) Functional properties of grasping-related neurons in the ventral premotor area F5 of the macaque monkey. J Neurophysiol 95:709-729. Rawa A (1885) Über die Nervennaht. Wiener Medizinische Wochenschrift 35:358-359. Ray WZ, Mackinnon SE (2010) Management of nerve gaps: autografts, allografts, nerve transfers, and end-to-side neurorrhaphy. Exp Neurol 223:77-85. Reed JL, Qi HX, Kaas JH (2011) Spatiotemporal properties of neuron response suppression in owl monkey primary somatosensory cortex when stimuli are presented to both hands. J Neurosci 31:3589-3601. Reed JL, Qi HX, Zhou Z, Bernard MR, Burish MJ, Bonds AB, Kaas JH (2010) Response properties of neurons in primary somatosensory cortex of owl monkeys reflect widespread spatiotemporal integration. J Neurophysiol 103:2139-2157. Reichenbach A, Bresciani JP, Peer A, Bulthoff HH, Thielscher A (2011) Contributions of the PPC to online control of visually guided reaching movements assessed with fMRI-guided TMS. Cereb Cortex 21:1602-1612. Ricciardi E, Vanello N, Sani L, Gentili C, Scilingo EP, Landini L, Guazzelli M, Bicchi A, Haxby JV, Pietrini P (2007) The effect of visual experience on the development of functional architecture in hMT+. Cerebral Cortex 17:2933-2939. Riehle AV, E. et al (2005) Motor cortex in voluntary movements : a distributed system for distributed functions. Boca Raton, Florida: CRC Press. Risling M, Aldskogius H, Hildebrand C, Remahl S (1983) Effects of sciatic nerve resection on L7 spinal roots and dorsal root ganglia in adult cats. Exp Neurol 82:568-580. Rizzolatti G, Matelli M (2003) Two different streams form the dorsal visual system: anatomy and functions. Exp Brain Res 153:146-157. Rizzolatti G, Cattaneo L, Fabbri-Destro M, Rozzi S (2014) Cortical mechanisms underlying the organization of goal-directed actions and mirror neuron-based action understanding. Physiol Rev 94:655-706. Rizzolatti G, Camarda R, Fogassi L, Gentilucci M, Luppino G, Matelli M (1988) Functional organization of inferior area 6 in the macaque monkey. II. Area F5 and the control of distal movements. Exp Brain Res 71:491-507. Roiser JP, Linden DE, Gorno-Tempinin ML, Moran RJ, Dickerson BC, Grafton ST (2016) Minimum statistical standards for submissions to Neuroimage: Clinical. Neuroimage Clin 12:1045-1047. Roland PE, Mortensen E (1987) Somatosensory detection of microgeometry, macrogeometry and kinesthesia in man. Brain Res 434:1-42. Romo R, Lemus L, de Lafuente V (2012) Sense, memory, and decision-making in the somatosensory cortical network. Curr Opin Neurobiol 22:914-919. Rosen B, Lundborg G (2001) The long term recovery curve in adults after median or ulnar nerve repair: a reference interval. J Hand Surg Br 26:196-200. Ruben J, Krause T, Taskin B, Blankenburg F, Moosmann M, Villringer A (2006) Sub- area-specific Suppressive Interaction in the BOLD responses to simultaneous finger stimulation in human primary somatosensory cortex: evidence for increasing rostral-to-caudal convergence. Cereb Cortex 16:819-826. Ruijs AC, Jaquet JB, Kalmijn S, Giele H, Hovius SE (2005) Median and ulnar nerve injuries: a meta-analysis of predictors of motor and sensory recovery after modern microsurgical nerve repair. Plast Reconstr Surg 116:484-494; discussion 495-486. Sadato N, Yonekura Y, Waki A, Yamada H, Ishii Y (1997) Role of the supplementary motor area and the right premotor cortex in the coordination of bimanual finger movements. J Neurosci 17:9667-9674.

Sadato N, Pascual-Leone A, Grafman J, Deiber MP, Ibanez V, Hallett M (1998) Neural networks for Braille reading by the blind. Brain 121 ( Pt 7):1213-1229. Sadato N, Pascual-Leone A, Grafman J, Ibanez V, Deiber MP, Dold G, Hallett M (1996) Activation of the primary visual cortex by Braille reading in blind subjects. Nature 380:526-528. Salvador R, Suckling J, Coleman MR, Pickard JD, Menon D, Bullmore E (2005) Neurophysiological architecture of functional magnetic resonance images of human brain. Cereb Cortex 15:1332-1342. Sammons RP, Keck T (2015) Adult plasticity and cortical reorganization after peripheral lesions. Curr Opin Neurobiol 35:136-141. Sanchez Panchuelo RM, Besle J, Schluppeck D, Humberstone M, Francis S (2018) Somatotopy in the Human Somatosensory System. Front Hum Neurosci 12:235. Sanes JN, Suner S, Lando JF, Donoghue JP (1988) Rapid reorganization of adult rat motor cortex somatic representation patterns after motor nerve injury. Proc Natl Acad Sci U S A 85:2003-2007. Sato K, Nariai T, Sasaki S, Yazawa I, Mochida H, Miyakawa N, Momose-Sato Y, Kamino K, Ohta Y, Hirakawa K, Ohno K (2002) Intraoperative intrinsic optical imaging of neuronal activity from subdivisions of the human primary somatosensory cortex. Cereb Cortex 12:269-280. Schenk T, Ellison A, Rice N, Milner AD (2005) The role of V5/MT+ in the control of catching movements: an rTMS study. Neuropsychologia 43:189-198. Schluter ND, Krams M, Rushworth MF, Passingham RE (2001) Cerebral dominance for action in the human brain: the selection of actions. Neuropsychologia 39:105- 113. Schmidlin E, Brochier T, Maier MA, Kirkwood PA, Lemon RN (2008) Pronounced reduction of digit motor responses evoked from macaque ventral premotor cortex after reversible inactivation of the primary motor cortex hand area. J Neurosci 28:5772-5783. Scholz J, Klein MC, Behrens TE, Johansen-Berg H (2009) Training induces changes in white-matter architecture. Nat Neurosci 12:1370-1371. Schroeder CE, Lindsley RW, Specht C, Marcovici A, Smiley JF, Javitt DC (2001) Somatosensory input to auditory association cortex in the macaque monkey. J Neurophysiol 85:1322-1327. Schurmann M, Caetano G, Hlushchuk Y, Jousmaki V, Hari R (2006) Touch activates human auditory cortex. Neuroimage 30:1325-1331. Seddon HJ (1942) A Classification of Nerve Injuries. Br Med J 2:237-239. Seeley WW, Menon V, Schatzberg AF, Keller J, Glover GH, Kenna H, Reiss AL, Greicius MD (2007) Dissociable intrinsic connectivity networks for salience processing and executive control. J Neurosci 27:2349-2356. Shenhav A, Cohen JD, Botvinick MM (2016) Dorsal anterior cingulate cortex and the value of control. Nat Neurosci 19:1286-1291. Shinder ME, Newlands SD (2014) Sensory convergence in the parieto-insular vestibular cortex. J Neurophysiol 111:2445-2464. Siedentopf CM, Heubach K, Ischebeck A, Gallasch E, Fend M, Mottaghy FM, Koppelstaetter F, Haala IA, Krause BJ, Felber S, Gerstenbrand F, Golaszewski SM (2008) Variability of BOLD response evoked by foot vibrotactile stimulation: influence of vibration amplitude and stimulus waveform. Neuroimage 41:504-510. Silva AC, Rasey SK, Wu X, Wall JT (1996) Initial cortical reactions to injury of the median and radial nerves to the hands of adult primates. J Comp Neurol 366:700-716. Sluming V, Barrick T, Howard M, Cezayirli E, Mayes A, Roberts N (2002) Voxel-based morphometry reveals increased gray matter density in Broca's area in male symphony orchestra musicians. Neuroimage 17:1613-1622.

Snyder PJ, Whitaker HA (2013) Neurologic heuristics and artistic whimsy: the cerebral cartography of Wilder Penfield. J Hist Neurosci 22:277-291. Sollerman C, Ejeskar A (1995) Sollerman hand function test. A standardised method and its use in tetraplegic patients. Scand J Plast Reconstr Surg Hand Surg 29:167- 176. Song S, Abbott LF (2001) Cortical development and remapping through spike timing- dependent plasticity. Neuron 32:339-350. Stagg CJ, Bachtiar V, Johansen-Berg H (2011) The role of GABA in human motor learning. Curr Biol 21:480-484. Staines WR, Graham SJ, Black SE, McIlroy WE (2002) Task-relevant modulation of contralateral and ipsilateral primary somatosensory cortex and the role of a prefrontal-cortical sensory gating system. Neuroimage 15:190-199. Sunderland S (1968) Nerves and nerve injuries. Edinburgh and London: Churchill Livingstone. Sur M, Wall JT, Kaas JH (1981) Modular segregation of functional cell classes within the postcentral somatosensory cortex of monkeys. Science 212:1059-1061. Sur M, Wall JT, Kaas JH (1984) Modular distribution of neurons with slowly adapting and rapidly adapting responses in area 3b of somatosensory cortex in monkeys. J Neurophysiol 51:724-744. Swan J (1820) A Dissertation on the Treatment of Morbid Local Affections of Nerves: To which the Jacksonian Prize was Adjudged by the Royal College of Surgeons. London. Tal Z, Geva R, Amedi A (2017) Positive and Negative Somatotopic BOLD Responses in Contralateral Versus Ipsilateral Penfield Homunculus. Cereb Cortex 27:962- 980. Talbot WH, Darian-Smith I, Kornhuber HH, Mountcastle VB (1968) The sense of flutter- vibration: comparison of the human capacity with response patterns of mechanoreceptive afferents from the monkey hand. Journal of Neurophysiology 31:301-334. Taylor KS, Anastakis DJ, Davis KD (2009) Cutting your nerve changes your brain. Brain 132:3122-3133. Thakur PH, Fitzgerald PJ, Hsiao SS (2012) Second-order receptive fields reveal multidigit interactions in area 3b of the macaque monkey. J Neurophysiol 108:243-262. Tommerdahl M, Simons SB, Chiu JS, Favorov O, Whitsel BL (2006) Ipsilateral input modifies the primary somatosensory cortex response to contralateral skin flutter. J Neurosci 26:5970-5977. Tommerdahl M, Hester KD, Felix ER, Hollins M, Favorov OV, Quibrera PM, Whitsel BL (2005) Human vibrotactile frequency discriminative capacity after adaptation to 25 Hz or 200 Hz stimulation. Brain Res 1057:1-9. Trendelenburg F (1923) Die ersten 25 jahre der Deutschen gesellschaft für chirurgie; ein beitrag zur geschischte der Chirurgie. Berlin: J. Springer. Vaesen K (2012) The cognitive bases of human tool use. Behav Brain Sci 35:203-218. Wall JT, Xu J, Wang X (2002) Human brain plasticity: an emerging view of the multiple substrates and mechanisms that cause cortical changes and related sensory dysfunctions after injuries of sensory inputs from the body. Brain Research Brain Research Reviews 39:181-215. Wall JT, Kaas JH, Sur M, Nelson RJ, Felleman DJ, Merzenich MM (1986) Functional reorganization in somatosensory cortical areas 3b and 1 of adult monkeys after median nerve repair: possible relationships to sensory recovery in humans. J Neurosci 6:218-233. Vallbo AB (2018) Microneurography: how it started and how it works. J Neurophysiol 120:1415-1427.

Vallbo AB, Johansson RS (1984) Properties of cutaneous mechanoreceptors in the human hand related to touch sensation. Hum Neurobiol 3:3-14. Vallbo AB, Olsson KA, Westberg KG, Clark FJ (1984) Microstimulation of single tactile afferents from the human hand. Sensory attributes related to unit type and properties of receptive fields. Brain 107 ( Pt 3):727-749. Waller AV (1850) Experiments on the Section of the Glossopharyngeal and Hypoglossal Nerves of the Frog, and Observations of the Alterations Produced Thereby in the Structure of Their Primitive Fibers. Philosophical Transactions of the Royal Society 140:423-429. van Kemenade BM, Seymour K, Wacker E, Spitzer B, Blankenburg F, Sterzer P (2014) Tactile and visual motion direction processing in hMT+/V5. Neuroimage 84:420-427. Wang X, Merzenich MM, Sameshima K, Jenkins WM (1995) Remodelling of hand representation in adult cortex determined by timing of tactile stimulation. Nature 378:71-75. Watkins KE, Paus T, Lerch JP, Zijdenbos A, Collins DL, Neelin P, Taylor J, Worsley KJ, Evans AC (2001) Structural asymmetries in the human brain: a voxel-based statistical analysis of 142 MRI scans. Cerebral Cortex 11:868-877. Weber EH (1851) Die Lehre vom Tastsinne und Gemeingefühle auf Versuche gegründet: Friedrich Vieweg und Sohn. Weir Mitchell SM, G.R. Keen, W.W. (1864) Gunshot Wounds and other Injuries of Nerves. Philadelphia: J.B. Lippincott & Co. Velpeau ALM (1846) New elements of operative surgery a treatise on minor surgery. New York: J & HG Langley. Verrillo RT, Capraro AJ (1974) Effect of simultaneous auditory stimulation on vibrotactile thresholds. Perception & Psychophysics 16:597-600. Whitney D, Ellison A, Rice NJ, Arnold D, Goodale M, Walsh V, Milner D (2007) Visually guided reaching depends on motion area MT+. Cereb Cortex 17:2644-2649. Winkler AM, Kochunov P, Blangero J, Almasy L, Zilles K, Fox PT, Duggirala R, Glahn DC (2010) Cortical thickness or grey matter volume? The importance of selecting the phenotype for imaging genetics studies. NeuroImage 53:1135-1146. Wise SP (2008) Forward frontal fields: phylogeny and fundamental function. Trends Neurosci 31:599-608. Witham CL, Baker SN (2011) Modulation and transmission of peripheral inputs in monkey cuneate and external cuneate nuclei. J Neurophysiol 106:2764-2775. Witzel C, Rohde C, Brushart TM (2005) Pathway sampling by regenerating peripheral axons. J Comp Neurol 485:183-190. von Bekesy G (1947) The recruitment phenomenon and difference limen in hearing and vibration sense. Laryngoscope 57:765-777. von Bekesy G (1959) Similarities between hearing and skin sensations. Psychological Review 66:1-22. Woolsey G (1907) Surgery: Its Principles and Practice; The surgery of the nerves. Philadelphia, Pennsylvania: Saunders. Vordemvenne T, Langer M, Ochman S, Raschke M, Schult M (2007) Long-term results after primary microsurgical repair of ulnar and median nerve injuries. A comparison of common score systems. Clin Neurol Neurosurg 109:263-271. Wu CW, Kaas JH (2003) Somatosensory cortex of prosimian Galagos: physiological recording, cytoarchitecture, and corticocortical connections of anterior parietal cortex and cortex of the lateral sulcus. J Comp Neurol 457:263-292. Xie S, Gong GL, Xiao JX, Ye JT, Liu HH, Gan XL, Jiang ZT, Jiang XX (2007) Underdevelopment of optic radiation in children with amblyopia: a tractography study. Am J Ophthalmol 143:642-646.

Zatorre RJ, Fields RD, Johansen-Berg H (2012) Plasticity in gray and white: neuroimaging changes in brain structure during learning. Nature Neuroscience 15:528-536. Zhou X, Panizzutti R, de Villers-Sidani E, Madeira C, Merzenich MM (2011) Natural restoration of critical period plasticity in the juvenile and adult primary auditory cortex. J Neurosci 31:5625-5634. Ziemann U, Hallett M (2001) Hemispheric asymmetry of ipsilateral motor cortex activation during unimanual motor tasks: further evidence for motor dominance. Clin Neurophysiol 112:107-113. Zola-Morgan S (1995) Localization of brain function: the legacy of Franz Joseph Gall (1758-1828). Annu Rev Neurosci 18:359-383.