INVESTIGATING CHANGES IN TACTILE AND PROPRIOCEPTIVE PERCEPTION AND MOTOR BEHAVIOUR AS A RESULT OF A MULTIMODAL ILLUSION INDUCED BY MONOCULAR BLINDNESS WITH AN OCCLUDING CONTACT LENS

PAULA MARIA DINOTO

A THESIS SUBMITTED TO THE FACULTY OF GRADUATE STUDIES IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS

DEPARTMENT OF PSYCHOLOGY, YORK UNIVERSITY, TORONTO, ONTARIO

MARCH 2012 Library and Archives Bibliotheque et Canada Archives Canada

Published Heritage Direction du Branch Patrimoine de I'edition

395 Wellington Street 395, rue Wellington Ottawa ON K1A0N4 Ottawa ON K1A 0N4 Canada Canada Your file Votre reference ISBN: 978-0-494-90028-4

Our file Notre reference ISBN: 978-0-494-90028-4

NOTICE: AVIS: The author has granted a non­ L'auteur a accorde une licence non exclusive exclusive license allowing Library and permettant a la Bibliotheque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, preter, telecommunication or on the Internet, distribuer et vendre des theses partout dans le loan, distrbute and sell theses monde, a des fins commerciales ou autres, sur worldwide, for commercial or non­ support microforme, papier, electronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats.

The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in this et des droits moraux qui protege cette these. Ni thesis. Neither the thesis nor la these ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent etre imprimes ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author's permission.

In compliance with the Canadian Conformement a la loi canadienne sur la Privacy Act some supporting forms protection de la vie privee, quelques may have been removed from this formulaires secondaires ont ete enleves de thesis. cette these.

While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n'y aura aucun contenu removal does not represent any loss manquant. of content from the thesis. Canada ABSTRACT

We present evidence for a visuotactile illusory paresthesia induced by monocular blindness by inserting a visually occluding contact lens into subjects' dominant eye. We induced sensory paresthesias in the region of the eye ipsilateral to the opaque contact lens in surrounding facial regions in all of 17 subjects during 66 experimental sessions. The proprioceptive aspect of this illusion was demonstrated by comparing perceived eyelid position with and without the occluding lens, and changes in extent of paresthesias was measured with the introduction of tactile, proprioceptive, motor, and visual feedback. Tactile sensitivity and tactile acuity were measured by stimulation of affected paresthetic regions and control regions using two devices: a Von Frey and Two-Point

Aesthesiometer, which respectively yielded results that suggest a potential impact of this illusion on sensory perception. A novel motor coordination task was performed to evaluate the observable influence of this illusion on motor behaviour, and we offer potential neural mechanisms that underlie this multimodal perceptual illusion.

iv DEDICATION

"0 de li altri poeti onore e lume, vagliami 'I lungo studio e 'I grande amore che m'hafatto cercar lo tuo volume.

Tuse'lo mio maestro e 7 mio autore, Tu se' solo colui da cu' io tolsi Lo bello stilo che m'hafatto onore."

(Alighieri, Inferno, Canto I, LXXXII-LXXXVII).

To the loving memory of my grandfather, Salvatore Di Noto, whose lasting spirit of benevolence has guided me on my path to goodness, truth, honesty, and knowledge. This work, and all others before and after it, I dedicate in his honour.

v ACKNOWLEDGEMENTS

The completion of this work could not be achieved without the support, love, and encouragement of my family; thank you to my mother and father for the rides to and from school, breakfast, lunch, and dinner, and for the constant source of affection and pride that has sustained me in all of my studies and endeavours; to my sisters Eva and Nana, my role models and inspirations that have taught me to follow my heart, persevere through even the most difficult times, and to seek refuge in God and family; to my brother Mikael, whose passion and strength of character inspire me to choose the right path in life, even if it may be the more difficult or less-travelled one; to my cousin Gianfranco for providing an outlet for my stress, encouraging me when morale was down, and for introducing me to the rest of my life; to Matthew, for giving me the courage to conquer insurmountable challenges and for making every dream easily achievable. To these angels in my life, I sincerely thank you for your love and support - this work could not have been achieved without you.

I would also like to thank Dr. Joe DeSouza for accepting me as a student in his lab and providing me the opportunity to pursue my Master of Arts degree in a fostering environment. Thank you for your guidance, advice, and support in all matters academic, professional, and personal. The lessons learned from you during these years will form the basis of how I evolve as a researcher, educator, and independent thinker. I would also like to thank my fellow lab members for their invaluable feedback, support, encouragement, and friendship; Sheng, Laura, Diana, Alexandria, Rachel, Daniela, Matin, Scott, - best of luck in your future pursuits, I hope to see you all happy, healthy, and successful many years from now!

Thank you to Dr. Gillian Einstein for your mentorship, and for teaching me the value in saying what I mean, and meaning what I say. Special thanks to Dr. Rob Cribbie for your patience and consultation on the statistics used in this work, and to Dr. Gerald Keith. Thank you to Dr. Denise Henriques for serving as my second committee member and for your valuable insights. Thank you very much to Alexandria West, Raymond Biastoch, and Christopher Luszczek for assisting in data analyses, and last but not least to Wolfpack Industries Ltd. for holding it down.

vi TABLE OF CONTENTS

ABSTRACT IV

DEDICATION V

ACKNOWLEDGEMENTS VI

TABLE OF CONTENTS VII

LIST OF FIGURES X

LIST OF TABLES XII

LIST OF APPENDICES XIII

LIST OF ABBREVIATIONS XIV

SECTION l: INTRODUCTION l

1.1: Anatomical Summary of Facial and Ocular Sensory and Motor Pathways 2

1.1.1 - Structure, innervation, and neural pathway s of the face 2

1.1.2. - Structure, innervation, and neural pathways of the face: motor control and proprioception 12

1.1.3. ~ Structure, innervation, and neural pathways of the eyes: sensory processing 14

1.1.4 ~ Structure, innervation, and neural pathways of the eye: motor control 19

1.1.5 ~ Structure, innervation, and neural pathways of the eye: visual processing 21

Vll 1.2 - Multisensory Integration: Neural Substrates 25

1.2.1 - Mechanisms for convergence of multiple sensory inputs 26

1.2.2 - Visuotactile processing areas 29

1.3 - Multisensory Incongruity: Phantom Illusions & Paresthesia 31

1.4 - Our Hypotheses 34

SECTION 2: METHODS 35

2.1 - Participants 35

2.2 - Apparatus 36

2.3 - Procedure 39

2.3.1. - Effect of Paresthesia on Sensory Perception of Eye Area 39

2.3.2. - Effect of Paresthesia on Sensory Perception of Facial Areas 40

2.3.3. ~ Effect of Paresthesia on Estimated Eyelid Position 40

2.3.4. Effect of Paresthesia on Motor Coordination 41

2.3.5. Changes in Perceived Paresthesias with Feedback Cues 42

2.3.6. - Extent of Paresthesia as Illustrated by Subjects 43

2.3.7. ~ Effect of Paresthesia on Tactile Perception 46

2.3.7.1.; Tactile Sensitivity 46

2.3.7.2.: Tactile Acuity 50

2.3.8. - Total experimental duration 53

SECTION 3: RESULTS 53

3.1 - Introduction 53

3.2 - Subjective descriptions following insertion of occluding lens 54

viii 3-3 - Extent of Paresthesia as Illustrated by Subjects 55

3.4 - Effect of Paresthesia on Sensory Perception of Eye Area 60

3.5 - Effect of Paresthesia on Sensory Perception of Facial Areas 62

3.6 - Effect of Paresthesia on Estimated Eyelid Position 66

3.7 - Effect of Paresthesia on Motor Coordination 69

3.8 - Changes in Perceived Paresthesias with Feedback Cues 72

3.9 - Effect of Paresthesia on Tactile Perception 75

3.9.1: Tactile Sensitivity 75

3.9.1.1: Tactile Sensitivity: New Controls 78

3.9.2: Tactile Acuity 83

SECTION 4: DISCUSSION 86

SECTION 5: FUTURE DIRECTIONS 106

REFERENCES 117

ix LIST OF FIGURES

Figure 1.1. Trigeminal nerve divisions 7

Figure 1.2. Trigeminal nerve pathway 11

Figure 1.3. Facial nerve pathway and facial muscles 15

Figure 1.4. Innervation and structure of the human cornea, iris, and lens 17

Figure 1.5. The extraocular muscles and nerves of innervation 21

Figure 2.1. Procedural setup 38

Figure 2.2. Participant preparation 38

Figure 2.3. Facial schematic 45

Figure 2.4. ArcMap Shapefile superimposed on subject's illustration 46

Figure 3.1. a. Subject illustration of extent of facial paresthesia: overall average

(11=17) 57

Figure 3.1. b. Subject illustration of extent of facial paresthesia: temporal averages (n=i7) 57

Figure 3.2. Temporal analysis of facial drawings of perceived paresthesias: area of facial illustration affected by paresthesias 58

Figure 3.3. Subject illustration of extent of facial paresthesias (n=i7) 59

Figure 3.4. Effect of paresthesia on sensory perception of eye area 61

Figure 3.5. Effect of paresthesia on facial and cranial sensory perception 65

Figure 3.6. Eyelid position estimates: ipsilateral versus contralateral eye

(n=i7) 67

x Figure 3.7. Eyelid position estimates: ipsilateral occluded versus unoccluded eye (n=n) 68

Figure 3.8. Coordinated winking task: unoccluded baseline versus average occluded condition performance time 71

Figure 3.9. Coordinated winking task: time analysis 72

Figure 3.10. Changes in perceived paresthesias with feedback 74

Figure 3.11. Changes to perceived paresthesias with visual feedback 75

Figure 3.12. Average stimulus intensity ratings for ipsilateral and contralateral sides of the face 77

Figure 3.13. New controls for tactile sensitivity analysis with Von Frey hairs 80

Figure 3.14. Average stimulus intensity ratings for ipsilateral zones obtained during experimental and control sessions 81

Figure 3.15. Average stimulus intensity ratings over time 82

Figure 3.16. Two-Point discrimination thresholds: main effects analysis 85

Figure 4.1. Cranial nerve nuclei in the brainstem 89

xi LIST OF TABLES

Table 1.1. Mechanoreceptor types: structure and function 4

Table 2.1. Subject summary table 37

Table 2.2. Von Frey Aesthesiometer parameters 47

xii LIST OF APPENDICES

APPENDIX A 110

APPENDIX B 112

APPENDIX C 114

xiii LIST OF ABBREVIATIONS

CNV - Fifth cranial nerve, or trigeminal nerve

Vi - Ophthalmic branch of trigeminal nerve

V2 - Maxillary branch of trigeminal nerve

V3 - Mandibular branch of trigeminal nerve

CNVII - Seventh cranial nerve, or facial nerve

VPM - Ventralposteromedial nucleus of thalamus

DTTr - Dorsal trigeminothalamic tract

VTTr - Ventral trigeminothalamic tract

SS - Somatosensory cortex

RF - Receptive field

LPS - Levator palpebrae superioris

00 - Orbicularis oculi

MLE - Maximum likelihood estimate fMRI- Functional magnetic resonance imaging

PPC - Posterior parietal cortex

VIP - Ventral intraparietal area

IPS - Intraparietal sulcus

SC - Superior colliculus

TPD - Two-point discrimination

xiv SECTION l: INTRODUCTION

Imagine a warm summer day, one free of emails, phone calls, and thesis writing, and you're enjoying a nap in the sunshine. Along comes a fly that catches your gaze, buzzing and circling fluidly closer and closer. You might start hoping that the fly doesn't land on you, and then start thinking about all of the (less- desirable) places that fly has visited today. Your mind quickly takes you to some other tangential thought and you drift off and close your eyes. Then imagine feeling something touch your face, and your reaction to this sudden stimulus: maybe you open your eyes and sit up, or touch your face on that spot to see if it was the dirty fly or something else. It really doesn't matter what touched your face; what is important about this scenario is the sub-text- the fundamental processes that allow us to see and feel the world around us, both separately and together as a coherent scene. The integration of multiple sensory inputs is a vital process that occurs constantly without our conscious awareness, and one that allows us to interact with our dynamic surroundings. An interesting way to understand how this integration occurs is to perturb or interrupt one of these sensory inputs and examine the perceptual and potential behavioural outcomes.

My thesis project will use a newly discovered, yet unpublished, visuotactile illusion which is induced by monocular blindness with a black occluding contact lens that results in unusual facial sensations, to investigate how multimodal information is combined in the brain. Specifically, by interrupting the congruity of visual input to both eyes and evaluating the nature of perceived changes in facial sensation and oculomotor behaviour, we can better understand the afferent bottom-up and top-down executive cortical processes that facilitate integration of these two modalities in the adult human brain.

1.1: Anatomical Summary of Facial and Ocular Sensory and Motor Pathways

1.1.1 - Structure, innervation, and neural pathway s of the face: tactile processing

In order to understand how we combine afferent signals from different sensory modalities, we must understand how these systems acquire and process incoming sensory stimuli. Beginning at the micro level, tactile inputs are first received by various receptor types housed within the layers of the skin, each of which are activated depending on the type of stimuli being administered.

Mechanical, non-painful touch to the skin activates mechanoreceptors, of which several subtypes exist. Merkel's disks are located in the small ridges of the skin and as a result can detect very fine pressure and stimulation. These are part of the slow adapting (SA) class of mechanoreceptors, whose selective and relatively slow firing rate provides very precise detail about the nature of tactile stimulation being applied to a small receptive field, which is typical of Type I mechanoreceptors. Ruffinian corpuscles detect stretching of the skin and process movement over a larger area and are classified as Type II SA receptors. Meissner's corpuscles lie in the most superficial layers of the skin and are part of the fast

2 adapting (FA) class of mechanoreceptors whose axons fire quickly to low thresholds of tactile stimulation in small receptive fields (Type I). Pacinian corpuscles are similar in that they are fast adapting, but they are located deeper in the skin and relay pressure and vibration information over larger receptive fields

(Type II, Table 1.1, Hensel, 1974; Johnson, 2002; Kennedy & Inglis, 2002; Purves et al., 2008). There is also a class of receptors called nociceptors, which are activated when painful stimulation is administered to the skin. They are divided into two classes: the fast-conducting A-delta fibres that signal acute pain, and the unmyelinated C fibres that conduct slow, lasting chronic pain (Purves et al.,

2008). Although this latter group of neurons has a very specialized pathway to the brain, mechanoreceptors send their axons to the dorsal root ganglia of the spinal cord, where they travel to the medulla and thalamus via the dorsal column, and are projected to the contralateral primary and secondary somatosensory cortices in the postcentral gyrus and parietal operculum, respectively.

3 Mechano Location Conduction Receptive Response Type Sensory -receptor velocity field size to function type sustained stimuli Merkel Tip of 40-65m/s 9mm2 Sustained SAI Form and epidermal firing (slow texture sweat adaptation) perception ridges Meissner Dermal 35-70m/s 22mm2 None (fast FAI Motion papillae adaptation) detection, (close to grip skin control surface) Pacinian Dermis 35-70m/s Entire None (fast FAN Vibration and finger or adaptation) deeper hand tissue Ruffinian Dermis 35-70m/s 60mm2 Sustained SAII Skin firing (slow stretch adaptation)

Table 1.1. Mechanoreceptor types: structure and function. (Hensel, 1974; Johnson, 2002; Kennedy & Inglis, 2002; Purves et al., 2008).

This sensory pathway for tactile information from the face is different from touch processing from the rest of the body. Mechanoreceptors, thermoreceptors, and nociceptors in the facial skin are innervated by the trigeminal nerve (cranial nerve V, CNV), which is further subdivided into three branches: ophthalmic (Vi), maxillary (V2), and mandibular (V3). Most superiorly, the ophthalmic division exits the skull at the superior orbital fissure and provides sensory information from the conjunctiva, cornea, orbit, dorsal aspect of nose, upper eyelid, forehead, and frontal sinuses, as well as proprioceptive information from extraocular and facial muscles of the eyelid and forehead. The maxillary branch exits the skull through the foramen rotundum and relays sensory inputs from the maxilla

4 (cranial bone spanning from the roof of the mouth to the ocular orbit) and overlying skin, including the upper lip, side of the nose, medial cheek, nasal cavity, palate, and nasopharynx. The most inferior branch of the trigeminal nerve, the mandibular branch, exits through the foramen ovale and relays sensation and also provides motor nerves to the buccal region, including the mucous membrane of the mouth, gums, the side of the head, scalp, and lower jaw including teeth, gums, anterior 2/3 of tongue, chin, and lower lip (Figure 1.1, Rubin & Safideh,

2007; Schuenke, Schulte, Schumacher & Rude, 2007).

Each of these three main divisions branch off into smaller nerves that innervate the facial muscles of expression. Branches of Vi include the lacrimal nerve (which innervates the gland of the same name), frontal nerve that divides into the supratrochlear and supraorbital nerves that innervates the upper eyelid, brow, forehead, and scalp. The nasociliary nerve gives rise to posterior and anterior ethmoidal nerves that supply the eyeball, and long and short ciliary nerves that innervate the pupil and mediate corneal reflexes. The V2 division exits the skull and further divides into three important nerves: zygomatic, ganglionic branches to the pterygopalatine ganglion, and the infraorbital nerve. The zygomatic nerve divides into the zygomaticofacial and zygomaticotemporal branches, which provide sensory innervation of the lateral cheek, temple, and lateral forehead. Ganglionic branches to the pterygopalatine sensory ganglion convey parasympathetic, postsynaptic information to the lacrimal nerve by a communicating branch that receives presynaptic inputs from the facial nerve

5 (cranial nerve VII, CNVII), which is largely responsible for conveying motor signals to facial muscles. Finally the V2 infraorbital nerve divides into fine branches that supply skin between lower eyelid and upper lip, including the medial cheek and side of nose, and forms the superior dental plexus that provides sensory innervation to the maxillary (or top row) of teeth. The V3 mandibular division of the trigeminal nerve is unique such that it receives sensory information and provides motor commands to various smaller branches. Sensory afferents are received by the auriculotemporal nerve from the side of the head, external auditory meatus and canal, and the tympanic membrane. The lingual nerve supplies the anterior 2/3 of the tongue and travels with gustatory fibres from the chorda tympani (part of the facial nerve), and from the inferior alveolar nerve that innervates the teeth, mental nerve (mucous membrane and skin of lower lip, chin, and mandible) and the buccal nerve, which provides sensory inputs from mucous membrane and buccinator muscle of the cheek. V3 supplies the following motor nerves: masseteric, deep temporal, and pterygoid to the muscles of the same name, the motor nerve of the tensor tympani tensor veli palatini muscles of the inner ear (Figure 1.1, Drake, Vogl & Mitchell, 2005;

Schuenke et al., 2007).

6 Mot Of fibers i Sentory fiber* Ophthalmic nerve Trigeminal (V) nerve and trigeminal (semilunar) ganglion Proprioceptive fiber* Tentorial (meningeal) branch • • • • Parasympathetic fiber* Nasociliary n, Motor nucleus of trigeminal n. — - • Sympathetic fiber* Sensory root of ciliary ganglion Mesencephalic nucleu* of Lacrimal n. trigeminal n, (proprioception) Frontal n. Principal sensory nucleus of Ciliary ganglion trigeminal n. (discriminatory sensation) Posterior ethmoidal n. Spinal tract and spinal Long ciliary n. nucleus of trigeminal n. Short ciliary nn. (pain and temperature! Supratrochlear n. Supraorbital n. (medial and lateral branches)' Anterior ethmoidal n. Infra trochlear n. External natal and internal nasal medial and lateral rami) branches oI anterior ethmoidal n.

**2 Facial VII i n Chorda tympani

Figure 1.1. Trigeminal nerve divisions. From Rubin & Safideh (2007).

7 Once sensory information is received from all of these trigeminal nerve branches, they are relayed to the trigeminal sensory ganglion located in the depression of the middle cranial fossa, which forms the larger trigeminal nerve that travels to the midbrain and forms the principal pontine nucleus rostrally and the mesencephalic trigeminal nucleus caudally that extends inferiorly to the medulla and substantia gelatinosa of the cervical spinal cord. A small number of nerves carry information about ipsilateral facial sensation to the ventralposteromedial (VPM) nucleus of the thalamus via the dorsal trigeminothalamic tract (DTTr), and onward to the somatosensory cortex (SS, postcentral gyrus, Brodmann areas l, 2, 3a and 3b). The great majority of nerves, however, entering the principal pontine nucleus decussate to the contralateral medial lemniscus and travel via the ventral trigeminothalamic tract (VTTr) to the

VPM, where sensory inputs are relayed to the somatosensory cortex (Figure 1.2,

Haines, 2004; Rubin et al., 2007). Thus, somatosensory processing areas for the face receive both ipsilateral and contralateral facial information and represent the face upside down on the cortex, with studies in both humans and non-human primates showing that the lips are represented medially to the localizations of the eyelid, brow, and forehead on more lateral portions of both primary and secondary somatosensory cortices (Kaas, Nelson & Sur, 1979; Nguyen, Tran, Inui,

Hoshiyama & Kakigi, 2004,2005; Servos, Engel, Gati & Menon, 1999).

There are several cell types within the trigeminal nerve pathway that are responsible for different aspects of perception and action; pseudo-unipolar cells,

8 whose cell bodies reside in the mesencephalic nucleus, are responsible for the

relay of sensory information from facial skin to ipsilateral and contralateral

sensory processing areas, and parasympathetic nerve fibres send motor

commands from the trigeminal motor nucleus (which is directly adjacent to the

principal sensory nucleus in the midbrain) to V3, which controls the muscles of

mastication. Efferent fibres from CNVII travel along with several branches and

sub-branches of the trigeminal nerve to various destinations; the lacrimal nerve

(Vi) and zygomatic nerve (V2) provide efferent signals to the lacrimal gland to

mediate the secretion of tears, the auriculotemporal nerve (V3) sends

glossopharyngeal fibres (CNVII) to the parotid salivary gland, and the lingual

nerve (V3) travels with the chorda tympani to the submandibular and sublingual

salivary glands (Schuenke et al., 2007).

Facial structures have differential representation in the SS depending on

how 'sensitive' these areas are. 'Sensitivity' itself is a term that has been defined in several ways with respect to tactile perception, and is often confused with acuity- the ability to discriminate two separate points. For the purpose of our study, we

will distinguish these two points. Sensitivity refers to the perceived strength of a given stimulus, which is based on the firing pattern of mechanoreceptors (the subtypes of which we mentioned previously). Thus, depending on the type and strength of tactile stimulus administered, and the type and firing pattern of

mechanoreceptor that is resultantly activated, the tactile stimulus will be

perceived (Table 1.1). Conversely, acuity is not based on the firing pattern of any

9 given mechanoreceptors, but rather on the size of the stimulus and how many mechanoreceptors are activated, each of which have a predefined receptive field

(RF) that occupies the skin surface. The smaller the RF size, the higher the acuity is for that area, as activation of adjacent RFs will allow the subject to perceive separate sites of tactile stimulation. Activation of one large receptive field, on the other hand, will be perceived as only one point. Receptive field size is also directly related to how densely a given area of skin surface is innervated by mechanoreceptors, and how many axons are sent from this area to somatosensory processing areas. The most highly innervated areas of the human body are the fingers, hands, tongue, and lips, which contain many mechanoreceptors of all subtypes (SAI & II, and FAI & II), have the smallest receptive fields, and consequently occupy the largest area in the somatosensory cortex (Purves et al.,

2008). The facial representation on the somatosensory cortex has been shown as inverted, and areas other than the lips have shown clustered activation on SS medial to the representation of the lips and lateral to the representation of the thumb (Nguyen et al., 2004; 2005; Servos et al., 1999). However, several microelectrode studies in adult humans have mapped the receptive fields of the inferior alveolar nerve including the lower lip, chin, and periodontal regions

(Trulsson & Essick, 2010), infraorbital nerve spanning the upper lip and cheek

(Johansson, Trulsson, Olsson, & Westberg, 1988; Nordin & Hagbarth, 1989), and the supraorbital nerve innervating the forehead (Nordin, 1990).

10 Cerebral cortex: postcentral gyrus

internal capsule Centromedian nucleus (intralaminar) Ventral posteromedial (VPM) nucleus of thalamus

Midbrain

'Dorsal trigeminal lemniscus .Trigeminal mesencephalic nucleus . «fcrval trigeminothalamic tract) /Trigeminal motor nucleus • Ventral trigeminal lemniscus ^Principal sensory trigeminal nucleus Aentral trigeminothalamic tract) ,Touch, pressure ^online reticular formation ,P»in, temperature 'Proprioception: from muscle spindles - , Trigeminal (semilunar) ganglion Pons . fuhalmicn. .'Maxillary n. -Sensory root and (tor root of mandibular n. Medullary reticular formation: Lateral reticular formation Medial reticular formation Ventral trigeminal lemniscus Spinal (descending) trigeminal tract Spinal (descending) trigeminal nucleus Dorsolateral fasciculus (of Lissauer) Cervical spinal cord ). Perkins Substantia gelatinosa (lamina II) MS, MFA

Figure 1.2. Trigeminal nerve pathway. From Rubin & Safideh (2007).

11 However, no studies have been found that systematically map receptive fields on the whole face or the upper or lower eyelids. Despite the fact that sensory innervation of facial areas is well known, there does not yet exist information on a) the spatial acuity of the whole human face and b) the corresponding extent of representation of individual facial areas in somatosensory cortices, and c) how these receptive field sizes and cortical regions are influenced by short-term changes in physiology (i.e., amputation or injury), environment (i.e., temperature, light or dark conditions), and multisensory interactions.

1.1.2 - Structure, innervation, and neural pathways of the face: motor control and proprioception

In addition to mediating tactile perception, the trigeminal nerve is responsible for facilitating motor control of the lower part of the face. Motor innervation of the facial muscles that control expression, however, is mediated by branchiogenic fibres of the facial nerve (CNVII). An understanding of how these cranial nerves intertwine and run together demonstrate just how intricate the facial sensorimotor system is, especially when considering the variety of functions these nerves serve; from emotional expression, to protection of sensory organs from noxious stimuli, the interplay of facial and trigeminal nerves is germane to the implications of our present study. The facial nerve is a collection of axons from several nuclei in the pons; upper motor neurons originating in the precentral gyrus project to the facial nucleus, which sends its axons looping

12 around the nucleus of the abducent nerve before exiting the lower pons. Axons from the superior salivatory nucleus contain parasympathetic neurons that innervate the lacrimal, nasal, submandibular, sublingual, and salivary glands, as well as the hard and soft palate of the mouth. The nucleus of the solitary tract consists of viscerosensory gustatory fibres from the tongue and the chorda tympani. Upon exiting the skull at the stylomastoid foramen efferents from the facial nucleus branch off into several divisions that innervate separate portions of the face. The first branch of CNVTI is the stapedial nerve that innervates the muscle of the same name within the temporal bone. The posterior auricular nerve innervates the posterior belly of the occipitofrontalis muscle (the back of the head near the base of the occipital bone), and also sends afferent sensory information about the external ear to the geniculate ganglion. Successive branches include the temporal, zygomatic, buccal and marginal mandibular. Smaller branches include diagastric and stylohyoid to muscles of the same name. The lowest branch of the facial nerve is the cervical branch, which joins with the anterior branch of C3 spinal nerve and innervates the neck (Figure 1.3).

In addition to mediating movement, expressions, and sensory processing, facial and ocular muscles contain muscle spindles and Golgi tendon organs responsible for relaying proprioceptive information. Proprioception is the sense of position within space that is often used to describe the limbs and body as a whole, with involvement of the vestibular system. With respect to the head and eyes specifically, however, proprioception is mediated by neurons whose axons

13 course with the trigeminal nerve but whose cell bodies reside in the

mesenscephalic trigeminal nucleus within the central nervous system, and not in

the trigeminal ganglion. These nerves innervate the muscles of the eyes and face,

especially the jaw, and are responsible for sending position information to the

thalamus and SS, as well as mediating reflexive and active control of muscles

(Cooper & Daniel,1949; Cooper, Daniel & Whitteridge, 1955; Purves et al., 2008).

1.1.3. - Structure, innervation, and neural pathways of the eyes: sensory

processing

Now that we have an understanding of the structural organization and

innervation of the face, we can examine the visual system. Briefly, the eyeball

itself is comprised of three concentric layers (outer to inner): the sclera, choroid,

and retina. The anterior, outer portion of the eye is the cornea, which bulges forward and connects posteriorly to the sclera. Tendons of the extraocular

muscles connect to this tough layer of connective tissue. The posterior border of the sclera, the lamina cribrosa, is pierced by axons of the optic nerve, and beneath it is the vascular coat of the eye that houses the iris, ciliary body (anterior) and choroid, which are distributed over the entire eyeball. The iris shields the eye from excessive light and covers the lens. Its root is continuous with the ciliary body, which contains the ciliary muscle for visual accommodation (refracts the lens to allow for visual accommodation).

14 —— Motor fibers Carotid plexus (on internal carotid a.) —— Semory fiber* Geniculate ganglion Parasympathetic fibers Greater petrosal a Facial (VII) nerve Sympathetic fibers Deep petrosal n, Motor root of facial n. Lesser petrosal n. Internal acoustic meatus Nerve of pterygoid canal Nervus Intermedius (of facial n.) Motor nucleus of facial n. Otic ganglion Superior salivatory nucleus Pterygopalatine ganglion Nucleus of the solitary tract Facial muscles Occipitofrontal!* m. (frontal belly) Orbicularis oculi m. Corrti gator supercllll m. Zygomatlcus major m. Zygomatics minor m.

Levator labii superior!* m. Levator labii superioris alaeque nasi m. Levator anguli oris m. Occipito­ Nasalis m. frontal is m. (occipital belly) Occipital branch Orbicularis oris m. of posterior auricular n.

Branches to auricular musclcs Mentalis m, Posterior auricular n. Nerve to stapedius m. Stylomastoid foramen Risorius m. Buccinator m. Sublingual gland Piatysma Submandibular gland Glossopharyngeal (IX) n. Submandibular ganglion Posterior belly of digastric m. Lingual n. Stylohyoid m. Chorda tympani W Figure 1.3. Facial nerve pathway and facial muscles. From Rubin & Safideh (2007).

15 The epithelium of the ciliary body produces the aqueous humour, which makes up the interior substance of the eyeball. Ciliary body is the most highly vascularized region of the body and serves to regulate the temperature of the eye and supply blood to outer layers of the retina. The retina itself is the inner layer of the eyeball, which includes an inner layer of photosensitive cells (the sensory retina) and an outer layer of pigment epithelium. The fovea centralis is a depressed area in the central retina that is approximately 4mm temporal to the optic disk (blind spot where the optic nerve leaves the posterior part of the eyeball), and light is normally focussed onto this site, which has the highest visual acuity (Purves et al., 2008; Rubin & Safideh, 2007; Schuenke et al., 2007).

Sensory processing of the eyeball is mediated by the ciliary ganglion, where preganglionic fibres from the oculomotor nerve and the nasociliary branch of the ophthalmic (CNVi) nerve pass through and innervate the eyeball through short ciliary nerves (Figure 1.4). In the apex of the cornea, ciliary nerves fibres have been shown to run in a superior-inferior direction, while they are oriented in a diagonal nasal-temporal direction in the periphery. In the human corneal epithelium it has been estimated that there are 7000 nociceptors per mm2, as these receptors typically relay the detection of painful or noxious stimuli and are incredibly sensitive (Belmonte, Garcia-Hirschfeld & Gallar, 1997; Drake et al.,

2005; Mxiller, Marfiirt, Kruse & Tervo, 2003).

16 Pupillary Anterior sphincter chamber Pupillary dilator

Chamber / Canal o( angle Schlemm

Ciliary musde

<«: Vfa'A-; •

Ciliary body Zonular fibers klera

Posterior chamber

Figure 1.4. Innervation and structure of the human cornea, iris, and lens. From Schuenke et al„ 2007.

About 20% of these nociceptors are called mechano-nociceptors and fire nerve impulses in response to brief or sustained stimulation of the corneal surface. They are known as 'phasic sensory receptors' that signal the presence and degree, but

not duration, of the stimulus. The threshold amount of force required to activate these receptors is about 0-6 mN, well below the force required to activate

mechano-nociceptor fibres of the skin but still potentially damaging to the corneal epithelium (Bessou and Perl, 1969). The remainder of corneal sensory fibres are known as polymodal nociceptors because they are equally activated by

mechanical energy as well as heat, exogenous chemical irritants, and endogenous chemical compounds such as neurotrophins released by damaged corneal tissue.

These nociceptors respond to stimuli with a continuous irregular discharge of

nerve impulses that persists for the duration of the stimulus and therefore not

17 only signals the presence and duration of a noxious stimulus, but also encodes its

intensity. Polymodal nociceptors also have an activation threshold slightly lower

than mechano-nociceptors, and this selectivity and high sensitivity of corneal

sensory nerves is responsible for the sharp, acute pain that results from

mechanical contact with the corneal surface by foreign agents like dust or an

eyelash (Belmonte et al., 1997,2004).

What happens when the cornea is exposed to repetitive stimulation, like

wearing a contact lens? Researchers have investigated this question and found

that continuous use of contact lenses reduces corneal sensitivity (Millodot, 1972,

1975, 1976, 1977)- Some of the mechanisms proposed for this desensitization include chronic hypoxia of the corneal epithelium associated with regular use of contact lenses, a reduction in corneal pH as a result of the depressed nerve function that results from hypoxia, or sensory adaptation to regular mechanical stimulation (Brennan & Bruce, 1991; Oliveira-Soto & Efron, 2003; Poise, 1978).

Sensitization occurs rapidly in the short-term, with reduced corneal touch thresholds (CTTs) in only 6 to 8 hours in subjects wearing hard contact lenses despite being adapted to them and regularly using them for at least three months

(Millodot, 1972; 1976). Millodot also conducted a study with soft contact lenses

(1975)) which were just introduced in his time and are the more contemporary form of contact lenses used today, but did not examine the effects of their

prolonged use on corneal sensitivity concluding only that soft contact lenses improve visual acuity better than hard contact lenses due to their reduction in

18 spherical aberration, which results from the less-malleable nature of hard contact lenses. That soft contact lenses fit the curvature of the eyeball better than hard lenses does not change the fact that they come into direct contact with ciliary nerves and likely are consistent with the literature that suggests short-term recovery of corneal sensitivity following removal of the contact lens, as the subjects in all of these studies were regular users of contact lenses. The compendium of research conducted by Michel Millodot and those succeeding him have shown that prolonged use of contact lenses does reduce corneal sensitivity overall, but also attest to the adaptability and plasticity of corneal nerve recovery that is observed immediately following removal and cessation of using contact lenses in the short term. This suggests a heightened sensitivity to corneal stimulation in novel contact lens wearers, but the ability of the corneal epithelium to adapt quickly to the sustained stimulation of ciliary nerves following insertion of a contact lens.

1.1.4 ~ Structure, innervation, and neural pathways of the eye: motor control

Each eye is connected by six extraocular muscles, and their coordinated interaction are essential in directing both eyes toward a single visual target. The lateral rectus controls abduction (lateral outward movement) and is innervated by the abducent nerve (CNVI), the medial rectus controls adduction (lateral inward movement) and is innervated by the oculomotor nerve (CNIII), which also

19 innervates the superior rectus (elevation, adduction, and medial rotation),

inferior rectus (depression, adduction, and medial rotation), and inferior oblique

(elevation, abduction, lateral rotation), as well as the ciliary muscles that control

refraction of the optic lens (Figure 1.4). The superior oblique muscle is

responsible for depression, abduction, and medial rotation, and is innervated by

the trochlear nerve (CNIV, Figure 1.5). These nerves convey efferent motor

signals to the muscles they innervate, and there has been some evidence to

support the presence of muscle spindles and palisade endings, both of which are

mechanoreceptors involved in proprioception. However, the mechanism by which

proprioceptive signals are processed from extraocular muscles, and how it

influences ocular motor control, is poorly understood (Steinbach, 1987; Weir,

2006). Studies in primates have identified a proprioceptive representation of the

contralateral eye in the depth of the central sulcus of primary somatosensory

cortex (Si) (Wang, Zhang, Cohen & Goldberg, 2007). This collection of neurons

was found to be consistent with the localization of proprioceptive information from the primate hand and forearm and near the localization of the eyebrow, suggesting that the diffuse representation of proprioceptive information from

muscle spindles exists in primate area 3a of the somatosensory cortex (SS) and

potentially sends information about the position of the eye to motor planning

regions of the cortex to guide subsequent eye movement (Wang et al., 2007; Xu,

Wang, Peck & Goldberg, 2011).

20 Medial rectus Superior Levator palpebral oblique superior)* Common tendinous ring Superior rectus Oculomotor nerve Lateral Trochlear rectus nerve

Internal Inferior carotid artery oblique

Inferior Abducent nerve rectus

Superior orbital fissure fissure ]

Maxii'd' Sphenoid bone smuv

Figure 1.5. The extraocular muscles and nerves of innervation. Adapted from Schuenke et al. (2007).

1.1.5 ~ Structure, innervation, and neural pathways of the eye: visual processing

The image that is projected onto the retina travels in two pathways: geniculate and nongeniculate. In the former, incident light activates a chain of different cell types in the retina beginning with light-sensitive photoreceptors, bipolar cells, and ganglion cells that leave the eye as the optic nerve. The two optic nerves meet at the diencephalon and form the optic chiasm, where the lateral and medial roots are formed. The geniculate pathway continues as approximately

90% of optic nerve axons from the lateral root terminate in the lateral geniculate nucleus and are further projected to the striate cortex in the primary visual processing areas (Vi). This pathway is responsible for processing conscious visual perception, whereas the remaining 10% of axons from the medial root of the optic nerve form the nongeniculate pathway. These nerves transmit information that is used to mediate non-conscious vision-related processes, and terminate in several

21 areas including the pulvinar of the thalamus, which in turn projects to the

superior colliculus where nongeniculate axons also terminate to process kinetic

information about head and eye position to follow moving visual targets.

Projections to the pretectal area serve as afferent signals that mediate pupillary

constriction and dilation via communicating branches to the oculomotor nerve.

The suprachiasmatic nucleus of the hypothalamus receives light information to

regulate circadian rhythms and accompanying visceral functions, vestibular and

thalamic nuclei in the tegmentum control smooth pursuit eye movements, and

the parvocellular nucleus of the reticular formation mediates autonomic

regulation and orofacial motor control, as shown in rodent studies (Purves et al.,

2008; Schuenke et al., 2007; Ter Horst, Copray, Liem & Van Willigen, 1991).

These pathways demonstrate the intricacy of the visual system and the sophistication with which we are able to consciously and unconsciously utilize visual information to interact with and adequately react to changing

environments.

Given the proper functioning of these extraocular muscles, the visual

processing areas of the brain will receive two retinal images, thereby constituting

binocular vision. If these muscles or any other structures of the visual system are

impaired, several visual deficiencies can arise. For the purposes of our

investigation, we will refer mainly to the occurrence of monocular vision. It is a

common belief that individuals impaired in one modality are enhanced in their

remaining, intact modalities. A more accurate description of monocularly blind

22 individuals, however, would be to say that their intact modalities have less competition for neural resources in uni- and multimodal processing areas. If inputs from both eyes compete for the area of the brain devoted to processing vision, the loss of one of those inputs leaves less 'traffic' for the remaining eye.

This is supported by evidence from studies on monocular viewers, whose performance in letter acuity, contrast sensitivity, and spatial ability tasks is either enhanced or intact when compared to binocular viewers (Gonzalez, Steeves,

Kraft, Gallie & Steinbach, 2002; Reed, Steeves, Kraft, Gallie & Steinbach, 1996;

Reed, Steeves & Steinbach, 1997). However, monocular viewers lack the ability to detect depth and are impaired in motion processing and oculomotor behaviours, especially those that have lost binocular vision later in life (i.e., following development of a binocular visual system) and have yet to develop strategies to accommodate for the sudden change in visual input (Steeves, Gonzalez, Gallie &

Steinbach, 2002; Steeves, Gonzalez & Steinbach, 2008). The study of monocular viewers has a substantial focus being on children born with one functioning eye or functionally lost in postnatal stages, but evidence in adult humans and cats have shown short-term restructuring of primary and association visual processing areas by the remaining functional eye (Gilbert & Wiesel, 1992; Hubel & Wiesel,

1962), or recruitment of multisensory processing areas by intact senses (Bavelier et al., 2001, 2006). However, no evidence exists for what phenomenological or perceptual effects are experienced immediately following enucleation in adulthood.

23 Several reflexes exist in this sensorimotor network that protect the visual system from harmful mechanical and visual stimuli. The corneal reflex, or blink reflex, results in closure of the palpebral fissure (opening of eye socket), which is surrounded by the orbicularis oculi muscle. This protects the eye from foreign matter, and also protects the cornea and conjunctiva, the latter of which is a thin vascularized mucous membrane that allows painless movement of the outermost layer of the eyeball (ocular conjunctiva) against the innermost layer of the eyelid

(palpebral conjunctiva). Normal blinking occurs approximately once every 2.5 seconds and keeps the eyes hydrated by distributing lacrimal fluid and other secretions from the lacrimal and Meibomian glands. It is facilitated by the antagonistic interactions of the levator palpebrae superioris (LPS) and orbicularis oculi (00) muscles, with the motor neurons of the former ceasing firing immediately prior to the onset of a blink and allowing 00 motor neurons to fire and lowering the upper eyelid. 00 neurons then cease firing and LPS motor neurons resume firing, raising the upper eyelid (Bour, Aramideh, & Ongerboer de

Visser, 2000). The pupillary light reflex is mediated by the ciliary muscle

(innervated by the oculomotor nerve) and afferent signals from the nongeniculate pathway of the optic nerve mentioned previously. It allows the eye to adapt to varying levels of brightness, or luminance, very quickly so as to protect photoreceptors from sudden potentially harmful increases in luminance. The afferent signals for this reflex are received from retinal photoreceptors (rods and cones), bipolar cells, and ganglion cells, the latter of whose axons form the optic

24 nerve that transmits luminance information to the pretectal area in the midbrain.

Once synapsing in this nongeniculate part of the visual pathway, an efferent signal is sent to the Edinger-Westphal accessory nuclei of the oculomotor nerve that projects to the short ciliary nerves that connect to the pupillary sphincter and allow it to contract or dilate. Because both eyes are innervated by bilateral nerves and nuclei, the light reflex is simultaneous and consensual (Drake et al., 2005;

Schuenke et al., 2007). These intricate processes are essential components to our study and from this discussion it is clear that the visuo-motor and visuo-sensory systems are incredibly sensitive and operate very rapidly and unconsciously. We aim to examine how interruption of one stream in this system can cause perturbations in the perception of other modalities within it.

1.2 - Multisensory Integration: Neural Substrates

That our nervous system is capable of processing afferent signals from six individual modalities, all equipped with their own specialized receptors, pathways, and cortical processing areas, is not enough to formulate a coherent percept of our surround. We will now try and understand how and where in the brain these multisensory signals converge and inform each other and our resulting perception.

25 1.2.1 - Mechanisms for convergence of multiple sensory inputs

Before we identify cortical regions involved in multisensory integration, we will briefly outline the computational theories of how these converging neural signals are processed. Inputs from multiple sensory modalities are integrated in a statistically optimal fashion, as demonstrated through computational models that estimate how our neural networks actually behave. One such model uses the maximum-likelihood estimate (MLE) to determine the most optimal means of estimating a perceptual outcome of combined sensory inputs by weighing the outcome estimate by the respective normalized variances of the unimodal inputs.

This model has accurately predicted that subjects using MLE have lower variances of bimodal estimates of object height (both visual and haptic), which result in a lower discrimination threshold in the bimodal condition than in the visual or haptic conditions alone (Ernst & Banks, 2002). Another premise assumes a linear combination of multimodal inputs depending on the time course and location of stimulus onset. When comparing the neural responses of unimodal and bimodal stimuli, there was an over-additive response to visuotactile bimodal stimuli (i.e., the neural activity of bimodal responses was greater than that of the summed unimodal responses) at the 75-iooms interval in the extraoccipital areas, a sub-additive bimodal response (i.e., bimodal response is less than the summed unimodal responses) in the right parietal lobe and over- additive bimodal response in the left anterior frontal cortex during the 100125ms interval, sub-additive bimodal response in midline anterior EEG localizations

26 during the 225-275111S interval, and overadditive bimodal responses in all electrode areas during the 275-300ms and 450-600ms windows. This pattern of activation is consistent with the stepwise activation of different brain areas recruited for visuotactile perception (Schiirmann, Kolev, Menzel & Yordanova,

2002).

The most common model of multisensory integration, however, is Bayes'

Rule, a statistical probability concept that has applications in a wide variety of fields, including artificial intelligence and human behavioural modelling. The use of Bayesian probability with respect to multisensory integration has been most commonly applied to the concept of enhancement, where the coupled presentation of a weak stimulus of one modality with stimulus in another modality will strengthen, or enhance, the effectiveness of the weak stimulus

(Anastasio, Patton & Belkacem-Boussaid, 2000; Patton, Belkacem-Boussaid &

Anastasio, 2002). Similarly, the presence of a second concurrent unimodal cue can supress the detection of a second unimodal stimulus via inhibiting converging signals upon multisensory processing areas (Meredith, 2002). Bayes' rule has been used to calculate the probability that multimodal processing areas of the brain will detect the presence of a target (i.e., object of interest that the larger neural network must signal an orienting response for) given the sensory inputs that converge from multiple sources. These calculations have most often been made based on the activity of neurons in the deep layers of the superior colliculus, an area where visual, auditory, and somatosensory inputs are received and

27 forwarded to motor areas to facilitate orienting responses (cat: Alvarado,

Stanford, Rowland, Vaughan & Stein, 2009; Harting, Feig & Van Lieshout, 1997;

Wallace & Stein, 1997, primate: Groh & Sparks, 1996,).

Studying the multimodal contributions of the superior colliculus in

humans is significantly more difficult given its subcortical nature, but its

retinotopy has been investigated with fMRI (Schneider & Kastner, 2005).

However, the behavioural application of these computational models have been

demonstrated in humans by a group of Italian researchers that examine the role

of peripersonal spatial representation with respect to detecting unimodal or

concurrently presented multimodal stimuli. Extinction is the pathological

inability following brain damage to detect a contralesional stimulus in a given

modality when presented with simultaneous bimodal stimuli. Elisabetta Ladavas

and colleagues (1998a) demonstrated cross-modal visuotactile inhibition and

interference in tactile stimulus detection in the ipsilesional hand in patients with

tactile extinction when presented with a visual stimulus presented near the

ipsilesional hand. Presentation of a visual stimulus near the contralesional hand

when both hands were simultaneously touched resulted in increased detection of

the tactile stimulus in the contralesional hand. This supports the notion that

bimodal neurons enhance the detection of a single damaged modality when simultaneously presented with the intact modality, referring once again to the

process of enhancement mentioned previously. Further support of bimodal

enhancement was shown by the lack of modulatory effects on tactile extinction

28 when the visual stimuli were presented far from the patients' contralesional hand, supporting the hand-centered nature of the visual receptive fields in primates

(Graziano, Yap & Gross, 1994; Ladavas et al., 1998a). Ladavas has also demonstrated bimodal modulation of extinction without patient's conscious awareness of an incoming tactile stimulus, demonstrating that visuotactile integration can occur as a bottom-up process (i.e., based on the presented stimulus only) without conscious top-down modulation of sensory processing

(Ladavas & Fame, 2004). Based on these results, we can see that vision overrides the other senses when competing for limited cortical resources in multimodal processing areas, a relationship which will play an important role in informing our present research questions.

1.2.2 - Visuotactile processing areas

Perhaps the greatest contributor to understanding multisensory regions of the brain in primates is Giacomo Rizzolatti. Through his work in direct stimulation of the periarcuate (premotor) cortex, or Brodmann areas 6 and 8, he provided a launching point for subsequent researchers from which to investigate how visuotactile interactions are processed and functionally organized in the brain (Rizzolatti, Scandolara, Matelli & Gentilucci, 1981a,b). This work has been continued by Michael Graziano, who has focussed his research on the overlap of visual and tactile receptive fields in primates for decades and has amassed an impressive volume of literature suggesting the ventral premotor cortex and the

29 putamen in macaques respond to both cutaneous touch and nearby visual stimuli around the face, arm, and hand. Additionally, the receptive fields around the arm move with the arm, but not when eye position is changed suggesting that visual receptive fields are arm or body-centered, and the adjustment of visual receptive fields with movement of the body helps inform subsequent reaching and motor behaviours mediated by the premotor cortex (Graziano & Gross, 1993; Graziano et al., 1994).

Both primate and human studies on multisensory convergence of the visual, tactile, and proprioceptive modalities have implicated the posterior parietal cortex (PPC) as the area responsible for the coding of peripersonal space, especially the ventral intraparietal area (VIP). In macaques Avillac et al. (2005,

2007), Colby et al. (1993) and Duhamel and colleagues (1998) have demonstrated an alignment of visual and tactile receptive fields, both in their spatial location and response properties such as direction of moving stimuli, immediately surrounding the face. By manipulating the presentation of visual, tactile, and proprioceptive information in humans with several experimental conditions,

Makin, Holmes, and Zohary (2007) found that visual information overrode proprioceptive signals in posterior VIP representation of the hand in peripersonal space, and proprioceptive- and tactile-specific activations were characteristic in anterior VIP. Visually dominant activations of peripersonal space were also demonstrated in the lateral occipital complex in the same study, implicating the

30 VIP, and specifically in the anterior portions, as a crucial region for multisensory

coding (Makin et al., 2007).

1.3 - Multisensory Incongruity: Phantom Illusions & Paresthesias

We conclude our discussion of multisensory integration with the description of what happens when the inputs received from separate modalities suddenly changes and becomes incongruent. These changes can arise from experimental manipulations, or physiological changes to the body that the brain has not yet had time to reorganize and redistribute 'unused' cortical receiving areas. In the case of the latter, phantom phenomena have been classified for a variety of body parts, most notably in the limbs following amputation or trauma.

The underlying mechanism for these lasting sensations after loss of the physical limb is the encroachment of adjacent sensory cortical areas on the newly deafferented cortex, resulting in spontaneous neural activity that is perceived by the patient as coming from their phantom limb (Ramachandran, Stewart &

Rogers-Ramachandran, 1992). These sensations can also be painful if the loss of limb was traumatic and free nerve endings continue to send afferent signals to the cortex. However, many sufferers can now find relief in experimental methods that have been adapted as therapy for painful phantom sufferers (Ramachandran &

Rogers-Ramachandran, 1996). Famously, Botvinick and Cohen (1998) were able to induce an illusory sense of ownership and elicit perceived touch to a rubber

31 hand and arm that was placed in a configuration that was anatomically congruent with the human subjects. Graziano (1999) was able to replicate this illusion in macaques with a taxidermic arm, and also found no activation on multisensory areas when the fake arm was positioned in an anatomically incongruent position.

This demonstrates the ability for our visuotactile systems to adapt our body schemas, but to a point: it must be anatomically possible for the illusory limb to be adopted into our own body schema, and the illusion is also facilitated if subjects are trained with simultaneous stimulation to their real (occluded) limbs and the visible phantom. These illusions have also been shown to modulate short- term activation in primary sensory receiving areas (Schaefer et al., 2006), demonstrating the immediate neurological impact of perceptual manipulations upon cortical activation.

These illusions are mediated by multisensory processing, specifically by manipulating physical conditions and 'fooling' our brains into believing what we see through congruent positioning and stimulation. But what happens when we change what we see? Specifically, what happens when the visual input is suddenly incongruous? This question was first investigated by Uta Wolfe and colleagues

(2007) using dark adaptation to create a disparity in the luminance that was processed by both eyes. Following a period of dark adaptation, which involves subjects being exposed to low levels of light for approximately 30 minutes, subjects were asked to cover one dark adapted eye, while the other was exposed to a higher level of luminance for 1 minute, effectually light adapting it. Next, the

32 luminance was decreased to the original low level used during the dark adaptation condition, and in the absence of tactile cues on the face to account for the disparity in light reaching both eyes, subjects reported unusual feelings in the light adapted eye and surrounding facial regions described as "numbness",

"sagging", and "drooping", termed as paresthesias (Wolfe et al., 2007, 2008).

These researchers hypothesized that the difference in luminance to both eyes induced the proprioceptive illusion of a drooping eyelid and face due to the lack of multisensory cues to attribute to the disparity. Indeed, when introducing various forms of tactile, proprioceptive, and motor feedback such as manually covering the light adapted eye and actively closing the light adapted eye, the paresthesias significantly decreased or disappeared.

My colleagues at York University collaborated with Dr. Wolfe and modified her paradigm from a light adapted eye to a completely visually occluded eye by using a black opaque contact lens. In their preliminary research, they were able to reliably replicate the illusory paresthesias in all subjects they had tested with the occluding lens, and were similarly able to reduce or eliminate the paresthesias when introducing the same forms of tactile, proprioceptive, and motor cues

(Jobst, Kucyi, Pynn & DeSouza, 2010).

33 1.4 - Our Hypotheses

Based on the precedents set forth by my colleagues (DeSouza et al.,

20iia,b; Di Noto et al., 2010, 2011a,b; Jobst et al., 2010; Wolfe et al., 2008), I

intend to replicate and expand upon their findings in the following ways:

• by inserting a black contact lens into the dominant eye of our subjects, we

expect to induce illusory sensations of paresthesia in all of them

• demonstrate a decrease in proprioceptive perception of subjects' estimate

of the position of their eyelids while experiencing paresthesias compared

to several control conditions without paresthesias

• reduce or eliminate perceptions of this illusion with introduction of

various forms of tactile, proprioceptive, and visual feedback that could

account for the disparity in binocular luminance by asking subjects, while

subjectively experiencing the illusion, to perform several tasks

• evaluate the impairment of facial paresthesias on tactile perception while

subjects experience paresthesia compared to control conditions by

measuring and comparing thresholds in tactile sensitivity and acuity

• observe reduction in winking behaviour using a novel motor coordination

task compared to controls.

By testing these hypotheses we hope to gain a better understanding of how

reduced visual signals result in an overall decrease in network activity in higher- order multimodal regions and in primary sensory and motor facial nuclei, which

34 result in facilitating this multimodal illusion and its various impacts upon perception and behaviour immediately and following occlusion of one eye in binocular adult viewers.

SECTION 2: METHODS

2.1 - Participants

A total of seventeen participants (6 male, n female, ages 21-52, M=277,

SD=8.3, Table 2.1) were involved in our study and were comprised of members of the university community (undergraduate and graduate students, postdoctoral fellows, and one faculty member within the Faculty of Health). All subjects had normal vision or vision corrected to normal with glasses or contact lenses. Ocular dominance was determined using a simple test (Crider, 1944): the subject extends one arm and with both eyes open aligns their index finger with a distant object.

The subjects then alternates closing the eyes to determine which single eye remains aligned with the object (i.e. the dominant eye). A total of 75 experimental sessions were conducted and paresthesias were reliably induced in all of them.

However usable data and our final analyses are based on 66 of these experimental sessions1. Eight of the seventeen subjects wore their prescription contact lens in

1 Experimental sessions were excluded based on early termination of the session due to flaws in preliminary experimental design.

35 their non-dominant eye during the experiment. Five of the participants are experienced with the occluding contact lens and have previously partaken in a similar experimental paradigm and have worn the experimental contact lens in the past. Four subjects had never worn or inserted a contact lens in their life prior to this experiment.

2.2 - Apparatus

All of the experimental sessions took place in a windowless room, 224" x

227" x 95". Subjects lay supine on a table in the center of the room and keep their gaze within a 25" x 25" central visual field (visual angle 21.450). Subjects were made comfortable with pillows and padding (Figure 2.1). Based on the previous literature involving facial paresthesias from our group (Wolfe et al. 2007, 2008;

Jobst et al., 2010), five predetermined facial zones were chosen for evaluation

(brow, eyelid, under eye, cheek, and upper lip, observed in Figure 2.2) and marked on both sides of the face with cosmetic eyeliner to identify where the experimenter would stimulate. Prior to onset of the experiment, subjects placed a black opaque contact lens (Alden Optical Laboratories, Model HP49 Sphere) in their dominant eye. For the evaluation of tactile sensitivity, subjects were stimulated in the five aforementioned facial zones with a Von Frey

Aesthesiometer (Lafayette Instrument Co., Model #16013). For the evaluation of

36 tactile acuity, subjects were stimulated in these same facial zones during separate experimental sessions with a Two-Point Aesthesiometer (Baseline (§)).

Subject Age Sex Dom­ #of Von jnd Two- Winki Con­ Initials inant times Frey contr Point ng tact eye opaque task ol Discrim motor lens in contact sessi ination task non- worn on of task domi­ prior to Von nant exper­ Frey eye iment task 1 31 F Right 0 Y N N N Y 2 23 M Left 5 Y N N N Y 3 23 F Left 3 Y N N N N 4 21 F Right 1 Y N N N N 5 41 M Right 2 Y Y N Y N 6 25 F Right 1 Y Y Y Y N 7* 52 M Left 0 Y Y Y Y N 8 36 F Right 0 Y Y Y Y Y 9 23 M Right 0 Y Y Y Y Y 10* 23 M Right 0 Y Y Y N N 11 22 F Right 0 N N Y Y Y 12 27 F Left 0 N N Y N Y 13 29 F Left 0 N N Y N Y 14 27 F Left 0 N N Y Y Y 15* 25 M Left 0 N N Y N N 16* 21 F Right 0 N N Y Y N 17 22 F Right 0 N N Y N N TOTAL 27.7 11F 10R 10 6 12 8 8 6M 7L

Table 2.1. Subject summary table. (*) Subject has never inserted a contact lens prior to this experiment.

37 Figure 2.1. Procedural setup. Subjects lay supine on an elevated table while the questionnaire and stimulation was administered by the experimenter, who stood at the head of the subject while applying the stimuli.

Figure 2.2. Participant preparation. Prior to insertion of the occluding black contact lens, five facial zone locations (brow, eyelid, under eye, cheek, and lip [superior-inferior]) were marked on both sides of subjects' faces with cosmetic eyeliner. All of these were the locations (ipsilateral and contralateral) were stimulated by the Von Frey Aesthesiometer, and only zones ipsilateral to the occluded dominant eye were stimulated with the Two-Point Aesthesiometer.

38 2.3 - Procedure

The following will describe each of the dependent variables (DVs) being measured as they are individually affected by the facial paresthesias. Each DV was tested three times for each subject in order to evaluate changes across time and to explore any measurable changes in perceived paresthesias. From beginning to end (insertion and removal of the contact lens, respectively) the duration of the experiment was 105 minutes (SEM=9 minutes).

2.3.1. - Effect of Paresthesia on Sensory Perception of Eye Area

Immediately following insertion of the occluding lens, subjects were given some time to habituate to wearing the contact lens to ensure that they were comfortable before proceeding with the experiment. The first form of evaluation

(n=i7) was in the form of a questionnaire; subjects were asked several questions regarding perception around their eye area (see Appendix A). Alternating between their occluded dominant eye and unoccluded non-dominant eye, subjects were asked to respond either 'Yes' or 'No' to the following questions:

"Does the eyelid feel saggy", "does the eyelid feel numb", "does the eyelid feel droopy", and "does the eye feel closed". Several catch questions were also included: "Does the eyelid feel painful", and "does it feel as though the eyelid is moving by itself?" (see Appendix A). Responses to these questions were coded as o for 'No' and 1 for 'Yes', and 0.5 for 'Maybe'. The order of which side was

39 evaluated was randomized and counterbalanced across subjects, who were also invited to provide subjective descriptions throughout the questioning period, which were recorded verbatim. This portion of the questionnaire took approximately 2 minutes.

2.3.2. - Effect of Paresthesia on Sensory Perception of Facial Areas

In the same questionnaire as above, all subjects (n=i7) were asked to respond either 'Yes' or 'No' to whether they felt any unusual sensations in regions of the face and head, including the forehead, brow, eyelid, under eye, cheek, temple, jaw, lip, chin, ear, nose, and top or back of head on both sides of the face ipsilateral to the occluding lens and contralateral to the occluding lens (as a control, see Appendix A). Similar to the evaluation of ocular sensation, responses were coded as 0 for 'No', 1 for 'Yes', and 0.5 for 'Maybe', the order of which side and head/face area was evaluated was randomized and counterbalanced across subjects, and they were free to provide any additional descriptions of unusual sensations throughout the experiment. This portion of the questionnaire took approximately 2 minutes.

2.3.3. ~ Effect of Paresthesia on Estimated Eyelid Position

All subjects were asked to estimate the position of their eyelids, both for their occluded and unoccluded eyes and during experimental sessions with the occluding lens (n=i7) and during control sessions without the occluding lens

40 (n=n) (see Appendix A). This estimate is on a o% to 100% scale (0% meaning the

eye was completely closed, 100% meaning the eye is completely open relative to

'normal' or 'average' eyelid position). These estimates were obtained three times

(once following evaluation of changes in perceived paresthesias) and compared in

two ways: comparing the occluded dominant eye to the unoccluded non-

dominant eye (n=i7), and comparing the dominant eye when occluded and when

unoccluded during a control session (n=n). These analyses will help determine if

the presence of paresthesias affect the perception of eyelid position, or

proprioceptive perception.

2.3.4. Effect of Paresthesia on Motor Coordination

The effect of paresthesias on motor coordination has been evaluated by asking subjects to perform an alternating winking task. Subjects who are able to

wink with each eye (n=8) were asked to make 20 winks alternating between both

eyes, thus winking 10 times with each eye. The task was performed three times when subjects wore the occluding contact lens, and once without the occluding lens to establish a baseline for task performance time. The first trial was recorded approximately 26 minutes following insertion of the occluding contact lens, the second trial was recorded 21.5 minutes later, and the third trial was recorded approximately 21.5 minutes later, on average a total of 69 minutes after subjects inserted the occluding lens into their dominant eye. Thus, a total of 640 individual winks were observed. Trials were recorded on an iPhone (3G, Apple

41 Inc.) and timed offline with the onset of the first wink indicating the start time, and the completion of the final wink indicating the end time. The total time it took to perform the winking task was divided by 20 to determine the average time per wink. Each eye was also examined individually to account for any errors in winking, which include closure of both eyes (i.e., blinking) or incomplete closure of the winking eye. A total of 132 such errors were observed and were not coded as successful winks. Averaged wink time for each condition will be compared, as will the number of errors performed during experimental and control conditions to determine if the presence of facial paresthesias affects the timing and/or accuracy of coordinated motor behaviour.

2.3.5. Changes in Perceived Paresthesias with Feedback Cues

Changes in perceived paresthesias were evaluated in all subjects (n=i7) during the 66 experimental sessions where paresthesias were reported by introducing various forms of somatosensory, proprioceptive, motor, visual and cognitive feedback, as shown in the procedures of Wolfe et al. (2007, 2008) and

Jobst et al. (2010). Specifically, subjects were asked to cover each of their eyes with their hands, actively close each eye, direct their eyes to their extreme left and right visual fields, and to examine themselves in a mirror in order to elucidate any changes in illusory sensations (see Appendix A). While performing each of these tasks, subjects were asked to report if they felt any changes in facial paresthesias.

Responses are coded such that a complete elimination of paresthesias is coded as

42 -loo, a decrease is -50, a slight decrease is -25, no change in paresthesias is o, slight increase is +25, increase is +50, and a strong increase or doubling in paresthetic sensations is coded as +100. These evaluations were conducted three times (once after each of the three stimulation trials) and coded values were averaged across time for each subject. The results have been analysed using one- sample t tests to determine significant changes in paresthesias following introduction of these various feedback cues.

2.3.6. - Extent of Paresthesia as Illustrated by Subjects

Following the previous series of tasks, all subjects (n=i7) were asked to draw or outline any area(s) with unusual sensations on a schematic diagram of a face superimposed with a grid (Figure 2.3) (see Appendix B). These drawings were scanned and saved as TIFF image files and then reformatted for processing using

Focus imaging software (Geomatica, version 10.3). The TIFF files were then individually loaded onto ArcMap (esri, version 10), a software system typically used for analysis of geographic and topographic data. For the purposes of our investigation we will use ArcMap to determine the extent of paresthesias for each individual subject's illustration. Once images were imported into ArcMap they were assigned spatial coordinates that match their actual size on the questionnaire. It is important to note that due to modifications in the questionnaire sheet, there are several different sizes of the facial schematic, and the spatial coordinates applied to each individual figure reflect these differences.

43 Next, a separate shape file (or image layer) was created which contained an outline of the area illustrated by each subject (Figure 2.4). The affected facial area illustrated by subjects on the schematic was calculated for each entry, and statistical analyses were completed using these values to determine the proportion of facial areas affected by paresthesias on the entire face as illustrated by subjects, as well as any significant changes in area over time and any differences in the extent of paresthesias based on the criteria of sex, ocular dominance (i.e., contact in left or right eye), and usage of contact lenses.

44 Figure 2.3. Facial schematic. Subjects were asked to illustrate the extent and perception of any unusual sensations on this schematic. This figure is to scale as it appeared on one version of the questionnaire, but the same image was resized in other versions. These differences have been accounted for in the analysis, and results will reflect the proportion, not actual size, of affected areas as illustrated by subjects.

45 ma Wk *.iAA» •/' "OT UL£JL

OQMMrtjfll! El - *)y OooMrtfVtnfilS1: b H if £J« ft: #Totfa»c ttf Q FdUr Cemefliew £ Vtoekoxw S I&MatwMSwvOT If; Q CWebwe Cwiwdem IS f0ass«rvan IF TracMng comwBom

fe 3D tndrit Tocfc I tottr* Took I C

Figure 2.4. ArcMap Shapefile superimposed on subject's illustration. By creating a shapefile, we were able to outline the area and extent of reported paresthesias as illustrated by subjects on the questionnaire sheet. Spatial coordinates were assigned to each drawing and the area of the outline was calculated for each image.

2.3.7. ~ Effect of Paresthesia on Tactile Perception

2.3.7.1.: Tactile Sensitivity

Based on the research by Wolfe et al. (2008) and Jobst et al. (2010), we have focused evaluations of the impact of tactile illusions on sensory perception to five

predetermined zones: brow, eyelid, under eye, cheek, and upper lip. In ten subjects (5 females, age=29.8, SD=IO.2, Table 1) these areas were marked on both sides of the face with cosmetic eyeliner (Figure 2.2) to identify where stimulation

46 would be applied with a Von Frey Aesthesiometer (Lafayette Instrument Co.,

Model #16013), comprised of nine monofilament nylon hairs of varying thicknesses, which exert varying intensities of tactile stimulus (Table 2.2). The thinnest Von Frey hairs elicit a very light touch, and the thickest hairs elicit a more intense, salient touch. Stimulation with this apparatus was used to quantitatively examine any disparities or changes in tactile sensitivity in facial zones ipsilateral to the occluded eye that subjects described as feeling unusual, and were also compared to contralateral, unaffected facial zones (for an example of how the apparatus is used, see Thompson & Lambert, 1994).

Von Frey Hair Pressure Used in Eyes Open Used in Eyes Closed Number Exerted (g) Condition Condition 1 0.05 Damaged and excluded Damaged and excluded from experiment from experiment 2 0.15 Damaged and excluded Damaged and excluded from experiment from experiment 3 0.4 Cheek Brow(1), Eyelid (1), Under eye (1), Lip (1) 4 0.8 Brow, Under eye Eyelid (2), Cheek (1), Lip (2) 5 1.2 Brow, Cheek, Lip Eyelid (3), Under eye (2) 6 3.2 Eyelid, Under eye Brow (2), Cheek (2), Lip (3) 7 4.5 Eyelid, Lip Brow (3), Under eye (3), Cheek (3) 8 6.5 Brow, Under eye, Eyelid (4) Cheek, Lip 9 10.2 Eyelid Brow (4), Under eye (4), Cheek (4), Lip (4)

Table 2.2. Von Frey Aesthesiometer parameters. The final 2 columns list the facial zones that were stimulated by each hair during both experimental and control conditions in each of the eyes open and closed conditions. The numerical code for each Von Frey hair used in the eyes closed condition, which will be used in subsequent analyses, is indicated in brackets.

47 Subjects were instructed to provide an intensity rating for stimuli administered in each of the five predetermined facial zones via the Von Frey hairs. While subjects lay supine on an elevated table (Figure 2.1), the experimenter stood at the head of the subject and randomized stimulation across the five facial zones, between the two sides of the face, and among seven of the nine varying densities of Von Frey hairs for a total of 70 stimulations (see

Appendix A). The stimulation period lasted on average 12 minutes. Subjects were instructed to base their ratings on a scale of -4 (no detection of stimulation), 0

(unsure if stimulated), and 1 to 4 (intensity of sustained stimulus), the same sensation scales utilized by Wolfe et al. (2007) and Jobst et al. (2010). Subjects were also permitted to respond in 0.5 increments. It is important to note that subjects were instructed to distinguish between the initial prick of the Von Frey hair and were asked only to rate the perceived strength of the sustained stimulation (analogous methods to Thompson et al., 1994). Subjects were trained with 3 to 5 sample stimulations prior to recording stimulus ratings. Subjects were also instructed during 40 of the 70 total stimulations to close their eyes in an attempt to control for any visibility of the experimenter and the expectation of an approaching stimulus near the face, thus reducing any confounding visual anticipatory cues to the touch (Drevets et al., 1995). However, due to a flaw in the experimental design stimulus intensity ratings were not obtained during both eyes open and eyes closed conditions for every facial zone with every level of Von

Frey hair thickness. The pattern of stimulation in eyes open and eyes closed

48 conditions is summarized in the third and fourth columns of Table 2.2 and were the same in both experimental and control facial zones. The stimulations were also randomized to switch between eyes open and eyes closed conditions so as to ensure that subjects were alert and paying attention to the experimenter's instructions. The experimenter said the word "close" as a prompt for the subject to close their eyes, and a randomized zone was stimulated with a randomized Von

Frey hair within 1 to 4 seconds. Immediately after detecting the stimulation the subject was instructed to open their eyes and provide an intensity rating. Any stimulus ratings provided by subjects opening their eyes prior to removal of the stimulus were excluded from analyses. Subjects were also carefully questioned and evaluated to ensure that any paresthetic sensations did not disappear or significantly attenuate within the brief eyes closed period. Subjects were asked to close their eyes and report when facial paresthesias disappeared completely while the experimenter timed it. On average, it took 9.75 seconds (SD=8.3 seconds) for paresthesias to disappear following closure of the eyes.

These instructions were provided to the subject prior to the onset of the experiment, and all subjects were required to understand and abide by them throughout the experiment. All subjects were stimulated in approximately the same zones for a total of 70 stimulations during each of the three trials during the experiment. During the eyes open conditions, subjects were asked to fixate on a

25" x 25" area of ceiling above them (visual angle 21.45"), and they were given reminders of the rating scale prior to the onset of the stimulation trial and

49 throughout as needed. Facial zone, density of Von Frey hair, and side of face

stimulated (ipsilateral or contralateral to occluding lens) was randomized within

and across subjects (see Appendix A).

Of the ten subjects who participated in the Von Frey stimulations, six

subjects (4 males, mean age= 33.3, SD=n.8) participated in a second

experimental session where they did not wear the occluding contact lens and

tactile sensitivity was evaluated in the same five facial zones as the experimental session where they did wear the occluding lens. The second session only

evaluated zones ipsilateral to the dominant eye, and was conducted in an attempt

to control for any effects that may have occurred on the contralateral side of the

face or due to any other unknown confounds. Statistical analyses of these new

control ratings will be conducted similarly to the control ratings obtained from

contralateral facial areas during the original experimental session.

2.3.7.2.: Tactile Acuity

In order to investigate any changes in tactile acuity (the ability to

discriminate fine touch) that could potentially result from the facial paresthesias,

12 subjects (8 female, mean age=27.7, SD=8.7) were tested during two separate

experimental sessions with a Two-Point Aesthesiometer (Baseline ®), a

calibrated ruler affixed with two moveable plastic points that can be used to

determine a two-point detection threshold. Five of the subjects who participated

in this evaluation were also participants in the Von Frey evaluation (3 males,

50 mean age=3i.8, SD=12.5, see Table 2.1). Two-point discrimination (TPD)

thresholds were determined for the same five facial zones as stimulated by the

Von Frey hairs (brow, eyelid, under eye, cheek, and upper lip) but only for zones

ipsilateral to the occluded eye (see Appendix C). In a second experimental

session, control TPD thresholds from ipsilateral facial zones were obtained when

subjects were void of the occluding lens and not experiencing facial paresthesias.

These two sessions were on average 10.5 days apart, and the order of which session occurred first (experimental or control) was counterbalanced across all subjects.

Both sessions took place in the same experimental room and stimuli were

administered in the same manner as the Von Frey hairs; subjects lay supine on a

table while the experimenter stood at the head of the table and stimulated subjects with the Two-Point Aesthesiometer on the side of the face ipsilateral to the occluding contact lens only (Figure 2.1). The stimulus was applied to each

zone both in an ascending order (beginning with one point and gradually moving the two points farther away from each other in 1 mm increments) an in a descending order (beginning with two points spaced far enough apart to be

perceived as two separate points and gradually moving closer together in imm increments- see Heft et al., 2010). Stimulation across zones was randomized, as was whether the order of stimulation was ascending or descending (see Appendix

C). Each stimulation session was 16 minutes on average, with each zone stimulated six times (ascending and descending thresholds obtained over three

51 trials). Subjects were asked to close their eyes immediately prior to application of the stimulation in order to minimize confounds associated with seeing the approaching stimulus, and were asked to open their eyes before providing a response: whether they felt one or two points. The distance at which subjects' changed their response (i.e., in the ascending condition, report feeling two points, and in the descending condition, the last distance at which subjects report feeling two points) were recorded as the TPD threshold. These distances were averaged for each zone across all six trial presentations to obtain a final threshold. The facial zone and whether the order of stimulus presentation was ascending or descending was randomized and counterbalanced across subjects. In order to minimize errors, subjects were informed of which zone was being stimulated and if the order was ascending or descending prior to each stimulus presentation. The experimenter ensured that subjects did not open their eyes prior to administration or removal of the stimulation, and all subjects were carefully questioned to ensure sustained paresthesias during the brief eyes closed period.

On average, it took 17.75 seconds (SD=14.87 seconds) for the paresthesias to completely disappear following closure of the eyes. TPD thresholds obtained during the experimental and control sessions were compared with a 2-way

ANOVA.

52 2.3.8 - Total experimental duration

The entire length of the experiment (WITH CONTACT: debriefing,

insertion of contact lens, evaluation of all dependent variables but only one of the

Von Frey or TPD evaluations, repeated three times, removal of contact lens) was

on average 86 minutes (SD=40.39 minutes) when the Von Frey analysis was

conducted and 121 minutes (SD=39.98 minutes) when the TPD analysis was

conducted. The control session (WITHOUT CONTACT LENS: debriefing,

evaluation of all dependent variables but only one of the Von Frey or TPD

evaluations, repeated three times) was on average 75 minutes (SD=22.72).

Subjects were free to take brief breaks from questionnaires and stimulations

throughout the experiment, to sit up, walk around the experimental room, and

take washroom breaks when needed. In total there were 66 experimental sessions

that were on average 105.1 minutes (SD= 43.2 minutes), and 51 control sessions

that were on average 85.9 minutes (SD=40.4 minutes).

SECTION 3: RESULTS

3.1 - Introduction

The following will summarize the findings of our experimental methods,

including all subject reports of paresthesias and changes in perception as well as

the statistical analyses conducted on the variables described in the previous

53 section. It is important to note once again that several evaluations were not conducted on all subjects due to time constraints within experimental sessions and revisions to our experimental design. As such, the subject number will be indicated in the results of each variable.

3.2 - Subjective descriptions following insertion of occluding lens

All 17 subjects evaluated reported unusual sensations in the eye and/or face area ipsilateral to the occluding contact lens during all 66 experimental sessions. On no occasion did subjects report unusual facial sensations on the side of the face contralateral to the occluding contact lens, or during control sessions, but 3 subjects did report unusual sensations near the midline of the forehead

(n=2), top of nose (n=i), top of head (n=i), and back of head (n=i). Subjects were not able to see through or in the periphery of the occluding lens, but three subjects did mention some perceived ability to detect a change in luminance through the contact lens although they were unable to perceive any form discrimination when tested. Six subjects reported an interference in normal vision of their contralateral unoccluded eye while wearing the occluding lens in their dominant eye, and two subjects displayed visible unilateral sagging of the occluded eyelid only during experimental sessions that displayed approximately

20-30% closure of the eyelid of the occluded eye only when both eyes were open and the unoccluded eye was open normally. Adjectives used to describe facial

54 sensation after insertion of the occluding lens include: "heavy", "numb",

"drooping", "sagging", "radiating", similar to what one feels when at the dentist

(n=2), "tingly", "edema", "vertigo", "confused", "weird", "aura", "white noise",

"headache", "something wrong", "tension", "puffy", "swollen", "odd", "heat",

"cooling", "squint", "tight", "pulling of skin", "skin thicker" or "something

underneath" area of paresthesia. Subjects also described the eye area as

"vacuous", "void", "gap', or "hole" where the eyeball should be. Subjects reported

unusual sensations on the following areas: eyelid (n=i7), under eye (n=i6), cheek

(n=i5), brow (n=io), forehead (n=7), temple (n=6), lip (n=6), top of head (n=5),

back of head (n=5), jaw (n=5), eyeball (n=s), ear (n=5), nose (n=4), chin (n=2),

jaw, upper and posterior teeth and gums, and back of neck (ipsilateral).

3.3 - Extent of Paresthesia as Illustrated by Subjects

We calculated the surface area and proportion of paresthesias illustrated

by each subject (n=i7) during experimental sessions on the facial schematic shown in Figure 2.3. Subjects were asked to illustrate the extent of any unusual sensations on the facial schematic three times during the experimental session following the questionnaire period2. This area was quantified in squared centimeters and divided by the total area of the schematic upon which it was

2 Five subjects (6, 7, 8, 9,10) participated in two separate experimental sessions where they inserted the occluding contact lens in both sessions and as such provided two sets of illustrations. The illustrations provided by these subjects were averaged for each time point across the two sessions, and the averaged values and images have been presented in these analyses.

55 drawn in order to obtain a value (in percent) that reflects the proportion of the facial drawing that the subjects perceived and illustrated as an 'affected area', or the parts of their face that they felt paresthesia. The average total illustrated area of paresthesia across all subjects is 5.88cm2 (SEM=0.03cm2), or 9.42%

(SEM=O.29%, Figure 3.1a) of the facial drawing, about one tenth of the total face, and one fifth of the side of the face ipsilateral to the occluded eye. Overlapping all subject illustrations in Figure 3.1a demonstrates the spread of subject localizations of perceived paresthesias, extending across the entire side of the face ipsilateral to the occluded eye, with darker areas reflecting the highest degree of overlap among subjects. One-sample t tests of the proportion of affected area as illustrated by subjects (in percent) reveal that the total and block average illustrations of paresthesias are significant compared to null reports of paresthesia (block 1: f(i6)=3.457, Pco.oi, block 2: f(i6)=3.728, Pco.oi, block 3: f(i6)=3786, Pco.oi, average: t(i6)=3.7ii, Pco.oi, Figure 3.2). A mixed ANOVA that included sex, ocular dominance (i.e., if the occluding contact lens was in the subjects' left or right eye), and regularity of contact lens use as between-subjects factorss revealed no significant difference in illustrated reports of paresthesia over time (F(2,i8)=o.68, P>.i, see below for time points), nor did subjects differ in any other factor (P>.i).

During block 1 (on average 30 minutes after inserting the occluding lens), subjects illustrated an average affected facial area of 8.86% (SEM=2.56%) of the

3 All ANOVAs in the remainder of the results will be mixed factor ANOVAs, with sex, ocular dominance, and regularity of contact lens use serving as between-subject factors, and others will be indicated as described.

56 face, 9.83% (SEM=2.64%) at block 2 (on average 53 minutes after insertion of the occluding lens), and 9.58% (SEM=2.53%) at block 3 (on average 74.1 minutes after inserting the occluding lens, Figure 3.1b). The affected area per subject averaged over time can be seen in Figure 3.3. The greatest extent of paresthesias was reported by Subject 7 and affected almost 40% of the face, and the least extent of paresthesias was reported by Subject 9, occupying an area of 0.8%

(SD=O.2%) of the face and limited to the upper eyelid of the occluded eye (Figure

3-3)-

Block 3: 9.58%

Figure 3.1. a. Subject illustration of extent of facial paresthesia: overall average (n=17). All subject illustrations of perceived paresthesias have been overlapped here and extend across the side of the face ipsilateral to the occluded eye. The average illustrated affected area and percentage of facial paresthesia across all subjects (n=17) over time can be seen as the darkest portion of the figure that reflects the area with the highest degree of overlap across every subject, b. Subject illustration of extent of facial paresthesia: temporal averages (n=17). The average illustrated area and percentage of facial paresthesia across all subjects (n=17) at each time point is shown here.

57 Subject Illustrations: Percentage of Face £ Affected by Paresthesias (n*17) |30

f 25

** 20 **

15

10

Block 1 Block 2 Block 3 Average

Figure 3.2. Temporal analysis of facial drawings of perceived paresthesias: percent of facial illustration affected by paresthesias. Significant reports of facial paresthesias were observed during each block of facial illustrations among all subjects compared to nil as indicated by asterisks, with no significant change in the area of illustrated paresthesias over time (Block 1 vs Block 2 vs Block 3) and compared to the overall average area of paresthesias. Error bars show standard error of the mean.

58 1:2.24cmi, 4.17% 2 161cm2, 2,13* 3:2.65cm'. 3.51* 4:20.51cm', 33.48* 5: 5.97cmJ, 9 97* 6: l OScnV, 1.77*

7: 23.81cm1, 36.18* 8: 9.11cm', 14.72* 9:0.50cm', 0.84* 10: 2.06cm', 3.40* 11: 7.18cm', 11.80* 12: 1.43cm', 2.35*

13:6.92cm', 11.37* 14:4.01cm', 6.59% 15:4.00cm', 6.58* 16:6.25cm', 10.28* 17:0.63cm', 1.04*

Figure 3.3. Subject illustration of extent of facial paresthesias (n=17). Subject illustrations of the extent of perceived facial paresthesias were scanned and outlined in ArcMap software. Each subject (n=17) is shown here, averaged across time with initials, area (in cm2) and percentage of total schematic that was identified as paresthetic. All illustrations of paresthesia were drawn on the left side of the diagram (i.e., corresponding to the right side of the face) to demonstrate overlap across subjects, but 7 of the subjects had a left dominant eye and thus wore the occluding lens in their left eye. The black pupil signifies the occluding contact lens.

59 3-4 - Effect of Paresthesia on Sensory Perception of Eye Area

In order to determine any effect of the illusory paresthesia on sensory

perception of the eye area, all 17 subjects were asked a series of questions

regarding the extent of 'sagging', 'drooping', 'numbness', and 'closure' of their

occluded and unoccluded eyes that required either a 'Yes' or 'No' answer. Subject

responses were coded as 1 for 'Yes', 0 for 'No', and any 'Maybe' or 'Little bit'

responses were coded as 0.25, 0.5, or 0.75 as clarified with the subject in order to

create an index for each of these variables. Both eyes (i.e., the subjects' occluded

dominant eye and unoccluded non-dominant eye) were evaluated to see if the

presence or lack of the occluding contact lens and any accompanying paresthesias affected ocular sensation perception. Subjects responded affirmatively (i.e., either

'Yes', 'Maybe' or 'Little bit') to the question "Does the (occluded) eyelid feel saggy"

63% of the time (SEM=0.055), to whether their occluded eye "feels closed" in 54% of trials (SEM= 0.058), to "eyelid feels numb" in 52% of trials (SEM= 0.059), and in 66% of trials to "eyelid feels droopy" (SEM=o.os6) (Figure 3.4). Subjects did

not respond affirmatively to any unusual sensation for their unoccluded eye,

resulting in all null responses for the control condition. As such, it forms the baseline value (nil) in Figure 3.4, which shows the difference in subjective reports of perceived sensations around the occluded and unoccluded eyes during the experimental session.

60 Difference in Reports of Paresthesia in Area Surrounding Occluded Eye versus Unoccluded Eye (n=17)

as 0.9 |0.8

§.0.7 v> $ 0.6

> 0.5 0, 1a. 0 0.3 +* 8 0.2 1 0.1

Saggy Closed Numb Droopy

"Does your eye(lid) feel..."

Figure 3.4. Effect of paresthesia on sensory perception of eye area. Subjects responded either 'Yes', 'Maybe', 'Little bit' or 'No' to the above questions, and responses were coded and compared between the occluded eye and non-occluded eye. Subjects responded 'No' in all cases for the unoccluded eye and the above plot shows the difference in reported sagging, closure, numbness, and drooping of the occluded eyelid. Error bars display standard error of the mean.

One-sample t tests reveal that these subject reports were all significant from nil, or from reports of these sensations in their unoccluded eye (sagging: t(i6)=8.29,

Pc.ooi, closure: t(i6)=6.37, Pc.ooi, numbness: f(i6)=6.o6, Pc.ooi, drooping: f(i6)=8.58, Pc.ooi, Figure 3.4). ANOVAs were performed for each of these indexes to determine if subject responses changed significantly over time, which was found to be the case only for subject reports of how closed they perceived

61 their eyelids to be (F(2,i8)=4.25, P-.031) and perceived drooping (F(2,i8)=5.93,

P=.on). Pairwise comparisons reveal that subject reports of both closure (t(i6)=

-3.139, P=.006) and droopiness (t(i6)=-2.447, P=.026) of their occluded eye decreased significantly between the first and third block of questioning, but not between the first and second (closure: t(i6)=-i.94i, P=.070, drooping: t(i6)=

-1.000, P=.332) or second and third blocks (closure: t(i6)=-i.376, P=.188, drooping: t(i6)=-i.922, P=.073). These results clearly indicate the presence of a salient proprioceptive illusion in the eyelid of the occluded eye that results in significant perceptions of sagging, closure, numbness, and drooping of the occluded eyelid only with symptoms that are consistent with the characteristics of a sensory paresthesia persisting over time. Based on our review of cortical processing areas, this illusion could be mediated by the feedback signals from multisensory association areas that receive visual, tactile, and proprioceptive information, and with the sudden cessation of visual inputs, may potentially be forwarding an inhibited or dampened signal to higher-order cortical tactile and proprioceptive processing regions for the eye and face areas surrounding the occluded eye.

3.5 - Effect of Paresthesia on Sensory Perception of Facial Areas

Similar to the previous section, all 17 subjects were evaluated on the effects of the paresthesias induced by wearing the occluding contact lens on facial

62 sensation perception. Subjects provided either 'Yes', 'Maybe', 'Little bit', or 'No' responses to the question "Does your X feel unusual", with the following facial areas serving as X: eyebrow, eyelid, under eye, temple, cheek, jaw, lip, ear, nose, top and back of head, and forehead. Responses were coded the same as in Section

3.4, with 'Maybe' or 'Little bit' coded as 0.25, 0.5, or 0.75 as clarified by the subject. Both sides of the face and head were evaluated and as with the evaluation of eye sensation, no subjects responded affirmatively (i.e., at least 'Little bit') to having unusual sensations of facial areas on the same side of their face as the unoccluded eye, and thus only indexes of sensations from facial regions ipsilateral to the occluded eye have been analysed. As such, subjects responded at least

'Little bit' to perceiving paresthesias in ipsilateral facial regions from 51 total trials (three trials for each of the 17 subjects) in the following proportions: the brow was affected in 64% of subject responses (SEM= 0.066), the eyelid in 91%

(SEM= 0.033) of responses, the under eye in 78% (SEM= 0.055) of responses,

28% (SEM= 0.063) in the temple, 81% (SEM= 0.054) in the cheek, 24% (SEM=

0.060) in the jaw, 26% (SEM= 0.060) in the ipsilateral lip, 26% (SEM= 0.061) in the ipsilateral ear, 24% (SEM= 0.058) on ipsilateral aspects of the nose, in 26%

(SEM= 0.060) of responses in the ipsilateral top and back of head, and the ipsilateral forehead in 30% (SEM= 0.064) of responses (Figure 3.5). This reveals the distribution of localizations for the paresthetic illusion reported by our subjects in the face and regions of the head ipsilateral to the occluding contact lens.

63 One-sample t tests reveal significant subject reports of paresthesia in all eleven facial regions when averaged over time compared to responses from the unoccluded side of the face, which were all nil (Figure 3.5). To correct for inflation of Type I error that could result from performing eleven one-sample t tests, we divided our statistical cut-off value (P=.05) by 11, giving us a new significance criterion of P=.oc>5. The red asterisks in Figure 3.5 reflect the indexes that remained significant with the new criteria, and represent the areas most reported as experiencing paresthesias across all subjects (n=i7). Separate ANOVAs were also conducted for each facial region to determine any significant changes in subject reports of paresthesia over time and for none of these main effects or interactions did we find a significant effect (J°>.1). These results provide further evidence of a salient sensory illusion present in facial regions ipsilateral to and surrounding the occluded eye.

64 Difference in Reports of Paresthesia in Facial and Cranial Areas Ipsilateral versus Contralateral to Occluded Eye (n»17)

2 1 i„. 8 oc 1 0.6 5 o CL o 0.4 * H o- 0.2 * * s y * *° <<° &&

Figure 3.5. Effect of paresthesia on facial and cranial sensory perception. Subjects (n=17) were asked whether the above areas felt "unusual" and responded either 'Yes', 'Maybe', 'Little bit', or 'No'. Subjects reported significant unusual sensations in all facial regions ipsilateral to the occluded eye, which are shown here as the difference in reports between the occluded and unoccluded eye, the latter of which received no reports of unusual sensations in any region at any point during the experiment and forms the baseline of the plot. Red asterisks reflect significance P<.005, *** (P<.001), ** (P<.01), * {P<.05). Error bars display standard error of the mean. 3.6 - Effect of Paresthesia on Estimated Eyelid Position

To examine the extent to which subjects perceived their eyelids as

drooping, and to quantify the extent of the subjects' perceived proprioceptive

illusion of the eyelid as described previously when dark adapted (Cassidy,

Schawerna & Wilkinson, 2010; Wolfe et al., 2007; Hubel et al., 1962), we

compared estimates of eyelid position of the occluded and contralateral

unoccluded eye during experimental conditions (n=i7), which revealed that subjects perceive their occluded eye as being significantly more closed than their

unoccluded eye (F(i,9)= 39.245, Pco.ooi), and this difference between occluded and unoccluded eye did not change significantly over time (F(2,i8)=2.364, P>.i)

(Figure 3.6), nor were there any other significant interactions with the condition of occlusion of the eyes (P>o.i). We confirmed these results with another ANOVA comparing eyelid position estimates of the occluded eye during a control session

(n=n), which also revealed no differences except for significant reports of eyelid closure during the experimental condition (F(i,6)= 29.063, Pco.oi) (Figure 3.7).

66 Estimated Eyelid Position Within Session: Contralateral Eye as Control (n=»17) -J00 •** £ 90 (P<0.001) & 80 ill

Figure 3.6. Eyelid position estimates: occluded versus control eye (n=17). When comparing estimates of eyelid position during the experimental condition, subjects perceived their occluded eye to be significantly more closed than their unocciuded eye. Error bars show standard error of the mean.

67 Estimated Eyelid Position Across Sessions: Ipsilateral Eye as Control (n=11)

100 ** £ 90 (P<0.01) £ 80 in •s 70 e 60 2 50 40 30 20 10 0 Occludeded Unoccluded (separate session)

Figure 3.7. Eyelid position estimates: ipsilateral occluded versus ipsilateral unoccluded eye (n=11). Subjects reported their dominant eye to be significantly more closed during the experimental condition while they were wearing the occluding lens compared to estimates obtained for the same eye during a separate control condition where subjects did not wear the occluding lens. Error bars show standard error of the mean.

68 3-7 - Effect of Paresthesia on Motor Coordination

In order to develop a measure that is independent of perceptual report we devised a facial motor coordination task. For the eight of the 17 subjects that were able to perform the alternating wink task, we recorded them performing ten winks alternating between both eyes and hypothesized that subjects would require increased sensorimotor faculties to perform the task while their dominant eye is affected by paresthesia induced by monocular occlusion, and would thus take longer to perform the alternating wink task compared to controls. It was surprising that only 8 of our 17 subjects could adequately perform a wink in each eye, as the remaining subjects were either unavailable for this evaluation or could not voluntarily close one eye without also closing the other. However, when considering just how closely related binocular eye movements are, and how intimately the blinking circuitry from the superior colliculus and cingulate motor areas are timed (Bour et al., 2000; Hanakawa, Dimyan & Hallett, 2007;

VanderWerf, Buisseret-Dalmas & Buisseret, 2002), it is understandable that so few of our subjects could supress the need to close both eyes when executing a voluntary closure of one eye. We wanted to see if there was a difference in task performance time between the two different conditions under which subjects were tested: with the occluding contact lens and experiencing paresthesias

(averaging across the three trials of the task performed during the experimental session), and without the occluding lens during normal conditions. Because the

69 previous analysis revealed no significant differences in any of our between-

subjects' factors (sex, ocular dominance, and presence of clear corrective lenses in

non-dominant eye during the experimental condition), we compared average task

performance time over the three experimental trials to the average baseline

timing of the task (both eyes unoccluded) and found that there is indeed a

significant difference between the two conditions (F(i,2)=20.332, Pc.oi, Figure

3.8). Next, we wanted to see if there was a difference in task performance over the

duration of the experiment and compared to baseline. The task was repeated

three times during the experimental session, each time after the questionnaire

and Von Frey or TPD stimulation trials, and a single control task was completed

either before or after the experimental session, which was counterbalanced across

subjects, to establish a non-contact rate of task performance. The first trial was

recorded approximately 26minutes following insertion of the occluding contact

lens, the second trial was recorded 21.5 minutes later, and the third trial was

recorded approximately 21.5 minutes later, a total of 69 minutes after subjects

inserted the occluding lens into their dominant eye. An ANOVA reveals no significant difference in task performance over time (F(3,6)=3.272, P>.1, Figure

3.9), nor was there a significant effect of any other variables (P>.i). The difference

in task performance between the two conditions could either be due to the

presence of paresthesias as reported by subjects or to the presence of the

occluding lens. This latter suggestion is unlikely, however, given that there was no difference in task performance between subjects who wore a clear corrective

70 contact lens in the unoccluded eye during the experimental session and in both eyes during the control session and are accustomed to wearing contact lenses compared to novice contact lens wearers (F(I,2)=.OOI, P>.I). These results demonstrate a lag in performance time of our alternating winking task when winking during the occluding contact-wearing (experimental) conditions compared to winking done without contact-wearing (baseline) (Figure 3.8).

Motor Coordination Task (n*8): Average Occluded Condition vs Unoccluded Baseline Condition

** PC0.01

0.6

0 + Occluded (Experimental Unoccluded (Control Condition) Condition)

Figure 3.8. Coordinated winking task: unoccluded baseline versus average occluded condition performance time. Occluded condition averages across the three trials of the task performed during the experimental condition when subjects wore the occluding lens and reported paresthesias. Error bars show standard error of the mean.

71 Motor Coordination Task (n«8): Change in Performance Over Time

Baseline Triad (Occluded) Trial2 (Ocduded) Trial3 (Occluded) (Unooduded)

Figure 3.9. Coordinated winking task: time analysis. Error bars show standard error of the mean.

3.8 - Changes in Perceived Paresthesias with Feedback Cues

Here we examined any subjective changes to perceived paresthesias with the introduction of feedback through various simple tasks that subjects were asked to carry out during the experiment. One-sample t tests were conducted on the coded responses provided by all subjects (n=i7) in response to whether they perceived any global changes in paresthesia or occlusion-induced symptoms on their faces following the introduction of various forms of feedback, which were averaged over time. We predicted that introducing these various forms of

72 feedback would reduce subjects' perceived paresthesias and one-tailed

significance values support our hypothesis.

Somatosensory feedback was shown to significantly reduce paresthesia

when subjects covered their occluded eye with their ipsilateral hand (f(i6)= -

10.029, P<0.001), when covering their unoccluded eye with their contralateral

hand (t(i6)= -3.596, Pco.oi), when asked to actively close their occluded eye

(t(i6)= -3.937, P<0.001), and when asked to actively close their unoccluded eye

(t(i6)= -2.580, P<0.05). Interestingly, a task originally intended as a control showed a significant increase in paresthesias when subjects were asked to blink

three times and report what they felt (t(i6)= 2.252,

P<.05). There was no significant change in paresthetic perception when subjects

were asked to direct both eyes to their extreme left visual field (f(i6)= -0.382,

P>.i) or when asked to direct both eyes to their extreme right visual field (f(i6)=

-1.739, P>.1, Figure 3.10). However, two (n=2) subjects did report a shift in the facial area experiencing the paresthesia when asked to move their eyes to their extreme left and right visual fields.

Visual and cognitive feedback for the monocular blindness was introduced

by asking subjects to look at themselves in a mirror and also resulted in significant decreases in sensations of "sagging" (t(i6)= -3.518, Pco.oi) and

"numbness" (t(i6)= -4.019, Pco.oi), but not "blindness", or sense of vision loss

(t(i6)= -1.971, P>0.05) (Figure 3.11). As with the analyses in Section 3.5, we corrected for Type I error by adjusting the significance criteria to Pc.005 by

73 dividing our original significance criterion (P<. 05) by the number of one-sample t tests conducted (eleven), and paresthesias were still significantly reduced when subjects were asked to cover their occluded and unoccluded eyes with their hands, actively close their occluded eye, and also significantly reduced perceptions of sagging and numbness when subjects had visual feedback from self-examination in a mirror (Figure 3.10, Figure 3.11).

Changes in Perception of Paresthesia with Feedback (n=17) 20

10

0 Loofftfl ht Blink 3x g-10

I -20 Of

£ -30 £ -40

-50

-60 Feedback Tasks

Figure 3.10. Changes in perceived paresthesias with feedback. Paresthesias were reduced significantly across all subjects (n=17) when feedback cues, involving the tactile and proprioceptive modalities, were introduced while subjects reported experiencing paresthesias. There was no change in reported paresthesias when subjects were asked to look left or right, and a significant increase in paresthesia was reported when subjects blinked three times. Red asterisks reflect significance P<.005, *** (P<.001), ** (Pc.01), * (P<.05). Error bars show standard error of the mean.

74 Changes in Perception of Paresthesia with Visual Feedback: Looking In Mirror (n«17)

0 -5 -10 1 -15 -20 K * -25 A -30 a.i -35 -40 Sensations

Figure 3.11. Changes to perceived paresthesias with visual feedback. Subjects reported significant decreases in paresthesia and accompanying symptoms when asked to look into a mirror compared to when they are not looking at themselves in a mirror. Red asterisks reflect significance P<.005, *** (P<.001), ** (F<.01), * (P<.05). Error bars show standard error of the mean.

3.9 - Effect of Paresthesia on Tactile Perception

3.9.1- Tactile Sensitivity

The effect of paresthesias on tactile sensitivity in affected facial areas was

measured by stimulation of the brow, eyelid, under eye, cheek, and lip with a Von

Frey Aesthesiometer (Table 2.2). Ten subjects were stimulated on these regions

both ipsilateral and contralateral to their occluded dominant eye and were asked

to provide stimulus intensity ratings on an ascending scale from o (no stimulation detected) to 4 (strong, salient stimulation). Stimulus intensity ratings for all

75 ipsilateral facial zones were compared to all contralateral stimulus intensity ratings, which were averaged for each subject over the five facial zones, three trials per experimental session, seven Von Frey hair thicknesses, eyes open and closed conditions, and paresthetic and non-paresthetic zones. Stimulus intensity ratings from the side of the occluded eye (Mean=2.o6, SEM=0.03) were not significantly different from ratings on the unoccluded side (Mean=2.o8,

SEM=O.C>3, F(I,4)=0.054, P>.05, Figure 3.12), nor were any other main effects or interactions significant. Although this may suggest no effect of paresthesias on tactile perception, this evaluation averages across many variables and any significant effect of paresthesias on tactile perception could be hidden.

To further elucidate any impact of paresthesias on tactile perception, we conducted a more detailed analysis of the previous investigation on stimulus intensity ratings from either side of the face. In addition to conducting an ANOVA that took into account the various facial zones and Von Frey hair intensities, all stimulus intensity ratings provided during the 'eyes open' conditions were excluded from analysis based on the finding that they may have been confounded by the subjects' ability to see the experimenter and approaching stimulus with their unoccluded eye. Due to a flaw in the original design of the experiment, not all zones were stimulated at all levels of the Von Frey hairs in the eyes open and

76 Average Stimulus Intensity Rating: Ipsiiateral vs. Contralateral Slda of Faca (n>10)

Ipsiiateral Contralateral

Figure 3.12. Average stimulus intensity ratings for ipsiiateral and contralateral sides of the face. There is no significant difference between the stimulus intensity ratings provided for facial zones ipsiiateral to the occluded eye and those provided for contralateral facial zones. These values average across all subjects (n=10), facial zones (brow, eyelid, under eye, cheek, and lip), time (3 trials per session), Von Frey hair thickness, eyes open and closed conditions, and paresthetic and non-paresthetic zones as reported by subjects. Error bars show standard error of the mean.

closed conditions, resulting in only 4 levels of Von Frey hairs included in the eyes closed condition and in subsequent analyses. Thus, in the following analyses Von

Frey hairs have been numbered 1 to 4 in order of ascending thickness, but actually represent different Von Frey thicknesses (see Table 2.2). Despite having significant differences between zones (F(2.5,9.8)=9.127, Pc.oi, with Greenhouse-

Geisser correction) and Von Frey hair thicknesses (F(I.3, 5.I)=66.65, P<.001, with Greenhouse-Geisser correction), and a significant interaction between these

77 two variables (F(2.3, 9)=6.6i5, P<.05, with Greenhouse-Geisser correction), tactile perception was still unaffected and not significntly different between the two sides of the face (F(I,4)= .119, P>.1 uncorrected), nor was there a significant interaction among the ratings from facial regions ipsilateral to the occluded eye and other factors included in analysis (P>.1). Although these results reveal significant nuances in tactile perception across individual facial zones and across varying degrees of touch intensity, the lack of significance when investigating facial zones that are known to be experiencing paresthesias on the same side of the face as the occluding lens may be due to our data, which combines stimulus intensity ratings from subjects who do not report paresthesias in these facial zones, which may in turn dampen any significant effects of the paresthesias on tactile perception.

3.9.1.1: Tactile Sensitivity: new controls

The above analyses were not based on perfectly matched controls; ideally, control ratings should be obtained from the exact same facial zones and sites of stimulation as during the experimental condition, and that is precisely what we did in this section. Six subjects from the previous investigation (4 males, M=33.3 years, SD=±n.8i) participated in a second session that was on average 308 days after the original experimental session where they did not wear the occluding contact lens4. Subjects were stimulated only on facial regions ipsilateral to the

4 This control session obtained control ratings for both the Von Frey analyses in Section 3.9 as well as the Two Point Discrimination thresholds that will be discussed in Section 3.10.

78 side where they wore the occluding contact lens in the experimental session with the same Von Frey apparatus and with eyes closed only (Figure 3.13). No subjects reported unusual facial sensations. The stimulus intensity ratings from these six subjects were compared during the experimental condition (with the occluding contact lens in their dominant eye) and control condition (second session void of the occluding contact lens). Initial comparisons of stimulus intensity ratings from experimental and control sessions revealed no difference in tactile perception

(F(I,2)=O.O8I, P>.1, Figure 3.14). However, this analysis still averages across zones, time, thickness of Von Frey hair, and paresthetic and normal facial zones, all of which may occlude any significant effects of paresthesias on tactile perception.

79 Figure 3.13. New controls for tactile sensitivity analysis with Von Frey hairs. Stimulus intensity ratings obtained during two sessions: experimental (with occluding contact lens in dominant eye, left) and control (without occluding contact lens, right). The same five facial zones were stimulated in both sessions with the same Von Frey hairs. The control sessions consisted of 'eyes closed' conditions only, as it was established that stimulus intensity ratings obtained while subjects had their eyes open may have been confounded by the possibility of seeing the approaching stimulus, which has been known to affect tactile perception (Drevets et al., 1995). Average Stimulus Intensity Ratings- Experimental vs. Control Sessions (n=6)

4

Experimental Control

Figure 3.14. Average stimulus intensity ratings for ipsilateral zones obtained during experimental and control sessions. There is no significant difference between the stimulus intensity ratings provided during the experimental and control sessions. These values average across all subjects (n=6), facial zones (brow, eyelid, under eye, cheek, and lip), time (3 trials per session), Von Frey hair thickness, and paresthetic and non-paresthetic zones as reported by subjects. The stimulus intensity ratings included in these analyses are from eyes closed conditions only. Error bars show standard error of the mean.

Similar to the analysis conducted with the original ten subjects in our evaluation of tactile sensitivity, a more detailed statistical analysis was conducted on the intensity ratings from the second control session void of the occluding contact lens to determine if tactile perception was influenced by the presence of paresthesias in the experimental condition. In addition to including the five facial zones, four levels of Von Frey hair thickness, condition (experimental or control), and time as factors in the ANOVA, sex, ocular dominance, and contact lens use

81 were included as between-subjects factors. In addition to an expected significant

main effect of Von Frey hair (F(i.i, 2.2)=6o.698, Pco.os, with Greenhouse-

Geisser correction), there was also a significant main effect for time (F(I.5,

3.i)=20.739, P<0.05, with Greenhouse-Geisser correction), which showed a significant decrease in stimulus intensity ratings between the first and second

trial during both the experimental and control conditions (Pc.os) (Figure 3.15).

This decrease in stimulus intensity perception following the initial stimulation

trial only may reflect a habituation of subjects' sensitivity to the nature of the Von

Frey hairs, as it averages across conditions.

Average Stimulus Intensity Ratings (n=6): Time

*

Trial 1 Trial 2 Trial 3

Figure 3.15. Average stimulus intensity rating over time. Stimulus intensity ratings, averaged over facial zone, Von Frey hair, time, and condition, decreased significantly following the first trial, but not the second. Error bars show standard error of the mean.

82 Additionally, an interaction with zone and condition was significant when sphericity was assumed (F(4,8)=6.096, P<.05), and found to be trending towards significance when the Greenhouse-Geisser adjustment was applied (F(i.8,

3.5)=6.096, P=.073). These results parse out as many confounding factors as possible in order to ascertain whether or not the presence of paresthesias indeed effect tactile perception in affected facial regions, and despite reaching a near- significant interaction between affected zones between the experimental and control conditions, there remains to be seen a significant decrease in tactile sensitivity in the presence of paresthesias. We must also be careful not to interpret the opposing sides of the face as analogous. Perhaps there are subtle individual differences in contralateral facial nerve innervation that occlude any significant differences in tactile sensitivity that could be attributed to the presence of paresthesias. This was our rationale for conducting a separate control session that examined the same side of the face that experienced paresthesias when subjects wore the occluding contact lens.

3.9.2 - Tactile Acuity

The following analyses examine the effect of paresthesias on two-point discrimination (TPD) in twelve subjects (Table 2.1), as measured by stimulation of the same five facial zones as the Von Frey analysis but instead with a Two-Point

Aesthesiometer (Baseline ®). Here, we wanted to use a more common method of evaluating tactile perception in affected facial areas with two separate sessions for our experimental and control conditions, which was shown in the previous

83 evaluation to yield the most conclusive evidence for dampened tactile perception

in the presence of paresthesias. The TPD thresholds obtained from all twelve

subjects (n=i2) were analyzed according to the five facial zones ipsilateral to their

occluded eye, condition (experimental or control), time and order of stimulus

application (i.e., either in an ascending or descending order) and only facial zone

was found to be a significant main effect (F(2.4,14.6)=37.513, Pc.ooi, with

Greenhouse-Geisser correction, Figure 3.16), with the TPD thresholds obtained for the lip being significantly lower than all other facial zones (brow: Pc.05, eyelid: Pc.01, under eye: Pc.ooi, cheek: Pc.ooi), thresholds from the brow significantly higher than the eyelid (Pc.05), and thresholds from the cheek are higher than those for the eyelid (Pc.05). No significant differences were found among TPD thresholds during the experimental and control conditions, or according to any other variables included in our analysis (P>.i). These results confirm previous findings of increased tactile acuity and sensitivity of the buccal

region (Purves et al., 2008; Trulsson et al., 2010), as subjects are able to discriminate two points significantly better in the lip region compared to other skin covered regions of the face. What this evaluation demonstrates, however, is a consistent albeit non-significant trend of decreased TPD and increase in tactile acuity perception during the control session across all facial zones evaluated

(Figure 3.16).

84 Two-Point Discrimination Thresholds (n=12): Main Effects Analysis (Zone) 1.4 ;

1.2 EC), 22 o JC a>W

0c 0.8 '•s(0 c (Experimental 1 0.6 IControl 5Cfl Q c o 0.4 a. 6 0.2

Brow Eyelid Undereye Cheek Lip

Figure 3.16. Two-Point discrimination thresholds: main effects analysis. Two- point discrimination thresholds for the lip were significantly lower than for all other facial zones across both conditions, and there is a clear trend for lower thresholds during the control condition compared to the experimental condition, despite being non-significant. Error bars show standard error of the mean.

85 SECTION 4: DISCUSSION

With respect to addressing our initial hypotheses, we were able to successfully replicate and demonstrate the multimodal illusion shown by Wolfe et al. (2007, 2008) and Jobst et al. (2010) in all 17 of our subjects. Sensory paresthesias were demonstrated in facial regions immediately surrounding the occluded eye and subject self-reports were consistent with the descriptive reports of previous investigations, classifying the paresthesias as 'heavy', 'saggy', 'droopy', and 'numb' (Figure 3.4). It is important to note that while the paresthesias induced by Wolfe et al. (2007) were by dark adaptation, the abstract by Wolfe et al. (2008) and further abstracts (DeSouza et al., 2011a,b; Di Noto et al., 2010,

2011a,b; Jobst et al., 2010) employed an occluding contact lens to completely eliminate vision in one eye. On average, the paresthesias occupied approximately

9.42% (SEM=o.29%) of the face, with the highest degree of overlap among subjects being in the eyelid, cheek, under eye, and brow (Figures 3.1a, 3.5), and were sustained for the entire duration of our experimental trials which lasted on average 1.5 hours (Figure 3.2). The paresthesias also induced a compelling proprioceptive perceptual illusion of a drooping eyelid, as subjects reported significant closure of their occluded eye when compared to their unoccluded eye during the first experimental session (Figure 3.6) and when compared to the occluded eye during control sessions conducted on average 11 days later for the

TPD evaluation (Figure 3.7), and an average of 306 days after the Von Frey

86 experimental session (Section 3.9.1.1). This collection of results is the first

evaluation of how the paresthetic illusion first described by Wolfe et al. (2007,

2008) and members of my lab (DeSouza et al, 2011a,b; Jobst et al, 2010) affects

tactile sensory perception, specifically tactile acuity and sensitivity, and motor

coordination and control in affected facial areas surrounding the occluded eye.

The most compelling behavioural evidence for multimodal sensorimotor

modulation caused by facial paresthesias is demonstrated by the results of our

motor task, which revealed a significant increase in the time to perform an

alternating winking task in the presence of wearing the occluding contact

compared to when subjects were not wearing the occluding lens. For this task, subjects were asked to close one eye then open it, then close and open the other

eye, repeating this alternative behaviour as quickly as possible ten times per eye

(consisting of 20 alternating winks in total). The task was performed three times over the course of the experimental session while subjects wore the occluding lens

in their dominant eye, and the average time it took to perform this task was significantly less than the baseline time when subjects did not wear the occluding lens or report facial paresthesias. We propose that the mechanism for this

impairment in motor coordination is facilitated by efference copy signals sent from primary visual cortex, which no longer receives visual information from the

occluded eye, to multisensory processing regions in the posterior parietal cortex

(specifically, VIP and IPS), putamen, and ventral premotor cortex and the superior colliculus (SC), which mediate integration of visual, tactile, and

87 proprioceptive information (Colby et al., 1993; Duhamel et al., 1998; Graziano et

1993,1994; Ladavas et al., 1998a,b, 2004; Makin et al., 2007; VanderWerf et al., 2002) and eye and eyelid movement (Dauvergne et al., 2004), respectively.

These regions send action potentials back to primary facial sensory receiving areas in the brainstem (i.e., the mesencephalic trigeminal nucleus) to try and account for the loss of visual input in the absence of any sensory information from the occluded eye, resulting in the perception of paresthesias in the eyelid and facial areas immediately surrounding the occluded eye. We propose further that the decreased activation in mesencephalic trigeminal nuclei at the end of this multisensory feedback loop is an essential step in facilitating the motor component of our illusory paresthesia. Axon collaterals from the mesencephalic trigeminal sensory nucleus, specifically in the pars caudalis and pars interpolaris, have been shown to project to the adjacent facial motor nucleus in rabbits and rats (Pinganaud, Bernat, Buisseret, & Buisseret-Delmas, 1999; Tsuboi, Kolta,

Chen & Lund, 2003). If neural activity in the trigeminal sensory nucleus is suppressed during paresthesia, the action potentials sent by these collaterals to facial nerve nuclei further supresses motor activity and in turn alters the speed of signalling the muscles involved in opening and closing the eyes- the levator palpebrae superioris (LPS) and orbicularis oculi (00) (Irwin, 2011; Rubin et al.,

2007; Schuenke et al., 2007), thus mediating the delayed performance time exhibited in our winking task (Figure 4.1). Thus, the cortical activity in midbrain nuclei of the trigeminal nerve, which mediate the sensory aspect of the

88 paresthesia illusion, adds an unknown variable to the neural processes underlying our motor task, and the results of our winking task support the notion that the induced paresthesias could contribute to slowing down the process of executing movement of affected facial muscles.

Medial direction

CP^Ul < PLK fv R« NU\|f 'K-'Vf V I ir.fcrr.ipj! "mt*

ni Xi*«Ux ! rti« frii -V .w

rnrx >jul nm kits ul f .!< I.U MIX It'll* I'lvifmru!

V«MltNii.K nt»t

rwr\ -Vttrfmr jf«| jmsft nxhk-jr nut It IJOJJ iMTVf \ I.

Smj w"iti» in

x jc »»nt ncrvr A V J Vilrt.iiv lt.li t m* lt*j*

If l«u< Hti*

Y4»;u% TTT^r*- 1 l

Sp.n.»l '»!;« Ifyu< C#"*!-

( Hit* f«b*n yul'*: irwj f rv* <4'tu»tTir.»l r««**r Afffrtfll nbvrt iff Miud Uher*

Figure 4.1. Cranial nerve nuclei in the brainstem. A potential site of interaction with trigeminal nerve afferents and facial nerve efferents could be in the brainstem, dissected in the diagram above. Collateral nerve fibres that travel between these nuclei (Plnganaud et al., 1999; Tsuboi et al., 2003) could potentially mediate the delayed motor performance in our alternating winking task in the presence of facial paresthesias. From Massey (2006).

89 We offer that the increase in task performance time of our coordinated winking task during conditions where subjects wore the occluding lens was not attributed to the presence of a contact lens in only the dominant eye of each subject, since there was no significant difference in task performance time for subjects who were regular contact lens wearers (4 subjects wore clear corrective lenses on a daily basis and wore a clear corrective contact lens in their unoccluded eye during the experiment) and novice contact lens wearers (n=4 who are inexperienced with contact lenses and did not wear a contact lens in their unoccluded eye during the experiment) in our initial omnibus analysis (Section

3.7, P>.i), although a significant difference between conditions was only obtained when performing analyses without between-subjects factors. This lends to our postulation that the increase in performance time is due to the suppression of facial motor neuron signals that innervate the 00 and LPS muscles that mediate voluntary blinking and winking in the presence of paresthetic sensory perceptions. This is supported by the anecdotal commentary of three out of eight subjects that stated winking and closing their occluded eye required more "force",

"gain", or "effort" than their unoccluded eye or their dominant eye when unoccluded during normal conditions. We propose that efference copy signals sent from multisensory processing areas or from adjacent sensorimotor nuclei to facial nerve nuclei in the brainstem cause an overall suppression of neural activity following suppression or inhibition of visual afferents that is induced by monocular occlusion with our contact lens. Evidence for voluntaiy and

90 involuntary blinking behaviour has been shown to activate the rostral cingulate area (Hanakawa et al., 2007) and is mediated by the motor signals of the facial nerve, which may be one of the potential multisensory regions inhibited or receiving less excitatory visual, proprioceptive, and tactile information from areas affected by the paresthesias induced in our experimental condition. Although we can only still speculate about the neural mechanism underlying the visual, proprioceptive, and tactile illusion demonstrated in our studies, and how it impacts our coordinated winking task, this is the first evidence of such an artificially-induced phantom affecting motor behaviour.

The proposed neural mechanism underlying this complex, multimodal illusion can be understood by examining how the individual unimodal afferents are affected by monocular blindness, and how these converging signals could potentially feed-forward to higher-level unimodal areas or feed-back to primary sensory receptors to induce paresthesias and proprioceptive and motor deficits.

The facial areas that subjects reported as paresthetic are all innervated by the trigeminal nerve, albeit by different branches (ophthalmic and maxillary). These afferents are forwarded from primary trigeminal nuclei in the brain stem to multisensory processing areas VIP and IPS in the PPC, putamen, and ventral premotor cortex, which are known sites of convergence for visual, somatosensory, and proprioceptive information, and project further to primary sensory processing areas as well as back to the midbrain nuclei (Colby et al., 1993;

Duhamel et al., 1998; Graziano et al., 1993,1994; Ladavas et al., 1998a,b, 2004;

91 Makin et al., 2007). The complete lack of vision from one eye following insertion of the occluding lens effectively halts afferents to primary visual cortex responsible for processing vision from the dominant eye. Additionally, there is a lack of touch signals from the face that would account for the occlusion of the dominant eye due to the lack of a physical occluder like a hand or eye patch, and following habituation of the ciliary nerves on the cornea to the occluding lens, which we know occurs very quickly (Millodot, 1972,1975,1976). This suggests an increase in the electric neural weight of proprioceptive signals in multisensory regions, which no longer compete for finite attentional resources with vision and touch and effectively mediates the salient proprioceptive illusion of a drooping eyelid that remains to account for the absence of visual input with supporting sensory signalling.

This proposed neural mechanism is supported by the rationale of the maximum-likelihood estimate (MLE) of multimodal convergence; if vision has a higher weight with respect to influencing multimodal perception, then eliminating this strong neural signal could result in significantly reduced overall activity in multimodal areas, which is perceived as a reduction in activity of the other, lesser-weighted modalities that are commonly processed in multisensory cortical regions (Ernst et al., 2002; Fame et al., 2000; Ladavas et al., 1998a,b,

2004), and in effect facilitating the perceived decrease in sensory input from facial areas surrounding the occluded eye, and the droop of eyelid muscles of the occluded eye only. This is supported by studies that use monocular patching in

92 cats to show stronger contralateral spatial orienting, known as the Sprague effect

(Sprague, 1966). They propose that this spatial bias is induced by the decrease in competing converging sensory afferents to the superior colliculus from both eyes that in turn results in a unilateral decrease in intracollicular activity that induces contralateral orienting (Chen, Erdahl & Barrett, 2009). Although these studies have been conducted on non-human species, similar adjustments in orienting responses have been demonstrated in human clinical populations, where concurrent presentation of stimuli in intact and damaged modalities will enhance detecting of the damaged modality (Ladavas et al., 1998a,b, 2004). Thus, the monocular blindness induced in our experiments could cause similar interferences in multimodal neural processing, exerting strong influences on tactile and proprioceptive perception due to the loss of visual afferents to this system and lack of explanatory somatosensory and proprioceptive feedback provided by using a contact lens instead of an eye patch.

The fact that the dominant eye is completely occluded provides no visual feedback that the eye is completely closed when subjects perform voluntary closure of their occluded eye; in fact, several subjects explicitly stated they were unsure if they were closing their occluded eye, both while performing the winking task, and also during the questionnaire period when they were asked to actively close their occluded eye to determine if this sensory and motor feedback changed perceived paresthesias (Figure 3.10). The lack of visual information from the occluded eye while both eyes are open during passive viewing and following

93 blinking and winking are vital to subjects' perceptual experience with the paresthetic illusion, as the introduction of cognitive and visual feedback (asking subjects to examine themselves in a mirror, and seeing with their unoccluded eye that their blind eye is actually occluded) significantly reduced subject reports of paresthesias (Figure 3.11), as did the introduction of somatosensory and proprioceptive feedback (Figure 3.10). The visual signals from the unoccluded eye now provide information as to why the occluded eye receives no visual input in the absence of tactile input from the surrounding facial regions that would account for the monocular occlusion, since sensory perception of the occluding contact lens on the cornea is habituated within seconds (as shown by work of

Millodot, 1972, 1975, 1976). This demonstrates the vitality of multisensory feedback and integration in sensory and proprioceptive perception of facial regions surrounding the eyes during normal conditions, and the immediate updating of this network following a perturbation of these conditions, as in our case when there is a disparity in vision, specifically luminance, to both of the eyes is induced following insertion of an occluding lens to one eye only. The lack of feedback from the lens following habituation of the ciliary nerves on the eyeball force the multisensory network to account for the lack of vision from an otherwise healthy and normally viewing eye, thus inducing the symptoms of illusory paresthesias, so-called because removal of the occluding lens and restoration of vision to the occluded eye effectively and immediately eliminates all perceptions of paresthesia in the affected eye and facial areas in all subjects.

94 Consistent with previous examinations of facial paresthesias and perceptual illusions induced by dark adapting one eye (Cassidy et al., 2010;

Schawerna et al., 2010; Wolfe et al., 2007), introducing various forms of tactile, proprioceptive, and motor cues that account for the difference in luminance reaching both eyes were found to significantly reduce paresthetic sensations in the eye and face area surrounding the occluded eye (Section 3.8). Interestingly, asking subjects to view themselves in a mirror, and providing them with visual and cognitive information that explains the monocular occlusion, resulted in significant reductions in subjects' sense of blindness (i.e., recovery of binocular vision and the sense of viewing normally through both eyes), which is similar to literature on perceptual illusions that effectively 'trick' the brain into believing what is seen and not otherwise perceived by the remaining intact senses (Figure

3.11). Other studies have shown that viewing a prosthetic arm in an anatomically congruent position to a subject's own real arm can facilitate the experience of illusory somatosensory perceptions (Botvinick et al., 1998; Fame et al., 2000).

Our subjects observing themselves in a mirror and seeing that their eye is occluded and not physically displaying the sensory or proprioceptive symptoms they perceive (i.e., drooping eyelid) significantly reduced or eliminated the illusory paresthesias altogether (Figure 3.11). This supports our hypothesis that the paresthesia illusion induced by the occluding contact lens is mediated and can be overridden by similar neural mechanisms that underlie phantom limb illusions

(Ramachandran et al., 1996).

95 With respect to the influence of this illusion on tactile perception, there is

no difference in either two-point thresholds or stimulus intensity ratings in facial

areas affected by the paresthesias (Section 3.9). The fact that tactile sensory

perception is not affected by the very salient reports of paresthesias in facial areas

(Sections 3.2 to 3.6) further supports our suggestion that a lack of visual inputs

from the occluded eye to multisensory regions in the PPC, where visual, tactile,

and proprioceptive afferents have been shown to converge (Colby et al., 1993;

Duhamel et al, 1998; Makin et al, 2007), or a decrease in overall activity among adjacent sensory and motor brainstem nuclei consequently dampen sensory signals from ipsilateral facial regions surrounding the occluded eye but are overridden and perceive normally when stimulated. We have already concluded that the neural mechanism of the illusion can be overridden by introducing tactile, proprioceptive, and visual feedback to the occluded eye (Section 3.8), and the fact that investigations of tactile perception with both Von Frey and Two-

Point Aesthesiometers find no significant differences in perception while experiencing paresthesias compared to control conditions (both within and across experimental sessions) support our conclusion that this illusion does not impede perception of touch stimuli applied to putatively affected facial regions (Section

3.9.1 and 3-9-2). Through these evaluations we are providing additional haptic signals to multisensory regions that may be involved in mediating the paresthesias, and these haptic perceptions could have potentially overridden the paresthesias by compensating for the lack of signals coming from the regions

96 around the occluded eye. A potential parallel in the literature may be evident in brain-damaged patients with extinction, such that the lack of sensory perception in one modality (in our case, vision) causes an interference in perception of another modality (touch) that is overridden when presented with concurrent bimodal stimulation (Fame et al., 2000; Ladavas et al., 1998a,b; 2004). Since subjects expected to be stimulated on their faces, even when instructed to close their eyes, the anticipatory and eventual touch signals to tactile processing areas could override any sensations of paresthesia, resulting in intact perception of applied touch stimuli.

From the tactile sensitivity evaluation with the Von Frey hairs, we can see that stimulus intensity ratings, and touch perception, was not affected by the presence of facial paresthesias (Section 3.9). As expected, significant differences in stimulus intensity ratings were found between facial zones; subjects reported significantly higher intensity ratings, and thus more salient tactile perception and lower tactile sensitivity thresholds, in the lip compared to other facial regions.

Although there has not been any study that systematically evaluates sensitivity thresholds in all of the presently examined facial areas under normal conditions in humans, the literature on tactile acuity has mapped the receptive field size in parts of the upper lip and cheek (Johansson et al., 1988; Nordin et al., 1989).

From these studies we can see that receptive fields have a very high degree of overlap between those that are very small (1.5cm2) and large (>25cm2 spanning the upper lip to lateral orbit). Based on our results of statistically non-different

97 two-point thresholds for paresthetic and normal facial areas, we can conclude that the touch signals from the two-point aesthesiometer activated separate receptive fields within each of the five facial zones evaluated, and that cutaneous signalling and resulting two-point perception remains unimpeded in the presence of sensory paresthesia in these same facial areas. We can also infer that the salient sensory signals sent from paresthetic facial areas following stimulation with the

Von Frey hairs activated slow-adapting receptors, which continuously fire action potentials during a sustained stimulus, resulting in an intact perception of touch when compared to control conditions (Hensel, 1974; Johnson, 2002; Kennedy &

Inglis, 2002; Purves et al., 2008). Additionally, two of the ten subjects stimulated by the Von Frey hairs and three of the twelve subjects stimulated with the Two-

Point Aesthesiometer reported some discomfort upon application of the stimulus.

Despite reporting that they ignored this discomfort when providing their stimulus intensity and two-point discrimination responses, these responses could be biased based on evidence of differential perception of noxious versus innocuous tactile stimuli, with the former putatively activating insular and limbic cortical areas like the anterior cingulate cortex in addition to primary and secondary somatosensory areas (Purves et al., 2008; Rainville, Duncan, Price, Carrier &

Bushnell, 1997), which provide an alternate pathway for tactile perception that does not involve the primary tactile areas (trigeminal nuclei in brainstem and thalamus, primary somatosensory cortex) whose neural activity may be suppressed following induction of our illusory paresthesia.

98 The only significant changes in paresthesias over time found in our entire

analyses were in Section 3.9.1.1 (Figure 3.15), where average stimulus intensity

ratings decreased overall (i.e., when averaged across both conditions, all facial

zones, and all Von Frey hairs) after the first of three trials during the experiment,

but not between the second and final trials, or the first and third trials. This may

reflect subjects' familiarization with and habituation to the nature of the Von Frey stimulation, as subject reports of paresthesias around the eye (Section 3.4), in the face (Section 3.5), and as reported in subject illustrations (Section 3.3) did not attenuate significantly over time.

On average, subjects most often reported sensations of paresthesia to be localized to the area of the face immediately surrounding the occluded eye (Figure

3.1a), which suggests that the tactile receptive fields most affected by the

monocular blindness overlap with the visual fields of the occluded eye. However,

2 of 17 subjects reported shifting of the overall area of the paresthesia when asked to move both eyes to the extreme right and left visual fields. That the paresthesias did not shift with eye position in the majority of 15 of 17 subjects, however, is consistent with the work of Graziano in macaques (et al., 1993; 1999), demonstrating that limb-centered somatic receptive fields don't move with eye

position, but rather move with the body part. However, this is difficult to apply to the face, as subjects cannot 'see' their own face and respond to its motion. This shift in paresthesia was not investigated further, but was noted when reported explicitly and spontaneously by the two subjects during the feedback task portion

99 of the experiment (see Section 2.3.5). Future investigations could involve subjects mapping the region affected by paresthesias while instructed to fixate on various areas in their visual field to determine whether eye position does indeed influence localization of the facial sensory paresthesia by explicitly asking subjects to focus on this aspect of the illusion.

In all of the present analyses we included three factors to ascertain if they had an impact on the presence and extent of the multisensory illusion we have observed: sex, ocular dominance (i.e., if the occluding contact lens is inserted into either the right or left eye), and whether subjects regularly wear contact lenses.

The latter variable was included to see if experienced contact lens wearers had any differential perceived paresthesias from novice contact lens wearers, who may be reporting paresthesias simply due to the unfamiliar feeling of wearing a contact lens in their eye. We also asked all regular contact lens wearers to wear their own clear corrective lenses in their unoccluded eye during the experimental condition to control for the fact that paresthesias may be reported due to the disparity in sensory feedback induced by the presence of a contact lens in only one eye, and also to wear both corrective lenses during control sessions. Neither this variable, nor any of the other between-subjects variables included in our analyses, yielded any significant conclusions in any of our investigations (P>.05).

This allows us to conclude that the paresthetic illusion is equally observed between males and females, in subjects with both left and right ocular dominance, in people who are experienced with wearing contact lenses just as much as those

100 who are inexperienced with wearing contact lenses, and that the paresthesias are

not simply the result of having a contact lens in only one eye. Ideally, however, future studies will include controls for differential sensory activation of ciliary

nerves of the unoccluded eye such that subjects will be asked to wear a contact lens (either clear or otherwise depending on the nature of the investigation) that is the same thickness and of the same dimensions as the occluding lens, although our results do not appear to be significantly impacted by this difference.

Similar symptomology and descriptions of facial numbness coupled with abnormal blink reflexes have been described in clinical populations afflicted with

Bell's Palsy, or idiopathic facial palsy. Damage to the facial nerve at various points along its course, from the brainstem to its exit points through the skull, result in paralysis and loss of sensory input from the face. Patients with Bell's Palsy have been shown to have dysfunctional trigeminal evoked potentials in addition to presenting with unilateral facial paralysis and asymmetry of facial movement and of facial expression at resting state, consistent with our hypothesis that facial paresthesias involve both sensory and facial motor brainstem nuclei and with our findings of impaired facial motor control in the presence of paresthesias (Adour,

Byl, Hilsinger, Kahn & Sheldon, 1978; Billue, 1997; Hanner, Badr, Rosenhall &

Edstrom, 1986). Although the first two of these studies found abnormal trigeminal sensory function in Bell's Palsy patients, our findings yielded non­ significant differences in tactile acuity and tactile perceptual thresholds with induced paresthesias. Consistent with Billue (1997), however, the results of our

101 alternating winking task demonstrate a significant impairment in the ability to

perform a voluntary wink in the presence of paresthesias in the occluded eye only,

consistent with the pathological symptoms of facial nerve palsy. The fact that

sensory inputs, motor outputs, and involuntary blink reflexes are all affected by

Bell's palsy and are also similarly influenced by our illusory paresthesia imply

that a similar brainstem network may be at play in both circumstances.

An anecdotal finding revealed during nine of the 66 total experimental

sessions from three of our 17 subjects was reporting of some ambient light

detection though the occluded lens, although they were unable to perceive any

form or movement. The subjects were carefully questioned about whether they

could see the experimenter's hand move in front of their occluded eye while they

also covered their unoccluded eye, and subjects could not detect this motion.

However, they were able to detect whether the lights in the experimental room

were turned off, as well as detect a change in luminance when they moved their

own hand in front of their occluded eye while covering their unoccluded eye. The former detection in luminance can be attributed to the fact that subjects likely

experienced dark adaptation in their occluded eye, which is a biphasic process

and initially occurs as quickly as 5 to 8 minutes following exposure to a darkened environment and then shifts into a slower process that takes approximately 30 to

40 minutes to fully adapt (Pirenne, 1962), and experimental sessions lasted 1.5

hours on average. The latter luminance detection may have been facilitated by

proprioceptive signals from the subjects' own arm and hand or from efference

102 copy signals to multisensory areas, signalling an orientation in front of the occluded eye that could account for the perceived change in luminance. This suggestion is supported by the fact that subjects were not able to detect a change in luminance when the experimenter waved their hand in front of the subjects' occluded eye. Nonetheless, these subjects reported paresthesias throughout the duration of the experiment and at no time did they report seeing any scattered light come through or around the edges of the occluding contact lens. Thus, their reports of detecting changes in luminance were not deemed to impede their ability to experience the paresthetic illusion and they remained included in our analysess.

A potential confound attributed to our mechanical stimulation methods is the activation of mechanoreceptors by moving facial hair. Any interference of facial hair (i.e., male beards or moustaches or contact with eyebrows or eyelashes in all subjects) reported by subjects resulted in a repetition of administering the stimulus until subjects did not report interference and provided a stimulus intensity rating or two-point detection response. Hairy skin is innervated by a special system of unmyelinated mechanoreceptors that are highly sensitive to perturbation and has been described in other mammals as well as in humans

(Vallbo, Olausson & Wessberg, 1999), and could pose a potential confound to our results on tactile sensitivity and acuity. However, only three male subjects had facial hair, and of the three female subjects who reported interference with

5 The ability to see light around the edges of the occluding contact lens was considered to be exclusion criteria, and resulted in one initial test subject (JS) to be excluded from the experiment and their experimental session was stopped immediately following insertion of the occluding lens.

103 eyelashes or eyebrows, stimulations were repeated until no interference was reported, and these were the responses included in our final analyses.

In sum, the subjective reports of paresthesias induced by monocular occlusion with a black opaque contact lens have solidified this perceptual illusion as a very salient phenomenon that is perceptible by all subjects examined (n=i7) in all 66 experimental sessions. We have offered several potential sites of action for this multimodal illusion, which encompasses the tactile and proprioceptive sensory systems, as well as imparting an influence on facial motor behaviour. We propose that the neural mechanism underlying illusory facial paresthesia is common to all of these modalities, and is likely similar to the mechanisms underlying "phantom" phenomena observed in amputees (Ramachandran et al.,

1992, 1996) and induced in non-human primates (Graziano, 1999), healthy

(Botvinick et al., 1998; Wolfe et al., 2007) and clinical (Fame et al., 2000) human populations. This supposition is supported by the reduction in perceived paresthesias with the introduction of tactile, proprioceptive, visual, and cognitive feedback cues that account for the difference in luminance reaching both eyes induced by the occluding contact lens. Significant reports of drooping and closure of the occluded eyelid provide compelling evidence for the proprioceptive component of this illusion, perhaps reflecting a reduction in overall neural activity in multisensory regions like the PPC that process visual, tactile, and proprioceptive perception. That these illusory sensations attenuate with the introduction of feedback cues also suggest that they are taking advantage of a

104 suppressed and/or diminished level of cortical activity in multisensory regions that are otherwise activated by our most dominant sense- vision. Our novel coordinated winking task also provides the first evidence in our line of research

(DeSouza et al., 2011a,b; Di Noto et al., 2010, 2011a,b; Wolfe et al., 2007) that demonstrates how this perceptual tactile and proprioceptive illusion impacts motor behaviour, specifically the ability to perform a winking task that involves the same facial structures affected by the multimodal paresthesias (see Section

3.7), which may be mediated by the neurons between adjacent sensory and motor facial nuclei in the brainstem. Similarly, the perception of salient tactile stimulation to affected facial areas is not hindered, suggesting that the paresthesia illusion can be overridden by incoming sensory signals and does not affect sensory perception or acuity (see Sections 3.9.1 and 3.9.2). It is also worthy to note that repeated administration of tactile stimulus does not result in a decrease in the paresthesias, as the illusion is sustained for the duration of our experimental sessions for an average of 1.5 hours. Together, these results supplement the current literature on perceptual changes immediately following monocular occlusion, which is very limited and more focused on long-term changes in visual processing areas. These paresthesias and proprioceptive perceptions are immediately apparent, persistent over time, attenuate significantly with the introduction of feedback from more salient modalities, and disappear immediately following removal of the occluding contact lens.

Understanding the neural mechanisms that mediate this complex multimodal

105 illusion will help us understand similar perceptual phenomena in patients experiencing phantoms and extinction, and will provide the framework for understanding the neural network that allows our plastic brains to account for sudden changes in inputs from a given modality by compensating for such shifts in perception through action in our remaining intact modalities.

SECTION 5: FUTURE DIRECTIONS

The most effective way to gather evidence for this facial phantom illusion, frankly, is to test as many subjects as possible and demonstrate that we can induce the illusory paresthesias in all of them. Together with members of my lab at York University and in collaboration with Uta Wolfe, we have successfully demonstrated this perceptual illusion in a total of 44 adult human subjects

(DeSouza et al., 2011a,b; Di Noto et al., 2010, 2011a,b; Jobst et al., 2010; Wolfe et al., 2008). The methods with which we evaluate subjects once wearing the occluding contact lens, however, should be defined prior to recruitment and experimentation. Although we do not yet have evidence to support an attenuation of the extent or presence of paresthesias with repeated exposure to the occluding lens (some subjects have worn the occluding lens in their dominant eye during multiple experimental sessions, the highest being 6 times prior to inclusion in this study, see Table 2.1), we cannot rule this out as a possibility until longitudinal evaluation of existing test subjects has been conducted.

106 Our results on the impediment of facial motor behaviour as a result of illusory facial paresthesias also provide a promising line of research on the interaction of multisensory perception and coordinated facial motor behaviour.

Some preliminary data has already been collected that is based on an unpublished behavioural model developed by a member of my lab at York University that predicts a shift in the overall area of the paresthesia in the same direction as lateral eye movements. This was also incidentally demonstrated in two subjects when asked to move both of their eyes to the extreme left and right visual fields.

But by utilizing an eyetracker, we can easily design tasks that will measure different aspects of eye movements, such as accuracy and latency, and compare whether the presence of paresthesias affect the output of these behaviours. The

Sherman Health Sciences Center at York University is already equipped with an eyetracker (SMI iView Red PT) within the functional neuromagnetic imaging

(fMRI) facility, and combining eyetracker-based tasks with functional neuroimaging will provide even more insight into the potential oculomotor influences and neural mechanisms underlying this perceptual illusion.

In order to validate the neural mechanisms that we hypothesize to mediate this illusory phantom phenomenon, we can utilize fMRI to identify regions of the brain that are functionally responsible for mediating the proprioceptive and haptic sensations and motor deficits observed in the present study. The occluding contact lens is safe for use in magnetic neuroimaging facilities, and some preliminary findings of exploratory nature have been presented at a poster

107 session at the Plastic Vision Conference held by the Centre for Vision Research at

York University (DeSouza et al., 2011b). As the first studies to examine the cortical activity of this specific phantom phenomenon, the regions of interest explored would be nonetheless well-informed based on what we already know about the neural mechanisms that underlie the individual sensory modalities that are affected by it. Based on multisensory literature and the nature of the illusion, we can look to contralateral primary sensory processing areas, such as the somatosensory cortex (areas 3a and 3b) and primary visual cortex

(striate/occipital lobe), as well as multisensory regions like the IPS and VIP within the posterior parietal cortex as starting points. By scanning subjects during separate experimental and control conditions (with and without the occluding contact lens, respectively), and performing similar tasks that evaluate tactile perception, facial and oculomotor behaviour, and incorporating an eyetracker to examine potential effects of this illusion on saccade generation, accuracy, and timing, we will be able to confirm our proposed mechanism and determine whether or not there is, indeed, a suppression of signalling in multisensory areas and a reduction in neural activation between sensory and motor brainstem nuclei and in primary somatosensory cortices while subjects perceive paresthesia. These investigations and collaborations with members of the Centre for Vision Research who work with clinical populations of monocular viewers and patients with oculomotor defects such as Jennifer Steeves and Martin Steinbach will allow us to better understand the neural mechanisms that facilitate multimodal integration

108 in normal populations, monocular viewers, and those affected by phantom phenomena. By investigating these adaptive and immediate processes with objective and innovative tasks and analytical methods, I intend to actively contribute to the advancement of perceptual research by untangling this complex yet elegant and efficient series of neural pathways.

109 Appendix A

SUBJECT: TIME INTERVAL: COPTCAGTIN MM / TlMMOtfl?A0riN/OUT: REGULAR CONTACT WEARER: # OF TIMES CONTACT WORN BEFORE: *Insert contact and give subject 1 -2 minutes to get used to wearing contact Once comfortable, subject will lie down on table and answer the following questions: TIME: Describe the following sensations, eit ler Yes or No, or descriptively Contact Control Subjective Reports Eye Eye EyeliMeels wesy?* Eye closed?* Byalfffitftufrhh?* HIV Can you see? i-' ••• .•••<, ,, IWSlSPlJHbytaelff Eye open wider than usual

Eyelid "painful" voi'M.ii

*.•' ••=• '

Do the following areas eel unusual, either Yes or No, or descriptively? Contact Control Subjective Reports Eye/Side Eye/Side BMkW :: • ;i '' Eyelid >' .. ' ' ; 1 .• MV.;; •. i 1- V" ; ' *•<'" . .... Cheek flu';-,': •' Ifll Jaw m- -! tm:,*» Ear

Top of head fi^'lfA ,» * v " ' • • 1 • , .ink'.' ».•* Forehead Von Frey detection threshold level: -4 [no detection at all] < 0 [unsure] < +4 [definite detection] *Hairs 3 (lightest) to 9 (densest) used, each zone tapped once with sustained pressure, randomized between left and right sides, and eyes open or closed. Left: CONTACT EYE/SIDE Right: CONTROL EYE/SIDE Shaded blocks: EYE CLOSED Hairs ->

A. Brow

B. Eyelid

c. Undereye

110 D. Cheek

E. Lip TIME:

Cover CI (hand)

WW*™ '4ii v.- Actively close CI lijlW "'•* ' . 1 1 !*.!•"< fH- 1 r . > ' >'!. ' " " Left Extreme Orbit

Blink x3;then open TIME: Left Eyelid Position (0% = Completely Closed ; 100% = Completely Open) udge your eyelid position Percentage (% ftrcenCBiin (%):

Right Eyelid Position (0% = Completely Closed ; 100% = Completely Open) Judge your eyelid position Percentage (%):

.i,. ^

TIME: Mirror Sensation vs. Gone Reduced Unchanged Increased Strong Detection Increase

: i .,y ^' 'v ' l jv. • . ,.s; , ..j; ...... j Numbness

.4, .. Blindness

TIME:

111 Appendix B

Please indicate on the drawing and changes in sensation.

112 113 Appendix C

SUBIECT: TIME INTERVAL: CONTACT IN RIGHT / LEFT EYE: TIME CONTACT IN / OUT: REGULAR CONTACT WEARER: # OF TIMES CONTACT WORN BEFORE: 'Insert (condition) and give subject 1-2 minutes to acclimate. Once comfortable, subject will lie down on table and answer the following questions: TIME: Describe the followin K sensations, e ther Yes or No, or descriptively Contact Eye Control Eye Subjective Reports Eyelid feels saggy?* Eye closed?* Eyelid oumb?* Can you see? Eyelid "droopy"* Eyelid "painful'' ByeHdcoi^iMy normal Do the following areas feel unusual, either Yes or N o, or descriptively? Contact Eye/Side Control Subjective Reports Eye/Side Brow Eyelid Undenan Y • ' • . • Cheek Temple law UP Ear NOM Top/Back of head

TIME:

Subjective Changes in Illusory Sensations Cover CI (hand)

CowrO«k«rlQuuid) v. Actively close CI

Left Extreme Orbit

WlfrtelriditOrbit Blink x3; then open

TIME:

114 Left Eyelid Position (0% = Completely Closed ; 100% = Completely Open) Judge your eyelid position Percentage (%): How certain are you of this estimate? Percentage (%):

Right Eyelid Position (0% = Completely Closed ; 100% = Completely Open) ludge your eyelid position Percentage (%): How certain are you of this estimate? Percentage T%):

Mirror Sensation vs. Gone Reduced Unchanged Increased Strong Detection Increase Sagging Numbness Tingling Blindness TIME:

Two-point Discrimination Subjects will close their eyes, and the aesthesiometer will be applied at a 90 angle to each of the 5 facial zones. 1=1 point, 2 = 2 points reported Ascending - Descending - AVERAGE CONTACT CONTACT BROW 1 point 1.0an 0.3cm 0.9cm o.4cm 0.8cm 0.5cm 0.7cm 0.6q» 0.6cm 0.7cm 0.5cm OJcW OtfcBl 0.9cm 0.3cm ijOcm Ipfltoit EYELID 1 point 1.0cm 0Jdfe:.;v O^teai 0.4cm 0.8cm OSC® '' 0.7cm 0.6cm 0.6cm 0.7c® 0.5cm 0.8cm 0.4cm 0.9cm 1.0cm 1 point UNDERETC lute 'IMsm v 0.3cm 0.9cm

0.5cm 0.7cm

0.6an ;..^£ . 0.7cm 0.5cm 03cm i" 0.9cm 0.3cm

115 1.0cm 1 point CHEEK 1 point 1.0cm 03cm 0.9cm 0.4cm 0.8cm 0.5cm 0.7cm 0.6cm 0.6cm 0.7cm 0.5cm 0.8cm 0.4cm 0.9cm 0.3cm 1.0cm 1 point UP 1 point 1.0cm 0.3cm 0.9cm 0.4cm 0.8cm 0.5cm 0.7 cm 0.6cm 0.6cm 0.7cm 0.5cm 0.8cm 0.4cm 0.9cm 0.3cm 1.0cm 1 point

116 REFERENCES

Adour, K.K., Byl, F., Hilsinger, R., Kahn, Z., and Sheldon, M. (1978). The true

nature of Bell's palsy: Analysis of 1000 consecutive patients. Laryngoscope, 88,

787-801.

Alighieri, Dante. (2003). Inferno: La Divina Commedia. Bologna: Zanichelli.

Alvarado, J.C., Stanford, T.R., Rowland, B.A., Vaughan, J.W., and Stein, B.E.

(2009). Multisensory integration in the superior colliculus requires synergy among corticocollicular inputs. The Journal ofNeuroscience, 29(20), 6580-6592.

Anastasio, T.J., Patton, P.E., and Belkacem-Boussaid, K. (2000). Using Bayes'

rule to model multisensory enhancement in the superior colliculus. Neural

Computation, 12,1165-1187.

Avillac, M., Deneve, S., Olivier, E., Pouget, A., and Duhamel, J.R. (2005).

Reference frames for representing visual and tactile locations in parietal cortex.

Nature Neuroscience, 8,941-949.

117 Avillac, M., Deneve, S., Olivier, E., Pouget, A., and Duhamel, J.R. (2007).

Multisensory integration in the ventral intraparietal area of the macaque monkey.

The Journal ofNeuroscience, 27(8), 1922-1932.

Bavelier, D., Brozinsky, C., Tomann, A., Mitchell, T.} Neville, H. and Liu, G.

(2001). Impact of early deafness and early exposure to sign language on the cerebral organization for motion processing, The Journal ofNeuroscience, 15,

8931-8942.

Bavelier, D., Dye, M.W. and Hauser, P. C. (2006). Do deaf individuals see better?

Trends in Cognitive Science, 10,512-518.

Belmonte, C., Garcia-Hirschfeld, J., and Gallar, J. (1997). Neurobiology of ocular

pain. Progress in Retinal and Eye Research, 16(1), 117-156.

Belmonte, C., Acosta, M.C., and Gallar, J. (2004). Neural basis of sensation in intact and injured corneas. Experimental Eye Research, 78,513-525.

118 Bessou, P., and Perl, E.R. (1969). Response of cutaneous sensory units with

unmyelinated fibers to noxious stimuli. Journal of Neurophysiology, 32,1025-

1043-

Billue, J.S. (1997). Bell's Palsy: an update on idiopathic facial paralysis. The Nurse

Practitioner, 22(8), 88-105.

Botvinick, M., & Cohen, J. (1998). Rubber hands 'feel' touch that eyes see. Nature,

391,756-

Bour, L. J., Aramideh, M., & Ongerboer de Visser, B. W. (2000).

Neurophysiological aspects of eye and eyelid movements during blinking in

humans. Journal of Neurophysiology, 83,166-176.

Brennan, N. A. and Bruce, A. S. (1991). Esthesiometry as an indicator of corneal health. Optometry and Vision Science, 68,699-702.

Cassidy, B.N., Schawerna, F.L., and Wilkinson, F. (2010). Phantom eye loss and

related illusions in asymmetric dark adaptation. Poster presented at meeting for

Canadian Society for Brain Behaviour and Cognitive Science, Halifax.

119 Chen, P., Erdahl, L., and Barrett, A.M. (2009). Monocular patching may induce ipsilateral "where" spatial bias. Neuropsychologia, 47(3), 711-716.

Colby, C.L., Duhamel, J.R., and Goldberg, M.E. (1993). Ventral intraparietal area of the macaque: anatomic location and visual response properties. Journal of

Neurophysiology, 69, 902-914.

Cooper, S., and Daniel, P.M. (1949). Muscle spindles in human extrinsic eye muscles. Brain, 72(1), 1-28.

Cooper, S., Daniel, P.M., and Whitteridge, D. (1955). Muscle spindles and other sensory endings in the extrinsic eye muscles; the physiology and anatomy of these receptors and of their connexions with the brain-stem. Brain, 78,564-583.

Crider, B. (1944). A battery of tests for the dominant eye. The Journal of General

Psychology, 31,179-190.

Dauvergne, C., Ndiaye, A., Buisseret-Delmas, C., Buisseret, P., VanderWerf, F., and Pinganaud, G. (2004). Projections from the superior colliculus to the trigeminal system and facial nucleus in the rat. The Journal of Comparative

Neurology, 478,233-247.

120 DeSouza, J.F.X., Kucyi, A., Pynn, L., Jobst, C., Di Noto, P., Keith, G., and Wolfe,

U. (2011a). A multisensory visuotactile illusion induced by monocular occlusion with a black contact lens does not depend on touch signals on the face: evidence from behavioural and modelling studies. Journal of Vision, 11(11), 780.

DeSouza, J.F.X., Bar, R., Di Noto, P., Jobst, C., Kucyi, A., Pynn, L.K., Mirzoev, A.,

West, A., and Wolfe, U. (2011b). Brain regions involved in a multisensory illusion of the face induced by occlusion of the dominant eye: evidence from fMRI. Poster session presented at International Conference on Plastic Vision, Toronto.

Di Noto, P.M., and DeSouza, J.F.X. A multisensory visuotactile illusion induced by monocular occlusion with a black contact lens: tactile thresholds. Program No.

579.1. 2010 Neuroscience Meeting Planner San Diego, CA: Society for

Neuroscience, 2010. Online.

Di Noto, P.M. and DeSouza, J.F.X. (2011a). Investigating a visuotactile illusion induced by monocular occlusion with a black contact lens: quantitative analyses.

Presentation session presented at Canadian Physiological Society-CAPNet Winter

Meeting, Ste. Adele, QC.

Di Noto, P.M., West, A., and DeSouza, J.F.X. Investigating the influence of visuotactile illusory paresthesias on tactile acuity (two-point discrimination) and

121 psychomotor performance (alternating winking task). Program No. 267.10. 2011

Neuroscience Meeting Planner. Washington, DC: Society for Neuroscience,

2011b. Online.

Drake, R.L., Vogl, W., and Mitchell, A.W.M. (2005). Gray's Anatomy for

Students. Philadelphia: Elsevier.

Drevets, W.C., Burton, H., Videen, T.O., Snyder, A.Z., Simpson Jr., J.R., and

Raichle, M.E. (1995). Blood flow changes in human somatosensory cortex during anticipated stimulation. Nature, 373, 249-252.

Duhamel, J.R., Colby, C.L., and Goldberg, M.E. (1998). Ventral intraparietal area of the macaque: congruent visual and somatic response properties. Journal of

Neurophysiology, 79,126-136.

Ernst, M.O., and Banks, M.S. (2002). Humans integrate visual and haptic information in a statistically optimal fashion. Nature, 415,429-433.

Fame, A., Pavani, F., Meneghello, F., & Ladavas, E. (2000). Left tactile extinction following visual stimulation of a rubber hand. Brain, 123, 2350-2360.

122 Gilbert, C.D., and Wiesel, T.N. (1992). Receptive field dynamics in adult primary

visual cortex. Nature, 356,150-152.

Gonzalez, E.G., Steeves, J. K. E., Kraft, S. P., Gallie, B. L. and Steinbach, M. J.

(2002). Foveal and eccentric acuity in one-eyed observers. Behavioural Brain

Research, 128, 71-80.

Graziano, M.S.A., and Gross, C.G. (1993). A bimodal map of space:

somatosensory receptive fields in the macaque putamen with corresponding

visual receptive fields. Experimental Brain Research, 97,96-109.

Graziano, M.S.A., Yap, G.S., and Gross, C.G. (1994). Coding of visual space by

premotor neurons. Science, 266,1054-1057.

Graziano, M. (1999). Where is my arm? The relative role of vision and

proprioception in the neuronal representation of limb position. PNAS, 96,10418-

10421.

Groh, J.M., and Sparks, D.L. (1996). Saccades to somatosensory targets. III. Eye-

position-dependent somatosensory activity in primate superior colliculus.

Journal of Neurophysiology, 75(1), 439"453-

123 Haines, D.E. (2004). Neuroanatomy: An Atlas of Structures, Sections, and

Systems (6th Ed.). Baltimore: Lippincott Williams & Wilkins.

Hanner, P., Badr, G., Rosenhall, U., and Edstrom, S. (1986). Trigeminal dysfunction in patients with Bell's Palsy. Acta Otolaryngol (Stockholm), 101, 224-

230.

Hanakawa, T., Dimyan, M.A., and Hallett, M. (2007). The representation of blinking movement in cingulate motor areas: a functional magnetic resonance imaging study. Cerebral Cortex, 18(4), 930-937.

Harting, J.K., Feig, S., and Van Lieshout, D.P. (1997). Cortical somatosensory and trigeminal inputs to the cat superior colliculus: light and electron microscopic analyses. The Journal of Comparative Neurology, 388,313-326.

Heft, M.W., and Robinson, M.E. (2010). Age differences in orofacial sensory thresholds. Journal of Dental Research, 89(10), 1102-1105.

Hensel, H. (1974). Thermoreceptors. Annual Review of Physiology, 36, 233-249.

124 Hubel, D. H. and Wiesel, T. N. (1962). Receptive fields binocular interaction and functional architecture in the cat's visual cortex. Journal of Physiology, 160,106-

154.

Irwin, D.E. (2011). Where does attention go when you blink? Attention,

Percpetion & Psychophysics, 73(5), 1374-1384.

Jobst, C.J., Kucyi, A., Pynn, L., & DeSouza, J.F.X. A multisensory visuotactile illusion induced by monocular occlusion with a black contact lens. Program No.

579.2. 2010 Neuroscience Meeting Planner San Diego, CA: Society for

Neuroscience, 2010. Online.

Johansson, R.S., Trulsson, M., Olsson, K.A., and Westberg, K.G. (1988).

Mechanoreceptor activity from the human face and oral mucosa. Experimental

Brain Research, 72, 204-208.

Johnson, K. (2002). Neural Basis of Haptic Perception. In Pashler, H., and

Yantis, S. (Eds.), Steven's Handbook of Experimental Psychology, 3rd Ed.,

Volume 1: Sensation and Perception (pp. 537-584). New York: John Wiley &

Sons, Inc.

125 Kaas, J.H., Nelson, R.J., Sur, M., Lin, C.S., and Merzenich, M.M. (1979). Multiple

representations of the body within the primary somatosensory cortex of primates.

Science, 204(4392), 521-523.

Kennedy, P.M., and Inglis, J.T. (2002). Distribution and behaviour of glabrous

cutaneous receptors in the human foot sole. Journal of Physiology, 538(3), 995-

1002.

Ladavas, E., di Pellegrino, G., Fame, A., and Zeloni, G. (1998a).

Neuropsychological evidence of an integrated visuotactile representation of

peripersonal space in humans. Journal of Cognitive Neuroscience, 10(5), 581-

589.

Ladavas, E., Zeloni, G., and Fame, A. (1998b). Visual peripersonal space centered

on the face in humans. Brain, 121,2317-2326.

Ladavas, E., and Fame, A. (2004). Visuo-tactile representation of near-the-body

space. Journal of Physiology-Paris, 98,161-170.

Makin, T.R., Holmes, N.P., and Zohary, E. (2007). Is that near my hand?

Multisensory representation of peripersonal space in human intraparietal sulcus.

The Journal of Neuroscience, 27(4), 731-740.

126 Massey, B.T. (2006). Physiology of oral cavity, pharynx and upper esophageal sphincter. GIMotility Online, doi:io.i()38/gimo2.

Meredith, M.A. (2002). On the neuronal basis for multisensory convergence: a brief overview. Cognitive Brain Research, 14,31 -40.

Millodot, M. (1972). Diurnal variation of corneal sensitivity. British Journal of

Ophthalmology, 56,844-847.

Millodot, M. (1975). Variation of visual acuity with soft contact lenses: a function of luminance. American Journal of Optometry & Physiological Optics, 52,541-

544-

Millodot, M. (1976). Effect of the length of wear of contact lenses on corneal sensitivity. Acta Ophthalmologica, 54,721-730.

Millodot, M. (1977). Does the long term wear of contact lenses produce a loss of corneal sensitivity? Experientia, 33,1475-1476.

Miiller, L.J., Marfurt, C.F., Kruse, F., and Tervo, T.M.T. (2003). Corneal nerves: structure, contents, and function. Experimental Eye Research, 76,521-542.

127 Nguyen, B.T., Tran, T.D., Inui, K., Hoshiyama, M., and Kakigi, R. (2004). Face representation in the human primary somatosensory cortex. Neuroscience

Research, 50, 227-32.

Nguyen, B.T., Inui, K., Hoshiyama, M., Nakata, H., and Kakigi, R. (2005). Face representation in the human secondary somatosensory cortex. Clinical

Neurophysiology, 116,1247-1253.

Nordin, M., and Hagbarth, K.E. (1989). Mechanoreceptive units in the human infra-orbital nerve. Acta Physiologica Scandinavica, 135,149-161.

Nordin, M. (1990). Low-threshold mechanoreceptive and nociceptive units with unmyelinated C fibres in the human supraorbital nerve. Journal of Physiology,

426, 229-240.

Oliveira-Soto, L., and Efron, N. (2003). Morphology of corneal nerves in soft contact lens wear: a comparative study using confocal microscopy. Ophthalamic and Physiological Optics, 23,163-174.

Patton, P., Belkacem-Boussaid, K., and Anastasio, T.J. (2002). Multimodality in the superior colliculus: an information theoretic analysis. Cognitive Brain

Research, 14,10-19.

128 Pinganaud, G., Bernat, I., Buisseret, P., and Buisseret-Delmas, C. (1999).

Trigeminal projections to hypoglossal and facial motor nuclei in the rat. The

Journal of Comparative Neurology, 415,91-104.

Pirenne, M.H. (1962). Dark Adaptation and Night Vision. In H. Davson (Ed.), The

Eue. Volume 2. London: Academic Press.

Poise, K. A. (1978). Etiology of corneal sensitivity changes accompanying contact lens wear. Investigative Opthalmology and Visual Science, 17,1202-1206.

Purves, D., Augustine, G.J., Fitzpatrick, D., Hall. W.C., LaMantia, A.S.,

McNamara, J.O., and White, L.E. (Eds.). (2008). Neuroscience. (4th Ed.).

Sunderland: Sinauer Associates.

Rainville, P., Duncan, G.H., Price, D.D., Carrier, B., and Bushnell, M.C. (1997).

Pain affect encoded in human anterior cingulate but not somatosensoiy cortex.

Science, 277,968-971.

Ramachandran, V.S., Stewart, M.I., and Rogers-Ramachandran, D. (1992).

Perceptual correlates of massive cortical reorganization. NeuroReport, 3,1159-

1160.

129 Ramachandran, V.S., and Rogers-Ramachandran, D. (1996). Synaesthesia in phantom limbs induced with mirrors. Proceedings of the Royal Society London,

263,377-386.

Reed, M. J., Steeves, J. K. E., Kraft, S. P., Gallie, B. L. and Steinbach, M. J. (1996).

Contrast letter thresholds in the non-affected eye of strabismic and unilateral eye enucleated children. Vision Research, 36,3011-3018.

Reed,M. J., Steeves, J. K. E. and Steinbach,M. J. (1997). A comparison of contrast letter thresholds in unilateral eye enucleated subjects and binocular and monocular control subjects. Vision Research, 37, 2465-2469.

Rizzolatti, G., Scandolara, C., Matelli, M. and Gentilucci, M. (1981a). Afferent properties of periarcuate neurons in macaque monkeys. I. Somatosensory responses. Behavioural Brain Research, 2,125-146.

Rizzolatti, G., Scandolara, C., Matelli, M. and Gentilucci, M. (1981b). Afferent properties of periarcuate neurons in macaque monkeys. II. Visual responses.

Behavioural Brain Research, 2,147-163.

Rubin, M., and Safideh, J.E. (2007). Netter's Concise Neuroanatomy.

Philadelphia: Saunders Elsevier.

130 Schaefer, M., Noennig, N., Heinze, H.J., and Rotte, M. (2006). Fooling your

feelings: Artificially induced referred sensations are linked to a modulation of the

primary somatosensory cortex. Neurolmage, 29, 67-73.

Schawerna, F.L., Cassidy, B.N., and Wilkinson, F. (2010). The role of initial light

adaptation in the phantom eye loss illusion. Poster presented at meeting for

Canadian Society for Brain Behaviour and Cognitive Science, Halifax.

Schneider, K.A., and Kastner, S. (2005). Visual responses of the human superior

colliculus: a high-resolution functional magnetic resonance imaging study.

Journal of Neurophysiology, 94,2491-2503.

Schuenke, M., Schulte, E., Schumacher, U., and Rude, J. (2007). Atlas of

Anatomy: Head and Neuroanatomy. Stuttgart: Thieme.

Schurmann, M., Kolev, V., Menzel, K., and Yordanova, J. (2002). Spatial

coincidence modulates interaction between visual and somatosensory evoked

potentials. NeuroReport, 13(6), 779-783.

Servos, P., Engle, S.A., Gati, J., and Menon, R. (1999). fMRI evidence for an

inverted face representation in human somatosensory cortex. NeuroReport, 10,

131 1393-1395-

Sprague, J.M. (1966). Interaction of cortex and superior colliculus in mediation of

visually guided behavior in cat. Science, 153(3743), 1544-1547.

Steeves, J. K. E., Gonzalez, E. G., Gallie, B. L. and Steinbach, M. J. (2002). Early

unilateral enucleation disrupts motion processing. Vision Research, 42,143-150.

Steeves, J.K.E., Gonzalez, E.G., and Steinbach, M.J. (2008). Vision with one eye:

a review of visual function following unilateral enucleation. Spatial Vision, 21(6),

509-529.

Steinbach, M.J. (1987). Proprioceptive knowledge of eye position. Vision

Research, 27(10), 1737-1744.

Ter Horst, G.J., Copray, J.C.V.M., Liem, R.S.B., and Van Willigen, J.D. (1991).

Projections from the rostral parvocellular formation to pontine and medullary

nuclei in the rat: involvement in autonomic regulation and orofacial motor

control. Neuroscience, 40(3), 735-758.

Thompson, P.B., and Lambert, J.V. (1994). Touch sensitivity through latex

examination gloves. The Journal of General Psychology, 122(1), 47-58.

132 Trulsson, M.,and Essick, G.K. (2010). Sensations evoked by microstimulation of

single mechanoreceptive afferents innervating the human face and mouth.

Journal of Neurophysiology, 103,1741-1747.

Tsuboi, A., Kolta, A., Chen, C.C., and Lund, J.P. (2003). Neurons of the trigeminal

main sensory nucleus participate in the generation of rhythmic motor patterns.

European Journal ofNeuroscience, 17, 229-238.

Vallbo, A.B., Olausson, H., and Wessberg, J. (1999). Unmyelinated afferents

constitute a second system coding tactile stimuli of the human hairy skin. Journal

of Neurophysiology, 81, 2753-2763.

VanderWerf, F., Buisseret-Delmas, C., and Buisseret, P. (2002). Afferent

Innervation of Eyelids and Their Connections to the Superior Colliculus.

Movement Disorders, 17, S8-S11.

133 Wallace, M.T., and Stein, B.E. (1997). Development of multisensory neurons and

multisensory integration in cat superior colliculus. The Journal ofNeuroscience,

17(7), 2429-2444-

Wang, X., Zhang, M., Cohen, I.S., and Goldberg, M.E. (2007). The proprioceptive

representation of eye position in monkey primary somatosensory cortex. Nature

Neuroscience, 10(5), 640-646.

Weir, C.R. (2006). Proprioception in extraocular muscles. Journal ofNeuro-

Ophthamology, 26,123-127.

Wolfe, U., Comee, J.A., and Sherman, B.S. (2007). Feeling darkness: A visually induced somatosensory illusion. Perception and Psychophysics, 69(6), 879-886.

Wolfe, U., & Carpinella, C. (2008). Psychophysical evidence for face-centered visuo-tactile neurons in humans. Perception, 37, S134-S135.

Xu, Y., Wang, X., Peck, C., and Goldberg, M.E. (2011). The time course of the tonic oculomotor proprioceptive signal in area 3a of somatosensory cortex.

Journal of Neurophysiology, 106,71-77.

134