Vibration Perception Thresholds of Skin Mechanoreceptors Are Inﬂuenced by Different Contact Forces

Journal of Clinical Medicine

Article Vibration Perception Thresholds of Skin Mechanoreceptors Are Inﬂuenced by Different Contact Forces

Claudio Zippenfennig * , Bert Wynands and Thomas L. Milani

Department of Human Locomotion, Faculty of Behavioral and Social Sciences, Institute of Human Movement Science and Health, Chemnitz University of Technology, 09107 Chemnitz, Germany; [email protected] (B.W.); [email protected] (T.L.M.) * Correspondence: [email protected]; Tel.: +49-371-531-34467

Abstract: Determining vibration perception thresholds (VPT) is a central concern of clinical research and science to assess the somatosensory capacity of humans. The response of different mechanoreceptors to an increasing contact force has rarely been studied. We hypothesize that increasing contact force leads to a decrease in VPTs of fast-adapting mechanoreceptors in the sole of the human foot. VPTs of 10 healthy subjects were measured at 30 Hz and 200 Hz at the heel of the right foot using a vibration exciter. Contact forces were adjusted precisely between 0.3 N–9.6 N through an integrated force sensor. Significant main effects were found for frequency and contact force. Furthermore, there was a significant interaction for frequency and contact force, meaning that the influence of an increasing contact force was more obvious for the 30 Hz condition. We presume that the principles of contrast enhancement and spatial summation are valid in Meissner and Pacinian corpuscles, respectively. In addition to spatial summation, we presume an effect on Pacinian corpuscles due to their presence in the periosteum or interosseous membrane.

Citation: Zippenfennig, C.; Keywords: vibration perception threshold; mechanoreceptors; sensory perception; spatial summa- Wynands, B.; Milani, T.L. Vibration tion; Meissner and Pacinian corpuscles Perception Thresholds of Skin Mechanoreceptors Are Inﬂuenced by Different Contact Forces. J. Clin. Med. 2021, 10, 3083. https://doi.org/ 1. Introduction 10.3390/jcm10143083 Mechanoreceptors in the skin of human feet provide important information about various environmental stimuli [1]. If this feedback system fails, e.g., due to diseases such as Academic Editor: José A. Vega diabetes mellitus or Parkinson’s disease, gait and balance often deteriorate [2,3]. It is therefore a central concern of clinical research and science to assess the somatosensory capacity Received: 29 May 2021 of humans, as a decrease in sensitivity may be an early sign of peripheral neuropathy [4]. Accepted: 11 July 2021 Published: 13 July 2021 Four different types of cutaneous mechanoreceptors are responsible for sensing and transmitting diverse pressures and vibrations at the sole of the foot: slowly adapting

Publisher’s Note: MDPI stays neutral (SA) type I (Merkel discs) and II (Ruffini corpuscles) receptors, which react to continuous with regard to jurisdictional claims in pressure, and fast-adapting (FA) type I (Meissner corpuscles) and II (Pacinian corpuscles) published maps and institutional affil- receptors, which respond to the onset and offset of stimulation. SA I and FA I receptors are iations. located near the skin surface and have small and well-defined receptive fields. SA II and FA II receptors lie deeper in the skin and have large receptive fields with obscure bound- aries [5,6]. In this study, we focused on Meissner (FA I) and Pacinian (FA II) corpuscles. Determining vibration perception thresholds (VPT) is a form of quantitative sensory testing. To produce reliable results, it is essential to standardize procedures. Numer- Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. ous influencing factors for determining VPTs have already been studied. These include This article is an open access article frequency [6–8], anatomical location [8], size of the probe [9,10], mechanical skin proper- distributed under the terms and ties [1,11], temperature [12], gender [13], and age [14]. One parameter that has received conditions of the Creative Commons little attention to date is the quantity of the load (pressure, acceleration, and contact area) Attribution (CC BY) license (https:// with which the indenting object stresses the corresponding area of the subject’s skin. creativecommons.org/licenses/by/ Over the last 15 years, research studies measuring VPTs have used different contact 4.0/). forces, ranging from no force offset to avoid pre-activation of mechanoreceptors before

J. Clin. Med. 2021, 10, 3083. https://doi.org/10.3390/jcm10143083 https://www.mdpi.com/journal/jcm J. Clin. Med. 2021, 10, 3083 2 of 12

the vibration stimulus [8], a preload of 2 N [11], and a preload of 4 N to simulate the pressure under the foot during walking [15]. These different contact forces may restrict the comparison of VPTs, since different studies have shown that increasing preloads from 0.1 N up to 8 N leads to an improvement in sensitivity [16–18]. In contrast, Hagander et al. [19] found decreased sensitivity at preloads of 1 N compared to 0.3 N and 0.5 N at the index finger. At the big toe, increasing preloads had no influence on VPTs [19]. Furthermore, Gregg [20] and Era and Hänninen [21] confirmed the results of Hagander et al. [19] at the big toe, finding no dependence between different contact forces and VPT. Two recently published studies evaluated the influence of standing compared to sitting while measuring VPTs at the sole of the foot [22,23]. Interestingly, these two studies also had contradictory results. While Mildren et al. [22] partially found a significant increase of VPT while standing in comparison to sitting, Germano et al. [23] found no significant differences. As both studies showed higher contact forces when standing compared to sitting, these two studies reflect the contradictions in the current literature. To our knowledge, only Gu and Griffin [24] have examined the influence of different contact forces on VPT at the sole of the foot [24]. Their results exhibit slight VPT decreases with increasing force at the hallux and the ball of the foot. VPT decreases were more pronounced at 160 Hz than at 20 Hz [24]. In contrast to Gu and Griffin [24], our study did not use a static probe surrounding, which would compress the skin around the probe as well. Instead, only the vibrating probe exerted the contact force. A wide range of contact forces (0.3–9.6 N) was examined, exceeding typically-used values (0–4 N) in previous studies [8,11,15], to detect possible ceiling effects. To evaluate the effect on Meissner and Pacinian corpuscles separately, VPTs were collected at the most sensitive frequency of each receptor type. We hypothesized that increasing contact forces leads to a decrease in VPTs at both mechanoreceptors, which results in improved foot sensitivity. Finally, we tested the hypothesis that there is an interaction between sensor type and increasing contact forces.

2. Materials and Methods Ten healthy young subjects participated in this study (8 /2 ; mean ± SD: 22.9 ± 1.4 years, 74.9 ± 9.7 kg, 178.8 ± 9.7 cm). All participants gave♂ their♀ written consent and were free of any diseases that could affect the sensory system (diabetes mellitus, peripheral neuropathy, neurological diseases, etc.). All procedures were performed in accordance with the recommendations of the Declaration of Helsinki. The measurement system and methodology were approved by the Ethics Committee of the Faculty of Be- havioural and Social Sciences of the corresponding university for a similar study in our work group (V-277-17-DS-KUS/WUS-22062018) [25]. The sensory measuring procedure has been validated and used in several published studies [1,23,26]. A modiﬁed vibration exciter (Typ 4180, Brüel & Kjaer Vibro GmbH, Darmstadt, Ger- many) powered by a powerbank (XTPower MP-3200, Batteries and Power Solutions GmbH, Ellwangen, Germany) was used to measure VPTs (Figure1). To exclude mechanical inﬂu- encing factors, the vibration amplitude (µm) was calculated using an external acceleration sensor (MMA2240KEG, NXP Semiconductors Netherlands B.V., Eindhoven, The Nether- lands) placed in series with the probe of the vibration exciter (Figure1). To measure the vibration amplitude as accurately as possible, the vertical movement of the vibration exciter’s probe (diameter 7.8 mm) was calibrated before the measurements using a high- precision capacitive position sensor (CS05, Micro-Epsilon Messtechnik GmbH & Co. KG, Ortenburg, Germany). This ensured direct readings of the vibration amplitude during VPT measurements. Using a swivel arm, the probe of the vibration exciter was placed precisely perpendicularly at the plantar heel. A force sensor (DS050A9, disynet GmbH, Brüggen- Bracht, Germany), also placed in series with the probe of the vibration exciter (Figure1), was used to adjust different contact forces: 0.3 N, 0.6 N, 1.2 N, 2.4 N, 4.8 N, and 9.6 N. This was done to cover an extensive range of contact forces. Previous pilot measurements showed that approximately 10 N was the upper limit, both from a practical point of view and for the wellbeing of the subjects. The contact force was continuously monitored and, if J. Clin. Med. 2021, 10, x FOR PEER REVIEW 3 of 12

J. Clin. Med. 2021, 10, 3083 3 of 12 measurements showed that approximately 10 N was the upper limit, both from a practical point of view and for the wellbeing of the subjects. The contact force was continuously necessary,monitored readjustedand, if necessary, between readjusted individual be measurements.tween individual The measurements. investigation startedThe investi- with thegation lowest started contact with force the lowest of 0.3 N.contact To give force the of skin 0.3 sufﬁcient N. To give recovery the skin time sufficient and to recovery prevent accumulatingtime and to prevent effects, accumulating the resting timeseffects, between the resting the times individual between contact the individual forces were contact also increasedforces were exponentially also increased (from exponentially 30 s to 8 min) (from to accommodate30 s to 8 min) possible to accommodate disadvantages possible of non-randomization.disadvantages of non-randomization.

acceleration sensor force sensor

vibration exciter

FigureFigure 1. MeasurementMeasurement setup. setup. The The 7.8 7.8mm mm diameter diameter probe probe (area (area ~48 mm ~482) mm of the2) ofmodified the modiﬁed vibration vibration exciter placed exciter perpen- placed perpendicularlydicularly at the plantar at the plantar heel of heelthe right of the foot. right To foot. keep To the keep test the person’s test person’s foot as foot still as as still possi asble, possible, the foot the was foot strapped was strapped into a intocushioned a cushioned splint. splint.The contact The contactforces (0.3 forces N, 0.6 (0.3 N, N, 1.2 0.6 N, N, 2.4 1.2 N, N, 4.8 2.4 N, N, and 4.8 9.6 N, N) and were 9.6 adjusted N) were and adjusted continuously and continuously monitored monitoredusing a force using sensor. a force The sensor. amplitude The amplitude of the vibration of the was vibration measured was measured directly in directly micromet in micrometersers using an accelerometer. using an accelerometer.

VPTs werewere measured measured at at the the heel heel of of the the right righ foott foot at two at two different different frequencies: frequencies: 30 Hz 30 and Hz 200and Hz.200 TheseHz. These two frequenciestwo frequencies are considered are considered ideal forideal measuring for measuring the vibration the vibration perception per- ofception Meissner of Meissner and Pacinian and Pacinian corpuscles corpuscles [6,7]. After [6,7]. a 10-minAfter a acclimation10-min acclimation period, participantsperiod, par- layticipants supine lay with supine their with right their foot right ﬁxed foot in a cushionedfixed in a cushioned splint (Figure splint1). (Figure VPTs were 1). VPTs measured were threemeasured times three for each times contact for each force contact using aforce self-written using a binaryself-written search binary method search[15, 22method]. Par- ticipants[15,22]. Participants had to press had a button to press as a soon button as they as soon felt aas vibration. they felt a Starting vibration. with Starting a sinusoidal with a vibrationsinusoidal burst vibration above burst the participants’above the participants’ individual individual thresholds thresholds (2 s duration (2 s duration followed fol- by alowed 2–7 s by pause), a 2–7 thes pause), program the reducedprogramor reduced increased or increased the stimulus the stimulus intensity intensity automatically, auto- dependingmatically, depending on whether on thewhether subject the felt subject the lastfelt the stimulus. last stimulus. The algorithm The algorithm stopped stopped four burstsfour bursts after after the ﬁrst the undetectedfirst undetected stimulus. stimulus The. meanThe mean of the oflast the recognizedlast recognized and theand lastthe unperceivedlast unperceived vibration vibration stimuli stimuli was was determined determined as VPT. as VPT. A graphical A graphical visualization visualization of the of measurementthe measurement algorithm algorithm can becan found be found in Appendix in AppendixA (Figure A A1(Figure). If the A1). subjects If the pressedsubjects thepressed button the more button than more twice than while twice no while stimulus no stimulus was present, was the present, measurement the measurement was repeated. was Outrepeated. of a total Out of a 360 total measurements, of 360 measurements, only four on hadly four to behad repeated to be repeated (~1%). (~1%). A single A single VPT measurementVPT measurement took abouttook about 2 min. 2 Themin. three The thr measurementsee measurements per contact per contact force thusforce resultedthus re- insulted a duration in a duration of approximately of approximately 6 min. The 6 min. skin The recovery skin recovery time lasted time a totallasted of a 15 total min of and 15 30min s. and Including 30 s. Including acclimatization, acclimatization, test trials, test and trials, storing and thestoring VPTs, the this VPTs, resulted this resulted in a total in durationa total duration of approximately of approximately one and one a half and hours a ha perlf hours testsubject. per testTherefore, subject. Therefore, the two different the two frequenciesdifferent frequencies were measured were measured on two different on two diff dayserent to days allow to subjects allow subjects to concentrate to concentrate for as long as possible. Since two different mechanoreceptors were addressed by the selected frequencies, learning effects were not expected.

J. Clin. Med. 2021, 10, x FOR PEER REVIEW 4 of 12

for as long as possible. Since two different mechanoreceptors were addressed by the selected frequencies, learning effects were not expected. Participants wore noise-cancelling headphones (QuietComfort 25 Acoustic Noise Cancelling headphones, Bose Corp., Framingham, MA, USA) to eliminate environmental noises. Temperature values of the foot were monitored using a non-contact infrared thermometer (UNI-T UT301C, Batronix GmbH & Co. KG, Preetz, Germany). The mean out of three VPT measurements per contact force condition was used for statistical analysis with R [27]. Data was checked for outliers, and the benefit of logarithmical transformation due to non-normal distribution was tested. After logarithmic trans- J. Clin. Med. 2021, 10, 3083 formation, the distribution of the data was still non-normal. The Aligned Rank Transform4 of 12 (ART) procedure was used to examine interaction effects [28]. A post hoc analysis was performed using interaction contrasts, looking at differences of differences [28]. Participants wore noise-cancelling headphones (QuietComfort 25 Acoustic Noise Cancelling3. Results headphones, Bose Corp., Framingham, MA, USA) to eliminate environmental noises.Plantar Temperature temperature values was ofwithin the foot acceptable were monitored ranges (30 usingHz pre a vs. non-contact post: 22.1 infrared± 1.5 vs. thermometer23.6 ± 2.7; 200 (UNI-T Hz pre UT301C, vs. post: Batronix 23.4 ± 2.9 GmbH vs. 24.2 & Co.± 2.4) KG, [12]. Preetz, VPTs Germany). decreased at both fre- quenciesThe meanwith increasing out of three contact VPT measurements forces (Table per1, Figures contact 2 force and condition3). was used for statistical analysis with R [27]. Data was checked for outliers, and the beneﬁt of logarithmical Table 1. Vibrationtransformation perception thresholds due to non-normal [µm] for each distribution contact force was (N) tested. and frequency After logarithmic (Hz). transformation, the distribution of the data was still non-normal. The Aligned Rank Transform (ART) n = 10 procedure 0.3 N was used 0.6 to examineN interaction 1.2 N effects [2.428 ].N A post hoc 4.8 analysis N was performed 9.6 N using interaction contrasts, looking at differences of differences [28]. VPT 30 Hz (µm) 10.9 ± 6.2 9.3 ± 5.3 7.5 ± 3.0 6.4 ± 4.5 4.1 ± 2.5 3.9 ± 2.5

VPT 200 Hz (µm) 3. Results 1.8 ± 2.8 1.5 ± 2.0 1.2 ± 1.6 1.4 ± 2.2 0.8 ± 0.9 0.7 ± 0.7 Plantar temperature was within acceptable ranges (30 Hz pre vs. post: 22.1 ± 1.5 vs. Mean ± SD of vibration perception thresholds (VPTs). 23.6 ± 2.7; 200 Hz pre vs. post: 23.4 ± 2.9 vs. 24.2 ± 2.4) [12]. VPTs decreased at both frequenciesA significant with increasing main effect contact was forces found (Table for ‘frequency’,1, Figures2 and indicating3). lower thresholds (higher sensitivity) for 200 Hz measurements (F(1,99) = 306.837, p < 0.001, ηp2 = 0.76). Fur- Table 1. Vibration perception thresholds [µm] for each contact force (N) and frequency (Hz). thermore, there was a significant main effect for ‘contact force’, meaning lower thresholds (highern sensitivity)= 10 with 0.3 increased N 0.6contact N forces 1.2 N(F(5,99) = 2.4 19.610, N p 4.8< 0.001, N ηp2 9.6 = 0.50). N Additionally, a significant interaction effect was observed for ‘frequency’ and ‘contact VPT 30 Hz (µm) 10.9 ± 6.2 9.3 ± 5.3 7.5 ± 3.0 6.4 ± 4.5 4.1 ± 2.5 3.9 ± 2.5 force’,VPT200 which Hz (meansµm) that 1.8 ±increased2.8 1.5 contact± 2.0 force 1.2 led± 1.6 to higher 1.4 ± reductions2.2 0.8 ± in0.9 VPTs 0.7 at± 300.7 Hz 2 Meancompared± SD of to vibration 200 Hz perception (F(5,99) thresholds= 12.947, (VPTs).p < 0.001, ηp = 0.40).

Figure 2. Cont.

J.J. Clin. Clin. Med. Med. 20212021,, 1010,, x 3083 FOR PEER REVIEW 55 of of 12 12 J. Clin. Med. 2021, 10, x FOR PEER REVIEW 5 of 12

Figure 2. Relationship between vibration perception thresholds (VPT) at 200 Hz and contact force for each subject. Each FigureFigure 2. 2. Relationship Relationship between between vibrationvibration perceptiperceptionon thresholds (VPT) at 200 HzHz andand cocontactntact forceforce for for each each subject. subject. Each Each scatterscatterscatter plot plot plot shows shows shows the thethe three threethree determined determined VPTs VPTsVPTs for for each each contactcontact force condition. AA smoothedsmoothed conditionalconditional mean mean was was added added usingusing the the R R function function “geom_smooth “geom_smooth ()” ()” in in combinationcombination with the argument “method “method = = loess”.loess”. SevenSeven Seven outout out of of of 10 10 subjects subjects showedshowedshowed a aa reduction reduction in in VPT VPT withwith increasing increasing contact contact force. force. The The mean mean from from the the respectiverespective threethree three VPTVPT measurementsmeasurements per per contactcontactcontact force force force was was used used for for stat statisticalstatisticalistical analysis analysis (Figure (Figure 4 4).4).). AnAn attemptattempt waswas mademade toto establishestablish aa psychophysicalpsychophysical relationship relationshiprelationship betweenbetween the the contact contact force force and and the the perceived perceived sensitivitysensitivitsensitivity acrossacross all subjects (Appendix B B,B,, FigureFigureFigure A2A2). A2).).

Figure 3. Cont.

J. Clin. Med. 2021, 10, x FOR PEER REVIEW 6 of 12

J. Clin.J. Clin. Med. Med.2021 2021, 10, 10 3083, x FOR PEER REVIEW 66 of of 12 12

Figure 3. Relationship between vibration perception thresholds (VPT) at 30 Hz and contact force for each subject. Each scatter plot shows the three determined VPTs for each contact force condition. A smoothed conditional mean was added using the R function “geom_smooth ()” in combination with the argument “method = loess”. All 10 subjects showed a reduction in VPT with increasing contact force. The mean from the respective three VPT measurements per contact force was used for statistical analysis (Figure 4). An attempt was made to establish a psychophysical relationship between the contact force and the perceived sensitivity across all subjects (Appendix B, Figure A2). Figure 3. Relationship between vibration perception thresholds (VPT) at 30 Hz and contact force for each subject. Each Figure 3. Relationship between vibration perception thresholds (VPT) at 30 Hz and contact force for each subject. Each scatter plot shows the three determined VPTs for each contact force condition. A smoothed conditional mean was added scatter plot shows the three determinedInteraction VPTs contrasts for each (Figure contact force4) were condition. found A for smoothed the difference conditional between mean was VPTs added at 200 using the R function “geom_smooth ()” in combination with the argument “method = loess”. All 10 subjects showed a using the R function “geom_smoothHz and 30 Hz ()” inin thecombination contact force with conditionthe argument 0.3 “methodN compared = loess”. to the All differences10 subjects showed at 2.4 N, a 4.8 reductionreduction in VPTin VPT with with increasing increasing contact contact force. force. The The mean mean from from the the respective respective three three VPT VPT measurements measurements per per contact contact force force N, and 9.6 N (all p < 0.001). Further significant differences of differences were found be- waswas used used for for statistical statistical analysis analysis (Figure (Figure4). An4). An attempt attempt was was made made to establishto establish a psychophysicala psychophysical relationship relationship between between the the tween 0.6 N and 2.4 N (p = 0.018), 4.8 N, and 9.6 N (both p < 0.001) and between 1.2 N and contactcontact force force and and the the perceived perceived sensitivity sensitivity across across all all subjects subjects (Appendix (AppendixB, FigureB, Figure A2 A2).). 4.8 N (p = 0.005) and 9.6 N (p = 0.001). Interaction contrasts (Figure 4) were found for the difference between VPTs at 200 Hz and 300.3N Hz in the 0.6Ncontact force1.2N condition2.4N 0.3 N compared4.8N to the9.6N differences at 2.4 N, 4.8 N, and 9.6 N (all p < 0.001). Further significant differences of differences were found be- 40 tween 0.6 N and 2.4 N (p = 0.018), 4.8 N, and 9.6 N (both p < 0.001) and between 1.2 N and 38 ** 4.836 N (p = 0.005) and 9.6 N (p = 0.001). ** 34 *** 32 0.3N 0.6N 1.2N*** 2.4N 4.8N 9.6N 30 * 28 40 *** 38 ** 26 *** ** 24 36 *** 22 34 *** 32 *** 20 * 18 30 16 28 *** 26 14 *** 24 *** 12 22 10 20 Vibration Perception Threshold [µm] Threshold Perception Vibration 8 18 6 16 4 14 2 12 0 10

Vibration Perception Threshold [µm] Threshold Perception Vibration 200Hz8 30Hz 200Hz 30Hz 200Hz 30Hz 200Hz 30Hz 200Hz 30Hz 200Hz 30Hz 6 FigureFigure 4. 4. Vibration4Vibration perception perception thresholds thresholds per per contact contact fo forcerce and and frequency. frequency. The The violin violin plots plots show show the the densitydensity of of2 the the distribution distribution for for the the vibration vibration perc perceptioneption thresholds thresholds (VPT) (VPT) for for the the frequencies frequencies 200 200 Hz Hz (blue)(blue) and and0 30 30 Hz Hz (green) (green) of of the the respective respective cont contactact forces forces based based on on the the width width of of the the plot plot ( (nn = 10 10 per per contactcontact force force200Hz condition). condition). 30Hz 200Hz The The individual 30Hz individual200Hz points points 30Hz wi within200Hzthin30Hz the the violin violin200Hz plots plots 30Hz represent represent200Hz 30Hz the the averaged averaged VPT VPT (mean(mean out out of of three three VPT VPT measurements measurements per per contact contact forc forcee condition) condition) of of each each subject. subject. The The black black dot dot is theis theFigure mean mean value4. Vibration value of all of subjects.allperception subjects. The thresholds The connecting connecting per line contact linebetween betweenforce the and black the frequency. black dots dotsrepresents The represents violin theplots mean the show mean dif- the ference.difference.density Superscripted of Superscripted the distribution symbols symbols for represent the represent vibration significan signiﬁcantperceptiont interaction interactionthresholds contrasts contrasts(VPT) and for andthedifferences frequencies differences of the 200 of dif- the Hz ferences (*** p < 0.001, ** p < 0.01, * p < 0.05). differences(blue) and (*** 30p Hz< 0.001,(green) ** ofp < the 0.01, respective * p < 0.05). contact forces based on the width of the plot (n = 10 per contact force condition). The individual points within the violin plots represent the averaged VPT (mean out of three VPT measurements per contact force condition) of each subject. The black dot is the mean value of all subjects. The connecting line between the black dots represents the mean difference. Superscripted symbols represent significant interaction contrasts and differences of the differences (*** p < 0.001, ** p < 0.01, * p < 0.05).

J. Clin. Med. 2021, 10, 3083 7 of 12

A significant main effect was found for ‘frequency’, indicating lower thresholds (higher 2 sensitivity) for 200 Hz measurements (F(1,99) = 306.837, p < 0.001, ηp = 0.76). Furthermore, there was a significant main effect for ‘contact force’, meaning lower thresholds (higher 2 sensitivity) with increased contact forces (F(5,99) = 19.610, p < 0.001, ηp = 0.50). Addition- ally, a significant interaction effect was observed for ‘frequency’ and ‘contact force’, which means that increased contact force led to higher reductions in VPTs at 30 Hz compared to 2 200 Hz (F(5,99) = 12.947, p < 0.001, ηp = 0.40). Interaction contrasts (Figure4) were found for the difference between VPTs at 200 Hz and 30 Hz in the contact force condition 0.3 N compared to the differences at 2.4 N, 4.8 N, and 9.6 N (all p < 0.001). Further significant differences of differences were found between 0.6 N and 2.4 N (p = 0.018), 4.8 N, and 9.6 N (both p < 0.001) and between 1.2 N and 4.8 N (p = 0.005) and 9.6 N (p = 0.001).

4. Discussion The aim of the present study was to investigate the effects of increasing contact forces on VPTs of two different mechanoreceptors (Meissner and Pacinian corpuscles) in healthy young subjects. We hypothesized that greater contact forces lead to lower VPTs at both mechanoreceptors and, based on their anatomical location, that there is an interaction between sensor type and increasing contact force. In accordance with the literature, we found a significant main effect for ‘frequency’, indicating lower thresholds for 200 Hz compared to 30 Hz [8,22]. Furthermore, we found a significant main effect for ‘contact force’, meaning that an increased contact force led to lower VPTs and, therefore, to better sensitivity. Furthermore, we found a significant interact- ing effect between ‘frequency’ and ‘contact force’, meaning that the influence of increasing contact force was different for Meissner than for Pacinian corpuscles. Figures2 and3 show a more pronounced decrease at 30 Hz than at 200 Hz. Pairwise comparisons exhibited significant differences between lower (0.3 N–1.2 N) and higher contact forces (2.4 N–9.6 N) (Figure4).

4.1. Meissner Corpuscles Meissner corpuscles have circular receptive fields with several zones of maximal sensitivity [6]. This area of high sensitivity lies in the middle of the circular receptive field. Towards the periphery, the Meissner corpuscle is increasingly less sensitive [6]. The increasing pressure caused increased skin deformation, because the area of skin surrounding the probe also increased [29]. This may not only have led to a reduction in the distance between the stimulus and the perception hotspots, but also to a general increase in the number of Meissner corpuscles being addressed [29]. Additionally, increasing contact forces led to mechanical deformation of the non-neuronal components. In the Meissner corpuscles, force transmission through collagen fibers results in bending of the axon terminals [5]. This compression generates action potentials during stimulus onset [5]. High contact forces might have pre-bent axon terminals before the actual stimulus. Thus, it is possible that the low-frequency vibrations can be transmitted more directly to the axon and, thus, trigger action potentials more easily. Furthermore, in the hallux, for example, Meissner corpuscles reach a density of approximately 10/mm2 [30]. Our probe stimulated a skin area of ~48 mm2. However, more surrounding skin was indented by the probe at higher pressures, which might have increased the affected area. Furthermore, skin mechanics change due to compression. The compressed skin may transmit vibrations farther across the skin [1]. Thus, more Meissner corpuscles can be used for stimulus detection. Interestingly, Gu and Griffin [10] challenge the theory of the increasing number of responding Meissner corpuscles [10]. They tested the spatial summation for vibrotactile stimuli of Meissner and Pacinian corpuscles at the foot. They used different probe sizes (1–10 mm in diameter) and vibration frequencies of 20 Hz and 160 Hz. They showed that the VPT decreased with increasing area only J. Clin. Med. 2021, 10, 3083 8 of 12

at a frequency of 160 Hz (representative for Pacinian corpuscles) [10]. Therefore, spatial summation only seems to occur at higher frequencies [9,10]. In addition to our approach regarding perception hotspots described above, the synergistic cooperation of slowly adapting type I receptors (Merkel discs) and rapidly adapting receptors (Meissner corpuscles) may provide a second explanatory model. Meissner corpuscles are relatively insensitive to static forces and have a high sensitivity at low spatial resolutions [31]. On the other hand, Merkel discs have a low sensitivity at high spatial resolutions [31]. According to Abraira and Ginty [31], it is conceivable that Merkel discs and Meissner corpuscles act together to encode a more complete picture of tactile space [31]. Merkel discs react to mechanical forces on the skin with a persistent and graded dynamic, followed by ‘bursting’ at irregular intervals, which correlates linearly with penetration depth [31,32]. We know that Merkel disc receptors also encode low frequency vibrations (5– 15 Hz) and that the transition between the frequency coding of the receptors overlaps [33]. Therefore, the increasing contact force could have led to an increased synergistic response behavior of Merkel discs. This synergistic interaction could have led to the increased sensitivity measured during the 30 Hz condition. Based on this synergistic interaction of these two mechanoreceptors, the increased sensitivity with increasing contact forces may also be attributed to contrast enhancement. Due to the receptor density and the overlapping receptive fields, sensory information is mediated simultaneously by several receptors [34]. As a result, the functional organization of the sensory processing networks is hierarchical [34]. Laterally inhibiting interneurons reduce or inhibit the excitation of weak signals at the receptors’ first interconnection level due to the activity in the neurons with strong signals [35]. As already mentioned, more receptors may be involved in stimulus detection with increasing contact force. To enable the neural network to precisely identify the stimulus source, i.e., the probe, adjacent weaker stimuli stemming from skin deformation may be more severely inhibited than for lower contact forces. At the same time, the signal of the strongly activated receptors directly under the probe would be amplified even more, so that the original stimulus (the vibration) could be perceived better.

4.2. Pacinian Corpuscles As already mentioned, the improvement in sensitivity of the Pacinian corpuscles with increasing contact force seems to be the result of spatial summation [10]. While Gu and Grifﬁn [10] used different probe sizes and vibration frequencies, Kekoni et al. [9] described another point which may have led to spatial summation in our study. They used the same vibration exciter with a similar measurement methodology [9]. By directly applying the probe of the vibration exciter to the measurement site, the vibration stimulus can spread to surrounding skin areas [9]. Supported by the possibility of changing skin characteristics from increasing compression and the high sensitivity of the Pacinian corpuscles, these residual vibrations, depending on their intensity, can also cause reactions in surrounding Pacinian corpuscles [36]. The altered and far-reaching proliferation of vibrations with altered contact forces of the probe represents a possible reason for the improving sensitivity of the Pacinian corpuscles. The smallest residual vibration may already lead to the excitation of surrounding Pacinian corpuscles. This is supported by the assumption that mechanical skin properties and the reaction of the Pacinian corpuscles may act together due to the skin’s own resonance and the optimal sensitivity of the Pacinian corpuscles [36]. Additionally, Pacinian corpuscles are not only located in the dermis, but also in the periosteum or interosseous membrane [32,37]. As the contact force increases, the heel fat pad is compressed, and the probe and calcaneus come closer together. The vibration stimulus may therefore be transmitted to the Pacinian corpuscles located in the periosteum/interosseous membrane and, thus, possibly have been the reason for the improved sensitivity of the Pacinian corpuscles. J. Clin. Med. 2021, 10, 3083 9 of 12

4.3. General Aspects and Limitations Many of our discursive approaches were interpretive, based on the interaction between mechanoreceptors, stimulus conduction, and stimulus processing. Unfortunately, we were limited by our technical possibilities and could only measure the outcome of mechanoreceptor stimulation in terms of VPTs. Using microneurography, including stimulus conduction and EEG-measurements, may provide information on stimulus processing. Furthermore, generalization of our discursive approaches is limited due to the small number of subjects, only one measured anatomical location, and one probe size. Based on a different receptor distribution and differences in the mechanical skin properties [1,38], an inclusion of further anatomical locations on the sole of the foot (e.g., first metatarsal head or big toe) and different probe sizes would be useful in further studies. An increase in the number of subjects is also conceivable. However, in sensory measurements, the neuronal responses that arise from stimulating receptors can be influenced by attention, cognitive processes, and other behavior patterns of the perceiver [39]. The success of sensory measurements depends to a large extent on the cooperation of the subject. To maintain high levels of concentration, the breaks described in the methods section for the skin to recover were also used as time for subjects to relax. In addition, the measurements at the two different frequencies were distributed over two days to ensure the active participation of the subjects at all times. Based on our results, we were able to establish a relationship between VPTs and contact forces considering different measurement frequencies. If we disregard the two highest contact forces tested in this study (4.8 N and 9.6 N), the lowest variability for determining VPTs was seen at 1.2 N (Figure4). The highest contact forces led to strong skin deformations, which could influence the response behavior of the mechanoreceptors and were unpleasant for the test subjects.

Author Contributions: C.Z., B.W. and T.L.M. participated in the design of the study; C.Z. and B.W. performed the experiments; C.Z. analyzed the data and designed figures; C.Z. and B.W. drafted the manuscript; C.Z., B.W. and T.L.M. edited and revised the manuscript and data analysis. All authors have read and agreed to the published version of the manuscript. Funding: C.Z. would like to thank the Sächsische Aufbaubank (SAB) for providing scholarship funding from the European Social Fund (ESF) and the Free State of Saxony. The publication of this article was funded by Chemnitz University of Technology. Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki. The measurement system and methodology were approved by the Ethics Committee of the Faculty of Behavioural and Social Sciences of the corresponding university in a similar study of our working group (V-277-17-DS-KUS/WUS-22062018). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to further analyses. Acknowledgments: The authors would like to thank Lisa Peterson for her linguistic assistance with the manuscript, as well as Ralph Dörfler, Philip Schulze, and Mirco Kaden for their technical support. Conflicts of Interest: The authors declare no conflict of interest.

Appendix A In the following, the measurement algorithm of a measurement trial to determine the VPT is visualized (Figure A1). The measurement algorithm adjusts the vibration amplitude according to the subjects’ perception. The vibration frequency (200 Hz or 30 Hz) remains constant during a measurement. J. Clin. Med. 2021, 10, x FOR PEER REVIEW 10 of 12

J. Clin. Med. 2021, J.10 Clin., x FOR Med. PEER2021 REVIEW, 10, 3083 10 of 12 10 of 12

Figure A1. Exemplary procedure for determining VPT. The algorithm is started by the examiner with a vibration burst above the participants’ individual threshold (a). The subject signals the per- Figure A1. Exemplary procedure for determining VPT. The algorithm is started by the examiner with a vibration burst Figure A1. Exemplaryception procedureof the initial for stimulus determining by pressing VPT. Th a ebu algorithmtton. The isalgorithm started bythen the automatically examiner halves the above the participants’with a vibration individualamplitude burst above threshold of the panext (articipants’). Thestimulus subject individual to be signals tested threshold the (b) perception until (thea). Thefirst of thesubjectvibration initial signals ca stimulusn no the longer per- by pressingbe perceived a button. Theception algorithm of thenthe initial automatically(c). The stimulus algorithm halves by pressingthen the averages amplitude a bu tton.the of last The the perceived nextalgorithm stimulus stimulus then to beautomatically (b tested) and the (b) untillast halves stimulus the ﬁrstthe that vibration was not can no longeramplitude be perceived of the (feltc ).next The(c) stimulusand algorithm continues to thenbe determiningtested averages (b) until the the last theVPT perceived first using vibration this stimulus averaged can no (b ) amplitudelonger and the be last perceived(d). stimulus After one that vibration was not felt (c) and(c). continues The algorithm determiningstimulus then averages is the not VPT felt the using (c ),last the this perceived algorithm averaged stimulus automatically amplitude (b) (andd ).tests Afterthe four last one more stimulus vibration stimuli that stimulus according was not is not to the felt proce- (c), felt (c) and continues determining the VPT using this averaged amplitude (d). After one vibration the algorithm automaticallydure tests just four described more stimuli (d–g according). VPT is then to the the procedure mean of the just last described recognized (d–g ().e VPTor g) isand then the the last mean unper- stimulus is not felt (c), the algorithm automatically tests four more stimuli according to the proce- of the last recognized (e or gceived) and the (f or last g) unperceivedvibration stimulus. (f or g ) vibration stimulus. dure just described (d–g). VPT is then the mean of the last recognized (e or g) and the last unperceived (f or g) vibrationAppendix stimulus. B Appendix B Appendix B Both mechanoreceptors have have one one common common reaction reaction to to increasing increasing contact contact forces: forces: a a doublingdoubling ofof thethe contactcontact forceforce doesdoes notnot leadlead toto aa vibrationvibration stimulusstimulus thatthat isis perceivedperceived twicetwice as Both mechanoreceptors have one common reaction to increasing contact forces: a well.as well. We We summarized summarized the the psychophysical psychophysical relationship relationship in in Figure Figure A2 A2.. To To create create a a sensation sensa- doubling of the contact force does not lead to a vibration stimulus that is perceived twice variabletion variable on the ony the-axis, y-axis, the VPT the VPT values values were were multiplied multiplied by − by1. This−1. This gave gave us theus the perceived per- as well. We summarized the psychophysical relationship in Figure A2. To create a sensa- sensitivityceived sensitivity in arbitrary in arbitrary units (a. units u.). The(a. u.). three The VPTs three obtained VPTs obtained for each for contact each forcecontact condition force tion variable on the y-axis, the VPT values were multiplied by −1. This gave us the per- werecondition summarized were summarized using the using median the for median each subject. for each subject. ceived sensitivity in arbitrary units (a. u.). The three VPTs obtained for each contact force condition were summarized using the median for each subject.

Figure A2. Psychophysical relationship between contact force and perceived sensitivity [VPT*(−1)] in arbitrary units (a. u.) Figure A2. Psychophysical relationship between contact force and perceived sensitivity [VPT*(−1)] in arbitrary units (a. at 200 Hz and 30 Hz. The median was calculated for the three determined vibration perception thresholds (VPT) for each u.) at 200 Hz and 30 Hz. The median was calculated for the three determined vibration perception thresholds (VPT) for contact force condition per subjects. Thus, the scatter plots represent the whole sample size (n = 10). A smoothed conditional Figure A2. Psychophysicalmean was added relationship using the between R function contact “geom_smooth force and perceived ()” in combination sensitivity with[VPT*( the−1)] argument in arbitrary “method units = (a. loess”. u.) at 200 Hz and 30 Hz. The median was calculated for the three determined vibration perception thresholds (VPT) for

J. Clin. Med. 2021, 10, 3083 11 of 12

References 1. Holowka, N.B.; Wynands, B.; Drechsel, T.J.; Yegian, A.K.; Tobolsky, V.A.; Okutoyi, P.; Mang’eni Ojiambo, R.; Haile, D.W.; Sigei, T.K.; Zippenfennig, C.; et al. Foot callus thickness does not trade off protection for tactile sensitivity during walking. Nature 2019, 571, 261–264. [CrossRef][PubMed] 2. Prätorius, B.; Kimmeskamp, S.; Milani, T.L. The sensitivity of the sole of the foot in patients with Morbus Parkinson. Neurosci. Lett. 2003, 346, 173–176. [CrossRef] 3. Simmons, R.W.; Richardson, C.; Pozos, R. Postural stability of diabetic patients with and without cutaneous sensory deficit in the foot. Diabetes Res. Clin. Pract. 1997, 36, 153–160. [CrossRef] 4. Garrow, A.P.; Boulton, A.J.M. Vibration perception threshold—A valuable assessment of neural dysfunction in people with diabetes. Diabetes. Metab. Res. Rev. 2006, 22, 411–419. [CrossRef][PubMed] 5. Zimmerman, A.; Bai, L.; Ginty, D.D. The gentle touch receptors of mammalian skin. Science 2014, 346, 950–954. [CrossRef] [PubMed] 6. Johansson, R.S.; Vallbo, Å.B. Tactile sensory coding in the glabrous skin of the human hand. Trends Neurosci. 1983, 6, 27–32. [CrossRef] 7. Johansson, R.S.; Landstro¨m, U.; Lundstro¨m, R. Responses of mechanoreceptive afferent units in the glabrous skin of the human hand to sinusoidal skin displacements. Brain Res. 1982, 244, 17–25. [CrossRef] 8. Wells, C.; Ward, L.M.; Chua, R.; Inglis, J.T. Regional Variation and Changes With Ageing in Vibrotactile Sensitivity in the Human Footsole. J. Gerontol. Ser. A Biol. Sci. Med. Sci. 2003, 58, B680–B686. [CrossRef][PubMed] 9. Kekoni, J.; Hämäläinen, H.; Rautio, J.; Tukeva, T. Mechanical sensibility of the sole of the foot determined with vibratory stimuli of varying frequency. Exp. Brain Res. 1989, 78, 419–424. [CrossRef][PubMed] 10. Gu, C.; Griffin, M.J. Spatial summation of vibrotactile sensations at the foot. Med. Eng. Phys. 2013, 35, 1221–1227. [CrossRef] 11. Strzalkowski, N.D.J.; Triano, J.J.; Lam, C.K.; Templeton, C.A.; Bent, L.R. Thresholds of skin sensitivity are partially influenced by mechanical properties of the skin on the foot sole. Physiol. Rep. 2015, 3.[CrossRef] 12. Schlee, G.; Sterzing, T.; Milani, T.L. Foot sole skin temperature affects plantar foot sensitivity. Clin. Neurophysiol. 2009, 120, 1548–1551. [CrossRef] 13. Hilz, M.J.; Axelrod, F.B.; Hermann, K.; Haertl, U.; Duetsch, M.; Neundörfer, B. Normative values of vibratory perception in 530 children, juveniles and adults aged 3–79 years. J. Neurol. Sci. 1998, 159, 219–225. [CrossRef] 14. Peters, R.M.; McKeown, M.D.; Carpenter, M.G.; Inglis, J.T. Losing touch: Age-related changes in plantar skin sensitivity, lower limb cutaneous reflex strength, and postural stability in older adults. J. Neurophysiol. 2016, 116, 1848–1858. [CrossRef][PubMed] 15. Perry, S.D. Evaluation of age-related plantar-surface insensitivity and onset age of advanced insensitivity in older adults using vibratory and touch sensation tests. Neurosci. Lett. 2006, 392, 62–67. [CrossRef][PubMed] 16. Cohen, L.H.; Lindley, S.B. Studies in Vibratory Sensibility. Am. J. Psychol. 1938, 51, 44. [CrossRef] 17. Cassella, J.P.; Ashford, R.L.; Kavanagh-Sharp, V. Effect of applied pressure in the determination of vibration sensitivity using the Neurothesiometer. Foot 2000, 10, 27–30. [CrossRef] 18. Lowenthal, L.M.; Derek, T.; Hockaday, R. Vibration Sensory Thresholds Depend On Pressure of Applied Stimulus. Diabetes Care 1987, 10, 100–102. [CrossRef] 19. Hagander, L.G.; Midani, H.A.; Kuskowski, M.A.; Parry, G.J. Quantitative sensory testing: Effect of site and pressure on vibration thresholds. Clin. Neurophysiol. 2000, 111, 1066–1069. [CrossRef] 20. Gregg, E.C. Absolute Measurement of the Vibratory Threshold. Arch. Neurol. Psychiatry 1951, 66, 403. [CrossRef] 21. Era, P.; Hanninen, V. The effects of intensity, frequency, static load, and shore hardness on different parameters in vibration sensitivity measurements. J. Neurol. Neurosurg. Psychiatry 1987, 50, 465–470. [CrossRef] 22. Mildren, R.L.; Strzalkowski, N.D.J.; Bent, L.R. Foot sole skin vibration perceptual thresholds are elevated in a standing posture compared to sitting. Gait Posture 2016, 43, 87–92. [CrossRef] 23. Germano, A.M.C.; Schmidt, D.; Schlee, G.; Milani, T.L. Plantar sensory vibration thresholds are not influenced by body position. Cogent Med. 2016, 3, 1238600. [CrossRef] 24. Gu, C.; Griffin, M.J. Vibrotactile perception thresholds at the sole of the foot: Effects of contact force and probe indentation. Med. Eng. Phys. 2012, 34, 447–452. [CrossRef][PubMed] 25. Schmidt, D.; Schlee, G.; Germano, A.M.C.; Milani, T.L. Larger contactor area increases low-frequency vibratory sensitivity in hairy skin. PeerJ 2020, 2020, 1–13. [CrossRef][PubMed] 26. Zippenfennig, C.; Niklaus, L.; Karger, K.; Milani, T.L. Subliminal electrical and mechanical stimulation does not improve foot sensitivity in healthy elderly subjects. Clin. Neurophysiol. Pract. 2018, 3, 151–158. [CrossRef][PubMed] 27. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2018; Available online: https://www.R-project.org/ (accessed on 12 July 2021). 28. Wobbrock, J.O.; Findlater, L.; Gergle, D.; Higgins, J.J. The aligned rank transform for nonparametric factorial analyses using only anova procedures. In Proceedings of the 2011 Annual Conference on Human Factors in Computing Systems—CHI ’11, Vancouver, BC, Canada, 7–12 May 2011; ACM Press: New York, NY, USA, 2011; p. 143. 29. Rhoades, R.A.; Pflanzer, R. Human Physiology, 4th ed.; Thomson: Stamford, CT, USA, 2003. 30. Bolton, C.F.; Winkelmann, R.K.; Dyck, P.J. A quantitative study of Meissner’s corpuscles in man. Neurology 1966, 16.[CrossRef] 31. Abraira, V.E.; Ginty, D.D. The sensory neurons of touch. Neuron 2013, 79, 618–639. [CrossRef] J. Clin. Med. 2021, 10, 3083 12 of 12

32. Rowe, M.; Tracey, D.; Mahns, D.; Sahai, V.; Ivanusic, J. Mechanosensory Perception: Are There Contributions From Bone- Associated Receptors? Clin. Exp. Pharmacol. Physiol. 2005, 32, 100–108. [CrossRef] 33. Alzheimer, C. Somatoviszerale Sensibilität. In Physiologie: Mit 92 Tabellen; Elsevier: München, Germany, 2008; pp. 50–67. 34. Gardner, E.P.; Martin, J.H. Coding of Sensory Information. In Principles of Neural Science; McGraw-Hill: New York, NY, USA, 2000. 35. Rösler, F. Psychophysiologie der Kognition. Eine Einführung in die Kognitive Neurowissenschaft; Spektrum Akademischer Verlag: Heidelberg, Germany, 2011; ISBN 978-3-8274-2598-0. 36. Manfredi, L.R.; Baker, A.T.; Elias, D.O.; Dammann, J.F.; Zielinski, M.C.; Polashock, V.S.; Bensmaia, S.J. The Effect of Surface Wave Propagation on Neural Responses to Vibration in Primate Glabrous Skin. PLoS ONE 2012, 7, e31203. [CrossRef] 37. Fleming, M.S.; Luo, W. The anatomy, function, and development of mammalian Aβ low-threshold mechanoreceptors. Front. Biol. 2013, 8, 408–420. [CrossRef][PubMed] 38. Strzalkowski, N.D.J.; Peters, R.M.; Inglis, J.T.; Bent, L.R. Cutaneous afferent innervation of the human foot sole: What can we learn from single-unit recordings? J. Neurophysiol. 2018, 120, 1233–1246. [CrossRef][PubMed] 39. Goldstein, E.B. Wahrnehmungspsychologie, 9th ed.; Gegenfurtner, K., Ed.; Springer: Berlin/Heidelberg, Germany, 2015; ISBN 978-3- 642-55073-7.