Cardiovascular and respiratory modulation of tactile ...

Translation and Integration

Cardiovascular and respiratory modulation of tactile afferents in the human finger pad Vaughan G. Macefield Prince of Wales Medical Research Institute and University of New South Wales, Sydney, NSW 2052, Australia (Manuscript received 9 January 2003; accepted 22 July 2003)

Experimental Physiology :

The human finger pad is highly vascularized so it might be expected that the on-going cardiac pulsations in the vicinity of a cutaneous mechanoreceptor would be reflected in its spontaneous or evoked discharge. The purpose of this study was to examine the prevalence of this potential cardiac rhythmicity in a random sample of tactile afferents from the finger pad. Unitary recordings were made from 21 rapidly (‘fast’) adapting (18 FA I; 3 FA II) and 44 slowly adapting (17 SA I; 27 SA II) afferents via tungsten microelectrodes in the median nerve. Skin blood flow was measured over the pulp of a contralateral finger; ECG activity, blood pressure and respiration were also recorded. Cardiac modulation, present either as a simple pulse rhythm or as modulation of an on-going discharge, was expressed by 44 % of the afferents. Only two out of 18 FA I units, and two out of three FA II units, exhibited cardiac rhythmicity, but their temporal coupling to the pulse was very tight. Modulation was more common for the slowly adapting afferents (57 %), and more prevalent among the SA II (65 %) than the SA I (47 %) classes. Nine spontaneously active SA II afferents exhibited respiratory rhythmicity, their background discharge falling during inspiration. It is concluded that cardiac modulation is common for both classes of slowly adapting tactile afferents (but less common for the rapidly adapting afferents), which may have implications for the sensory signalling of tactile information. Experimental Physiology (2003) 88.5, 617–625.

Tactile mechanoreceptors in the glabrous skin of the primate hand provide the brain with sensory information on the shape, size and surface features of an explored object and also provide the CNS with signals of the load and tangential forces experienced by the digits during manipulation of held objects (Macefield, 1998). Four types of low-threshold mechanoreceptor have been identified in human glabrous skin: the Meissner corpuscles and Merkel cell–neurite complexes, both of which are located at the interface between the epidermis and dermis, and the Pacinian corpuscles and Ruffini endings, both of which are located more deeply within and below the dermis (Miller et al. 1958; Johansson & Vallbo, 1983). Because the glabrous skin is well vascularized, particularly in the pulp of the finger pads, one might expect that the constant pulsation of blood vessels would excite the mechanoreceptors located in the vicinity of the vessels. Indeed, such pulse rhythmicity has been observed in the firing of cutaneous afferents associated with hair follicles in the skin of the cat and rabbit (Brown & Iggo, 1967). In addition, while it has been shown that human muscle spindles are modulated by the arterial pulse (McKeon & Burke, 1981), as are muscle spindles in the rabbit (Ellaway & Furness, 1977), the same

has not been shown formally for human cutaneous receptors. Hence, the purpose of the present study was to determine whether cardiac modulation is common in a random sample of tactile afferents supplying the finger pads.

METHODS Data were obtained from 23 experiments in 11 healthy male and 10 healthy female subjects, ranging in age from 20 to 43 years. Each subject provided informed written consent to the procedures, which were conducted according to the Declaration of Helsinki and had the approval of the Committee on Experimental Procedures Involving Human Subjects, University of New South Wales. Recording procedures The hand and forearm were supported by a table and the median nerve located by electrical stimulation through a 1 mm diameter surface probe between the flexor carpi radialis and palmaris longus tendons, approximately 2 cm proximal to the wrist. An insulated tungsten microelectrode (type 25-10-1, Frederick Haer Co., Brunswick, ME, USA or type TM33B20, World Precision Instruments, Sarasota, FL, USA) was inserted through the skin and manually directed into the nerve while delivering weak electrical pulses (0.01–1.0 mA, 0.2 ms, 1 Hz) through the micro-

Publication of The Physiological Society 2548

Downloaded from Exp Physiol (ep.physoc.org) by guest on May 24, 2011

Email: [email protected]



618

V. G. Macefield

electrode; an adjacent non-insulated microelectrode inserted subdermally served as the reference. Once paraesthesiae were elicited below 25 mA, the microelectrode was switched to an amplifier (DAM 80i, World Precision Instruments) and the position of the microelectrode tip manually adjusted until unitary action potentials could be discriminated. Units were classified as rapidly adapting (referred to as fast adapting, FA) or slowly adapting (SA) to a sustained indentation and further subclassifed as type I or type II according to their possession of small and well-defined receptive fields or large and poorly defined fields, respectively. Characteristically, SA II afferents also respond to remote planar skin stretch and FA II afferents to remote percussion stimuli or to blowing over their receptive field. Mechanical thresholds were measured with calibrated von Frey hairs (Simms-Weinstein Aesthesiometers, Stoelting, Chicago, IL, USA) and receptive field areas marked on the skin. The nail of the receptor-bearing digit was glued using cyanoacrylate cement to an aluminium strip, shaped to accommodate the nail, and fixed to a 30 deg wedge that held the digit extended at 150 deg. Neural activity was amplified (w 104), filtered (100 Hz–10 kHz) and sampled by computer at 12.8 kHz (16 bits) using the SC/ZOOM data acquisition and analysis system (Department of Physiology, University of Umeå, Sweden). Electrocardiographic (ECG) activity was recorded with standard Ag–AgCl chest electrodes, respiratory movements with a strain-gauge transducer (Pneumotrace) attached to a strap around the chest, and continuous blood pressure from a finger by pulse plethysmography (Finapres, Ohmeda, Louisville, CO, USA). Skin blood flow was measured from the pulp of a contralateral finger by an infrared photoelectric pulse plethysmograph (ADInstruments, Sydney, Australia). ECG activity was sampled at 3.2 kHz and the other signals at 400 Hz. In some experiments static indentations (4 N, 10 s) of the finger pulp were made with a 5 cm diameter aluminium disc applied normal to the centre of the finger pad via a servo-controlled force motor. Compression force was recorded with a sensitive transducer (Nano F/T transducer, Assurance Technology Incorporated, NC, USA) and the fluctuations in force due to the underlying arterial pulse waves measured at high gain. Data analysis The unitary integrity of each recording was confirmed off-line using the spike recognition feature of SC/ZOOM, which features an algorithm that extracts a spike morphology signature of spikes that pass through dual-amplitude and dual time-window discriminators (Edin et al. 1988). Spike superimposition at high temporal resolution confirmed the unitary integrity of the recording. Instantaneous frequency was generated as the inverse of the interspike interval from the accepted spikes. Fast-Fourier transforms (FFTs) were computed from the instantaneous frequency, ECG, blood flow (or its first time-derivative), respiration or force records (0.1–2.0 Hz window) and used to demonstrate cardiac or respiratory rhythmicity in the firing of individual tactile afferents. For some units spike onset latency was measured from the peak of the first time-derivative of the blood flow signal. Linear regression analysis was performed between this latency and the magnitude of the pulsatile blood flow signal. All statistical analyses were performed with Statistica v5.0 (Tulsa, OK, USA).

RESULTS Afferent sample Unitary recordings were made from 65 myelinated sensory axons within the median nerve at the wrist. Eighteen units

Exp Physiol 88.5

were classified as rapidly adapting type FA I and three as type FA II. Of the 44 slowly adapting receptors 17 were classified as type SA I and 27 as type SA II. Sixteen of the SA II afferents (59 %) were spontaneously active in the absence of external stimulation. With the exception of four afferents located more proximally on the hand, all afferents had receptive fields on the distal phalanx of digits I–IV and responded to stimuli applied to the finger pad. Evidence of cardiac rhythmicity in the discharge of tactile afferents For the 10 experiments in which the majority (66 %) of tactile afferent recordings were made, indices of cardiovascular activity were obtained from the ECG, blood pressure and cutaneous blood flow recordings. For 13 experiments the only index was provided by the systolic pressure pulses detected during a static compression (4 N) of the finger pulp. Twenty-seven out of 61 (44 %) afferents exhibited cardiac rhythmicity, present either as a simple pulse rhythm or (more commonly) as modulation of an ongoing discharge. Only two out of 18 FA I units and two out of three FA II units fired with a cardiac rhythm. Figure 1 shows experimental records from one of the FA II

Figure 1 Raw data showing the spontaneous discharge of a single FA II afferent. This unit (the large spike) was driven by the arterial pulse, generating two spikes with each heartbeat. Note the tight temporal coupling to skin blood flow. Superimposed spikes are illustrated in the inset. dflow/dt; first time-derivative of the skin blood flow signal; BP, blood pressure.


Cardiovascular modulation of tactile afferents

afferents. This ending was spontaneously active, generating two spikes per heartbeat that essentially coincided with the major inflexion points of the blood flow signal. This is illustrated graphically in the ECG-triggered histogram shown in Fig. 2A, in which the occurrence of spikes (4) and the next R-wave (5) are plotted relative to the R-wave of the ECG (time zero). The other FA II ending only discharged with the pulse during static compression of the finger pad. The two FA I endings each fired once per heartbeat, as shown for one of these in the ECG-triggered histogram of Fig. 2B. The records of Fig. 3 show that their discharge was tightly coupled to skin blood flow. Nevertheless, there was some jitter in the occurrence of the spikes, but this could largely be explained by differences in the rate of skin blood



Exp Physiol 88.5

619

flow from beat to beat. This is illustrated graphically for two units in Fig. 4, in which spike onset latency is plotted against the first time-derivative of the blood flow signal; the latency was inversely related to the magnitude of blood flow – the higher the flow the earlier the unit was recruited. Cardiac rhythmicity was more common for the slowly adapting (57 %) than for the rapidly adapting (19 %) afferents, and more prevalent among the SA II (17/27; 63 %) than the SA I (8/17; 47 %) classes. It was usually present as modulation of a background discharge, either spontaneous or evoked by compression of the finger pad. An example of a spontaneously active SA II ending is shown in Fig. 5. The discharge of this unit increased transiently during the primary and secondary peaks of the flow rate signal, indicating its close proximity to a blood vessel. However, note that it also generated a third peak in its discharge that was not coupled to any apparent event in the blood flow trace; presumably, the sensory ending was responding to some hidden oscillation of the blood vessel (e.g. reflection of the systolic pulse from the vessel wall) that could not be detected by the blood flow sensor. As noted above, for 13 experiments cardiac modulation was examined by looking for coupling between the discharge

Figure 2 Cycle-triggered histograms of the occurence of spikes (4) and the next R-wave (R-wave n+1; 5) for one FA II afferent (A) and one FA I afferent (B) relative to the R-wave of the ECG (time zero). The FA II afferent (same unit as in Fig. 1) generated two spikes per heartbeat, whereas the FA I afferent generated only one spike.

Figure 3 Spontaneous discharge of a single FA I afferent (same unit as in Fig. 2B) that generated a single spike with a cardiac rhythm. Superimposed sweeps (n = 10), synchronized to the rising phase of the first timederivative of the skin blood flow signal (dflow/dt) show the tight coupling between blood flow and afferent firing. Superimposed spikes are shown in the inset.


V. G. Macefield

Exp Physiol 88.5



620

Figure 4 Correlations between the magnitude of the skin blood flow (measured in arbitrary units; a.u.) and the spike onset latency, measured from the peak of the first time-derivative of the blood flow signal. The data shown in A are from the same unit illustrated in Fig. 2. Both units showed a significant inverse relationship between blood flow and spike onset latency.

of an afferent and the arterial pressure pulses superimposed on the force record during static compression (4 N) of the finger pulp. Across subjects the average amplitude of these pulsations was 6.2 ± 0.5 mN (n = 19; mean ± S.E.M.). Cardiac modulation was present for neither of two FA I endings, but was present for one FA II, two out of six SA I and five out of nine SA II afferents. Experimental records from one SA I afferent during a static compression are shown in Fig. 6. The discharge of this ending became entrained to the pulse during the late stage of adaptation to the compression stimulus – any modulation during the earlier phase of the response would have been masked by the high discharge rates. Cardiac modulation was easier to see in the SA II afferents: the unit shown in Fig. 7 exhibited clear modulation throughout the plateau phase of the stimulus. For this unit, cardiac modulation was confirmed by the presence of a clear peak in the power spectrum of the instantaneous frequency record corresponding to the cardiac frequency (not shown), as measured from the force record. Three SA II endings in this sample also fired a solitary spike at the cardiac rhythm (confirmed by palpation of the radial artery) in the absence of any applied stimulus.

Figure 5 Modulation of the background firing rate of one SA II afferent by the underlying arterial pulsations. This unit increased its discharge during the primary and secondary peaks in the blood flow signal, but also exhibited a third peak that was not related to any observable component in the blood flow signal.

On the whole, the magnitude of the cardiac modulation was higher for the SA I than for the SA II afferents. Calculated from the tonic discharge (spontaneous or evoked) of 10 SA I and 16 SA II endings, the mean modulation ( ± S.E.M.) was 172 ± 51 and 33 ± 10 %, respectively. This was due both to the lower static firing rates (10.3 ± 2.7 vs. 17.7 ± 2.1 Hz) and higher pulserelated increases in discharge (15.0 ± 4.7 vs. 4.4 ± 0.9 Hz) of the SA I than of the SA II endings. All of these differences were statistically significant (P < 0.05, paired t test). The tonic discharges of five SA I and eight SA II afferents, measured from a subsample of these afferents



621



Exp Physiol 88.5

Figure 6 Modulation of the static discharge of one SA I afferent during a 4 N static compression of the finger pad. The discharge of the receptor was phasically coupled to the vascular pulsations in the finger pad when it had adapted to the compression stimulus.

studied during the final 1 s of a static compression (4 N) of the finger pulp, were 6.6 ± 0.9 and 22.2 ± 7.5 Hz, respectively, whereas their phasic increases were 10.9 ± 3.8 and 5.1 ± 4.0 Hz, respectively. Respiratory modulation Although the firing rates of SA II afferents are typically very regular, the background discharge of nine spontaneously active SA II afferents showed periodic fluctuations. Spectral analysis of the instantaneous frequency record revealed that this periodicity corresponded to that of respiration. Experimental records from two SA II afferents are illustrated

in Fig. 8. It is clear that the firing rate declined during inspiration, as shown by the inverted and scaled respiratory record superimposed on the instantaneous frequency trace. It is also clear that the amplitude of this respiratory modulation could be large, some 20 % for the unit in Fig. 8A and 50 % for the unit in Fig. 8B. The depression in firing during inspiration could not be accounted for by an inspiration-related fall in blood volume; low-pass filtering of the flow signal (to provide an index of changes in blood volume) did not demonstrate such a decrease. For three SA II units, a spectral peak was present at the cardiac as well as the respiratory frequency, indicating that their

Figure 7 Modulation of the static discharge of a single SA II afferent during a 4 N static compression of the finger pad. Amplification of the force signal revealed arterial pulsations to which the discharge of the receptor was phasically coupled.


622

V. G. Macefield



background discharge was modulated by both the cardiovascular and respiratory systems. An example from one such unit is shown in Fig. 9.

DISCUSSION In this study we have shown that almost half of a sample of low-threshold cutaneous mechanoreceptors (44 %) in the finger pads of the human hand respond to the normal arterial pulsations occurring within the finger at rest. In addition, some spontaneously active SA II endings exhibited respiratory modulation of their background discharge. The finger pulp is highly vascularized (Chaudakshetrin et al. 1988); accordingly, it is very compliant at compression forces below 1 N (Westling & Johansson, 1987; Serina et al. 1997), so arterial pulsations will be readily transmitted through this viscoelastic medium. Moreover, the finger pad has the highest density of tactile afferents in the hand (Johansson & Vallbo, 1979), which are used both for fine tactile discrimination of surface features and for encoding the forces associated with manipulation of held objects. Because the cardiovascular (and respiratory) fluctuations in the local environment of the receptors are always present, it is reasonable to conclude that these events are an obligatory source of ‘physiological noise’ that may influence the signalling capacities of tactile afferents.

Exp Physiol 88.5

Methodological considerations In the present study, data were reported from cutaneous afferents sampled sequentially during the course of microneurographic studies designed to address other issues of tactile afferent physiology. There was no selection bias towards those possessing cardiac or respiratory rhythmicity over those that did not, and analysis of such rhythmicities was applied to those units with and without overt rhythmicity. Nevertheless, the proportion of slowly adapting afferents (particularly the SA II endings) in this sample was relatively high, which can be explained by a searching strategy (slow finger movements and skin stretch rather than brisk tapping over the digits) that favoured such afferents. For the majority of recordings, direct measurements of ECG activity, blood pressure, skin blood flow and respiration were available, which allowed detailed analysis of the parameters to which afferent firing was coupled. For the remainder, previously recorded but unreported data obtained during static compression of the finger pad were analysed to examine the presence of cardiac modulation. As such, these latter experiments were limited in that cardiac modulation could only be quantified by measuring the arterial pulsations in the force record at high gain (ECG was not recorded). Although measuring the common peaks in the force and instantaneous frequency records by Fourier analysis did confirm that the cardiac modulation

Figure 8 Respiratory modulation of the background firing rate of two different SA II endings (A and B). The respiratory signal (inspiration upwards) was inverted, scaled and superimposed on the instantaneous frequency trace to better illustrate the modulation. Downloaded from Exp Physiol (ep.physoc.org) by guest on May 24, 2011

Exp Physiol 88.5




was real, this approach necessarily excludes identification of those spontaneously active units possessing cardiac modulation in the absence of a compression stimulus (other than those three units for which firing to the arterial pulse was confirmed by palpation of the radial artery). Cardiac modulation of rapidly adapting cutaneous afferents Very few (2/18; 11 %) rapidly adapting afferents with small receptive fields (FA I afferents), which supply the Meissner corpuscles located at the interface between the epidermis and dermis (Darian-Smith, 1984), exhibited cardiac modulation, but the discharge of the two that did was tightly coupled to skin blood flow. Two of the three FA II afferents, which supply the Pacinian corpuscles located subdermally, fired at a cardiac rhythm. Given their exquisite mechanical sensitivity and their location near blood vessels (Stark et al. 1998) one would expect that many Pacinian corpuscles would be driven by the arterial pulsations. Unfortunately, owing to their low density in the human hand (Johansson & Vallbo, 1979; Stark et al. 1998) and their relative paucity in microneurographic recordings, it is difficult to say that two-thirds of all Pacinian corpuscles would discharge with the arterial pulse. Indeed, because a spontaneously active FA II afferent does stand out as being unusual, one would normally suspect that such an afferent would fire with the cardiac rhythm and confirm this by monitoring the arterial pulse. So it is most certainly not the case that two-thirds of all FA II afferents in the hand fire with a pulse rhythm; the proportion of Pacinian corpuscles that do so is probably similar to that of the Meissner corpuscles (FA I afferents). According to calculations by Johansson & Vallbo (1979), the density of FA I endings in the distal phalanx is 77 units cm_2 and the total area of skin on all five distal phalanges is 33 cm2. For an average area of skin on each distal phalanx of 6 cm2 , the total number of FA I units in this area would be 460. Based on the percentage of FA I afferents with cardiac rhythmicity (11 %), this would mean that about 50 FA I afferents in each finger pad would fire with the pulse. If one assumes that a similar proportion of FA II endings fires with the pulse then, based on their density in the terminal phalanx (15 units cm_2, Johansson & Vallbo, 1979), only 10 (i.e. 15 w 6 w 0.11) Pacinian corpuscles in each finger pad would be expected to respond to on-going arterial pulsations. Cardiac modulation of slowly adapting cutaneous afferents Unlike the rapidly adapting afferents, the presence of cardiac rhythmicity was quite high in both classes of slowly adapting afferent, yet was more common for the SA II (63 %) than the SA I (47 %) classes. Whether this is due to the fact that the SA II afferents, which innervate the Ruffini endings, are located more deeply (within the dermis) than the SA I afferents (which innervate the Merkel cell–neurite complexes in the dermal papillae; Johansson & Vallbo, 1983) is not known. For the spontaneously active SA II endings, the cardiac rhythmicity was expressed as a modulation of the background discharge. This may well facilitate detection of

623

pulse rhythmicity if one considers the background firing rate a ‘carrier frequency’ on which the (cardiac) frequency modulation is imposed. Based on the density of SA I and SA II afferents in the distal phalanges reported to be 42 and 28 units cm_2, respectively (Johansson & Vallbo, 1979), each distal phalanx is supplied by some 250 SA I endings and 170 SA II endings (including those clustered around the nails). This means that about 120 (i.e. 250 w 0.47) SA I and 107 (i.e. 170 w 0.63) SA II endings in each distal phalanx respond to the arterial pulsations. Respiratory modulation of slowly adapting cutaneous afferents The background discharge of nine spontaneously active SA II endings was modulated by respiration; falling during inspiration. This was not an indirect effect of changes in skin blood flow but was, evidently, a consequence of the mechanical coupling of the skin in the digits to the torso – Ruffini endings (SA II afferents) are very sensitive to forces in the plane of the skin (Chambers et al. 1972). However, it must be pointed out that the fingernail of the receptorbearing digit was held to the supporting table, thereby allowing a fixation point about which relative movement of the skin could occur. Respiratory modulation has also

Figure 9 Fast-Fourier transform (FFT) of the background firing rate, ECG activity and respiration signal for the same SA II afferent illustrated in Fig. 8A. Note that the firing rate of the afferent was modulated by both the cardiac rhythm and by respiration.


624

V. G. Macefield



been observed in the background discharge of a muscle spindle ending in soleus, for which respiratory movements were picked up by a force transducer used to measure ankle torque (Fig. 5 in Gandevia & Burke, 1985). Implications Given that a total of 60 rapidly adapting (50 FA I; 10 FA II) and 230 slowly adapting (120 SA I; 110 SA II) tactile afferents in the distal phalanx of each finger would be carrying some cardiac modulation of their discharge, this begs the question: does this physiological noise have an impact on the signalling capacities of our tactile afferents? This is difficult to answer. We know that the arterial pulsations in the finger pad are very small, on average only 6 mN, which, when superimposed on a 4 N static compression, amounts to a modulation of only 0.15 %. However, the modulation of individual tactile afferents, particularly the SA I endings, can be very high, with instantaneous frequencies associated with the pulse being nearly two times higher than the background discharge. Even the tonic discharge of SA II afferents carried a sizable cardiac modulation, amounting to one-third of the background discharge. The effect of this modulation would be to increase the variability in the afferent discharge. Recently, the effects of different types of noise on the responses of SA I afferents in the finger pad of the monkey have been modelled (Goodwin & Wheat, 1999). Although cardiovascular noise was not considered, according to their model this type of noise could be described as ‘additive noise’, the effect of which would be to decrease the resolution of the afferent signal. Given that these are the only class of tactile afferent capable of encoding a wide range of local curvatures, and are hence important in shape discrimination (Goodwin et al. 1995, 1997) and are considered to play the primary role in encoding roughness (Blake et al. 1997), any imposed noise may be expected to influence their capacity to signal these important tactile features. Moreover, we know that the input from a single SA I afferent can be perceived, as can the input from individual FA I and FA II afferents (Ochoa & Torebjörk, 1983; Vallbo et al. 1984; Torebjörk et al. 1987; Macefield et al. 1990). With respect to the SA II afferents, we also know that the input from a single ending generally does not reach perceptual levels (Ochoa & Torebjörk, 1983; Vallbo et al. 1984; Torebjörk et al. 1987; Macefield et al. 1990), but it is possible that the input from many SA II afferents does. This is important when one considers the fact that the arterial pulsations in the finger pulp serve as a source of synchronization, in which many SA I and SA II afferents, and some FA I and FA II afferents, will increase their discharge (or be recruited) within a relatively narrow window of time. Unless it is filtered out at higher levels of the CNS, the pulse-related discharge of cutaneous afferents may also be perceived (i.e. as a signal). That we do not perceive the ongoing pulsations of the finger pad, even when attention is directed at it, may suggest that this ‘cardiovascular noise’ is indeed filtered out. Yet the fact that we can feel arterial pulsations when a finger pad contacts a rigid surface (or

Exp Physiol 88.5

squeezes against the thumb) means that enough cutaneous afferents in the finger pads respond to these mechanical events to reach perceptual levels. Given the need for highfidelity transduction of a mechanical stimulus in tasks requiring fine tactile discrimination, it is important to know the potential sources of physiological noise that may influence the sensory signals required by the CNS. It is concluded that cardiovascular noise is common in the discharge of primary afferents from the glabrous skin of the hand, and that this must either be filtered out or ignored by the CNS.

Blake DT, Hsiao SS & Johnson KO (1997). Neural coding mechanisms in tactile pattern recognition: the relative contributions of slowly and rapidly adapting mechanoreceptors to perceived roughness. J Neurosci 17, 7480–7489. Brown AG & Iggo A (1967). A quantitative study of cutaneous receptors and afferent fibres in the cat and rabbit. J Physiol 193, 707–733. Chambers MR, Andres KH, Duering MV & Iggo A (1972). The structure and function of the slowly adapting type II mechanoreceptor in hairy skin. Q J Exp Physiol 57, 417–445. Chaudakshetrin P, Kumar VP, Satku K & Pho RW (1988). The arteriovenous pattern of the distal digital segment. J Hand Surg [Br] 13, 164–166. Darian-Smith I (1984). The sense of touch: performance and peripheral neural processes. In Handbook of Physiology, section 1, The Nervous System, ed. Darian-Smith I, pp. 739–788. American Physiological Society, Bethesda. Edin B, Bäckström PA & Bäckström LO (1988). Single unit retrieval in microneurography: a microprocessor-based device controlled by an operator. J Neurosci Methods 24, 137–144. Ellaway PH & Furness P (1977). Increased probability of muscle spindle firing time-locked to the electrocardiogram in rabbits. J Physiol 273, 92P. Gandevia SC & Burke D (1985). Effect of training on voluntary activation of human fusimotor neurones. J Neurophysiol 54, 1422–1429. Goodwin AW, Browning AS & Wheat HE (1995). Representation of curved surfaces in responses of mechanoreceptive afferent fibers innervating the monkey’s finger pad. J Neurosci 15, 798–810. Goodwin AW, Macefield VG & Bisley JW (1997). Encoding of object curvature by tactile afferents from human fingers. J Neurophysiol 78, 2881–2888. Goodwin AW & Wheat HE (1999). Effects of nonuniform fiber sensitivity, innervation geometry, and noise on information relayed by a population of slowly adapting type I primary afferents from the finger pad. J Neurosci 19, 8056–8070. Johansson RS & Vallbo ÅB (1979). Tactile sensibility in the human hand: relative and absolute densities of four types of mechanoreceptive units related to psychophysical thresholds in the human hand. J Physiol 297, 405–422. Johansson RS & Vallbo ÅB (1983). Tactile sensory coding in the glabrous skin of the human hand. Trends Neurosci 6, 27–31.

Macefield G, Gandevia SC & Burke D (1990). Perceptual responses to microstimulation of single afferents innervating joints, muscles and skin of the human hand. J Physiol 429, 113–129. Downloaded from Exp Physiol (ep.physoc.org) by guest on May 24, 2011

Exp Physiol 88.5




Macefield VG (1998). The signalling of touch, finger movements and manipulation forces by mechanoreceptors in human skin. In Neural Aspects of Tactile Sensation, ed. Morley JW, pp. 89–130. North Holland Elsevier, Amsterdam. McKeon B & Burke D (1981). Component of muscle spindle discharge related to arterial pulse. J Neurophysiol 46, 788–796. Miller MR, Ralston HJ & Kasahara M (1958). The pattern of cutaneous innervation of the human hand. Am J Anat 102, 183–217. Ochoa JL & Torebjörk HE (1983). Sensations evoked by intramural microstimulation of single mechanoreceptor units innervating the human hand. J Physiol 342, 633–654. Phillips JR, Johansson RS & Johnson KO (1992). Responses of human mechanoreceptive afferents to embossed dot arrays scanned across finger pad skin. J Neurosci 12, 827–839. Serina ER, Mote CD Jr & Rempel D (1997). Force response of the fingertip pulp to repeated compression – effects of loading rate, loading angle and anthropometry. J Biomech 30, 1035–1040. Stark B, Carlstedt T, Hallin RG & Risling M (1998). Distribution of human Pacinian corpuscles in the hand. A cadaver study. J Hand Surg [Br] 23, 370–372. Vallbo ÅB, Olsson KA, Westberg KG & Clark FJ (1984). Microstimulation of single tactile afferents from the human hand. Sensory attributes related to unit type and properties of receptive fields. Brain 107, 727–749. Torebjörk HE, Vallbo ÅB & Ochoa JL (1987). Intraneural microstimulation in man. Its relation to specificity of tactile sensations. Brain 110, 1509–1529. Westling G & Johansson RS (1987). Responses in glabrous skin mechanoreceptors during precision grip in humans. Exp Brain Res 66, 128–140.

Acknowledgements This work was supported by the National Health and Medical Research Council of Australia (Program Grant 002306).


625