- Email: [email protected]
Evidence that tactile stimulation inhibits nociceptive sensations produced by innocuous contact cooling
Behavioural Brain Research 162 (2005) 90–98
Research report
Evidence that tactile stimulation inhibits nociceptive sensations produced by innocuous contact cooling Barry G. Green a,b,∗ , Kate L. Schoen a b
a The John B. Pierce Laboratory, Yale University School of Medicine, 290 Congress Avenue, New Haven, CT 06519, USA Department of Surgery (Otolaryngology), Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06510, USA
Received 19 December 2004; received in revised form 3 March 2005; accepted 6 March 2005 Available online 13 April 2005
Abstract It was recently shown that stinging, pricking or burning is reliably perceived by some individuals when the skin is cooled to temperatures as mild as 25–30 ◦ C. These seemingly paradoxical sensations, which have been termed innocuous-cold nociception (ICN), were significant only when cooling was produced by a thermode resting statically on the skin (static contact); touching an already cooled thermode to the skin (dynamic contact) produced reports of only coolness and cold. The present study investigated the hypothesis that ICN is inhibited by tactile stimulation produced when a thermode contacts the skin. Experiment 1 pitted the tactile hypothesis against an alternative explanation that inhibition results from higher rates of skin cooling during dynamic contact. ICN was measured at three different cooling rates (−1.0, −2.5, −5.0 ◦ C/s) when the thermode was resting on the skin or was touched to the skin at the moment cooling began. The results supported the tactile hypothesis: faster cooling rates during static contact led to stronger rather than weaker nociceptive sensations, and ICN was suppressed even when dynamic contact was coincident with the onset of cooling, and thus could not affect cooling rate. Experiment 2 confirmed the latter result and showed that suppression was greatest at 28 ◦ C, less at 24 ◦ C, and not significant at 18 ◦ C. We conclude that dynamic tactile stimulation produced by contact with a surface inhibits the nociceptive component of innocuous but not noxious cooling. The implications of this conclusion for the role of cold perception in behavioral thermoregulation versus haptic perception, and for theories of cold perception in general, are discussed. © 2005 Published by Elsevier B.V. Keywords: Cold; Psychophysics; Pain; Touch; Inhibition; Individual differences
1. Introduction Mild cooling has been assumed to produce only sensations of coolness and cold. Contrary to this view, it was recently discovered that when subjects are asked to rate the intensity of nociceptive as well as thermal sensations, they often report stinging or burning at skin temperatures between 25 and 30 ◦ C [1]. The occurrence of these unexpected sensations, which are referred to as ‘innocuous-cold nociception’ (ICN), raises new questions about the sensory processes that underlie perception of nonpainful cold.
∗
Corresponding author. Tel.: +1 203 458 2262; fax: +1 203 624 4950. E-mail address: [email protected] (B.G. Green).
0166-4328/$ – see front matter © 2005 Published by Elsevier B.V. doi:10.1016/j.bbr.2005.03.015
First, ICN implies that temperatures as mild as 30 ◦ C can stimulate the nociceptive pathway. Such stimulation cannot be accounted for by C-polymodal nociceptors (CPNs), which have cooling thresholds from about 18 to 22 ◦ C and are believed to mediate the cold pain threshold [2,3]. Instead, afferent fibers must exist that respond to noxious cold yet have cooling thresholds in the innocuous temperature range. Although cold-sensitive fibers with these general characteristics have been reported in the past [4,5], their responses to temperatures above the cold pain threshold were not considered a significant factor in cold perception. Second, the occurrence of ICN in only about half of all individuals tested [1] means people differ in their perception of innocuous cold. The possibility that these individual differences might be attributable to cognitive factors or to response bias was ruled out by show-
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
ing that ICN was perceived only under limited conditions of stimulation when cooling occurred as a thermode was resting statically against the skin. If instead the thermode was cooled and then touched to the skin, ICN was virtually absent. The absence of ICN when a thermode is touched to the skin, the large individual differences in its occurrence, and the common practice in psychophysical studies of asking subjects to attend only to cold sensations, probably explains why ICN had not been reported in prior studies of cold perception. The purpose of the present study was to investigate further the stimulus conditions necessary for both induction and suppression of ICN. Particular emphasis was placed on evaluating the hypothesis that suppression is caused by tactile stimulation produced when a thermode first contacts the skin. If this hypothesis were supported, suppression of ICN would provide the first evidence that cold perception does not depend solely on the parameters of the thermal stimulus. 2. Experiment 1 In their discussion of ICN, Green and Pope [1] noted that in addition to producing dynamic tactile stimulation, touching a cooled thermode to the skin undoubtedly lowers skin temperature more rapidly than does cooling via a Peltier thermode already in contact with the skin. In the latter case, cooling rate is slowed by the thermal inertias of both the thermode and the skin. Because the stimulus parameters that influence ICN were unknown, it was possible that higher rates of cooling during dynamic contact might work against perception of ICN. Green and Pope provided an initial test of this possibility by collecting data on the frequency of burning, stinging or pricking sensations perceived during different rates of cooling (between −0.5 and −5.0 ◦ C/s). Although the incidence of ICN did not change with cooling rate, it remained possible that its intensity did. In the present experiment, we investigated this possibility by measuring the perceived intensity of ICN at three different cooling rates. A condition was also included in which cooling was started at the moment of skin contact, thereby producing dynamic tactile stimulation that was correlated with the onset of cooling but did not affect cooling rate. Significant suppression of ICN in the latter condition would support the tactile inhibition hypothesis. 2.1. Methods 2.1.1. Subjects A total of 44 subjects (28 females and 16 males, average age 23.3 years) served in Experiment 1. Subjects were recruited on and around the Yale University campus and were screened by questionnaire to exclude individuals with a history of neurological injury or pathology as well as those with cutaneous lesions or disorders affecting the site of stimulation. Each subject gave informed consent and was paid for his/her participation.
91
2.1.2. Equipment Temperature stimuli were produced using a 4 × 4 array of 0.64 cm2 Peltier thermoelectric modules (hereafter referred to collectively as ‘the thermode’) designed and built in The John B. Pierce Laboratory machine and electrics shop. The 16 independently controlled modules are individually bonded to a water-circulated heat sink. Epoxied to the top of each ceramic module is a machined copper plate (8 mm × 8 mm × 0.96 mm) with a 40-ga copper constantin thermocouple recessed in a 0.5-mm deep groove in the center of the plate. The thermocouple provides constant temperature feedback at the skin–thermode interface. The parameters of thermal stimulation for each Peltier module are controlled via a LabView program that enables the experimenter to specify the base (adapting) temperature, rate of temperature change, and target temperature (all with resolutions of ±0.1 ◦ C), as well as dwell time at the target temperature (±0.1 s). The surface temperature of each thermode was periodically checked using a thermocouple thermometer (Physitemp, BAT-12) to ensure an accuracy of ±0.2 ◦ C. The thermode is mounted on a three-way microscope stage which, in turn, is attached via a lockable ball joint to a floor-mounted positioning arm. The thermode and its positioning system stand beside a modified dental chair in which the subject sits. Stimulation occurred on the subject’s right forearm as it rested volar side up on a foam-padded armrest. Placement of the thermode on the subject’s forearm was achieved manually. Its location on the skin was adjusted before testing began by changing the angle of the positioning arm in three dimensions until the stimulating face of the thermode was aligned parallel to the skin’s surface just above the desired area of stimulation. For static conditions of stimulation the thermode was pressed lightly down onto the skin with a force sufficient to make full contact (the bearing force was not measured). It was then locked in place by flipping a switch that activated a pneumatic brake on the positioning arm. Locking the thermode in place both stabilized its position and prevented its full weight from bearing on the subject’s arm. For conditions that required dynamic contact of the thermode against the arm, the thermode was first positioned approximately 1/2 in. above the arm and locked in place. At the desired time the experimenter released the brake and brought the thermode directly down to the skin with sufficient pressure to make full contact. All testing was conducted in a laminar-flow environmental chamber with air temperature and relative humidity controlled at 26 ◦ C and approximately 30%, respectively.
2.1.3. Psychophysical practice procedure Each of the experiments included a separate practice session for new subjects who had not participated in any prior thermal perception experiments in our laboratory. The practice session was completed before the first test session and consisted of two brief (10–15 min) exercises designed to train individuals to use the Labeled Magnitude Scale (LMS [6]) to rate the intensity of thermal sensations. The LMS is a type of “category-ratio” scale [7] in which labeled intensity descriptors (e.g., “weak”, “strong”) are spaced according to their empirically-derived semantic magnitudes. The scale is bounded at the bottom by “no sensation” and at the top by “strongest imaginable sensation of any kind” [8]. A computer mouse is used to move an arrow along the scale to the point that corresponds with the perceived strength of a given sensation; double-clicking the mouse saves the response. To familiarize subjects with the meaning and use of the LMS, they were first asked to imagine 16 commonly experienced thermal
92
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
sensations (e.g., washing hands in cold tap water; walking barefoot on hot pavement) and to rate the estimated intensity of the sensations on the scale. In a second exercise subjects rated the intensity of 11 thermal stimuli produced by the thermode as it rested on the right forearm. Starting from a base temperature of 33 ◦ C, the experimenter presented an alternating series of warming (38, 36, 40, 42 ◦ C) and cooling (22, 30, 28, 20, 24, 26, 18 ◦ C) stimuli on two rows of the thermode (i.e., a total of eight thermodes). The rate of temperature change for the practice stimuli was ±3.0 ◦ C/s, with a 5.0 s dwell at the target temperature and a 30-s inter-trial interval (ITI). Subjects performed three tasks during every trial: a rating of the intensity of thermal sensation (cool-cold, warm-hot), a rating of the intensity of nociceptive sensations (burning, stinging or pricking), and a checklist task in which they clicked on one or more descriptors (nothing, cool, cold, warm, hot, burning, stinging/pricking, aching, and painful) to indicate the sensation qualities they had perceived. Subjects were told to choose as many or as few descriptors as necessary to describe each sensation fully, and to respond with “nothing” on trials when no sensation was felt.
2.1.4. Experimental procedure The effect of cooling rate on ICN was measured under two conditions, static and dynamic + static. In the static condition the thermode was cooled to the target temperature as it rested on the skin. In the dynamic + static condition the experimenter lifted the thermode from the skin before triggering cooling, then returned it to the skin at the moment cooling began (approximately 1 s later). Skin contact occurred as close to the moment cooling began as was physically possible under manual control. The results from these two conditions were compared to those from a third condition, dynamic, in which the thermode was set to the target temperature before it was touched to the skin. Thus, in the dynamic condition skin cooling occurred at a fast but unknown rate that was determined by the rate of heat transfer between the skin and the already cooled thermode. Three different cooling rates were tested in the static and dynamic–static conditions: −1.0, −2.5 or −5.0 ◦ C/s. The target temperature in all three conditions was 28 ◦ C, with a dwell time of 5 s. The orders of conditions and cooling rates were counterbalanced across subjects and over two experimental sessions. Half of the subjects received the static condition first and had ascending rates of cooling in the first session, then received the static–dynamic condition first with descending rates of cooling in the second session; the other half received the conditions and cooling rates in reverse order. The dynamic condition was always presented between the other two conditions. Within each session two different strategies were used to avoid sensory adaptation from repeated testing at the same site: (1) on each trial testing was conducted on only two of the four rows of the thermode (a total of eight, 0.64 cm2 thermodes); and (2) placement of the thermode was alternated between two sites on the volar surface of the right forearm after each pair of trials. At the beginning of a session the thermode was placed on Site 1 for a 3-min adaptation to the 33 ◦ C base temperature. The first combination of condition and cooling rate was then presented using rows 1 and 2 of the thermode. After the thermodes returned to base temperature there was a 1-min ITI before the same combination of condition and rate was presented to rows 3 and 4. The experimenter then moved the thermode to Site 2 and repeated the procedure, beginning with the 3-min adaptation at 33 ◦ C. This sequence was followed until each condition and cooling rate had been presented twice.
On static trials, in which the thermode remained on the subject’s forearm throughout cooling, the experimenter cued the subject that a stimulus was about to begin by saying, “ready, attend now”, and indicated when to rate sensation intensity and quality by saying “ready, rate”. On dynamic and dynamic + static trials, in which the thermode was lifted and then placed back on the arm, the experimenter told the subject to attend to the sensation as soon as the thermode was replaced and to make intensity and quality ratings after the experimenter said “rate”. For all conditions, the subjects were instructed to rate the maximum sensations felt up to the time of the experimenter’s signal, which was given just prior to termination of the 28 ◦ C cooling stimulus (i.e., at the end of the 5 s dwell time).
2.2. Results To assess differences in ICN across conditions, only those subjects (n = 25) who rated burning/stinging/pricking (hereafter referred to as nociceptive sensations) as more than ‘barely detectable’ in the static condition during the slowest rate of cooling (−1.0 ◦ C/s) were included in the statistical analyses. An ANOVA conducted on the thermal ratings (coolness, cold) for all subjects prior to this segregation showed that the subset of individuals who reported nociceptive sensations did not rate the stimuli to be any colder than did individuals who reported no nociceptive sensations [main effect group; F(1,42) = 0.42, p = 0.52]. The main results of the experiment are shown in Fig. 1. On the left side of the figure are the log-mean ratings for cold and nociceptive sensations in the static condition, which both increased with cooling rate. A multivariate repeated measures
Fig. 1. The log-mean ratings of cold and nociceptive sensations as a function of cooling rate for the three conditions of Experiment 1. In the static condition, both types of sensations increased in intensity with cooling rate, ruling out cooling rate as a cause of contact-suppression of ICN. In the dynamic condition, only nociceptive sensations were significantly suppressed relative to all cooling rates in the static condition. In the dynamic + static condition, suppression of nociceptive sensations increased with cooling rate, indicating that cooling rate has opposite effects on cold and nociceptive sensations. Vertical bars represent the standard errors of the means (S.E.M.).
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
ANOVA calculated for these data only, with rate of cooling and sensation quality as factors, confirmed that there was a main effect of cooling rate [F(2,48) = 7.4, p < 0.005]. There was no interaction between cooling rate and sensation quality. Because both cold and nociceptive sensations increased in intensity with rate of cooling, the results rule out cooling rate as a cause of contact-suppression of ICN. An overall ANOVA conducted on the data for all three stimulus conditions from the same group of subjects revealed significant main effects of condition [F(6,144) = 13.2, p < 0.0001] and sensation quality [F(1,24) = 47.5, p < 0.0001], and an interaction between these factors [F(6,144) = 10.7, p < 0.0001]. The latter interaction was driven primarily by a much stronger suppression of nociceptive sensations compared to cold in the dynamic condition (center of Fig. 1), and by diverging effects of cooling rate for the different qualities in the dynamic + static condition. The stronger suppression of nociceptive sensations was confirmed by Tukey HSD tests (p < 0.05), which indicated that suppression was significant in the dynamic condition relative to all cooling rates in the static condition, whereas cold ratings were significantly suppressed relative only to the highest cooling rate (−5.0 ◦ C/s) in the static condition. In terms of sensation magnitude, nociceptive sensations in the dynamic condition were reduced by 69% from the level reported for −1.0 ◦ C/s cooling, whereas cold sensation was reduced by only 13%. Relative to −5 ◦ C/s cooling in the static condition, nociceptive sensations were reduced by 77%, and cold by only 37%. In contrast to the static condition, nociceptive sensations tended toward lower rather than higher intensities at the fastest cooling rate in the dynamic + static condition. An ANOVA that included only the dynamic + static data showed an interaction between cooling rate and sensation quality [F(2,48) = 9.2, p < 0.001], confirming the visual impression that cooling rate had opposite effects on cold and nociceptive sensations. The latter sensations were significantly lower in the dynamic + static condition than in the static condition for the two fastest cooling rates, but not for the slowest rate of −1.0 ◦ C/s (Tukey HSD, p < 0.05). The implication of this result is that the suppressive effect of dynamic contact can be significant even when skin contact occurs before the thermode has been cooled, but only when the delay between contact and peak cooling is short. For cooling rates of −2.5 and −5.0◦ C/s, the delay was approximately 2.5 and 1 s, respectively. Fig. 2 shows the frequencies at which the primary sensation qualities were reported in the different conditions and for different rates of cooling. The two most frequently reported qualities were cold and stinging/pricking. Both sensations show a slight trend toward higher frequencies of occurrence at higher rates of cooling in the static condition. More interesting, however, is the large decrease in stinging/pricking in the dynamic condition. ICN was predominantly a stinging and/or pricking sensation, the occurrence of which fell from 66% of trials at the fastest rate of cooling in the static
93
Fig. 2. Frequencies at which thermal and nociceptive sensation qualities were reported for three conditions at three rates of cooling in Experiment 1. The data indicate that ICN has predominantly stinging and/or pricking qualities that are suppressed by dynamic contact.
condition to only 20% of trials in the dynamic condition. In contrast, reports of burning sensations never exceeded 25% of trials in any condition, and remained essentially unchanged across cooling rate and condition.
3. Experiment 2 Having ruled out the rate-of-cooling hypothesis and having found support for the tactile inhibition hypothesis, we next investigated whether suppression of nociceptive sensations by dynamic contact was unique to ICN or extended to lower, potentially noxious temperatures. As mentioned above, the peripheral source of ICN remains speculative, but near and below the cold pain threshold nociceptive sensations are thought to be mediated primarily by CPNs [2,3,9]. If suppression occurs in the temperature range served by CPNs, it would suggest that it is part of a general inhibitory mechanism by which tactile stimulation inhibits pain [10–13]. In addition, we investigated whether ICN is still suppressed when dynamic tactile stimulation occurs after the target temperature has been reached. Significant suppression after the skin has been cooled would provide further support for the tactile inhibition hypothesis. 3.1. Methods 3.1.1. Subjects Thirty-one individuals (16 females and 15 males, average age 25 years) served in Experiment 2. Subjects were recruited and screened in the same manner as Experiment 1. Each subject gave informed consent and was paid for his/her participation. 3.1.2. Equipment and practice procedure The thermal stimulation system and psychophysical practice procedures were the same as in Experiment 1. Only subjects who had
94
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
not participated in Experiment 1 were required to serve in the practice session. 3.1.3. Experimental procedure There were four conditions in the experiment. Three of the conditions – static, dynamic, and dynamic + static – were the same as in Experiment 1. In the fourth condition, which we called static + dynamic, cooling began as in the static condition with the thermode resting against the skin, but as soon as the target temperature was reached, the thermode was lifted from the skin and then immediately replaced (within approximately 1 s). Subjects were instructed to rate the sensation intensity and quality just after the thermode was returned to the skin, and it was emphasized that their ratings should reflect the sensations experienced at the exact moment of the experimenter’s signal. This was a critical feature of the psychophysical procedure, because the static + dynamic condition made it possible for maximum sensations to occur when the target temperature was reached during the static phase of stimulation. Yet, the primary object of the study was to measure the sensations felt immediately after the thermode re-contacted the skin (i.e., during the dynamic phase of stimulation). Three stimulus temperatures were used: 28, 24 and 18 ◦ C. Cooling rate and target temperature durations (dwell) were set to −5 ◦ C/s and 45 s, respectively. The dwell time was extended in this experiment to allow a second set of ratings 30 s after initial cooling. These data were collected but will not be reported; nearly every subject rated nociceptive sensations as ‘barely detectable’ or below within the 30-s time period, so no meaningful comparisons could be made between conditions (adaptation of ICN relative to cold sensation is being systematically investigated in a separate study). Like Experiment 1, stimulation was alternated between pairs of rows of the thermode from trial to trial, and placement of the thermode was alternated between the two sites on the forearm from one condition to the next. Thus, each condition was presented twice (with a 1-min ITI) using separate pairs of rows, after which the thermode was moved to the second site so that the skin could adapt to 33◦ C for 3 min before testing resumed. The 3-min adaptation period and 1-min ITI prevented retesting the same skin site for at least 4 min. In practice, the time required to move the thermode between conditions lengthened this interval to ≥5 min. Subjects served in three sessions and were randomly assigned to one of four groups in which the conditions and temperatures were presented in counterbalanced orders across sessions. All four conditions were tested using a single temperature in each session. For example, subjects in group 1 received the four conditions in a fixed order with 18 ◦ C in the first session, 24 ◦ C in the second session, and 28 ◦ C in the third session. Other groups received the conditions in different fixed orders and with the temperatures either ascending or descending across sessions.
tendency for subjects who reported nociceptive sensations to give slightly higher cold ratings in every condition. Fig. 3 shows the log-mean ratings of nociceptive sensations (a) and of temperature (b) for the 14 subjects who reported more than barely detectable nociceptive sensations during static stimulation at 28 ◦ C. The results in Fig. 3a show that reduction in nociceptive sensations caused by dynamic contact was less at colder temperatures. A multivariate repeated measures ANOVA confirmed that there was a significant interaction among the effects of condition, sensation quality and temperature [F(6,78) = 2.79, p < 0.05]. Tukey HSD tests (p < 0.05) further showed that nociceptive sensations were significantly lower in all three dynamic conditions at 28 ◦ C, in only the dynamic + static condition at 24 ◦ C, and in none of the dynamic conditions at 18 ◦ C. As in previous experiments, temperature sensations were not reduced significantly by dynamic contact. The same ANOVA revealed an interaction between condition and sensation quality [F(3,39) = 6.85, p < 0.001], which confirmed
3.2. Results As in Experiment 1, only subjects who reported more than barely detectable nociceptive sensations were included in the analyses of nociceptive sensations. Also as in Experiment 1, an ANOVA conducted on the ratings of thermal sensations for the full subject sample revealed no significant main effect of group [F(1,29) = 3.76, p = 0.062)]. However, in this experiment there was a consistent though nonsignificant
Fig. 3. The log-mean perceived intensity ratings of nociceptive (a) and thermal (b) sensations for three target temperatures under three conditions of contact cooling in Experiment 2. Suppression of ICN (a) occurred in all dynamic conditions at 28 ◦ C, in only the dynamic + static at 24 ◦ C, and in no conditions at 18 ◦ C. Dynamic contact did not significantly suppress cold sensations (b) in any conditions at any of the target cooling temperatures. Vertical bars represent the S.E.M.
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
95
Experiment 2 provided additional support for the tactile hypothesis by showing that dynamic contact at the moment of peak cooling could still suppress ICN. In addition, the lesser reduction in ICN at temperatures approaching the threshold for cold pain indicates that the effect of mechanical contact does not extend to noxious temperatures, and thus is limited to ICN. 4.1. Possible mechanisms of tactile-mediated suppression
Fig. 4. Percentage of trials on which nociceptive and cold sensations were reported at three temperatures for each testing condition in Experiment 2. Nociceptive sensations were more frequently reported as stinging (a) at 28 ◦ C, whereas reports of burning (b) increased with colder temperatures. Details of the graph are the same as Fig. 3.
that the effect of condition was greater for nociceptive sensations than for thermal sensations. Tukey tests confirmed that temperature sensations were not significantly reduced relative to the static condition in any of the dynamic conditions, even at 28 ◦ C. Fig. 4 displays the percentages of trials on which stinging/pricking, burning or cold were reported at each temperature and for each condition. Once again, the more frequently experienced nociceptive quality was stinging/pricking which, consistent with the ratings of perceived intensity in Fig. 3a, was most often reported in the static condition. At 28 ◦ C, burning was reported only about half as often as stinging/pricking, but became more frequent at the two lower temperatures, where it was reported at about the same frequency across all conditions. Consistent with Fig. 3b, cold was also more frequently reported in the static condition than in any of the dynamic conditions. As in Experiment 1, cold ratings began to give way to “cool” ratings (not shown in Fig. 4) under conditions of dynamic tactile stimulation.
4. Discussion The results of both experiments support the hypothesis that a reduction in ICN during dynamic stimulation is caused by tactile stimulation produced as the thermode first contacts the skin. The data from Experiment 1 ruled out rate of cooling as a significant factor, showing that faster cooling rates produced stronger rather than weaker nociceptive sensations.
The present results indicate that the temporal attributes of tactile stimulation are critical for inducing suppression. First, the failure of steady mechanical pressure to produce suppression means that stimulation of slowly-adapting (SA) mechanoreceptor afferent fibers alone is insufficient. Stimulation of rapidly-adapting (RA) mechanoreceptors, which respond best to dynamic mechanical stimulation and vibration, appears essential. Previous studies of vibrotactile inhibition of both heat and electrocutaneous pain have yielded conflicting evidence as to which type of RA fiber (i.e., those associated with Meissner or Pacinian corpuscles) is more important for inhibition [10,12], but high-frequency vibration (≥100 Hz) has consistently been found to be an effective analgesic stimulus. Second, the finding that the fastest cooling rate in the dynamic + static condition of Experiment 1 produced the strongest suppression is also indicative of the importance of the dynamic phase of tactile stimulation. More rapid cooling in that condition, in which cooling began at the moment of skin contact, shortened the lag between the onset of tactile stimulation and the time of maximum cooling. Cooling to 28 ◦ C at −5.0 C◦ /s produced a lag of only 1 s, compared to 2.5 and 5 s at rates of −2.5 and −1.0◦ /s, respectively. If the important factor is the time between skin contact and initiation of nociceptive stimulation, which previous work indicates occurs at a temperature of 30–31 ◦ C [1], the window over which inhibition is optimal may be on the order of a few hundred milliseconds. To study this possibility will require temporal control of the onset of cooling relative to mechanical contact with a precision that cannot be achieved with the current manual system of thermode placement. The ability of touch and vibration to inhibit pain has most often been explained in terms of the Gate Control Theory [14–17], which postulates that an inhibitory circuit in the spinal cord causes afferent activity conducted in small, mostly unmyelinated fibers that have high thresholds (e.g., CPNs) to be blocked by activity conducted in large, myelinated afferent fibers that have low thresholds (e.g., tactile and cold fibers). Although still accepted in principle, the details of the gating mechanism remain in question, and descending inhibitory processes are also known to play a role in tactile analgesia [17–19]. For example, Yarnitsky et al. [20] invoked the Gate Theory as a possible explanation for the ability of vibration to attenuate cold pain evoked by a 2 ◦ C stimulus, even though the effect was significant only when vibration was delivered either contralaterally or to a different dermatome. The ability
96
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
of the theory as it stands now to account for the present data is limited both by the absence of a mechanism that would explain why dynamic tactile stimulation produces stronger inhibition than static pressure, and by the failure of suppression to extend to cold temperatures (<24 ◦ C) that stimulate CPNs [2,3,9]. ICN and its suppression appear to involve an as yet unidentified type of cold-sensitive nociceptor which, based on sensation quality, may belong to the class of A␦-fibers rather than C-fibers. Past research has associated stinging and pricking with stimulation of A␦ nociceptors and burning with stimulation of C-fiber nociceptors [21,22]. Figs. 2 and 4 show that ICN has a predominantly stinging and/or pricking quality, and that this quality is suppressed by dynamic contact. The incidence of burning sensations, which in Experiment 1 was only 20–25% of trials, was not reduced by dynamic contact even though the incidence of stinging/pricking fell by more than two-thirds. In Experiment 2, the incidence of burning sensations at 28 ◦ C was again very low (10–25% of trials), but increased roughly two-fold in all conditions at 24 and 18 ◦ C (Fig. 4b), where suppression was either weak or nonexistent. This pattern of results implies that burning sensations caused by stimulation of CPNs may be more resistant to tactile suppression than stinging or pricking caused by stimulation of A␦-fibers that have a lower threshold to cooling. Complicating this interpretation are the results of an earlier study from this laboratory in which mechanical contact appeared able to suppress nociceptive sensations produced by application of the sensory irritant methyl salicylate to the forearm [23]. The sensations produced by methyl salicylate, which have a burning quality [24], decreased when the forearm touched a surface that contained warmed or cooled thermodes that bracketed the site of chemical stimulation. The reduction was strongest when the thermodes were cooled to 25 ◦ C, but was significant even at skin temperature (∼33 ◦ C). Because methyl salicylate was recently shown to stimulate the cold-sensitive ion channel TRPA1 [25,26], which is expressed by CPNs [27], it is likely that dynamic tactile stimulation similar to that produced in the present study is capable of inhibiting stimulation from CPNs. An alternative explanation for the resistance to suppression at colder temperatures might be that tactile stimulation from mere contact with the thermode is too weak to inhibit stronger nociceptive stimulation caused by more extreme cooling. However, the increase in perceived intensity of stinging and pricking at the two lowest temperatures (Fig. 3a) was relatively modest and did not reach statistical significance (Tukey tests, all p’s > 0.05). In addition, the evidence from Experiment 1 (Fig. 1) that suppression was greatest at the cooling rate that produced the strongest nociceptive sensation also does not support the idea that suppression fails when nociceptive sensations intensify. Perhaps neither intensification of nociceptive stimulation nor recruitment of CPNs can alone account for resistance to suppression. Instead, the overall pattern of stimulation may be the critical factor—that is, stimulation of CPNs in the context of ongoing stimulation
of the sensory fibers that mediate ICN. This pattern of stimulation uniquely defines noxious or near-noxious cold temperatures in a way that stimulation of CPNs alone (which also respond to heat) does not. This hypothesis raises the question of whether dynamic contact can also suppress burning sensations produced by heat. The ability of vibration to raise the heat pain threshold [12,28] demonstrates that at least one form of dynamic tactile stimulation is capable of inhibiting heat-induced nociception. 4.2. Functional implications of tactile suppression Cutaneous temperature sensitivity serves at least three basic functions: (1) behavioral thermoregulation; (2) avoidance of dangerously cold (freezing) surfaces; and (3) haptic assessment of the temperature of surfaces and materials. These three functions place different demands on the sensory system. Whereas (1) and (3) require relatively fine discrimination of temperature, (2) does not. Protection from freezing surfaces requires only that rapid cooling to skin temperatures 20 ◦ C or more below normal skin temperature be perceived as unpleasant or painful. This function seems well served by the cold sensitivity of CPNs and A␦ cold-sensitive nociceptors that express TRPA1, which has a cold threshold near 17 ◦ C [2,26,29,30]. In contrast, behavioral thermoregulation, arguably the most vital of the three functions, requires signaling of skin temperatures that are only slightly below normal. Because even small changes in skin temperature can reflect a significant amount of heat loss in a cold environment, it is important that such reductions be perceived as uncomfortable [31]. For example, whole-body exposure to ambient air at 15 ◦ C for about 30 min lowers average skin temperature to only about 30 ◦ C [32], yet total heat loss during this brief time can be significant. It has, therefore, been proposed [1] that in the absence of contact with a surface, ICN may contribute a nociceptive component to cold perception that increases discomfort and serves to motivate behavioral strategies to reduce heat loss. On the other hand, experiencing discomfort from mild cooling during exploration or manipulation of objects would be less advantageous. Touching surfaces that have temperatures between 20 and 30 ◦ C ought not signal danger, even though surfaces composed of a thermally conductive material (e.g., metal or stone) will cool the skin much more effectively than convective and radiative cooling in air of the same temperature. Evidence from nerve-block experiments suggests that stimulation of low-threshold cold fibers normally exerts an inhibitory pressure on the nociceptive component of cold [9,33–36]. Tactile stimulation during dynamic mechanical contact may sum with the low-threshold cold signal to reinforce inhibition of the “alarm” component of cooling that would otherwise intensify during rapid conductive cooling. Complicating this functional argument is the fact that not all subjects report ICN. In Experiment 1, 58.1% of subjects reported more than barely detectable nociceptive sensations during static stimulation at a cooling rate of −1 ◦ C/s, whereas
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
in Experiment 2, 45.1% did so in the same condition. Less than universal perception of ICN implies that it is not an essential part of behavioral thermoregulation. It may nevertheless provide an adaptive advantage to those individuals who perceive it. It is also possible that many more people would experience ICN under more ecologically valid conditions of partial- or whole-body cooling. Measurements of spatial summation in the first study of ICN showed only modest increases in intensity for contact stimulation on a small region of the forearm [1], but whether ICN becomes more salient during convective and radiative cooling on larger areas of skin remains to be learned.
5. Summary and conclusions The present study confirms that dynamic contact can suppress nociceptive sensations produced by innocuous cooling [1] and supports the hypothesis that suppression is caused by dynamic tactile stimulation produced when a thermode first contacts the skin. The data further indicate that for significant suppression to develop, tactile stimulation must occur at or near the moment cooling becomes sufficient to stimulate nociceptive sensations (i.e., when skin temperature falls to 28–30 ◦ C). An important additional finding was that suppression is limited to cold temperatures above the threshold for cold pain. Taken together these results imply that innocuous cooling can be perceived differently depending upon the physical conditions that produce it. Perceptions of cooling via convection, radiation or evaporation are more likely to include nociceptive qualities of stinging and/or pricking than are perceptions of cooling from conduction, as during haptic exploration of objects and surfaces. Thus, theories of nonpainful cold perception and behavioral thermoregulation cannot be based on the response characteristics of cold fibers and the perception of cold sensation alone. For temperatures between about 20 and 30 ◦ C, the potential for nociceptive and tactile stimulation to affect the quality and intensity of perceptions of skin cooling must be considered.
Acknowledgments The authors thank Jennifer Pope for collecting a portion of the data reported here. This study was supported in part by a grant from the National Institutes of Health (RO1 NS038463).
References [1] Green BG, Pope JV. Innocuous cooling can produce nociceptive sensations that are inhibited during dynamic mechanical contact. Exp Brain Res 2003;148:290–9. [2] Campero M, Serra J, Ochoa JL. C-polymodal nociceptors activated by noxious low temperature in human skin. J Physiol 1996;497(Part 2):565–72.
97
[3] Simone DA, Kajander KC. Excitation of rat cutaneous nociceptors by noxious cold. Neurosci Lett 1996;213:53–6. [4] LaMotte RH, Thalhammer JG. Response properties of high-threshold cutaneous cold receptors in the primate. Brain Res 1982;244:279– 87. [5] Georgopoulos AP. Functional properties of primary afferent units probably related to pain mechanisms in primate glabrous skin. J Neurophysiol 1976;39:71–83. [6] Green BG, Shaffer GS, Gilmore MM. Derivation and evaluation of a semantic scale of oral sensation magnitude with apparent ratio properties. Chem Senses 1993;18:683–702. [7] Borg G. A category scale with ratio properties for intermodal and interindividual comparisons. In: Geissler H-G, Petxold P, editors. Psychophysical Judgment and the Process of Perception. Berlin: VEB Deutxcher Verlag der Wissenschaften; 1982. p. 25–34. [8] Bartoshuk LM, Duffy VB, Fast K, Green BG, Prutkin J, Snyder DJ. Labeled scales (e.g., category, Likert, VAS) and invalid across-group comparisons: what we have learned from genetic variation in taste. Food Qual Pref 2003;14:125–38. [9] Wasner G, Schattschneider J, Binder A, Baron R. Topical menthol—a human model for cold pain by activation and sensitization of C nociceptors. Brain 2004;127:1159–61. [10] Pertovaara A. Modification of human pain threshold by specific tactile receptors. Acta Physiol Scand 1979;107:339–41. [11] Wall PD, Cronly-Dillon JR. Pain, itch and vibration. Arch Neurol 1960;2:365. [12] Hollins M, Roy EA, Crane SA. Vibratory antinociception: effects of vibration amplitude and frequency. J Pain 2003;4:381–91. [13] Roy EA, Hollins M, Maixner W. Reduction of TMD pain by high-frequency vibration: a spatial and temporal analysis. Pain 2003;101:267–74. [14] Craig JC, Rollman GB. Somesthesis. Annu Rev Psychol 1999;50:305–31. [15] Doehring KM. Relieving pain through touch. Adv Clin Care 1989;4:32–3. [16] Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971–9. [17] Handwerker HO, Iggo A, Zimmermann M. Segmental and supraspinal actions on dorsal horn neurons responding to noxious and non-noxious skin stimuli. Pain 1975;1:147–65. [18] Sandkuhler J. The organization and function of endogenous antinociceptive systems. Prog Neurobiol 1996;50:49–53. [19] Wall PD. The gate control theory of pain mechanisms. A reexamination and re-statement. Brain 1978;101:1–18. [20] Yarnitsky D, Kunin M, Brik R, Sprecher E. Vibration reduces thermal pain in adjacent dermatomes. Pain 1997;69:75–7. [21] Price DD, Dubner R. Mechanisms of first and second pain in the peripheral and central nervous systems. J Invest Dermatol 1977;69:167–71. [22] Zotterman Y. Studies in the peripheral nervous mechanisms of pain. Acta Med Scand 1933;80:185–242. [23] Green BG. Measurement of sensory irritation on the skin. Contact Dermatitis 2000;11:170–80. [24] Green BG, Flammer LJ. Methyl salicylate as a cutaneous stimulus: a psychophysical analysis. Somatosens Mot Res 1989;6:253– 74. [25] Bandell M, Story GM, Hwang SW, Viswanath V, Eid SR, Petrus MJ, et al. Noxious cold ion channel TRPA1 is activated by pungent compounds and bradykinin. Neuron 2004;41:849– 57. [26] Story GM, Peier AM, Reeve AJ, Eid SR, Mosbacher J, Hricik TR, et al. ANKTM1, a TRP-like channel expressed in nociceptive neurons, is activated by cold temperatures. Cell 2003;112:819–29. [27] Tominaga M, Caterina MJ. Thermosensation and pain. J Neurobiol 2004;61:3–12. [28] Kakigi R, Watanabe S. Pain relief by various kinds of interference stimulation applied to the peripheral skin in humans: pain-related
98
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
brain potentials following CO2 laser stimulation. J Peripher Nerv Syst 1996;1:189–98. [29] Khasabov SG, Cain DM, Thong D, Mantyh PW, Simone DA. Enhanced responses of spinal dorsal horn neurons to heat and cold stimuli following mild freeze injury to the skin. J Neurophysiol 2001;86:986–96. [30] Beise RD, Carstens E, Kohll¨offel LUE. Psychophysical study of stinging pain evoked by brief freezing of superficial skin and ensuing short-lasting changes in sensations of cool and cold pain. Pain 1998;74:275–86. [31] Stevens JC, Marks LE, Gagge AP. The quantitative assessment of thermal discomfort. Environ Res 1969;2:149–65.
[32] Ebaugh FG, Thauer R. Influence of various environmental temperatures on the cold and warmth thresholds. J Appl Physiol 1950;3:173–82. [33] Fruhstorfer H, Harju EL, Lindblom UF. The significance of Adelta and C fibres for the perception of synthetic heat. Eur J Pain 2003;7:63–71. [34] Yarnitsky D, Ochoa JL. Release of cold-induced burning pain by block of cold-specific afferent input. Brain 1990;113:893–902. [35] Wahren LK, Torebjork E, Jorum E. Central suppression of coldinduced C fibre pain by myelinated fibre input. Pain 1989;38:313–9. [36] Fruhstorfer H. Thermal sensibility changes during ischemic nerve block. Pain 1984;20:355–61.
Research report
Evidence that tactile stimulation inhibits nociceptive sensations produced by innocuous contact cooling Barry G. Green a,b,∗ , Kate L. Schoen a b
a The John B. Pierce Laboratory, Yale University School of Medicine, 290 Congress Avenue, New Haven, CT 06519, USA Department of Surgery (Otolaryngology), Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06510, USA
Received 19 December 2004; received in revised form 3 March 2005; accepted 6 March 2005 Available online 13 April 2005
Abstract It was recently shown that stinging, pricking or burning is reliably perceived by some individuals when the skin is cooled to temperatures as mild as 25–30 ◦ C. These seemingly paradoxical sensations, which have been termed innocuous-cold nociception (ICN), were significant only when cooling was produced by a thermode resting statically on the skin (static contact); touching an already cooled thermode to the skin (dynamic contact) produced reports of only coolness and cold. The present study investigated the hypothesis that ICN is inhibited by tactile stimulation produced when a thermode contacts the skin. Experiment 1 pitted the tactile hypothesis against an alternative explanation that inhibition results from higher rates of skin cooling during dynamic contact. ICN was measured at three different cooling rates (−1.0, −2.5, −5.0 ◦ C/s) when the thermode was resting on the skin or was touched to the skin at the moment cooling began. The results supported the tactile hypothesis: faster cooling rates during static contact led to stronger rather than weaker nociceptive sensations, and ICN was suppressed even when dynamic contact was coincident with the onset of cooling, and thus could not affect cooling rate. Experiment 2 confirmed the latter result and showed that suppression was greatest at 28 ◦ C, less at 24 ◦ C, and not significant at 18 ◦ C. We conclude that dynamic tactile stimulation produced by contact with a surface inhibits the nociceptive component of innocuous but not noxious cooling. The implications of this conclusion for the role of cold perception in behavioral thermoregulation versus haptic perception, and for theories of cold perception in general, are discussed. © 2005 Published by Elsevier B.V. Keywords: Cold; Psychophysics; Pain; Touch; Inhibition; Individual differences
1. Introduction Mild cooling has been assumed to produce only sensations of coolness and cold. Contrary to this view, it was recently discovered that when subjects are asked to rate the intensity of nociceptive as well as thermal sensations, they often report stinging or burning at skin temperatures between 25 and 30 ◦ C [1]. The occurrence of these unexpected sensations, which are referred to as ‘innocuous-cold nociception’ (ICN), raises new questions about the sensory processes that underlie perception of nonpainful cold.
∗
Corresponding author. Tel.: +1 203 458 2262; fax: +1 203 624 4950. E-mail address: [email protected] (B.G. Green).
0166-4328/$ – see front matter © 2005 Published by Elsevier B.V. doi:10.1016/j.bbr.2005.03.015
First, ICN implies that temperatures as mild as 30 ◦ C can stimulate the nociceptive pathway. Such stimulation cannot be accounted for by C-polymodal nociceptors (CPNs), which have cooling thresholds from about 18 to 22 ◦ C and are believed to mediate the cold pain threshold [2,3]. Instead, afferent fibers must exist that respond to noxious cold yet have cooling thresholds in the innocuous temperature range. Although cold-sensitive fibers with these general characteristics have been reported in the past [4,5], their responses to temperatures above the cold pain threshold were not considered a significant factor in cold perception. Second, the occurrence of ICN in only about half of all individuals tested [1] means people differ in their perception of innocuous cold. The possibility that these individual differences might be attributable to cognitive factors or to response bias was ruled out by show-
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
ing that ICN was perceived only under limited conditions of stimulation when cooling occurred as a thermode was resting statically against the skin. If instead the thermode was cooled and then touched to the skin, ICN was virtually absent. The absence of ICN when a thermode is touched to the skin, the large individual differences in its occurrence, and the common practice in psychophysical studies of asking subjects to attend only to cold sensations, probably explains why ICN had not been reported in prior studies of cold perception. The purpose of the present study was to investigate further the stimulus conditions necessary for both induction and suppression of ICN. Particular emphasis was placed on evaluating the hypothesis that suppression is caused by tactile stimulation produced when a thermode first contacts the skin. If this hypothesis were supported, suppression of ICN would provide the first evidence that cold perception does not depend solely on the parameters of the thermal stimulus. 2. Experiment 1 In their discussion of ICN, Green and Pope [1] noted that in addition to producing dynamic tactile stimulation, touching a cooled thermode to the skin undoubtedly lowers skin temperature more rapidly than does cooling via a Peltier thermode already in contact with the skin. In the latter case, cooling rate is slowed by the thermal inertias of both the thermode and the skin. Because the stimulus parameters that influence ICN were unknown, it was possible that higher rates of cooling during dynamic contact might work against perception of ICN. Green and Pope provided an initial test of this possibility by collecting data on the frequency of burning, stinging or pricking sensations perceived during different rates of cooling (between −0.5 and −5.0 ◦ C/s). Although the incidence of ICN did not change with cooling rate, it remained possible that its intensity did. In the present experiment, we investigated this possibility by measuring the perceived intensity of ICN at three different cooling rates. A condition was also included in which cooling was started at the moment of skin contact, thereby producing dynamic tactile stimulation that was correlated with the onset of cooling but did not affect cooling rate. Significant suppression of ICN in the latter condition would support the tactile inhibition hypothesis. 2.1. Methods 2.1.1. Subjects A total of 44 subjects (28 females and 16 males, average age 23.3 years) served in Experiment 1. Subjects were recruited on and around the Yale University campus and were screened by questionnaire to exclude individuals with a history of neurological injury or pathology as well as those with cutaneous lesions or disorders affecting the site of stimulation. Each subject gave informed consent and was paid for his/her participation.
91
2.1.2. Equipment Temperature stimuli were produced using a 4 × 4 array of 0.64 cm2 Peltier thermoelectric modules (hereafter referred to collectively as ‘the thermode’) designed and built in The John B. Pierce Laboratory machine and electrics shop. The 16 independently controlled modules are individually bonded to a water-circulated heat sink. Epoxied to the top of each ceramic module is a machined copper plate (8 mm × 8 mm × 0.96 mm) with a 40-ga copper constantin thermocouple recessed in a 0.5-mm deep groove in the center of the plate. The thermocouple provides constant temperature feedback at the skin–thermode interface. The parameters of thermal stimulation for each Peltier module are controlled via a LabView program that enables the experimenter to specify the base (adapting) temperature, rate of temperature change, and target temperature (all with resolutions of ±0.1 ◦ C), as well as dwell time at the target temperature (±0.1 s). The surface temperature of each thermode was periodically checked using a thermocouple thermometer (Physitemp, BAT-12) to ensure an accuracy of ±0.2 ◦ C. The thermode is mounted on a three-way microscope stage which, in turn, is attached via a lockable ball joint to a floor-mounted positioning arm. The thermode and its positioning system stand beside a modified dental chair in which the subject sits. Stimulation occurred on the subject’s right forearm as it rested volar side up on a foam-padded armrest. Placement of the thermode on the subject’s forearm was achieved manually. Its location on the skin was adjusted before testing began by changing the angle of the positioning arm in three dimensions until the stimulating face of the thermode was aligned parallel to the skin’s surface just above the desired area of stimulation. For static conditions of stimulation the thermode was pressed lightly down onto the skin with a force sufficient to make full contact (the bearing force was not measured). It was then locked in place by flipping a switch that activated a pneumatic brake on the positioning arm. Locking the thermode in place both stabilized its position and prevented its full weight from bearing on the subject’s arm. For conditions that required dynamic contact of the thermode against the arm, the thermode was first positioned approximately 1/2 in. above the arm and locked in place. At the desired time the experimenter released the brake and brought the thermode directly down to the skin with sufficient pressure to make full contact. All testing was conducted in a laminar-flow environmental chamber with air temperature and relative humidity controlled at 26 ◦ C and approximately 30%, respectively.
2.1.3. Psychophysical practice procedure Each of the experiments included a separate practice session for new subjects who had not participated in any prior thermal perception experiments in our laboratory. The practice session was completed before the first test session and consisted of two brief (10–15 min) exercises designed to train individuals to use the Labeled Magnitude Scale (LMS [6]) to rate the intensity of thermal sensations. The LMS is a type of “category-ratio” scale [7] in which labeled intensity descriptors (e.g., “weak”, “strong”) are spaced according to their empirically-derived semantic magnitudes. The scale is bounded at the bottom by “no sensation” and at the top by “strongest imaginable sensation of any kind” [8]. A computer mouse is used to move an arrow along the scale to the point that corresponds with the perceived strength of a given sensation; double-clicking the mouse saves the response. To familiarize subjects with the meaning and use of the LMS, they were first asked to imagine 16 commonly experienced thermal
92
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
sensations (e.g., washing hands in cold tap water; walking barefoot on hot pavement) and to rate the estimated intensity of the sensations on the scale. In a second exercise subjects rated the intensity of 11 thermal stimuli produced by the thermode as it rested on the right forearm. Starting from a base temperature of 33 ◦ C, the experimenter presented an alternating series of warming (38, 36, 40, 42 ◦ C) and cooling (22, 30, 28, 20, 24, 26, 18 ◦ C) stimuli on two rows of the thermode (i.e., a total of eight thermodes). The rate of temperature change for the practice stimuli was ±3.0 ◦ C/s, with a 5.0 s dwell at the target temperature and a 30-s inter-trial interval (ITI). Subjects performed three tasks during every trial: a rating of the intensity of thermal sensation (cool-cold, warm-hot), a rating of the intensity of nociceptive sensations (burning, stinging or pricking), and a checklist task in which they clicked on one or more descriptors (nothing, cool, cold, warm, hot, burning, stinging/pricking, aching, and painful) to indicate the sensation qualities they had perceived. Subjects were told to choose as many or as few descriptors as necessary to describe each sensation fully, and to respond with “nothing” on trials when no sensation was felt.
2.1.4. Experimental procedure The effect of cooling rate on ICN was measured under two conditions, static and dynamic + static. In the static condition the thermode was cooled to the target temperature as it rested on the skin. In the dynamic + static condition the experimenter lifted the thermode from the skin before triggering cooling, then returned it to the skin at the moment cooling began (approximately 1 s later). Skin contact occurred as close to the moment cooling began as was physically possible under manual control. The results from these two conditions were compared to those from a third condition, dynamic, in which the thermode was set to the target temperature before it was touched to the skin. Thus, in the dynamic condition skin cooling occurred at a fast but unknown rate that was determined by the rate of heat transfer between the skin and the already cooled thermode. Three different cooling rates were tested in the static and dynamic–static conditions: −1.0, −2.5 or −5.0 ◦ C/s. The target temperature in all three conditions was 28 ◦ C, with a dwell time of 5 s. The orders of conditions and cooling rates were counterbalanced across subjects and over two experimental sessions. Half of the subjects received the static condition first and had ascending rates of cooling in the first session, then received the static–dynamic condition first with descending rates of cooling in the second session; the other half received the conditions and cooling rates in reverse order. The dynamic condition was always presented between the other two conditions. Within each session two different strategies were used to avoid sensory adaptation from repeated testing at the same site: (1) on each trial testing was conducted on only two of the four rows of the thermode (a total of eight, 0.64 cm2 thermodes); and (2) placement of the thermode was alternated between two sites on the volar surface of the right forearm after each pair of trials. At the beginning of a session the thermode was placed on Site 1 for a 3-min adaptation to the 33 ◦ C base temperature. The first combination of condition and cooling rate was then presented using rows 1 and 2 of the thermode. After the thermodes returned to base temperature there was a 1-min ITI before the same combination of condition and rate was presented to rows 3 and 4. The experimenter then moved the thermode to Site 2 and repeated the procedure, beginning with the 3-min adaptation at 33 ◦ C. This sequence was followed until each condition and cooling rate had been presented twice.
On static trials, in which the thermode remained on the subject’s forearm throughout cooling, the experimenter cued the subject that a stimulus was about to begin by saying, “ready, attend now”, and indicated when to rate sensation intensity and quality by saying “ready, rate”. On dynamic and dynamic + static trials, in which the thermode was lifted and then placed back on the arm, the experimenter told the subject to attend to the sensation as soon as the thermode was replaced and to make intensity and quality ratings after the experimenter said “rate”. For all conditions, the subjects were instructed to rate the maximum sensations felt up to the time of the experimenter’s signal, which was given just prior to termination of the 28 ◦ C cooling stimulus (i.e., at the end of the 5 s dwell time).
2.2. Results To assess differences in ICN across conditions, only those subjects (n = 25) who rated burning/stinging/pricking (hereafter referred to as nociceptive sensations) as more than ‘barely detectable’ in the static condition during the slowest rate of cooling (−1.0 ◦ C/s) were included in the statistical analyses. An ANOVA conducted on the thermal ratings (coolness, cold) for all subjects prior to this segregation showed that the subset of individuals who reported nociceptive sensations did not rate the stimuli to be any colder than did individuals who reported no nociceptive sensations [main effect group; F(1,42) = 0.42, p = 0.52]. The main results of the experiment are shown in Fig. 1. On the left side of the figure are the log-mean ratings for cold and nociceptive sensations in the static condition, which both increased with cooling rate. A multivariate repeated measures
Fig. 1. The log-mean ratings of cold and nociceptive sensations as a function of cooling rate for the three conditions of Experiment 1. In the static condition, both types of sensations increased in intensity with cooling rate, ruling out cooling rate as a cause of contact-suppression of ICN. In the dynamic condition, only nociceptive sensations were significantly suppressed relative to all cooling rates in the static condition. In the dynamic + static condition, suppression of nociceptive sensations increased with cooling rate, indicating that cooling rate has opposite effects on cold and nociceptive sensations. Vertical bars represent the standard errors of the means (S.E.M.).
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
ANOVA calculated for these data only, with rate of cooling and sensation quality as factors, confirmed that there was a main effect of cooling rate [F(2,48) = 7.4, p < 0.005]. There was no interaction between cooling rate and sensation quality. Because both cold and nociceptive sensations increased in intensity with rate of cooling, the results rule out cooling rate as a cause of contact-suppression of ICN. An overall ANOVA conducted on the data for all three stimulus conditions from the same group of subjects revealed significant main effects of condition [F(6,144) = 13.2, p < 0.0001] and sensation quality [F(1,24) = 47.5, p < 0.0001], and an interaction between these factors [F(6,144) = 10.7, p < 0.0001]. The latter interaction was driven primarily by a much stronger suppression of nociceptive sensations compared to cold in the dynamic condition (center of Fig. 1), and by diverging effects of cooling rate for the different qualities in the dynamic + static condition. The stronger suppression of nociceptive sensations was confirmed by Tukey HSD tests (p < 0.05), which indicated that suppression was significant in the dynamic condition relative to all cooling rates in the static condition, whereas cold ratings were significantly suppressed relative only to the highest cooling rate (−5.0 ◦ C/s) in the static condition. In terms of sensation magnitude, nociceptive sensations in the dynamic condition were reduced by 69% from the level reported for −1.0 ◦ C/s cooling, whereas cold sensation was reduced by only 13%. Relative to −5 ◦ C/s cooling in the static condition, nociceptive sensations were reduced by 77%, and cold by only 37%. In contrast to the static condition, nociceptive sensations tended toward lower rather than higher intensities at the fastest cooling rate in the dynamic + static condition. An ANOVA that included only the dynamic + static data showed an interaction between cooling rate and sensation quality [F(2,48) = 9.2, p < 0.001], confirming the visual impression that cooling rate had opposite effects on cold and nociceptive sensations. The latter sensations were significantly lower in the dynamic + static condition than in the static condition for the two fastest cooling rates, but not for the slowest rate of −1.0 ◦ C/s (Tukey HSD, p < 0.05). The implication of this result is that the suppressive effect of dynamic contact can be significant even when skin contact occurs before the thermode has been cooled, but only when the delay between contact and peak cooling is short. For cooling rates of −2.5 and −5.0◦ C/s, the delay was approximately 2.5 and 1 s, respectively. Fig. 2 shows the frequencies at which the primary sensation qualities were reported in the different conditions and for different rates of cooling. The two most frequently reported qualities were cold and stinging/pricking. Both sensations show a slight trend toward higher frequencies of occurrence at higher rates of cooling in the static condition. More interesting, however, is the large decrease in stinging/pricking in the dynamic condition. ICN was predominantly a stinging and/or pricking sensation, the occurrence of which fell from 66% of trials at the fastest rate of cooling in the static
93
Fig. 2. Frequencies at which thermal and nociceptive sensation qualities were reported for three conditions at three rates of cooling in Experiment 1. The data indicate that ICN has predominantly stinging and/or pricking qualities that are suppressed by dynamic contact.
condition to only 20% of trials in the dynamic condition. In contrast, reports of burning sensations never exceeded 25% of trials in any condition, and remained essentially unchanged across cooling rate and condition.
3. Experiment 2 Having ruled out the rate-of-cooling hypothesis and having found support for the tactile inhibition hypothesis, we next investigated whether suppression of nociceptive sensations by dynamic contact was unique to ICN or extended to lower, potentially noxious temperatures. As mentioned above, the peripheral source of ICN remains speculative, but near and below the cold pain threshold nociceptive sensations are thought to be mediated primarily by CPNs [2,3,9]. If suppression occurs in the temperature range served by CPNs, it would suggest that it is part of a general inhibitory mechanism by which tactile stimulation inhibits pain [10–13]. In addition, we investigated whether ICN is still suppressed when dynamic tactile stimulation occurs after the target temperature has been reached. Significant suppression after the skin has been cooled would provide further support for the tactile inhibition hypothesis. 3.1. Methods 3.1.1. Subjects Thirty-one individuals (16 females and 15 males, average age 25 years) served in Experiment 2. Subjects were recruited and screened in the same manner as Experiment 1. Each subject gave informed consent and was paid for his/her participation. 3.1.2. Equipment and practice procedure The thermal stimulation system and psychophysical practice procedures were the same as in Experiment 1. Only subjects who had
94
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
not participated in Experiment 1 were required to serve in the practice session. 3.1.3. Experimental procedure There were four conditions in the experiment. Three of the conditions – static, dynamic, and dynamic + static – were the same as in Experiment 1. In the fourth condition, which we called static + dynamic, cooling began as in the static condition with the thermode resting against the skin, but as soon as the target temperature was reached, the thermode was lifted from the skin and then immediately replaced (within approximately 1 s). Subjects were instructed to rate the sensation intensity and quality just after the thermode was returned to the skin, and it was emphasized that their ratings should reflect the sensations experienced at the exact moment of the experimenter’s signal. This was a critical feature of the psychophysical procedure, because the static + dynamic condition made it possible for maximum sensations to occur when the target temperature was reached during the static phase of stimulation. Yet, the primary object of the study was to measure the sensations felt immediately after the thermode re-contacted the skin (i.e., during the dynamic phase of stimulation). Three stimulus temperatures were used: 28, 24 and 18 ◦ C. Cooling rate and target temperature durations (dwell) were set to −5 ◦ C/s and 45 s, respectively. The dwell time was extended in this experiment to allow a second set of ratings 30 s after initial cooling. These data were collected but will not be reported; nearly every subject rated nociceptive sensations as ‘barely detectable’ or below within the 30-s time period, so no meaningful comparisons could be made between conditions (adaptation of ICN relative to cold sensation is being systematically investigated in a separate study). Like Experiment 1, stimulation was alternated between pairs of rows of the thermode from trial to trial, and placement of the thermode was alternated between the two sites on the forearm from one condition to the next. Thus, each condition was presented twice (with a 1-min ITI) using separate pairs of rows, after which the thermode was moved to the second site so that the skin could adapt to 33◦ C for 3 min before testing resumed. The 3-min adaptation period and 1-min ITI prevented retesting the same skin site for at least 4 min. In practice, the time required to move the thermode between conditions lengthened this interval to ≥5 min. Subjects served in three sessions and were randomly assigned to one of four groups in which the conditions and temperatures were presented in counterbalanced orders across sessions. All four conditions were tested using a single temperature in each session. For example, subjects in group 1 received the four conditions in a fixed order with 18 ◦ C in the first session, 24 ◦ C in the second session, and 28 ◦ C in the third session. Other groups received the conditions in different fixed orders and with the temperatures either ascending or descending across sessions.
tendency for subjects who reported nociceptive sensations to give slightly higher cold ratings in every condition. Fig. 3 shows the log-mean ratings of nociceptive sensations (a) and of temperature (b) for the 14 subjects who reported more than barely detectable nociceptive sensations during static stimulation at 28 ◦ C. The results in Fig. 3a show that reduction in nociceptive sensations caused by dynamic contact was less at colder temperatures. A multivariate repeated measures ANOVA confirmed that there was a significant interaction among the effects of condition, sensation quality and temperature [F(6,78) = 2.79, p < 0.05]. Tukey HSD tests (p < 0.05) further showed that nociceptive sensations were significantly lower in all three dynamic conditions at 28 ◦ C, in only the dynamic + static condition at 24 ◦ C, and in none of the dynamic conditions at 18 ◦ C. As in previous experiments, temperature sensations were not reduced significantly by dynamic contact. The same ANOVA revealed an interaction between condition and sensation quality [F(3,39) = 6.85, p < 0.001], which confirmed
3.2. Results As in Experiment 1, only subjects who reported more than barely detectable nociceptive sensations were included in the analyses of nociceptive sensations. Also as in Experiment 1, an ANOVA conducted on the ratings of thermal sensations for the full subject sample revealed no significant main effect of group [F(1,29) = 3.76, p = 0.062)]. However, in this experiment there was a consistent though nonsignificant
Fig. 3. The log-mean perceived intensity ratings of nociceptive (a) and thermal (b) sensations for three target temperatures under three conditions of contact cooling in Experiment 2. Suppression of ICN (a) occurred in all dynamic conditions at 28 ◦ C, in only the dynamic + static at 24 ◦ C, and in no conditions at 18 ◦ C. Dynamic contact did not significantly suppress cold sensations (b) in any conditions at any of the target cooling temperatures. Vertical bars represent the S.E.M.
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
95
Experiment 2 provided additional support for the tactile hypothesis by showing that dynamic contact at the moment of peak cooling could still suppress ICN. In addition, the lesser reduction in ICN at temperatures approaching the threshold for cold pain indicates that the effect of mechanical contact does not extend to noxious temperatures, and thus is limited to ICN. 4.1. Possible mechanisms of tactile-mediated suppression
Fig. 4. Percentage of trials on which nociceptive and cold sensations were reported at three temperatures for each testing condition in Experiment 2. Nociceptive sensations were more frequently reported as stinging (a) at 28 ◦ C, whereas reports of burning (b) increased with colder temperatures. Details of the graph are the same as Fig. 3.
that the effect of condition was greater for nociceptive sensations than for thermal sensations. Tukey tests confirmed that temperature sensations were not significantly reduced relative to the static condition in any of the dynamic conditions, even at 28 ◦ C. Fig. 4 displays the percentages of trials on which stinging/pricking, burning or cold were reported at each temperature and for each condition. Once again, the more frequently experienced nociceptive quality was stinging/pricking which, consistent with the ratings of perceived intensity in Fig. 3a, was most often reported in the static condition. At 28 ◦ C, burning was reported only about half as often as stinging/pricking, but became more frequent at the two lower temperatures, where it was reported at about the same frequency across all conditions. Consistent with Fig. 3b, cold was also more frequently reported in the static condition than in any of the dynamic conditions. As in Experiment 1, cold ratings began to give way to “cool” ratings (not shown in Fig. 4) under conditions of dynamic tactile stimulation.
4. Discussion The results of both experiments support the hypothesis that a reduction in ICN during dynamic stimulation is caused by tactile stimulation produced as the thermode first contacts the skin. The data from Experiment 1 ruled out rate of cooling as a significant factor, showing that faster cooling rates produced stronger rather than weaker nociceptive sensations.
The present results indicate that the temporal attributes of tactile stimulation are critical for inducing suppression. First, the failure of steady mechanical pressure to produce suppression means that stimulation of slowly-adapting (SA) mechanoreceptor afferent fibers alone is insufficient. Stimulation of rapidly-adapting (RA) mechanoreceptors, which respond best to dynamic mechanical stimulation and vibration, appears essential. Previous studies of vibrotactile inhibition of both heat and electrocutaneous pain have yielded conflicting evidence as to which type of RA fiber (i.e., those associated with Meissner or Pacinian corpuscles) is more important for inhibition [10,12], but high-frequency vibration (≥100 Hz) has consistently been found to be an effective analgesic stimulus. Second, the finding that the fastest cooling rate in the dynamic + static condition of Experiment 1 produced the strongest suppression is also indicative of the importance of the dynamic phase of tactile stimulation. More rapid cooling in that condition, in which cooling began at the moment of skin contact, shortened the lag between the onset of tactile stimulation and the time of maximum cooling. Cooling to 28 ◦ C at −5.0 C◦ /s produced a lag of only 1 s, compared to 2.5 and 5 s at rates of −2.5 and −1.0◦ /s, respectively. If the important factor is the time between skin contact and initiation of nociceptive stimulation, which previous work indicates occurs at a temperature of 30–31 ◦ C [1], the window over which inhibition is optimal may be on the order of a few hundred milliseconds. To study this possibility will require temporal control of the onset of cooling relative to mechanical contact with a precision that cannot be achieved with the current manual system of thermode placement. The ability of touch and vibration to inhibit pain has most often been explained in terms of the Gate Control Theory [14–17], which postulates that an inhibitory circuit in the spinal cord causes afferent activity conducted in small, mostly unmyelinated fibers that have high thresholds (e.g., CPNs) to be blocked by activity conducted in large, myelinated afferent fibers that have low thresholds (e.g., tactile and cold fibers). Although still accepted in principle, the details of the gating mechanism remain in question, and descending inhibitory processes are also known to play a role in tactile analgesia [17–19]. For example, Yarnitsky et al. [20] invoked the Gate Theory as a possible explanation for the ability of vibration to attenuate cold pain evoked by a 2 ◦ C stimulus, even though the effect was significant only when vibration was delivered either contralaterally or to a different dermatome. The ability
96
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
of the theory as it stands now to account for the present data is limited both by the absence of a mechanism that would explain why dynamic tactile stimulation produces stronger inhibition than static pressure, and by the failure of suppression to extend to cold temperatures (<24 ◦ C) that stimulate CPNs [2,3,9]. ICN and its suppression appear to involve an as yet unidentified type of cold-sensitive nociceptor which, based on sensation quality, may belong to the class of A␦-fibers rather than C-fibers. Past research has associated stinging and pricking with stimulation of A␦ nociceptors and burning with stimulation of C-fiber nociceptors [21,22]. Figs. 2 and 4 show that ICN has a predominantly stinging and/or pricking quality, and that this quality is suppressed by dynamic contact. The incidence of burning sensations, which in Experiment 1 was only 20–25% of trials, was not reduced by dynamic contact even though the incidence of stinging/pricking fell by more than two-thirds. In Experiment 2, the incidence of burning sensations at 28 ◦ C was again very low (10–25% of trials), but increased roughly two-fold in all conditions at 24 and 18 ◦ C (Fig. 4b), where suppression was either weak or nonexistent. This pattern of results implies that burning sensations caused by stimulation of CPNs may be more resistant to tactile suppression than stinging or pricking caused by stimulation of A␦-fibers that have a lower threshold to cooling. Complicating this interpretation are the results of an earlier study from this laboratory in which mechanical contact appeared able to suppress nociceptive sensations produced by application of the sensory irritant methyl salicylate to the forearm [23]. The sensations produced by methyl salicylate, which have a burning quality [24], decreased when the forearm touched a surface that contained warmed or cooled thermodes that bracketed the site of chemical stimulation. The reduction was strongest when the thermodes were cooled to 25 ◦ C, but was significant even at skin temperature (∼33 ◦ C). Because methyl salicylate was recently shown to stimulate the cold-sensitive ion channel TRPA1 [25,26], which is expressed by CPNs [27], it is likely that dynamic tactile stimulation similar to that produced in the present study is capable of inhibiting stimulation from CPNs. An alternative explanation for the resistance to suppression at colder temperatures might be that tactile stimulation from mere contact with the thermode is too weak to inhibit stronger nociceptive stimulation caused by more extreme cooling. However, the increase in perceived intensity of stinging and pricking at the two lowest temperatures (Fig. 3a) was relatively modest and did not reach statistical significance (Tukey tests, all p’s > 0.05). In addition, the evidence from Experiment 1 (Fig. 1) that suppression was greatest at the cooling rate that produced the strongest nociceptive sensation also does not support the idea that suppression fails when nociceptive sensations intensify. Perhaps neither intensification of nociceptive stimulation nor recruitment of CPNs can alone account for resistance to suppression. Instead, the overall pattern of stimulation may be the critical factor—that is, stimulation of CPNs in the context of ongoing stimulation
of the sensory fibers that mediate ICN. This pattern of stimulation uniquely defines noxious or near-noxious cold temperatures in a way that stimulation of CPNs alone (which also respond to heat) does not. This hypothesis raises the question of whether dynamic contact can also suppress burning sensations produced by heat. The ability of vibration to raise the heat pain threshold [12,28] demonstrates that at least one form of dynamic tactile stimulation is capable of inhibiting heat-induced nociception. 4.2. Functional implications of tactile suppression Cutaneous temperature sensitivity serves at least three basic functions: (1) behavioral thermoregulation; (2) avoidance of dangerously cold (freezing) surfaces; and (3) haptic assessment of the temperature of surfaces and materials. These three functions place different demands on the sensory system. Whereas (1) and (3) require relatively fine discrimination of temperature, (2) does not. Protection from freezing surfaces requires only that rapid cooling to skin temperatures 20 ◦ C or more below normal skin temperature be perceived as unpleasant or painful. This function seems well served by the cold sensitivity of CPNs and A␦ cold-sensitive nociceptors that express TRPA1, which has a cold threshold near 17 ◦ C [2,26,29,30]. In contrast, behavioral thermoregulation, arguably the most vital of the three functions, requires signaling of skin temperatures that are only slightly below normal. Because even small changes in skin temperature can reflect a significant amount of heat loss in a cold environment, it is important that such reductions be perceived as uncomfortable [31]. For example, whole-body exposure to ambient air at 15 ◦ C for about 30 min lowers average skin temperature to only about 30 ◦ C [32], yet total heat loss during this brief time can be significant. It has, therefore, been proposed [1] that in the absence of contact with a surface, ICN may contribute a nociceptive component to cold perception that increases discomfort and serves to motivate behavioral strategies to reduce heat loss. On the other hand, experiencing discomfort from mild cooling during exploration or manipulation of objects would be less advantageous. Touching surfaces that have temperatures between 20 and 30 ◦ C ought not signal danger, even though surfaces composed of a thermally conductive material (e.g., metal or stone) will cool the skin much more effectively than convective and radiative cooling in air of the same temperature. Evidence from nerve-block experiments suggests that stimulation of low-threshold cold fibers normally exerts an inhibitory pressure on the nociceptive component of cold [9,33–36]. Tactile stimulation during dynamic mechanical contact may sum with the low-threshold cold signal to reinforce inhibition of the “alarm” component of cooling that would otherwise intensify during rapid conductive cooling. Complicating this functional argument is the fact that not all subjects report ICN. In Experiment 1, 58.1% of subjects reported more than barely detectable nociceptive sensations during static stimulation at a cooling rate of −1 ◦ C/s, whereas
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
in Experiment 2, 45.1% did so in the same condition. Less than universal perception of ICN implies that it is not an essential part of behavioral thermoregulation. It may nevertheless provide an adaptive advantage to those individuals who perceive it. It is also possible that many more people would experience ICN under more ecologically valid conditions of partial- or whole-body cooling. Measurements of spatial summation in the first study of ICN showed only modest increases in intensity for contact stimulation on a small region of the forearm [1], but whether ICN becomes more salient during convective and radiative cooling on larger areas of skin remains to be learned.
5. Summary and conclusions The present study confirms that dynamic contact can suppress nociceptive sensations produced by innocuous cooling [1] and supports the hypothesis that suppression is caused by dynamic tactile stimulation produced when a thermode first contacts the skin. The data further indicate that for significant suppression to develop, tactile stimulation must occur at or near the moment cooling becomes sufficient to stimulate nociceptive sensations (i.e., when skin temperature falls to 28–30 ◦ C). An important additional finding was that suppression is limited to cold temperatures above the threshold for cold pain. Taken together these results imply that innocuous cooling can be perceived differently depending upon the physical conditions that produce it. Perceptions of cooling via convection, radiation or evaporation are more likely to include nociceptive qualities of stinging and/or pricking than are perceptions of cooling from conduction, as during haptic exploration of objects and surfaces. Thus, theories of nonpainful cold perception and behavioral thermoregulation cannot be based on the response characteristics of cold fibers and the perception of cold sensation alone. For temperatures between about 20 and 30 ◦ C, the potential for nociceptive and tactile stimulation to affect the quality and intensity of perceptions of skin cooling must be considered.
Acknowledgments The authors thank Jennifer Pope for collecting a portion of the data reported here. This study was supported in part by a grant from the National Institutes of Health (RO1 NS038463).
References [1] Green BG, Pope JV. Innocuous cooling can produce nociceptive sensations that are inhibited during dynamic mechanical contact. Exp Brain Res 2003;148:290–9. [2] Campero M, Serra J, Ochoa JL. C-polymodal nociceptors activated by noxious low temperature in human skin. J Physiol 1996;497(Part 2):565–72.
97
[3] Simone DA, Kajander KC. Excitation of rat cutaneous nociceptors by noxious cold. Neurosci Lett 1996;213:53–6. [4] LaMotte RH, Thalhammer JG. Response properties of high-threshold cutaneous cold receptors in the primate. Brain Res 1982;244:279– 87. [5] Georgopoulos AP. Functional properties of primary afferent units probably related to pain mechanisms in primate glabrous skin. J Neurophysiol 1976;39:71–83. [6] Green BG, Shaffer GS, Gilmore MM. Derivation and evaluation of a semantic scale of oral sensation magnitude with apparent ratio properties. Chem Senses 1993;18:683–702. [7] Borg G. A category scale with ratio properties for intermodal and interindividual comparisons. In: Geissler H-G, Petxold P, editors. Psychophysical Judgment and the Process of Perception. Berlin: VEB Deutxcher Verlag der Wissenschaften; 1982. p. 25–34. [8] Bartoshuk LM, Duffy VB, Fast K, Green BG, Prutkin J, Snyder DJ. Labeled scales (e.g., category, Likert, VAS) and invalid across-group comparisons: what we have learned from genetic variation in taste. Food Qual Pref 2003;14:125–38. [9] Wasner G, Schattschneider J, Binder A, Baron R. Topical menthol—a human model for cold pain by activation and sensitization of C nociceptors. Brain 2004;127:1159–61. [10] Pertovaara A. Modification of human pain threshold by specific tactile receptors. Acta Physiol Scand 1979;107:339–41. [11] Wall PD, Cronly-Dillon JR. Pain, itch and vibration. Arch Neurol 1960;2:365. [12] Hollins M, Roy EA, Crane SA. Vibratory antinociception: effects of vibration amplitude and frequency. J Pain 2003;4:381–91. [13] Roy EA, Hollins M, Maixner W. Reduction of TMD pain by high-frequency vibration: a spatial and temporal analysis. Pain 2003;101:267–74. [14] Craig JC, Rollman GB. Somesthesis. Annu Rev Psychol 1999;50:305–31. [15] Doehring KM. Relieving pain through touch. Adv Clin Care 1989;4:32–3. [16] Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971–9. [17] Handwerker HO, Iggo A, Zimmermann M. Segmental and supraspinal actions on dorsal horn neurons responding to noxious and non-noxious skin stimuli. Pain 1975;1:147–65. [18] Sandkuhler J. The organization and function of endogenous antinociceptive systems. Prog Neurobiol 1996;50:49–53. [19] Wall PD. The gate control theory of pain mechanisms. A reexamination and re-statement. Brain 1978;101:1–18. [20] Yarnitsky D, Kunin M, Brik R, Sprecher E. Vibration reduces thermal pain in adjacent dermatomes. Pain 1997;69:75–7. [21] Price DD, Dubner R. Mechanisms of first and second pain in the peripheral and central nervous systems. J Invest Dermatol 1977;69:167–71. [22] Zotterman Y. Studies in the peripheral nervous mechanisms of pain. Acta Med Scand 1933;80:185–242. [23] Green BG. Measurement of sensory irritation on the skin. Contact Dermatitis 2000;11:170–80. [24] Green BG, Flammer LJ. Methyl salicylate as a cutaneous stimulus: a psychophysical analysis. Somatosens Mot Res 1989;6:253– 74. [25] Bandell M, Story GM, Hwang SW, Viswanath V, Eid SR, Petrus MJ, et al. Noxious cold ion channel TRPA1 is activated by pungent compounds and bradykinin. Neuron 2004;41:849– 57. [26] Story GM, Peier AM, Reeve AJ, Eid SR, Mosbacher J, Hricik TR, et al. ANKTM1, a TRP-like channel expressed in nociceptive neurons, is activated by cold temperatures. Cell 2003;112:819–29. [27] Tominaga M, Caterina MJ. Thermosensation and pain. J Neurobiol 2004;61:3–12. [28] Kakigi R, Watanabe S. Pain relief by various kinds of interference stimulation applied to the peripheral skin in humans: pain-related
98
B.G. Green, K.L. Schoen / Behavioural Brain Research 162 (2005) 90–98
brain potentials following CO2 laser stimulation. J Peripher Nerv Syst 1996;1:189–98. [29] Khasabov SG, Cain DM, Thong D, Mantyh PW, Simone DA. Enhanced responses of spinal dorsal horn neurons to heat and cold stimuli following mild freeze injury to the skin. J Neurophysiol 2001;86:986–96. [30] Beise RD, Carstens E, Kohll¨offel LUE. Psychophysical study of stinging pain evoked by brief freezing of superficial skin and ensuing short-lasting changes in sensations of cool and cold pain. Pain 1998;74:275–86. [31] Stevens JC, Marks LE, Gagge AP. The quantitative assessment of thermal discomfort. Environ Res 1969;2:149–65.
[32] Ebaugh FG, Thauer R. Influence of various environmental temperatures on the cold and warmth thresholds. J Appl Physiol 1950;3:173–82. [33] Fruhstorfer H, Harju EL, Lindblom UF. The significance of Adelta and C fibres for the perception of synthetic heat. Eur J Pain 2003;7:63–71. [34] Yarnitsky D, Ochoa JL. Release of cold-induced burning pain by block of cold-specific afferent input. Brain 1990;113:893–902. [35] Wahren LK, Torebjork E, Jorum E. Central suppression of coldinduced C fibre pain by myelinated fibre input. Pain 1989;38:313–9. [36] Fruhstorfer H. Thermal sensibility changes during ischemic nerve block. Pain 1984;20:355–61.