Figure 1. Effects of absolute skin temperature and rate of change in skin temperature on thermal sensations. Panel A shows changes in the absolute temperature threshold (i.e., the relative change in skin temperature required to induce a thermal sensation; solid squares, solid curves) and in the just noticeable difference threshold (open squares, broken curves), resulting from different adapting skin temperatures (curve 1: males; curve 2: females). It can be observed that, when detecting decreases in skin temperature, the warmer the baseline temperature, the greater the change in temperature required to induce a cold sensation. A similar effect is noticeable when detecting increases in skin temperature [from Kenshalo et al. (176); reprinted with permission from AAAS]. Panel B shows the effect of the rate of the stimulus temperature change upon the warm and cool thresholds of three male subjects measured at normal skin temperature. The ordinate shows the change from normal skin temperature necessary to produce threshold warm and cool sensations at each of the rates of change in temperature. Rates of change as low as 0.1°C/s failed to increase either threshold. Rates of temperature change slower than 0.1 °C/s resulted in increases in both warm and cool thresholds although this effect was more pronounced upon the warm threshold [from Kenshalo et al. (175); with kind permission from Springer Science and Business Media]. Panel C shows the relationship between the intensity of cooling pulses (duration: 4 s) applied to the hand and related subjective magnitude estimates of thermal sensation. Skin had been previously adapted at 34°C [From Darian‐Smith (76); © The American Physiological Society].

Figure 2. Relationships between intensity of radiant heating applied to the (A) cheek, (B) forearm, (C) calf, and (D) back and subjective magnitude estimates of thermal sensation, as resulting from stimulation of different areal extents [reprinted from Stevens et al. (274) with permission from Elsevier].

Figure 3. Body maps showing regional distribution of (A) skin cooling (°C), (B) absolute mean votes for thermal sensation and (C) frequency of wetness perception, as resulting from 10‐s application of a relative cold‐dry stimulus (15°C lower than local skin temperature) to each of 12 skin sites (i.e., 6 on the front and 6 on the back torso). Data were collected on the left side of the body and the body maps presented were developed assuming left‐right symmetry. Regions showing greater skin cooling, colder sensations, and more frequent wetness perceptions are represented in darker colors. The rating scale used by the participants to score their absolute thermal sensations is reported next to the respective body map. Two main tendencies are shown. First, the regional differences in thermal and wetness sensations present a similar pattern across the torso (e.g., as opposed to the chest, the lateral and lower back appears more sensitive to cold and wetness). Second, these sensory patterns seem independent from the regional variations in skin cooling (i.e., regions which show greater skin cooling, such as the lateral chest, are not necessarily the ones in which the stimulus was perceived as colder and more often wet) [from Filingeri et al. (99); © The American Physiological Society].

Figure 4. A cold thermoreceptors in the glabrous skin of the cat's nose. Panel A shows a vertical semithin section (650×) through a marked area in the cat's nose, where a cold thermoreceptors is identified (black arrows). A lamellate encapsulated receptor with a small capsule space (lc) is located in the marked field too, but lies about 80 μ deeper in the skin near the epidermal columnar ridge (cr). Capillary (cp). Panel B shows the same cold receptor (electron micrograph; 20,000×) with its axon (tax) entering into the basal layer of the epidermis. The axoplasm contains numerous mitochondria (mi) and some scattered glycogen particles. The basal lamina (bl) of the epidermis fuses with the basal lamina of the receptor axon (arrows). Schwann cell (sc), collagen fibers of the papillary connective tissue (cf), tonofibrils (tf) of the epithelial cells are also marked. Panel C shows a schematic representation of the marked cold receptor in panel A and various mechanoreceptors in the glabrous skin of the cat's nose. The cold receptor (cld) is located in the top of a dermal papilla and its terminals penetrate into the basal layer of the epidermis. Merkel cell neurite complexes (m) are sometimes located in the epithelium of the epidermal columnar ridge (cr). Lamellate encapsulated receptors with a capsule space (lc) are present in the connective tissue below. Tip of the wire hole (w), used to slice the sample. Adhesive ridge (at). Panel D shows a schematic representation of the cold receptor axon shown in panel B. The terminal protrudes into a basal epidermal cell. The Schwann cell cover and basal lamina fuse with the epidermis. The typical structures of the receptor matrix are present below the receptor membrane. Receptor axon (tax), Schwann cell (sc), epidermis (e), papillary connective tissue (pct), and basal lamina (bl) [from Hensel (145); with kind permission from Springer Science and Business Media].

Figure 5. Responses of human cold sensitive (A, B, and E) unmyelinated C‐fibers and (C) myelinated Aδ‐fibers to a (D) standard cooling staircase. Panels A and B show two C‐fibers discharging upon cooling all the way down to 0°C. Panel C shows one Aδ‐fiber discharging upon cooling, but stopping firing at about 14°C [Reprinted from Campero et al. (39) with permission from John Wiley & Sons, Inc.]. Panel E shows responses of a single cold sensitive C‐fiber to 10 s temperature pulses. Top trace: full response to a temperature pulse from a baseline skin temperature of 35°C to 30°C, with the temperature profile shown below. Underneath traces: expanded 1‐s sections of responses showing peak phasic (left) and tonic (right) discharges, for temperature steps (rate of change: 10°C s⁻¹) from 35°C to the temperatures indicated [reprinted from Campero et al. (40) with permission from John Wiley & Sons, Inc.].

Figure 6. Paradoxical discharge of human and primate cold sensitive fibers in response to heat stimuli. Panel A shows activity recorded in a human cold sensitive C‐fiber in response to skin warming (10‐s pulse) from 35°C to 40°C. Panel B shows activity in the same fiber in response to skin warming from 35°C to 45°C. It can be observed that, while innocuous warming (35°C‐40°C) does not activate the unit, skin warming beyond the noxious threshold (<42°C) results in a paradoxical discharge in this cold sensitive unit. Panel C shows a 1‐s expansion of the period marked by the open bar in panel B, highlighting the bursting pattern of response for this human C‐fiber [Reprinted from Campero et al. (40) with permission from John Wiley & Sons, Inc.]. Panel D shows paradoxical discharge in primate cold sensitive A‐fibers in response to heat stimuli (i.e., >47°C) and their modulation by different core temperatures. The graph shows the proportion of sampled cold fibers, which responded to stimuli of different intensities. It can be observed that the skin warming threshold for this paradoxical discharge decreases (from 49°C to 41°C) as core temperature increases (from 37°C to 39°C) [from Long (201); © The American Physiological Society].

Figure 7. Static thermal sensitivity of primate and human warm sensitive C‐fibers. Panel A shows average static impulse frequency of two populations of single warm fibers from hairy skin of foot in rhesus monkey as function of constant temperatures. The dashed line indicates shift from regular go irregular discharge or bursts. It can be observed that in one population, the function of discharge rate versus steady state stimulus temperature follows a bell‐shaped curve, while in the other population, steady frequency increases continuously with rising skin temperature up to 44°C [From Hensel and Iggo (147); with kind permission from Springer Science and Business Media]. Panel B shows total number of spikes for a human warm sensitive C‐fiber during the first 3 s after the application of thermal stimuli (filled circles) in comparison with the averaged total impulses during corresponding periods for 15 polymodal nociceptive C‐fibers (triangles) [from Hallin et al. (134); adapted by permission from BMJ Publishing Group Limited].

Figure 8. Impulse frequency as a function of time in a single cold sensitive A‐fiber isolated from human radial nerve during skin cooling and rewarming [from Hensel and Boman (146); © The American Physiological Society].

Figure 9. Dynamic thermal sensitivity of primate cold sensitive A‐fibers and warm sensitive C‐fibers. Panel A shows peristimulus time histograms for a cold sensitive A‐fiber computed using spike trains evoked by skin cooling. The cool‐stimulus intensity‐rate series consisted of decreases in temperature of 0.5°C, 1°C, 2°C, and 5°C from an adapting temperature of 30°C and at rates of (i) 2 and (ii) 0.4°C·s⁻¹ [From Kenshalo and Duclaux (174); © The American Physiological Society]. Panel B shows peristimulus time histograms for a warm sensitive C‐fibers computed using spike trains evoked by skin warming. (i) Responses to warming pulses of intensities 2°C, 4°C, 6°C, and 8°C from a baseline skin temperature of 34°C. The profile of the warming pulses is shown in black at the bottom of the graph. It can be observed that, although the fibers' responses increased in an orderly fashion with increasing stimulus intensity, the temporal profile of the response was remarkably constant during the first 4 s of stimulation. This can be further observed in (ii) and (iii), in which the fiber's responses to stimuli of intensities of (ii) 4°C and 2°C and of (iii) 4°C and 8°C, have been normalized by equating the 4‐s cumulative impulse counts of the two responses. Such normalization show how the temporal profile of the responses to warming pulses of different intensities match well during the first 4 s of stimulation. As that, indicating that the rate of frequency increase in these fibers appears to be the same regardless of the intensity of the temperature change. Such observation implies that this parameter is not the main one used by the brain to determine the magnitude of temperature change [from Darian‐Smith et al. (74); © The American Physiological Society].

Figure 10. Cold sensitive lamina I neurons in the cat's spinal cord. Panel A shows a photomicrograph of a lesion marking the recording site (black arrow) of a thermosensitive lamina I spinothalamic neuron in the lumbar‐7dorsal horn of the cat' spinal cord. Bar = 0.5 mm. Panel B shows two histogram records (i and ii) highlighting the reproducibility of the response of this cold‐sensitive lamina I thermosensitive neuron to the standard cooling/warming stimulus sequence (shown by the traces at the bottom of each record indicating skin‐thermode temperature). (iii) Summary of the mean discharge rates at each temperature for 4 applications of the same stimulus [From Craig et al. (67); © The American Physiological Society]. Panel C shows camera lucida reconstructions of the full dendritic extent of three different categories of lamina I neurons identified in the cat' spinal cord: (i) fusiform nociceptive (NS) specific; (ii) pyramidal (COLD) thermoreceptive specific; (iii) multipolar heat‐pinch‐cold (HPC) specific (note: rostral is left, medial is up) [reprinted by permission from Macmillan Publishers Ltd: Nature Neuroscience (Han et al., copyright 1998) (135)].

Figure 11. Cold‐sensitive Lamina I spinothalamic neuron in the monkey's spinal cord. Panel A shows an antidromic response in this spinothalamic neuron to stimulation pulses applied from an electrode in the VMpo of the thalamus. Panel B shows two periods of inhibition of the spontaneous firing of this neuron upon radiant warming applied on the unit's receptive field (i.e., glabrous skin of the foot), which is illustrated in panel C (black region indicates maximal sensitivity). Panel D shows dynamic responses of this spinal cold sensitive neuron to progressive cooling steps of 4°C, from 34.5°C (baseline) to 12.5°C. Responses to warming pulses are shown in the second half of the record, where immediate reductions in tonic firing upon warming are observable, as well as a characteristic paradoxical discharge upon noxious heating (>47°C). Panel E shows relationship between dynamic and static responses of this neuron and skin‐thermode interface temperature. For warming pulses, the values plotted represent the mean firma rate during the total duration of the heat pulse. Note that response at 40°C is inhibition (less‐than baseline firing rate at 34%) [from Dostrovsky and Craig (88); © The American Physiological Society].

Figure 12. Warm sensitive lamina I neurons in the cat's spinal cord. Panel A shows responses of a single warm sensitive neuron to innocuous (38.7°C) and noxious (53.0°C) skin warming respectively. Specifically, firing rate (1‐s bins), temperature profile, and single unit recording are shown. Panel B shows individual temperature stimulus‐response curves for eight different group warm sensitive lamina I neurons and panel C shows group mean (±1 standard deviation) for the same group (n = 8) [reprinted from Andrew and Craig (5) with permission from John Wiley & Sons, Inc.].

Figure 13. Microstimulation of the posterior and inferior core of the ventral caudal nucleus (a portion of the VPL) of the human thalamus evokes nonpainful thermal sensations in humans. Eleven patients undergoing stereotactic surgery for the treatment of movement disorders and pain were trained to describe sensations evoked intraoperatively by thalamic microstimulation. The figure shows maps of receptive and projected fields for trajectories in the region of the ventral caudal nucleus in a single patient. The stereotactic coordinates of the anterior commissure (AC) and posterior commissure (PC) were determined by computer‐assisted tomography. These coordinates were used to generate maps of the human thalamus in sagittal section that had been transformed to match the AC‐PC line in that patient. Panels A shows positions of the trajectories relative to nuclear boundaries as predicted radiologically from the position of the anterior commissure‐posterior commissure (AC‐PC) line. The AC‐PC line is indicated by the horizontal line in the panel; the trajectories are shown by the two oblique lines. The positions of nuclei are inferred from the AC‐PC line and therefore are only an approximate indicator of nuclear location (anatomic names; Vc: ventralis caudalis; Vcpor: ventralis caudalis portae; Vcpc: ventralis caudalis parvocellularis; Lim: limitans; MG: medial geniculate; ML: medial lemniscus; WM: white matter below the ventral nuclear group). Panel B shows location of single neurons, stimulation sites, and trajectories (S 1 and S2) relative to the PC. The locations of stimulation sites are indicated by ticks to the left of the trajectory; the locations of single neurons are indicated by ticks to the right of the trajectory. Single neurons with receptive fields (RFs) are indicated by long ticks; those without are indicated by short ticks. The core thalamic region is defined as the area where the majority of ticks to the right of the trajectory are long. The quality of the evoked sensation is indicated by the symbol at the end of the tick to the left of the trajectory. Square: paresthetic sensation was evoked at that site. Open circle: sensation of nonpainful heat. No symbol: no sensation was evoked. Scales are as indicated. Each site where a neuron was recorded or stimulation was carried out or both is indicated by the same number in panels B and C. In panel C, S1 and S2 show the site number, projected field (PF), and RF for that site. The threshold (in microamperes) is indicated below the PF diagram [from Lenz and Seike (195); © The American Physiological Society].

Figure 14. Positron emission topography imaging performed during tonic cooling of the hand (range of 33°C‐22°C) in humans and activation of the insular cortex. Panel A shows the relationship between: (top graph) magnitude estimate of the perceived intensity of cold (following each scan using an open‐ended scale of 0‐10) and stimuli temperatures; (bottom graph) average regional cerebral blood flow activity in left (contralateral) insular cortex and stimuli temperatures. Panel C shows the location of “thermosensory cortex” in the dorsal margin of left (contralateral) insular cortex (identified by regression analysis of regional cerebral blood flow activation with stimulus temperature) in frontal, axial, and sagittal views (note that in the axial view, this site seems parietal simply because the insula cannot be seen). The activation in right anterior insula is also visible in the axial view. It can be observed that cortical activity in such region strongly correlates with human discrimination of the intensity of the thermal stimulus [reprinted by permission from Macmillan Publishers Ltd: Nature Neuroscience (Craig et al., copyright 2000) (64)].

Figure 15. Evoked potentials recorded directly from the brain in response to an innocuous cold stimulus applied to the contralateral hand of two female patients who had subdural grids implanted for surgical treatment of medically intractable complex partial seizures. Panel A shows: (top) temperature profile of the cold stimulus used; (bottom) the maximal potential recorded over the parietal operculum. Panel B shows the location of the recording electrodes. Cold‐evoked potentials were recorded from three rows of six electrodes labeled C (1‐6), D (1‐6), and E (1‐6) (CS: central sulcus; IFS: inferior frontal gyrus; SFS: superior frontal sulcus; PreCS: precentral sulcus; PostCS post central sulcus; SF: sylvian fissure). Panels C‐E shows cold‐evoked potentials recorded from each electrode in each row. Overlays of potentials recorded from two adjacent electrodes with large cold‐evoked potentials for each row are shown as the bottom tracing in the corresponding C‐E panels. It can be observed that the largest cold‐evoked potentials were recoded from electrodes adjacent to the sylvian fissure [from Greenspan et al. (129); © The American Physiological Society].

Figure 16. Tridimensional structure of the TRPV1 ion channel (determined by electron cryo‐microscopy). Panel A present a linear diagram depicting major structural domains, color coded to match ribbon diagrams in panel B. The TRPV1 channel contains six transmembrane spanning regions (S) with a pore‐forming reentrant loop between the fifth (S5) and the sixth (S6). Both the carboxyl (C‐) and amino (containing three to five ankyrin repeats) termini are intracellular (dashed boxes denote regions for which density was not observed, for example, first two ankyrin repeats, or where specific residues could not be definitively assigned, for example, C‐terminal b‐strand). Panel B presents diagrams showing three different tridimensional views of a TRPV1 channel [reprinted by permission from Macmillan Publishers Ltd: Nature (Liao et al., copyright 2013) (197)].

Figure 17. The capsaicin receptor vanilloid receptor 1, VR1 (subsequently renamed TRPV1) activated by noxious thermal stimuli. Panel A shows Xenopus (i.e., clawed frog) oocytes cells transiently transfected with TRPV1 which exhibit a pronounced increase in cytoplasmic free calcium when transiently exposed to a peak temperature of 45°C. The same cells did not respond to heat when transfected with vector alone (pcDNA3). Relative calcium concentrations are indicated by the color bar. Panel B shows results from whole‐cell patch‐clamp analysis of TRPV1‐transfected Xenopus oocytes cells. It can be observed that inward currents are generated in response to both heat and capsaicin. The temperature of the bath medium was raised from 22°C to 48°C (heat), and then restored to 22°C, after which capsaicin (0.5 mmol/L) was added to the bath. Stimulus‐induced current‐voltage relations are shown on the right. Panel C shows that TRPV1 expressed in Xenopus oocytes is activated by noxious, but not innocuous warm temperatures. Two‐electrode voltage‐clamp was performed in oocytes injected with either TRPV1 cRNA or water, while the perfusate temperature was raised from 22°C to ∼45°C. The asterisk indicates a significant difference from water‐injected oocytes. Panel D shows the inhibitory effect of Ruthenium red (RR) (a potent inhibitor of intracellular calcium release) on heat‐ and capsaicin‐evoked responses in TRPV1‐expressing oocytes [reprinted by permission from Macmillan Publishers Ltd.: Nature (Caterina et al., copyright 1997) (44)].

Figure 18. TRPM8 ion channel as the principal detector of environmental cold. Results are reported from analysis of cold‐evoked responses in cultured sensory neurons and intact sensory nerve fibers from TRPM8‐deficient mice. Also, behavioral discriminatory ability between cold and warm surfaces, and responses to evaporative cooling, were analyzed in TRPM8 mutant mice. Panel A shows immunostaining of trigeminal ganglia (left), corneal afferents (middle), and spinal cord dorsal horn (right) with anti‐TRPM8 (green) and anti‐TRPV1 (red) antibodies in TRPM8‐deficient mice. It can be observed that these animals present a selective loss of TRPM8 expression (scale bars, 50 mm). Panel B shows: (i) responses of trigeminal neurons to menthol (TRPM8 agonist), capsaicin (TRPV1 agonist) and potassium chloride (KCl) in TRPM8 expressing (+/+) and lacking (−/−) mice (note: no menthol‐sensitivity is observable in TRPM8‐deficient neurons); (ii) prevalence of sensory neurons responding to capsaicin, mustard oil, menthol, and icillin in TRPM8 expressing (+/+) and lacking (−/−) mice (note: significantly reduced proportion of neurons responding to menthol and icillin in TRPM8‐deficient neurons); (iii) responses of trigeminal neurons to cold and menthol (green bar) in TRPM8 expressing (+/+) and lacking (−/−) mice (note: no cold and menthol‐sensitivity is observable in TRPM8‐deficient neurons; dotted line: menthol‐insensitive neurons); and (iv) comparison of cold and heat sensitivity in WT, TRPM8‐ and TRPV1‐deficients mice (note: cold and heat sensitivities are selectively reduced in TRPM8‐ and TRPV1‐deficient mice, respectively). Panel C shows: (i) cooling‐induced responses in cutaneous C‐fibers; (ii) percentage of cold‐activated C‐fibers; (iii) cooling‐induced responses in cutaneous A‐mechanoreceptive fibers; and (iv) percentage of cold‐activated fibers A‐mechanoreceptive fibers, in WT and TRPM8‐deficient mice [reprinted by permission from Macmillan Publishers Ltd: Nature (Bautista et al., copyright 2007) (11)].

Figure 19. A schematic summary of the molecular, neuroanatomical and neurophysiological bases of skin thermal sensations in humans. Human hairy skin comprises of a number of first order neurons innervating the epidermal layer of the skin. Innocuous warmth is encoded by unmyelinated C‐fibers expressing TRPV1 ion channels. Expression of warm sensitive potassium channels (TREK/TRAAK) likely contribute to membrane resting potentials. The presence of warmth sensitive TRPV3 and TRPV4 ion channels expressed in skin keratinocytes, contributing to warmth transduction via ATP‐mediated signaling mechanisms to C‐fibers could also contribute to warmth detection. Innocuous cold is encoded primarily by myelinated Aδ‐fibers expressing TRPM8 ion channels and cold sensitive sodium channels (Nav1.8; contributing to membrane resting potentials). Myelinated Aβ‐fibers (mechanoreceptors) expressing TRPM8 ion channels show uncorrelated responses to cold temperatures. Unmyelinated cold and heat sensitive C‐2 fibers, expressing cold sensitive TRPM8, heat sensitive TRPV1 and potentially cold sensitive TRPM3 and TRPA1 ion channels, encode noxious cold (<15°C) and heat (>45°C). In panel A and B, temperature‐dependent changes in discharge frequency of first‐order thermosensitive neurons and in TRP‐mediated ionic conductance, are shown. Neuroanatomically, the specific ascending thermosensory pathway which allows humans to peripherally encode and centrally process skin thermal sensations comprises of: (i) first‐order thermosensitive Aδ‐, C‐, and C2‐fibers, terminating in the spinal cord Lamina I, and synapsing with cold‐ and warm‐sensitive and heat‐pinch‐cold sensitive second order neurons; (ii) second‐order lamina I spinothalamic neurons, ascending contralaterally along the anterolateral columns of the spinothalamic tract, and terminating in the posterior part of the VMpo of the thalamus; and (iii) third‐order thermosensitive neurons, located in posterior part of the VMpo of the thalamus and projecting to the posterior insular cortex, that is, the main thermosensory cortex. Functionally, this neurophysiological pathway sub serves peripheral and central mechanisms underpinning our ability to characterize both the discriminative and affective components of skin thermal sensations in the context of thermal behavior. Potential interactions with other cortical (e.g., somatosensory cortex) and sub cortical regions (e.g., nucleus of the tractus solitarius, hypothalamus), receiving sensory inputs from both thermoreceptive and mechanoreceptive fibers (note: these ascend to the somatosensory cortex via the medial lemniscus and the thalamic ventroposterior nuclei), likely contribute to modulate and enrich thermal processing in the context of multimodal (e.g., thermal‐tactile) somatosensory interactions and human thermosensory experience.

Figure 20. Specific ranges and rates of local skin cooling drive the perception of local skin wetness. This figure presents psychophysical results related to the application of six progressively more intense cold‐dry stimuli (via a thermode) on the forearm of blindfolded females, who reported their stimulus‐driven wetness perception upon contact cooling. Panel A shows relative drops in skin temperature from baseline and corresponding cooling rates as a result of each of the six cold‐dry stimuli. Panel B shows wetness perception scores recorded as a result of each of the six cold‐dry stimuli (phase B) and after removal of the stimulus (i.e., bare skin phase, C). Skin cooling rates corresponding to each stimulus are reported between brackets. The point “1” of the wetness perception scale corresponds to the threshold set to identify perceived skin wetness. It can be observed that cold‐dry stimuli producing skin cooling rates in the range of 0.14 to 0.41°C·s⁻¹ induced a clear illusion of skin wetness [reprinted from Filingeri et al. (102) with permission from Elsevier].

Figure 21. Neurophysiological model of cutaneous wetness sensitivity. Mechanosensitive Aß‐, cold sensitive Aδ‐ and warm sensitive C‐fibers and their projections from the skin, through peripheral nerve, spinal cord (via the dorsal‐column medial lemniscal pathway and the spinothalamic tract), thalamus and cerebral cortex (including the primary and secondary somatosensory cortices SI and SII, the insular cortex and the posterior parietal lobe) are shown. Panel A and B shows the neural model of wetness sensitivity (consisting of Aδ and Aß afferents) under normal and under selective reduction in the activity of A‐nerve fibers, respectively. Panel C, E, and G show the pathways for wetness sensitivity during static contact with warm, neutral, and cold moisture. Panel D, F, and H shows the pathways for wetness sensitivity during dynamic contact with moisture [from Filingeri et al. (98); © The American Physiological Society].

Figure 22. Ocular surface wetness as regulated by TRPM8‐dependent cold thermoreceptors in the cornea. This figure presents responses of mice corneal neurons to cooling ramps and menthol, as well as dependence of tear secretion rate on corneal temperature, as recorded in WT mice expressing TRPM8 ion channels (TRPM8^+/+), in mutant mice presenting a reduced expression of TRPM8 ion channels (TRPM8^+/−), and in TRPM8‐deficient mice. Panel A shows mean firing frequency as a result of a cooling ramp (left) and menthol application (right). It can be observed that TRPM8^+/− and TRPM8^−/− corneal neurons present reduced and absent response to cooling and menthol respectively. Panel B shows mean firing frequency as a result of the same cooling ramp in corneal neurons from mice expressing (TRPA1^+/+) or deficit of (TRPA1^−/−) the noxious‐cold sensitive TRPA1 ion channel. It can be observed that no changes in innocuous cold sensitive of corneal neurons occur as a result of the absence of TRPA1, indicating a lack of involvement of this channel in the transduction of innocuous corneal cooling. Panel C shows basal tearing rate (mean wetted length of the phenol red thread in mm) measured in the eyes of anesthetized mice exposed to environmental temperatures of 25°C and 42°C that modified their corneal surface temperature to the values indicated. It can be observed that increases in tearing rate during corneal cooling occurring in WT mice (black bar), is abolished in TRPM8 lacking mice (TRPM8^−/−; red bar), while conserved in TRPA1 lacking mice (TRPA1^−/−; blue bar). Panel D shows changes in tearing rate as a result of capsaicin (1 μmol/L) and AITC (500 μmol/L) (two well‐known chemo irritants) application to the cornea of WT mice (black bar) and TRPM8 lacking mice (TRPM8^−/−; red bar). It can be observed that in both TRPM8 expressing and lacking mice, irritation‐induced tearing is unaltered, indicating that while TRPM8 likely controls basal tearing, this is not involved in reflex responses to noxious agents. Panel D shows tearing rates at different corneal temperatures in 11 healthy humans. It can be observed that, as corneal temperature increases, tearing rate decreases [reprinted by permission from Macmillan Publishers Ltd: Nature Medicine (Parra et al., copyright 2010) (240)].

Figure 23. Conceptual model of human hygrosensation. The model comprises biophysical (i.e., thermal and tactile inputs induced by the presence of moisture on the skin), neurophysiological (i.e., central integration of afferents inputs from thermosensitive TRP ion channels and nerve fibers and mechanosensitive Degenerin/Epithelial sodium channels (DEG/ENaC) ion channels and nerve fibers) and psychophysiological mechanisms (i.e., perceptual inference operated by cortical and subcortical somatosensory and association areas), which allow humidity and wetness detection in humans. The skin's contact with moisture generates thermal and tactile inputs, which are peripherally integrated by specific nervous structures. These inputs evoke thermal and tactile sensations, which, in the absence of specific hygroreceptors, are associated to the perception of skin wetness. Repeated exposures to these stimuli (i.e., sensory experience) contribute to generate a neural representation of a typical wet stimulus via learning mechanisms. At this point, only if the learnt combination of stimuli (i.e., coldness and stickiness), as coded by the specific neural afferents (i.e., A‐nerve fibers) is presented, wetness will be sensed. In the occurrence of physical wetness on the skin, the bottom‐up processes (i.e., combination of thermal and mechanical sensory afferents) as well as the top‐down ones (i.e., inference of the potential perception based on the neural representation of a typical wet stimulus) might, therefore, interact in giving rise (or not) to the perception of wetness [from Filingeri et al. (101)].