Neuroimaging of pleasantness and unpleasantness induced by thermal stimuli

ABSTRACT

Functional brain imaging techniques provide unique insight into the process of human thermal regulation and its associated hedonics. Similar neuroimaging techniques have predominantly focused on the neural characterization of thermal response separately from hedonics. In this instance, there is a gap in the understanding of hedonics related to regional brain activations. Responses to localized, thermal stimuli are yet to be characterized, but it would appear that thermoregulatory regions are widely distributed throughout the hemispheres of the human brain. The distributed nature of neural activations related to temperature responses is consistent with multiple related functions contributing to thermoregulation. Estimating hedonics of thermal stimulation includes a cognitive process that could potentially interfere with identifying activation specific to hedonics. A future challenge for brain imaging studies is to more accurately dissect the functional neuroanatomy of thermoregulation and related hedonics in hemispheric regions.

KEYWORDS:

• Functional magnetic resonance imaging
• contact heat
• hedonics
• posterior cingulate cortex
• human

Introduction

Thermal comfort is generally defined as the “condition of mind, which expresses satisfaction with the thermal environment” [1]⁠. Thermal comfort is a key parameter for a healthy and productive workplace and living environment, and so has been commonly investigated to assist the ergonomics of building design [2,3]⁠. Research has mostly focused on developing thermal models of the human body and on body–environment interactions. Fanger used data from climate chamber studies and combined the theories of heat balance with the physiology of thermoregulation [4,5]⁠. His work has motivated a series of mathematical models to describe and predict thermal comfort⁠. Other researchers have used field surveys to directly measure the subjective acceptability of thermal environments [5]⁠. Both approaches aim to establish thermal comfort standards for the workplace and general living.

Research that extends beyond the pragmatic assessment of optimum working and living conditions to consider the nature of any pleasant or unpleasant experience due to the thermal environment or the neural mechanism of thermal experience is less common. Cabanac proposed the concept of “alliesthesia,” which describes a pleasant or unpleasant sensation following a thermal stimulus that depends on the subject’s internal state [4]⁠. Attia and Engel further demonstrated that comfort is often perceived as a state of neutrality or indifference in relation to the external environment. They went on to show that the sensation of positive comfort could only be induced as a sense of relief when transitioning out of hyper or hypothermic conditions [6, 7], used alliesthesia as a logical framework to differentiate positive thermal comfort, which depends on an extreme internal state of thermal discomfort, from thermal neutrality, which depends on a steady internal thermal state [7]⁠. The term “hedonic” refers to a perspective that focuses on the spectrum of experiences ranging from pleasure to pain and includes biological, social, and phenomenological aspects and their relationship to motivation and action (APA Dictionary of Psychology).

Kingma et al.have recently contributed to the concept of thermal neutrality by taking thermal comfort into account by showing that ambient thermal neutrality does not guarantee that the body is in thermal balance. Instead, the body increases heat production or loses heat based on a combination of the body’s core temperature, skin temperature, and environmental temperature. These findings have considerably affected the understanding of the thermoneutral zone. It could be further inferred that provided animals seek thermal comfort, thermal adaptive behaviors could be initiated before the boundaries of thermoneutral zones are reached [8].

There is currently increased interest in customizing thermal comfort to individual preferences and improving existing thermoregulatory models. For example, the Predicted Mean Vote or adaptive model, which predicts the thermal comfort of a group of people is not able to precisely predict the thermal comfort of individuals. [9], found that variability of thermal sensation in the same ambient temperature is about ±3°C difference from the neutral temperature.

Understanding the underlying mechanisms of thermal comfort will likely inform the design and operations of indoor environments.

Here, neuroimaging studies that have investigated the neural correlates of innocuous temperature, hedonics, defined as the sensory pleasures as well as other higher types of pleasure, and the interaction of the two are discussed in this narrative review [10,11]⁠. Thermoregulatory behavior inevitably includes a hedonic dimension, and so studies that explore hedonic brain representation are also reviewed and discussed [12]⁠. The possibility of exploring thermal comfort using neuroimaging is discussed in the final section [11,13]⁠.

The neural bases of innocuous heat (warmth) in the brain

To understand the neural representation of innocuous thermal stimuli may be useful to better comprehend how the thermal environment directs and modulates behavior. When people are too hot or too cold, they will interrupt their activities to remove or add clothing or find another way to influence the source of thermal regulation such is the case for air-conditioning systems. Those behaviors are often driven by slight changes in perception that may not be entirely conscious until the disruption begins, but they may be detectable via changes in brain activity before the urge to disrupt ongoing activities becomes apparent.

Positive affective evaluation of physical stimuli sensations such as temperature is commonly described as pleasantness [10]⁠.

The exact neural correlates of thermoregulatory control remain nowadays elusive. Prior animal studies and human lesion studies have shown that the preoptic area (POA) and the hypothalamus play an important role in the thermoregulatory control [14]⁠. However, a more recent consensus points out that the view of the neural source of thermoregulatory control is far from being unified. Indeed, it is proposed that body temperature might be regulated by independent thermoeffector loops (with its own afferent and efferent branches) including multiple feedback, feedforward and open-loop components contributing to specific aspects of thermal balance [15].

When separating between early, intermediate, and late stages of heat intensity processing, different patterns of brain activations have been shown to reflect different aspects of the experience of pain and stimulus processing. The early phase of painful heat perception has been associated with activations in the anterior insula and anterior cingulate cortex (ACC) activation, while persistent heat responses have been associated with activations in the dorsolateral prefrontal cortex (DPC) and inferior parietal lobule [16,17]⁠, have focused on examining the neural mechanisms related to the transition from innocuous to painful thermal stimuli (ranging from 44°C to 49°C). Activation of dorsal cingulate and both anterior and posterior insula were shown to be consistent with similar activations observed in studies examining the transition from innocuous to painful sensation [17–19]⁠.

There are several brain imaging studies have examined changes due to non-noxious thermal sensation, but many more have examined noxious heat leading to the painful experience. Early responses to noxious heat include activation of the AI and ACC. Similar to the early responses to noxious heat, the transition from non-noxious to noxious heat involves activation of the anterior and posterior insula and the ACC. With persistent noxious heat, DPC and the inferior parietal lobe are additionally activated.

One of the differences between brain response to innocuous versus noxious stimuli is the habituation of the brain response to noxious but not innocuous stimuli [20]⁠ – a finding that has been consistently shown with both near-infrared spectroscopy and functional magnetic resonance imaging (fMRI) methodology [21,22]⁠. In addition, it has been demonstrated that noxious stimuli produce a bilateral response (though the contralateral side appears to show a stronger activation than the ipsilateral side), while innocuous stimuli produce only a contralateral response [20,23–25,26]⁠, showed that innocuous and noxious stimuli produced responses with different time courses in medial ACC and contralateral primary and secondary somatosensory cortex. Later work has also provided support for temporal differences, indicating that the late stages of heat processing are correlated with heat perception, rather than pain [16]⁠.

Studies on the neural underpinnings of innocuous heat processing have been relatively scarce. In an early electroencephalography study [27], suggested the existence of second-order neurons specific for transferring innocuous sensations as a slow-conducting pathway, most likely processed by the anterior cingulate and opercular insular cortices (IC)⁠. Additionally, fMRI studies showed contralateral primary somatosensory, mid anterior cingulate, and secondary somatosensory early activation to an innocuous temperature [26]⁠. More consistently, it has been shown that activity in the right anterior IC is associated with innocuous temperature perception [28]⁠. However, Becerra reports the same brain regions to be activated by noxious stimuli with the only difference being that innocuous warmth is associated with weaker, less reliable responses [21,29]⁠. Probable activation in the right anterior IC in response to feelings of warmth is consistent with the hypothesis that visceral, thermal, and pain sensations are processed in a shared neural network [30–32]⁠.

In another positron emission tomography study, Rolls showed that reports of warmth were associated with activations in the orbitofrontal cortex (OFC) [29]⁠.

Recent findings of [33] further reinforce the idea that the right anterior IC is important for temperature perception.

Insula and operculum are thought to function as a relay region, where visceral sensations are translated into emotions and responsible for visceral awareness [31,34,35]⁠. Given the importance and ubiquity of thermal sensations for all the human experience, further inquiry into the relationship between thermal sensation, thermoregulation, and hedonics of temperature perception is warranted [36]⁠. The [37], draw attention to physiological and psychological factors, such as body composition, metabolic rate, adaptation to certain thermal environments, and perceived control, to differences in thermal perception.

The neural bases of positive hedonic in the brain

Hedonic refers to sensory pleasures as well as other higher types of pleasure (e.g., cognitive, social, esthetic, and moral). Evolutionary pleasure can be explained as motivation to pursue rewards that can increase the likelihood of ones’ survival [38]⁠.

Brain imaging techniques have shown that pleasantness is mediated by a distributed neural network that is suggested to process different modalities of pleasantness. However, affective evaluation of stimuli does not entirely depend on the modality of stimuli itself, but also on internal body state (e.g. core, surface body temperature) [4]⁠. Temperature intensity is perceived as skin detectors translate thermal signals into neuronal messages describing local temperature and its changes resulting in the skin feeling “cold”, “warm,” or “hot” according to a given stimulus. However, affective evaluation of thermal stimuli as pleasant or unpleasant occurs whenever the cutaneous thermal signal is outside the optimal range. At that point, and after the initial behavioral responses, the autonomic response, which helps establishing internal homeostasis is triggered [4,7]⁠.

It is important to emphasize that primarily the behavioral responses are triggered, as the first line of defense in the displacement of body temperature [39,40], further emphasized that both core body temperature and skin temperature contribute equally toward thermal comfort, whereas core temperature predominates in the regulation of the autonomic and metabolic responses.

Pleasantness is not merely a sensation. For physical stimuli to be perceived as pleasant, multiple specialized pleasure-generating neural circuits need to be activated to elicit “liking” reactions [41]⁠. Brain mechanisms involved in positive affective evaluation and feelings of pleasantness are most commonly reported in the basal ganglia, particularly in the nucleus accumbens (Nac) and ventral pallidum [38,42,43]⁠ as well as in the orbitofrontal region. Basal ganglia have been specifically indicated in reward prediction important in reinforcement learning models [44]⁠. The OFC has been involved in learning about rewards [45]⁠ as well as representation of rewards [46]⁠ e.g. pleasantness related to listening to music or enjoying a favorite food is experienced differently and there are different orbitofrontal areas activated in response to different stimuli, e.g. pleasant touch is activating right hemisphere of the OFC, the pleasant taste is more prominent laterally, and the olfactory area is moderately activated far lateral area of the OFC [40]⁠.

Neuroimaging studies show that there appears to be an overlap between brain activations evoked by different sensory stimuli. These include the OFC [47]⁠, ACC [48]⁠ and IC [49]⁠ as well as subcortical structures nucleus accumbens (Nac) [50]⁠, ventral pallidum [51]⁠, amygdala [52]⁠ ,and mesolimbic tegmentum [53]⁠. These perceptual overlaps suggest the possibility that the same hedonic-generating circuit, embedded in larger mesocorticolimbic systems, could give a pleasurable activation to all such rewards, even when the final experience of each seems otherwise unique [54]⁠.

Intense subjective pleasantness related to the consumption of chocolate or sweetened drinks is related to brain activity within the mid-anterior zone of the OFC [41]⁠. It has been demonstrated that immediate pleasantness sensations increase endogenous opioid signals recruited broadly in limbic structures [55]⁠. Secondary recruitment of hedonic mechanisms, which are slower in activation, is predominantly related to addictions, where dopamine-related sensitization has enhanced circuit reactivity [56]⁠.

In contrast, primary and sensory somatosensory cortices are activated more by neutral touch than by affective touch. However, the rostral part of the cingulate cortex is activated exclusively by pleasant touch. Rolls provided evidence for separate brain systems related to the processing of pleasantness, unpleasantness, and intensity [57]. Brain activation elicited almost exclusively by the intensity of temperature applied to the skin occurs in the ventral posterior IC and somatosensory cortex; however, activity in those regions is not synchronized with changes in the hedonic value of stimuli [57]⁠. The findings on these orbitofrontal and pregenual cingulate regions are consistent with reports of pleasantness related to other stimulus modalities – pleasant touch [58]⁠, pleasant odor [59]⁠, and pleasant taste [60]⁠. Orbitofrontal and pregenual cingulate cortex and ventral striatum are correlated with activations in primary sensory cortex. These findings on the other hand, show that subjective feelings are also represented in these brain areas, but their results have shown separate brain mechanisms for processing the affective value of stimuli as opposed to their sensory value [61]⁠. Regions of the somatosensory cortex, including SI and part of SII in the mid-insula, were activated more by the neutral touch than by the pleasant and painful stimuli. Part of the posterior insula was activated only in the pain condition and different parts of the brainstem, including the central gray, were activated in the pain, pleasant and neutral touch conditions. The results provide evidence that different areas of the human orbitofrontal cortex are involved in representing both pleasant touch and pain, and that dissociable parts of the cingulate cortex are involved in representing pleasant touch and pain.

A comprehensive review conducted by Kuhn shows that correlations of subjective pleasantness were found in mOFC, ventromedial prefrontal cortex, left ventral striatum, pregenual cortex, right cerebellum, left thalamus, and the midcingulate cortex. Pleasantness reports are related to brain activity in regions described as part of the reward circuitry (mOFC, ventral striatum). Whether subjective pleasantness reports were made during or after scanning revealed no significant differences in brain activation; hence, evaluation of likability or pleasure is an automatic process that is neither elicited nor enhanced by instructions to report the outcome of these judgments.

There is an emerging realization that diverse pleasure sensations share overlapping brain substrates and that there are focal points and separable brain mechanisms for generating the feelings of “liking” and “wanting” for the same reward [62]⁠. The identification of specific activations related to desire and fear could provide a better understanding of normal pleasures and affective wellbeing as well as affective psychopathologies.

Neuroimaging of thermal comfort

There are currently a limited number of studies using fMRI to study brain responses associated with the hedonic evaluation of thermal stimuli in healthy subjects [13], conducted the first fMRI experiment to estimate neural responses related to thermal comfort while applying whole-body cooling. They showed amygdala activation associated with increased discomfort perception as indicated in Figure 1. There was no activation in the thalamus, somatosensory, cingulate, or insula cortices [13]⁠.

Figure 1. (Top) The areas that showed significant covariation of BOLD signals with cold discomfort in every subject, superimposed on the high-resolution MRI of the individual unrelated to the study. In all subjects, only the bilateral amygdala showed a negative correlation with the comfort score. (Bottom) Significant correlation of MR signal change by individual analysis. Normalized MR signal (relative to the global signal which was set to 100) in the right amygdala was plotted against the comfort score. The regression line was y = −2.46x + 144.8, X = 0.63, F(1,68) = 115.7 [P < 0.0001, F test). Adapted from [13]

[63], conducted an fMRI experiment exploring the neural correlates of thermal comfort using a water circulating tube that covered the whole hand. Continuous warm or cold stimuli were applied in an innocuous range from 26°C to 39°C. Participants provided reports of the intensity of thermal stimulation but not its hedonic properties. Activation was observed in the contralateral S2 and bilaterally in the IC in response to both unilateral warm and cold stimulation of the hand. Although previous studies have shown that amygdala activation is associated with cold-induced discomfort, these current findings indicate that separate neural pathways are responsible for hedonic perception and thermal regulation [13, 63]⁠.

In a PET study, Farrell et al., showed as indicated in Figure 2 that whole-body warming and cooling (estimated as pleasant) were associated with posterior mid-cingulate cortex (pMCC) further supporting the role of pMCC in hedonic responses during temperature regulation as well as integration of temperature-rated signals [64]⁠. In a later study Farrell et al., showed that several brain regions including the dorsal cingulate cortex, anterior insula, and midbrain were related to the activity in the preoptic area more strongly during heating than during thermal neutral state. He further suggests that during thermal stress, the preoptic area communicates to several other brain regions with known relevance to the control of autonomic effectors [65].

Figure 2. Thermal comfort activations included the midcingulate cortex (MCC) (y = 16 and −4), inferior frontal gyrus (IFG) (y =16), superior frontal gyrus (SFG) (y = 16), supplementary motor area (SMA) (y = −4), premotor cortices (y = −4), primary somatosensory (Post CG) (y = −20 and −30), and motor cortices (Pre CG) (y =−20), thalamus (y = −20), bilateral superior temporal gyri (STG) (y = −20 and −30) and dentate nucleus of the cerebellum [y = −56). y – Distance from anterior commissure in mm. Positive values are anterior, and negative values are posterior to the anterior commissure. Adapted from “Brain activation associated with ratings of the hedonic component of thermal sensation during whole-body warming and cooling.” By [64]

[66], report pMCC to be more frequently implicated in the hedonic dimension of pain and is a region potentially responsible for the processing of alliesthesia – that is, the subjective response to an external stimulus that reflects the internal homeostasis. Primarily behavioral responses are triggered, known as the first line of defense in the displacement of body temperature. Thermoregulatory behavior is modified by changes in thermal comfort, while skin temperature is playing a vital signaling role along with autonomic heat loss responses [39]. The aversive nature of unpleasant thermosensation can be attenuated by behaviors that facilitate thermoregulatory processes such as seeking shelter or moving away from unpleasant ambient conditions. Conversely, pleasant thermal experiences encourage the persistence of actions that are likely to maintain body temperatures, such as remaining close to a radiant source of heat in a cold environment [67]⁠.

Concluding remarks

There is a gap in the understanding of hedonics related to regional brain activations. Responses to localized, thermal stimuli are yet to be characterized. The distributed nature of neural activations related to temperature responses is consistent with multiple related functions contributing to thermoregulation. Estimating hedonics of thermal stimulation includes a cognitive assessment process that could potentially interfere with identifying activation specific to hedonics. It would be of interest in future research to explore brain representation of temperature sensations with a positively hedonic valence, as well as at the circuitry involved in the integration of environmental, peripheral, and core temperature signals as they could introduce a shift in how we design and operate indoor spaces and the effects they could potentially have on overall livability of spaces.

As mentioned, although thermoregulatory systems have traditionally been conceptualized as serving primarily homeostatic functions, increasing evidence suggests that neural pathways responsible for regulating body temperature may be linked more closely with emotional states than previously recognized [68–75].

Abbreviations

• POA: preoptic area

• pMCC: posterior mid-cingulate cortex

• ACC: anterior cingulate cortex

• fMRI: functional magnetic resonance imaging

• IC: insular cortex

• OFC: orbitofrontal cortex

Disclosure statement

No potential conflict of interest was reported by the author.