Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Neuroscience. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 18 April 2024

Pain and Its Modulationfree

Pain and Its Modulationfree

  • Asaf KellerAsaf KellerSchool of Medicine, University of Maryland


Sensory perceptions are inherently subjective, being influenced by factors such as expectation, attention, affect, and past experiences. Nowhere is this more commonly experienced than with the perception of pain, whose perceived intensity and emotional impact can fluctuate rapidly. The perception of pain in response to the same nociceptive signal can also vary substantially between individuals. Pain is not only a sensory experience. It also involves profound affective and cognitive dimensions, reflecting the activation of and interactions among multiple brain regions. The modulation of pain perception by such interactions has been most extensively characterized in the context of the “descending pain modulatory system.” This system includes a variety of pathways that directly or indirectly modulate the activity of neurons in the spinal dorsal horn, the second-order neurons that receive inputs directly from nociceptors. Less understood are the interactions among brain regions that modulate the affective and cognitive aspects of pain perception. Emerging data suggest that certain pain conditions result from dysfunction in pain modulation, suggesting that targeting these dysfunctions might have therapeutic value. Some therapies that are thought to target pain modulation pathways—such as cognitive behavior therapy, mindfulness-based stress reduction, and placebo analgesia—are safer and less expensive than pharmacologic or surgical approaches, further emphasizing the importance of understanding these modulatory mechanisms. Understanding the mechanisms through which pain modulation functions may also illuminate fundamental mechanisms of perception and consciousness.


  • Sensory Systems

Pain goes by when you just don’t think about it.

—Josh A.


Pain ranges in intensity, quality, and duration and has diverse pathophysiologic mechanisms. The perception of pain may reflect activation of nociceptors or their axons by either brief (e.g., pinprick) or lasting (e.g., inflammation) insults. Pain can also arise from insults to the central nervous system (CNS) itself—“central pain”—such as in multiple sclerosis, cerebrovascular accidents, or spinal cord injury. In many cases, pain is perceived in the absence of clear evidence of actual or threatened tissue damage, disease process, or lesion of the somatosensory system. This “nociplastic pain” is thought to be common and is likely involved in many prevalent chronic pain conditions such as low back pain, headache, and fibromyalgia (Raja et al., 2020).

Pain and nociception are different phenomena, and pain cannot be inferred from activity in sensory neurons. Pain is not only a sensory-discriminative experience. The perception of pain integrates information that may arise from activation of nociceptors, with emotional, affective, aversive, and cognitive constructs (Price, 2000). Pain is a learned experience, such that the perception of pain—and pain relief—are affected by expectation. There is growing evidence that the negative affective, cognitive, and psychosocial state of chronic pain is universal in different chronic pain states (Gustin et al., 2011). It is also emerging that therapies targeting the aversive-affective dimensions of pain may be more promising than previous attempts to target the sensory-discriminative dimension (Auvray et al., 2010).

Pain is an individual experience that cannot be reliably quantified and is strongly affected by, for example, social and personal expectations, experiences, affective states, and competing needs (such as hunger, thirst, or sex drive). The neurobiological mechanisms driving the modulation of pain perception are beginning to be revealed. These efforts not only provide insights into mechanisms of perception but can also guide research into harnessing these intrinsic modulator mechanisms to achieve pain relief.

Chronic Pain

The reciprocal influences of the affective and sensory components of pain are relevant to both acute and chronic pain, which are fundamentally and mechanistically different conditions. Acute pain is essential for survival, initiating immediate action by retreating from harm or by suppressing movement to promote healing. Acute nociceptive pain, triggered by nociceptor activation, is a symptom of an underlying medical condition, tends to correlate with the severity of that condition, and ends with the termination of the medical condition (Elman & Borsook, 2016).

Chronic pain, on the other hand, is thought to have no obvious survival value. However, emerging evidence suggests that the plasticity related to chronic pain—usually referred to as “maladaptive plasticity”—has important survival advantages (Crook et al., 2014; Oshima et al., 2016; Price & Dussor, 2014). In general, chronic pain is recognized as pain that persists past the normal time of healing (Bonica, 1991) or pain that persists beyond a particular length of time determined by common medical experience. The transition from acute to chronic pain is difficult to define but is thought to involve the engagement of central nervous system structures (Okubo et al., 2013).

Chronic pain affects over 100 million Americans—more than affected by heart disease, cancer, and diabetes combined. In the United States alone, pain costs up to $650 billion/year in medical treatment and lost productivity (Institute of Medicine Committee on Advancing Pain Research, Care, and Education, 2011). Chronic pain is the most common complaint of patients in outpatient clinics (Upshur et al., 2006). Common chronic pain complaints include headache, low back pain, cancer pain, arthritis pain, and neurogenic pain and can result from a variety of conditions and insults at any level of peripheral and central nervous systems. In most patients, chronic pain starts within weeks or months after the original insult and includes increased pain with noxious stimulation (hyperalgesia) and pain in response to previously innocuous stimuli (allodynia; Apkarian et al., 2009). Perhaps most debilitating is the presence, in nearly all patients, of ongoing or spontaneous pain, which occurs in the absence of a stimulus (Bennett, 2012; Boivie, 2006; Greenspan et al., 2004).

Pain is Multidimensional

Pain not only hurts. It is a multidimensional experience, composed of unpleasant sensory, affective, and cognitive experiences (Auvray et al., 2010; Fields, 1999; Melzack & Casey, 1968; Price, 2000; Treede et al., 2000). Although sensory characteristics of pain are tightly coupled to activation of nociceptors (Gold & Gebhart, 2010), nociceptor activation does not always produce pain, and pain can occur without an identifiable nociceptive input (Lee & Tracey, 2010). Indeed, in a laboratory setting, subjects can readily dissociate the degree of unpleasantness from the perceived intensity of different noxious stimulus modalities (Rainville et al., 1992). A dramatic demonstration of the fact that the affective and sensory components of pain are not only dissociable but are subserved by different neuronal pathway is a report that transecting the corpus callosum eliminates sensation ipsilateral to the stimulus while leaving intact the bilateral unpleasantness evoked by noxious stimuli (Stein et al., 1989).

A Distributed Network for Pain

The dissociable nature of pain components indicates that pain is subserved by a distributed neuronal network that includes parallel somatosensory, limbic, and other components (Melzack, 1999). Perhaps because pain perception is so critical for survival, and perhaps because nociception is one of the earliest neural processes to evolve (Broom, 2001; Walters & Williams, 2019), pain and nociception are associated with a large number of nervous system pathways and centers. For example, nociceptive stimuli can be transduced and relayed not only by specific peripheral afferents but also by “wide-dynamic range” somatosensory fibers. These afferents engage multiple converging and diverging central nervous system (CNS) pathways in the spinal cord, hindbrain, midbrain, and forebrain. In the neocortex, for example, several cortical areas are often referred to as belonging to a so-called pain matrix, a network of areas in which pain is generated from nociception (Ingvar, 1999; but see Legrain et al., 2011). Pain can also be perceived in the absence of activation of peripheral nociceptors, a phenomenon known as central pain syndrome, defined as “pain initiated or caused by a primary lesion or dysfunction in the central nervous system” (Edinger, 1891; Merskey & Bogduk, 1994).

This extensive redundancy in neuronal circuits of nociception and pain implies that modulating the activity in only one of these circuits may regulate pain but will fail to completely abolish pain perception. However, the surprising conclusion of thousands of preclinical studies is that increasing or decreasing neuronal activity in any of these “pain centers” can result in pain relief. That these findings have never translated into effective pain therapies might have to do more with our ability to reliably measure “pain,” or with limitation of preclinical pain models, than with fundamental differences in pain processing in preclinical versus clinical studies (Sadler et al., 2022).

Human and animal studies have shown that the different dimensions of the pain experience may arise from activity in different components of this matrix (Bowsher, 1957). The “lateral system,” including the somatosensory thalamus and cortex, is thought to be involved primarily in the sensory-discriminative dimension of pain (Price et al., 2006). This sensory-discriminative dimension reports the location and intensity of pain. The “medial system”—including the parabrachial complex, amygdala, hypothalamus, mesolimbic structures, medial thalamic nuclei, and the anterior cingulate and the prefrontal cortex—is thought to be involved primarily in the affective-motivational-cognitive dimensions of pain. These relate to feelings of unpleasantness and negative affect, as well as a determination of the appropriate or possible response in a particular situation.

Pain and Negative Affect

Because of the multidimensionality of the pain experience, patients with chronic pain commonly have comorbid emotional disorders and cognitive deficits, and they exhibit alterations in the function of brain networks linked to emotional and cognitive deficits (Borsook, 2012; Bushnell et al., 2013). Therefore, treating pain, especially chronic pain, requires addressing not only the sensory aspects of pain but also its affective and cognitive components since all these influence pain perception. Further, pain perception is strongly modulated by affect, and the converse is also true.

For example, patients who have depression and anxiety or a tendency to catastrophize report more intense pain experiences (Haythornthwaite et al., 1991; Sullivan, Rodgers, & Kirsch, 2001). Similarly, chronic pain shows significant comorbidity with clinical depression (Bair et al., 2008; Bushnell et al., 2013). Most patients with depression report at least one pain complaint, and depression is present in 5%–85% (depending on the study setting) of patients with pain conditions (Bair et al., 2003). Patients with chronic pain often have affective disorders, such as anxiety (Bair et al., 2008; Bushnell et al., 2013), and anxiety is a risk factor for developing chronic pain (Bushnell et al., 2013; Dimova et al., 2013). Several large studies of individuals, including twins, found a greater than chance association between chronic pain conditions and affective disorders, including major depression, panic attacks, and posttraumatic stress disorder, suggestive of a common etiology for these conditions (Kato et al., 2006; Schur et al., 2007). The comorbidity of affective disorders and chronic pain often results in misdiagnosis and undertreatment of depression and similar disorders. As Bair et al. (2003) remind us, more than 75% of patients in primary care settings who have depression present exclusively with physical complaints, and their affective disorders are rarely diagnosed (Kroenke et al., 1997; Simon et al., 1999).

The persistency of chronic pain, with its accompanying negative affective symptoms, may create a self-amplifying stressor, in which pain increases fear (De Peuter et al., 2011), depression (Fishbain et al., 1997), and catastrophizing (Quartana et al., 2009), and these negative affects, in turn, amplify pain perception (Elman & Borsook, 2016). Pain catastrophizing is an important construct, defined as a set of negative emotional and cognitive processes, involving amplification of pain-related symptoms, rumination about pain, feelings of helplessness, and pessimism about pain-related outcomes (Edwards et al., 2006; Sullivan, Rodgers, & Kirsch, 2001; Sullivan, Thorn, et al., 2001).

Pain Can Be Modulated

The different dimensions of pain are dissociable, and they influence each other. Most individuals have experienced the effect of mood swings on their perception of pain or on their pain thresholds. Popular literature refers to a “runner’s high” that provides athletes not only with a sense of euphoria but also a suppression of pain perception. More exotic descriptions of dissociation between affective and sensory pain components include reports of yogi who can consciously modulate their pain perception; this behavioral feat is apparently associated with changes in brain activity (Peper et al., 2006). The phenomenon of initial painlessness described by wounded soldiers is also often described (Beecher, 1946).

The conscious experience of pain represents an interpretation of nociceptive stimuli influenced by memories and by emotional, pathological, and genetic factors (Tracey & Mantyh, 2007). It is also strongly influenced by cognitive factors, including attentional state, emotional context, attitudes, expectations, hypnotic suggestions, or anesthesia-induced changes in consciousness (Bushnell et al., 2013; Fields, 2000; Villemure & Bushnell, 2002). This explains why the perception of pain cannot be predicted from analysis of the nociceptive drive or input. Accordingly, the International Association for the Study of Pain defines pain as “an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage” (Raja et al., 2020).

Following are examples of experiences that can strongly affect pain.

That pain can be perceived in the absence of noxious stimuli can be demonstrated in the thermal grill illusion, originally demonstrated by Thunberg (1896). This illusion of burning heat and pain is created by placing on the skin an interlaced grill of warm and cool bars. The illusion is thought to represent an unmasking phenomenon that reveals the central inhibition of pain by thermosensory integration (Craig & Bushnell, 1994; Shin & Chang, 2021). Increased activity in anterior cingulate cortex (ACC) appears to be selectively associated with the perception of this illusion (Craig et al., 1996). Increasing evidence in humans suggests that the thermal grill illusion shares mechanisms with nociplastic chronic pain syndromes and might represent a clinical marker of central sensitization (Adam et al., 2022; Osumi et al., 2022).

Stress and anxiety can enhance nociception and exacerbate pain perception, a phenomenon referred to as stress-induced hyperalgesia (SIH; Jennings et al., 2014). For example, Traub et al. (2014) demonstrated, in rats, that muscle inflammation followed by stress induces visceral hypersensitivity that persists for months, modeling these human comorbid pain conditions. This SIH phenomenon is accompanied by increased activation of brain regions associated with the affective component of pain (Hubbard et al., 2016). Visceral SIH may involve the endocannabinoid system (Hong et al., 2009). Chronic stress in rodent models results in not only visceral SIH but also thermal and mechanical SIH (Shi et al., 2010), a form of SIH that also involves the endocannabinoid system (Lomazzo et al., 2015).

The converse of SIH is stress-induced analgesia (SIA), a form of adaptive pain suppression, an evolutionarily conserved response to stress that has survival value (Amit & Galina, 1986; Butler & Finn, 2009; Ford & Finn, 2008). SIA may be mediated by both opioid and nonopioid mechanisms, the latter including the endocannabinoid system (Hohmann et al., 2005; Neugebauer, 2015; Valverde et al., 2000). SIA is critically dependent on supraspinal sites, including the periaqueductal gray and the rostroventral medulla, key components of the descending pain pathway (Neugebauer et al., 2009; Woodhams et al., 2017). As discussed below, these regions are thought to mediate the interactions between affect and pain perception.

Thus, stress can evoke both analgesia and hyperalgesia, and both phenomena appear to involve the endocannabinoid system and descending pain modulatory pathways. It is possible that endocannabinoid signaling in key components of the descending pain pathway mediates SIA, while a deficit in endocannabinoid signaling may underlie SIH (Woodhams et al., 2017).

The relationship between stress and pain perception is likely related to the phenomenon of the social transfer of pain. Langford et al. (2010) showed that mice given identical noxious stimuli and tested together display increased pain behaviors, compared to being tested alone or compared with mice that have not received the noxious stimulus. Similarly, mice housed with mice that have peripheral nerve injury exhibit enhanced pain responses to acetic acid (Baptista-de-Souza et al., 2015). This behavior appears to represent SIH, because the cage-mates of the nerve-injured animals displayed anxiety-like behavior on elevated plus maze and open-field tests. Smith et al. (2016) reported that naive, “bystander” mice housed and tested in the same room as mice subjected to inflammatory pain develop corresponding hyperalgesia. This form of social transfer of pain appears to be mediated by olfactory cues and appears to occur without anxiety or SIH (Smith et al., 2016). These phenomena involve anterior cingulate inputs to the nucleus accumbens (Smith et al., 2021). It is likely that social transfer of pain, as a social cue, provides a recognition of another’s pain that can lead to the avoidance of harm or trigger empathy and caregiving behavior (Smith et al., 2016).

Pain perception is also strongly modulated by competing need states, such as hunger, in that individuals must prioritize adaptive responses that protect against dangerous stimuli versus those required for caloric intake. Alhadeff et al. (2018) found that hunger inhibits responses to inflammatory pain in rodents, a phenomenon mediated by inputs to the parabrachial nucleus from the hypothalamus. They also showed that acute pain reduces food seeking and suppresses activity in hypothalamic hunger circuits. Phua et al. (2021) showed that the reciprocal pathway—from parabrachial neurons to the thalamus—suppresses the motivational drive for feeding.

The effects of affect on pain perception can be leveraged to ameliorate pain. For example, Davis et al. (2014) showed that, in depressed women with pain, improving mood facilitates pain recovery. Light therapy, which has been used to control depression (Golden et al., 2005; Moscovici, 2006), can also alleviate pain (at least in animals), possibly by acting through central opioid mechanisms and descending pain modulation (Ibrahim et al., 2017). Thus, interventions that address the negative affect of chronic pain may provide effective pain relief.

Rainville et al. (1997) showed that hypnotic suggestion can affect the perception of the affective component of pain while leaving the perception of the intensity of pain constant. In this condition, imaging reveals changes in the ACC but not in the primary somatosensory cortex, suggesting that these cortical areas are differentially involved in the affective and sensory components of pain, respectively. Related to this, other imaging studies suggest that the regions related to the affective components of pain, but not its sensory-discriminative aspects, are crucial to the empathy for others’ pain (Morrison et al., 2004; Singer et al., 2004).

Particularly promising are noninvasive, low-cost pain treatments that are based on behavioral interventions (Garland et al., 2019). These approaches include mindfulness meditation, which appears to be effective in a number of chronic pain conditions, including migraine headaches (Burrowes et al., 2021; Kabat-Zinn et al., 1985; Seminowicz et al., 2020). Cognitive behavioral therapy (CBT) is a psychosocial intervention focused on reducing symptoms of various conditions. There exists growing evidence that CBT effectively reduces pain burden in a variety of chronic pain conditions (Bonatesta et al., 2022; Cherkin et al., 2016; Darnall, 2021; Hadley & Novitch, 2021; Palermo et al., 2010). Behavioral interventions that improve coping with pain are emerging as particularly powerful tools. For example, pain reprocessing therapy helps patients reconceptualize their pain as due to nondangerous brain activity rather than peripheral tissue injury. Nearly 70% of individuals with chronic back pain were pain free or nearly pain free 1 year after treatment (Ashar et al., 2021). Pain relief was correlated with lasting changes in cortical activation and connectivity. Even a single session of a specialized pain management class, “empowered relief,” results in improvements in pain catastrophizing, intensity, and pain interference that last at least 3 months (Darnall et al., 2021).

Conditioning processes can produce analgesia in a variety of clinical and experimental pain conditions (Fanselow, 1998). Placebo analgesia, the most studied form of the placebo effect, is a prominent example of the phenomenon by which initially innocuous cues can acquire salience to cause a physiologically beneficial effect (Colloca, 2020). Nocebo hyperalgesia is a phenomenon that is opposite to placebo analgesia, whereby expectation of pain increases because of negative psychosocial contexts (Colloca & Barsky, 2020). Progress has been made in uncovering the neurobiological mechanisms of placebo analgesia and, to a much lesser extent, nocebo hyperalgesia (Colloca, 2020; Zunhammer et al., 2021). These mechanisms involve pain-modulating networks described below.

Acupuncture has long been applied to treating disease, and its potential for relieving chronic pain has received considerable attention. The clinical efficacy of acupuncture for a number of chronic pain conditions remains controversial, in part because of study design issues (Liu et al., 2021; Wang, Yin, et al., 2021). The difficulty in dissociating placebo from treatment effects also complicates interpretation of acupuncture studies (Musial, 2019). There is some evidence, from animal models, that acupuncture may relieve pain by acting on descending pain inhibitory systems, as described below (Lv et al., 2019).

Virtual reality is emerging as another tool for reducing pain burden. This refers to a computer technology that provides a distraction method through interaction with computer-simulated entities in a pseudo-natural immersion, by engaging multiple senses, including vision, hearing, and touch. It is used to substantially reduce pain in children with severe burn injuries (Smith et al., 2022; Won et al., 2017).

Darnall and collaborators (Das et al., 2005; Garcia et al., 2021; Khadra et al., 2018; Smith et al., 2022) have shown that an 8-week self (at home) treatment with virtual reality can substantially reduce pain burden in individuals with chronic low back pain. This virtual reality system for chronic pain reduction has recently received approval by the Food and Drug Administration. The mechanisms of this pain relief remain to be determined, but there is evidence that it is associated with reduced pain-related activity in the cortex and thalamus (Hoffman et al., 2007).

Mechanisms of Pain Modulation

That pain perception, and the affective component of pain, can be modulated by such a variety of psychosocial and biological phenomena reinforces the notion that pain processing is distributed throughout the nervous system. This suggests also that anatomical substrates exist for direct communication with and influence on pain-related centers from brain regions involved in a variety of sensory, affective, and cognitive functions. Perhaps the most extensively studied—and most potent—of these interactions comprise the descending pain modulatory system. This somewhat vaguely defined system encompasses inputs from “higher-order” brain areas to “lower-order” brain centers to modulate the activity of neurons that directly process nociceptive inputs. These “descending” inputs synapse on and directly modulate second-order nociceptive neurons in the spinal cord (and medullary dorsal horn). These pathways can even indirectly modulate first-order neurons—the nociceptors that relay inputs from the periphery (or viscera) to the spinal cord (Eccles et al., 1963; Jiménez et al., 1987; Zimmerman et al., 2019).

Growing evidence indicates that chronic pain is related to abnormalities in top-down pain modulatory brain circuits (Heinricher et al., 2009; Lewis et al., 2012; Ossipov et al., 2014; Ren & Dubner, 2002). These descending pathways exert bidirectional control over nociception; imbalance in this circuitry toward facilitation of postsynaptic targets may promote and maintain chronic pain (Ossipov et al., 2014; Vanegas & Schaible, 2004; You et al., 2010). Therefore, engaging these descending systems to suppress pain signals at early stages of neural processing may be a highly effective strategy for treating pain, especially chronic pain. There is increasing evidence that risk for certain chronic pain conditions is determined by individual differences in descending pain modulation (Edwards, 2005). Several chronic pain conditions are associated with reduced descending inhibition, such as fibromyalgia (Lautenbacher & Rollman, 1997) and chronic tension–type headache (Pielsticker et al., 2005). Further, disruptions in descending pain inhibition may predict the development of chronic postsurgical pain (Yarnitsky, 2010). In rodents, there is evidence that the expression of nerve injury pain depends on descending modulation and that engagement of descending inhibition protects in the transition from acute to chronic pain (De Felice et al., 2011).

Periaqueductal Gray/Rostroventral Medulla

The most completely characterized descending pain-modulating circuit is the periaqueductal gray–rostroventral medulla (PAG–RVM) system (Chen & Heinricher, 2019; Dubner & Ren, 1999; Fields, 2000; Heinricher et al., 2009). PAG integrates negative emotions with inputs from sensory, autonomic, neuroendocrine, and immune systems to drive responses to threat (Behbehani, 1995; George et al., 2019). It receives nociceptive inputs directly from the spinal cord and participates in pain modulation through its projections to the RVM, the thalamus and cortex, and the limbic system. The RVM includes the raphe magnus, nucleus reticularis gigantocellularis–pars alpha, and the nucleus paragigantocellularis lateralis (Fields et al., 2006). It is a final common relay in descending modulation of pain, integrating inputs from PAG and other subcortical and cortical structures to the spinal dorsal horn as well as the trigeminal nucleus caudalis (spinal subnucleus caudalis; the medullary dorsal horn; Fields et al., 2006; Heinricher et al., 2009). There is growing evidence that imbalance between facilitatory and suppressive outputs from the RVM to spinal neurons contributes to chronic pain states (reviewed by Denk et al., 2014; Heinricher et al., 2009; Ossipov et al., 2014).

RVM contains several classes of neurons, including ON cells, whose activity increases prior to behavioral responses to noxious stimuli, and OFF cells, whose activity is suppressed before such responses (Fields et al., 1983). Respectively, these two cell classes amplify and suppress nociceptive transmission, such that a shift in the balance between ON and OFF cell population activity can enhance or suppress pain behaviors (Heinricher et al., 2009).

PAG and RVM are also implicated in mediating the affective component of pain perception. For example, stress-induced analgesia is critically dependent on both the PAG and the RVM (Woodhams et al., 2017). In rats, prolonged electrical stimulation of PAG produces lasting and profound increases in measures of negative affect (Wright & Panksepp, 2011). Buhle et al. (2013) showed that two conditions known to elicit strong emotional responses—physical pain and negative image viewing—both enhance negative affect and PAG activity in humans. Human imaging suggests that patients with fibromyalgia have significant disruptions in the functional connectivity of the PAG, particularly with brain regions implicated in negative affect, and that these reductions are associated with worse fibromyalgia impact scores. These findings suggest that the PAG is a site of dysfunction contributing to the clinical manifestations and pain (Coulombe et al., 2017).


A significant proportion of rostroventral medulla neurons—which are thought be neither ON nor OFF cells—express serotonin (5HT), which they release upon dorsal horn neurons. There is conflicting evidence regarding the role of 5HT in chronic pain, with some suggesting that 5HT is pronociceptive and others that it is antinociceptive (Okubo et al., 2013; Ossipov et al., 2014; Suzuki & Dickenson, 2005; Viguier et al., 2013).

Like serotonin, norepinephrine is a monoamine with known involvement in pain modulation (Llorca-Torralba et al., 2016; Pertovaara, 2013). It arises from a number of brainstem and pons nuclei. These include the locus coeruleus that projects upon spinal neurons to directly regulate nociception and forebrain structures, where it is thought to regulate the affective component of pain (Wang, Tobe, et al., 2021). The nucleus of the solitary tract sends dense norepinephrine-containing projections to the parabrachial nucleus, where it is thought to regulate both sensory and affective components of pain (Kawai, 2018; Milner et al., 1986).

The effects of monoamines—both norepinephrine and serotonin—on nociception and pain appear to depend on disease states (Carter & Sullivan, 2002; Llorca-Torralba et al., 2016). They may have only a modest role in maintaining basal pain thresholds in healthy individuals but acquire analgesic functions at the onset of persistent injury. However, in prolonged or chronic pain conditions, this inhibitory control appears to lose efficacy and may even transform to an excitatory, hyperalgesic function (Aby et al., 2022).

Determining if and when monoamines are pro- or antinociceptive is key to developing evidence-based therapeutics for chronic pain, as existing monoamine modulators have had inconsistent clinical efficacies, and some have resulted in significant side effects (Bardin, 2011; Patel & Dickenson, 2016; Tuveson et al., 2011).

Thalamus and Cortex

Thalamocortical and corticothalamic interactions gate, modulate, and process information related to nearly all aspects of sensation, perception, affect, cognition, and motor control (Sherman & Guillery, 2013). It is, therefore, not surprising that thalamocortical-corticothalamic pathways are critically involved in pain–affect interactions. Neuroimaging studies demonstrate that nociceptive inputs almost always results in activation in primary (SI) and secondary (SII) somatosensory cortex, insular cortex, anterior cingulate cortex (ACC), and related thalamic nuclei. It appears that SI is associated with sensory-discriminative aspects of pain, SII has both sensory and affective/cognitive functions, and the insula and ACC are important for affective-motivational and certain cognitive aspects of pain, including anticipation, attention, and evaluation (Neugebauer et al., 2009). The medial prefrontal cortex has important interactions with both the amygdala and descending pain modulatory pathways through which pain and affect can modulate each other. Such interactions have been demonstrated in human imaging studies. These studies have shown, for example, that negative affect, pain, and cognitive control activate overlapping regions in the cingulate cortex (Shackman et al., 2011). Emotional states affect pain unpleasantness, and the magnitude of this effect often correlates with altered pain-evoked ACC activations (Villemure & Bushnell, 2009).

Corticospinal Pathways

Cognitive influences on pain perception are attributed to cortical circuits whose descending outputs modulate information processing at brainstem levels, including via the periaqueductal gray and locus coeruleus described above (Apkarian et al., 2005; Tracey & Mantyh, 2007). Besides these indirect pathways, the neocortex provides dense projections that directly target second-order neurons in the spinal cord and the trigeminal nuclei. Brodal et al. (1956) provided one of the first descriptions of direct projections from cortical areas, in cats, to sensory trigeminal nuclei. Subsequent work in cats showed direct inputs from SI (Dunn & Tolbert, 1982) and SII (Tashiro et al., 1983) to the spinal subnucleus caudalis (SpVc), the target of primary nociceptive afferents from the head and neck (Dubner & Ren, 2004). In rats, direct inputs to SpVc arise from SI, SII, and the insula (Gojyo et al., 2002; Malmierca et al., 2014; Noseda et al., 2010; Wang et al., 2015), and the inputs from SI are somatopically organized (Wise et al., 1979). Efferents from SI and SII in rats diverge to target overlapping regions in SpVc (Smith et al., 2015). SI projects directly to trigeminal nuclei also in mice (Hattox & Nelson, 2007). Some corticotrigeminal axons collateralize in the spinal cord or tectum (Killackey et al., 1989).

That these corticotrigeminal pathways affect sensory processing was demonstrated more than half a century ago (Hernandez-Peon & Hagbarth, 1955). Darian-Smith and Yokota (1966) showed that corticotrigeminal inhibitory influences may occur through both presynaptic and postsynaptic mechanisms. We have demonstrated that corticotrigeminal inputs from SI directly target SpVc to potently suppresses its responses to noxious stimuli and to suppress pain responses (Castro et al., 2017).

Cortical Stimulation

The role of direct and indirect pathways from the neocortex in pain modulation has led to attempts to activate these pathways to alleviate pain. The ability of electrical stimulation of the motor cortex to reduce symptoms of central pain in humans was fortuitously discovered in 1991 (Tsubokawa et al., 1991). Since then, its use has been extended for the treatment of several neuropathic pain conditions, including trigeminal deafferentation pain, postherpetic neuralgia, brachial plexus, and phantom limb pain. Stimulation of other cortical areas, such as SI, can ameliorate pain in animals and humans (Canavero & Bonicalzi, 2002; Canavero et al., 2002, 2003; Lee et al., 2017).

While motor cortex stimulation provides relief to some patients, the overall efficacy of this approach remains relatively low and inconsistent (Canavero & Bonicalzi, 2002; Garcia-Larrea & Peyron, 2007; Lima & Fregni, 2008). Key to improving these approaches is increased knowledge of the most efficacious cortical stimulation sites and the fundamental neural mechanisms that mediate cortical regulation of pain (O’Connell et al., 2014). Some progress has been made in this area. Masri et al. have shown that motor cortex stimulation produces analgesia by activating zona incerta and thereby restoring inhibition in the thalamus (Lucas et al., 2011), as well as by suppressing activity in SI and prefrontal cortex (Jiang et al., 2014). Other studies have implicated activation of other brain areas distant from the stimulation site, synaptic plasticity, and endogenous opioids (Moisset et al., 2016).

The Mesolimbic System

There is a growing realization that pain, particularly its affective facets, may be causally related to impaired reward and motivation functions (Becker et al., 2012; DosSantos et al., 2017; Mitsi & Zachariou, 2016; Navratilova et al., 2016). As reviewed by Elman and Borsook (2016), the notion of unity of negative and positive rewards dates as early as the early Greek and Chinese scholars and physicians and was later refined by Fichte and Hegel. Elman and Borsook also remind us that “Dostoevsky and Nietzsche expanded this concept to the holistic and indivisible pain-pleasure amalgamation, while Spinoza upheld the pain-pleasure continuum by designating them opposite anchors of the perfection scale.” In this context, behaviors that result in pain relief or in the prevention of painful states are rewarding.

These rewarding behaviors appear to depend on dopamine transmission in mesolimbic centers, particularly in the nucleus accumbens (NAc). This nucleus is a key node in the reward circuitry, as it integrates inputs from mesencephalic dopaminergic neurons and from neurons in the ventral hippocampus, amygdala, and frontal cortical areas, all structures that process affective information (Britt et al., 2012; Bromberg-Martin et al., 2010). Anatomical and functional changes in these reward/motivation circuits in chronic pain may lead to the comorbid affective and cognitive disorders observed in these patients (Navratilova & Porreca, 2014; Neugebauer et al., 2009). NAc activity in humans appears to encode predicted value and anticipates its analgesic potential on chronic pain (Baliki et al., 2010). In patients with fibromyalgia, for example, dopaminergic responses to pain appear to be abnormal (Wood et al., 2007): Patients with fibromyalgia experience noxious stimuli as more painful than healthy controls. Control subjects release dopamine in the basal ganglia during the painful stimulation, whereas patients with fibromyalgia do not. Hypersensitivity to pain and high rates of comorbid chronic pain are common in several disorders linked with deficits in dopamine system function, including disorders of mood and affect, substance abuse, and Parkinson disease. In contrast, hyposensitivity to pain is common in patients with schizophrenia, which is linked to excessive dopamine neurotransmission (Jarcho et al., 2012).

Nerve injury, in either rats or mice, selectively increases the excitability of NAc neurons, and this amplified activity appears to be causally related to the injury-induced pain (Ren et al., 2016). These neurons can drive descending pain modulatory pathways and to regulate aversive responses. The activity of dopaminergic neurons is also profoundly affected by painful stimuli, which depress most mesolimbic dopamine neurons (value-coding neurons), and increases activity in a subset of neurons (salience-coding neurons; Brischoux et al., 2009).

In rats with chronic pain, peripheral nerve block results in conditioned place preference (a positive-reward behavior) and evokes dopamine release in the NAc (Navratilova et al., 2012). Similar behavioral and neurochemical events occur after pharmacological pain relief (Xie et al., 2014). The relief of ongoing pain requires opioid signaling in the cingulate cortex and subsequent downstream activation of dopamine activity in the NAc, mediating the reward of pain relief (Navratilova et al., 2015). Indeed, dopamine release in the NAc may emerge as a biomarker of pain relief reward that reflects analgesic efficacy (Xie et al., 2014).

Extensive evidence for an overlap in neuronal circuits subserving both pain perception and reward/motivation strongly support the notion that pain may be related to impaired reward functions (Mitsi & Zachariou, 2016; Potvin et al., 2009). For example, brain areas involved in the reward-aversion neuronal circuitry that are important for decision-making are also implicated in pain processing: They respond to noxious stimuli, and their activation or inhibition modulates the level of perceived pain (Becerra et al., 2001; Ploghaus et al., 2003). Several lines of evidence suggest that chronic pain leads to a hypodopaminergic state that impairs motivated behavior (Taylor et al., 2016). The resulting decreased responsiveness to rewards may be related to the anhedonia and depression common with chronic pain. Thus, strategies to restore dopamine signaling may represent a novel approach to manage the affective sequelae of chronic pain.


The amygdala is one of the most likely sites for interactions between chronic pain and negative affect. It is well established that the amygdala has an important role in emotions and affective disorders (Maren, 2005; Paré et al., 2004; Phelps & LeDoux, 2005). Anatomical, neurochemical, electrophysiological, and behavioral studies support its role in the emotional–affective dimension of pain (Gauriau & Bernard, 2002; Heinricher et al., 2009; Janak & Tye, 2015; Neugebauer, 2015; Neugebauer et al., 2004, 2020; Rhudy et al., 2008). This almond-shaped brain area in the medial temporal lobe is closely associated with cortical and subcortical structures relevant to both pain processing and emotions. The amygdala affects the insular, orbital, and medial prefrontal cortex, basal forebrain nuclei, bed nucleus of the stria terminalis, and medial dorsal thalamus, as well as the hypothalamus and brainstem areas (Bourgeais et al., 2001). The amygdala projects to key structures in the descending modulation of pain (described below), including the periaqueductal gray, parabrachial nucleus, reticular formation, dorsal nucleus of the vagus, solitary tract nucleus, and ventrolateral medulla (Davis et al., 1998; LeDoux, 1998; Price et al., 2003). Carrasquillo et al. (Wilson et al., 2019) have shown that the amygdala can modulate pain bidirectionally—either amplifying or suppressing pain perception—by activating specific neuronal classes in this nucleus.

Amygdala inputs to the medial prefrontal cortex are thought to provide emotion and value-based information to guide decision-making and behavior control (Holland & Gallagher, 2004; Laviolette & Grace, 2006; McGaugh, 2004). A complex network of connections intrinsic to the amygdala regulates the outputs from this structure, thereby modulating emotional responses and pain-related outputs and behaviors (Neugebauer et al., 2004, 2009) and, through interactions with cortical areas, also contributes to cognitive aspects such as pain-related decision-making deficits (Ji et al., 2010). Support for the role of these amygdala-related interactions comes also from findings demonstrating that activation of the amygdala differentiates patients with fibromyalgia with and without major depression (Giesecke et al., 2005).

Thus, the amygdala interacts with brain regions and systems involved in nociception and pain perception, fear and anxiety, attention and cognition, and autonomic function. Neugebauer and collaborators have shown that impaired cortical cognitive control leading to amygdala disinhibition results in the persistence of pain and its affective dimension (Neugebauer, 2015; Woodhams et al., 2017).

Parabrachial Complex

A key structure for encoding the affective component of pain is the parabrachial complex (PB). The PB comprises the lateral and medial parabrachial nuclei, as well as the Kölliker–Fuse nucleus, and plays a prominent role in pain processing (Chiang et al., 2019; Gauriau & Bernard, 2002; Roeder et al., 2016; Uddin et al., 2018). The PB receives dense inputs from lamina I nociceptive spinal neurons, a pain-related projection far denser than the spinothalamic pathway (Polgár et al., 2010; Spike et al., 2003). As a result, PB neurons can respond robustly—and preferentially—to noxious stimuli (Chen et al., 2017; Gauriau & Bernard, 2002; Roeder et al., 2016). The PB projects, often reciprocally, to several regions linked to pain and affect, including periaqueductal gray, rostroventral medulla, thalamus, insula, amygdala, and zona incerta (Bianchi et al., 1998; Roeder et al., 2016). Thus, the PB appears to serve as a key nidus for pain and its affective perception.

The PB is involved also in a variety of other functions relevant to homeostasis and the state of the body, including visceral malaise, taste, temperature, pain, and itch. A view is emerging of the PB as providing a general alarm to various central nervous system (CNS) structures about aversion and threats, both interoceptive and exteroceptive (Chiang et al., 2019; Palmiter, 2018).

PB is reciprocally connected with the amygdala (Neugebauer, 2015; Neugebauer et al., 2020) and the insular cortex (Grady et al., 2020; Huang et al., 2021; Yasui et al., 1989). Nociceptive inputs to the amygdala originate primarily from the PB, whose afferents form giant—presumably highly efficacious—perisomatic synapses in the “nociceptive amygdala” (CeLC) (Delaney et al., 2007). We have shown that the CeLC projects back to the PB, exerting potent inhibition upon PB neurons (Raver et al., 2020). The CeLC integrates nociceptive inputs with aversive inputs (Neugebauer, 2015; Neugebauer et al., 2003).

The PB is critically involved in chronic pain. Bernard and collaborators (Matsumoto et al., 1996) reported that PB neuronal activity is increased in a rat model of arthritic pain. Expression of immediate early genes in the PB increases after chronic constriction injury (CCI) of the sciatic nerve in rats (Jergova et al., 2008). Our findings—in both rats and mice with chronic pain after CCI of the infraorbital nerve (CCI-Pain)—confirm that CCI-Pain is associated with amplified activity of PB neurons (Raver et al., 2020; Uddin et al., 2018).

Sexual Dimorphism in Descending Pain Modulation

The neurobiology of nociception and pain is remarkably different in females and males. As a result, women are more sensitive to pain and less tolerant of pain than men (Mogil, 2012). Women are also much more likely to develop chronic pain as a result of a variety of conditions (Berkley, 1997; Greenspan et al., 2007). Similarly, nearly all preclinical studies report that female rodents display quantitative sex differences in acute and chronic pain sensitivity and behaviors (Mogil, 2020).

Differences in sex-related behaviors and prevalence involve genetic, epigenetic, physiological, and anatomical dimorphism that affects several peripheral and central circuits. These include descending pain modulatory pathways that exhibit sexual dimorphism (Fullerton et al., 2018). For example, the periaqueductal gray of female rodents has more rostroventral medulla–projecting neurons than in males, but more of these neurons are activated in males in persistent pain (Loyd & Murphy, 2006; Loyd et al., 2007). These neurons are more strongly disinhibited by systemic morphine in male rats (Loyd et al., 2007), likely explaining why most studies, in both humans and rodents, report that morphine administration produces a greater degree of analgesia in males (Loyd & Murphy, 2009).

The differences in descending pain modulation may explain why women exhibit greater adaptation and habituation to painful stimuli (Hashmi & Davis, 2009). These differences may relate to the greater activation of the salience network in human males, supporting greater sustained attention to pain and prevention of pain habituation (Wang et al., 2014). In contrast, in females, there is greater engagement of the descending modulation system mediating pain habituation (Wang et al., 2014).

Studies in humans report conflicting results on metrics of descending pain modulation. Conditioned pain modulation (CPM) describes a psychophysical paradigm in which central pain inhibition is tested by means of “pain inhibits pain.” It tests the ability of a conditioning painful stimulus to inhibit a different, subsequent painful stimulus. (In animal models, this paradigm is referred to as diffuse noxious inhibitory control, or DNIC. Some use the term DNIC in human studies, too.) A literature review of studies examining sex differences concluded that fewer than half of these studies found that males showed greater CPM than females (Hermans et al., 2016; van Wijk & Veldhuijzen, 2010). Interestingly, women show more stable and reproducible CPM scores between tests, such that CPM is recommended for bedside evaluation of chronic pain only in women (Martel et al., 2013).