Cortical Imaging of Pain in Humans: A Journey from Nociception to Compassion

FIGURE 4-1 Interacting “pain matrices” activated following a noxious somatic stimulus. Distribution of regions of the “full pain matrix” according to their putative functional role in pain perception and memory. Note that although the flux of information tends to be from left to right, continuous interaction among matrices exists, creating dynamic and continuously reconstructing patterns. Although not indicated with further arrows, pain memories at the far right can also change pain perception by affecting both internal affective states (third-order matrix) and attentional modulation (second-order matrix). STT, spinothalamic tract; PIMO, posterior insula–medial operculum; PPC, posterior parietal cortex; ACC, anterior cingulate cortex; MCC, midcingulate cortex; DLPFC, dorsolateral prefrontal cortex; ALPFC, anterolateral prefronta cortex.

1. None of them is a direct main target of the STS.

2. Their direct stimulation does not evoke pain.

3. Their selective destruction does not induce analgesia.

4. They can all be also activated in contexts not involving pain.

5. Their contribution to the PM, from nil to predominant, depends on the context where noxious stimuli are applied.

Two of these second-order areas participate almost constantly to the PM, namely the anterior insula (AI) and the dorsal anterior cingulate (ACC). Mid- and anterior insular activation probably reflects the posterior-to-anterior information flux within this structure [16], supporting the transformation of sensory events into internal feelings and vegetative reactions [9, 61]. The “cognitive section” of the anterior cingulate (rostral to the anterior commissure) is also consistently activated by painful stimuli [23, 59], and together with prefrontal and posterior parietal areas is thought to sustain attentional and evaluative processes of anticipation, learning, and cognitive control. The contribution of these areas to the PM varies enormously with contextual factors, and their activation can be dissociated from actual stimulus intensity (reviews in references [2, 56]). This second-order contextual network receives input from the nociceptive matrix, but can also greatly influence the nociceptive areas via top–down projections. For instance, attending actively to noxious stimuli enhances activity in areas receiving spinothalamic tract (STT) afferents, such as the posterior insula, S1, and midcingulate, whereas distraction tends to suppress such activities ([5, 44, 46, 58]; review reference [63] and see Chapter 14). Such top–down influences modify perception by changing the sensory gain “at the source,” that is, in cortical receiving areas, thalamus, and even at brain stem and spinal cord [53, 57, 64]. Further, vegetative peripheral reactions in viscera driven by anterior insular networks generate new ascending information through splanchnic and vagus nerves [55], thus providing new input to cortical and subcortical nociceptive targets. Hypnotic suggestions can influence the nociceptive, the second-order, or both matrices, depending on the instructions given to the subject (e.g., reexperiencing pleasant autobiographical memories [15], versus up and down modulation of the unpleasantness [48]). A dissociation between preserved activity in sensory cortices (posterior insula) but abated response in second-order parietal and temporal networks has been described under hypnosis (see Chapter 7).

The regions constituent of this second-order matrix can be activated in many contexts other than physical pain. For instance, the AI and ACC pertain to a “salience network” responding to behaviorally relevant stimuli, which they integrate into perceptual decision-making. These two regions are jointly activated not only in response to painful stimuli, but also when subjects are confronted to unpleasant situations such as observing expressions of disgust [61], feeling guilt [51], experiencing social and moral suffering [14, 28], or seeing/imagining other people in pain [25, 52]. This has led to considerable controversy as to whether subjects confronted to such unpleasant situations do physically feel pain [14, 28] and further, how similar the experiences of vicarious pain are to actual experiences of pain (see discussion in Chapter 14). Obviously, feeling pain and representing someone else’s suffering is different, at least in physiological conditions^†: neither the mid-AI nor the rostral ACC are nociceptive sensory areas, and therefore they cannot support by themselves the corporal specificity that characterizes somatic pain, as shown in both stimulation and lesion studies [21, 38]. Their activation in response to unpleasant stimuli appears to reflect their belonging to a broad system for salience detection which may also include the ventral striatum, and which (a) generates internal state modifications, (b) alters the individual responsiveness to specific stimuli, and (c) may contribute in fine to modulate the subjective value of rewards. In this vein, responses in both AI and ACC can be triggered by pleasant stimuli to the same degree as unpleasant ones [50], and direct stimulation of the mid and AI in monkeys was able to elicit not only disgusting, but also socially affiliative behaviors [9].

The conscious perception of the noxious stimulus needs also the contribution of regions in this second-order matrix. Sensory inputs become conscious only if they create distributed brain activation: activity constrained to sensory areas does not produce reportable (i.e., declarative) conscious perceptions, which emerge only when sensory responses are associated to distributed fronto–parieto–temporal activation [11, 12], that is, areas integrating the second-order PM. Functional connectivity within the frontoparietal network appears crucial for declarative consciousness [29], and abundant evidence shows that the contribution of dorsolateral frontal and posterior parietal networks separates undetected versus detected stimulus changes, masked versus unmasked words, extinguished versus seen visual objects, and missed versus reported stimuli during attentional blinks (e.g., [7, 11, 49], and see Chapter 2). The functional coupling between stimulus-specific areas and such higher-order networks is currently considered an essential signature of the access of sensory information to consciousness, making it available to high-level processes, including perceptual categorization, long-term memorization, evaluation, and intentional action.^‡

To sum up, while the nociceptive matrix provides the sensory specificity of the pain experience, the joint activity of the nociceptive and second-order matrices is essential to ensure (a) the modulation of vegetative reactions and internal feelings via anterior insular networks, (b) the attentional modulation of sensory gain by bottom–up/top–down transaction between dorsolateral frontal networks and sensory areas, and (c) the access of nociceptive information to declarative consciousness by activation of a distributed fronto–parieto–temporal network and top–down reentrance in sensory areas.

FROM IMMEDIATE PERCEPTION TO PAIN MEMORIES: THIRD-ORDER NETWORKS

While nociceptive and attentional/perceptual networks may suffice to achieve an immediate conscious perception of noxious stimuli, impressive changes of the pain experience can occur without changes in the matrices described above. For instance, the enhancement of subjective pain during the observation of other people’s suffering [10, 20, 36] can develop in the absence of significant changes in thalamus, insula, operculum, or anterior cingulate, and has been associated instead with activity in high-level polymodal regions outside the “classical” PM, falling mostly within limbic/paralimbic systems, and including the perigenual cingulate, orbitofrontal cortex, temporal pole, and anterolateral prefrontal areas [19, 20]. Similarly, the pain-relieving effects derived from expectations and beliefs, including placebo effects, self–control over the stimulus, or strong religious values [42, 62, 63] have been associated with activity changes in the orbitofrontal, perigenual cingulate, and anterolateral prefrontal cortices, rather than in the classical PM. Orbitofrontal, perigenual, and anterolateral prefrontal activities were also described when the subjective perception of identical noxious stimuli changed from highly unpleasant to moderately pleasant by hedonic context manipulation [31]. A full dissociation between “core PM” activity and subjective pain reports was recently described during meditation-related analgesia, which was associated with enhanced activity in basic PM areas, but decreased activity in high-order prefrontal regions (see Chapter 8).

Only gold members can continue reading. Log In or Register to continue