Neural correlates of the prolonged salience of painful stimulation

Abstract

Pain is a unique class of sensory experience from the perspective of salience. Nonpainful somatosensory stimuli usually require behavioral relevance or voluntary attention to maintain salience. In contrast, painful stimuli tend to have sustained salience even without explicit behavioral relevance or voluntary attention. We have previously identified a frontal–parietal–cingulate network of regions responding transiently to nonpainful sensory events. This network is sensitive to the task relevance and novelty of sensory events and likely represents the salience of events in the sensory environment. Since pain can remain salient for a prolonged period, we hypothesized that this network should show transient responses to the onset or offset of a nonpainful stimulus, but sustained responses throughout the duration of a painful stimulus. To test this hypothesis, we used functional MRI to examine the response of these regions to sustained (60-s) periods of painful and nonpainful transcutaneous electrical nerve stimulation. As predicted, the temporoparietal, inferior frontal, and anterior cingulate cortex showed only transient responses to the onset or offset of nonpainful stimulation, but a sustained response throughout the duration of painful stimulation. These regions therefore show tonic responses to stimuli with tonic salience, supporting a general role for these areas in representing stimulus salience. The thalamus and putamen also responded tonically throughout painful but not nonpainful stimulation. Previous studies have implicated the basal ganglia in supporting voluntary sustained attention. Our findings suggest that the basal ganglia may play a more general role in supporting sustained salience, whether through voluntary or involuntary mechanisms.

Introduction

Pain is a unique class of sensation from an attentional perspective. Attentional control is often subdivided into a voluntary, endogenous mechanism and an involuntary, exogenous mechanism Posner et al., 1980, Jonides and Irwin, 1981, Desimone and Duncan, 1995, Frith, 2001. Nonpainful stimuli can engage involuntary attention through sensory-driven mechanisms when they appear, disappear, or change Muller and Rabbitt, 1989, Swick and Knight, 1998, Downar et al., 2000. However, involuntary attention is usually only transient. To capture attention tonically, stimuli must usually engage voluntary attention Fox and Raichle, 1984, Muller and Rabbitt, 1989, Coull, 1998, Nobre, 2001. Painful stimuli are unusual in that they can engage involuntary attention over a sustained period. Even when not involved in current behavior, pain demands and holds attention Eccleston and Crombez, 1999, Casey, 2000. The unique attentional properties of pain raise the question of how painful stimuli maintain their prolonged salience.

In neuroimaging studies in humans, painful stimuli reliably activate the primary somatosensory cortex (S1), secondary somatosensory cortex (S2), anterior cingulate cortex (ACC), and anterior insula. Pain-related activations have also been observed in the thalamus, basal ganglia, prefrontal, posterior parietal (BA 5/7), and inferior parietal (BA 39/40) cortices, striatum, cerebellum, periaqueductal gray, and supplementary motor area (SMA) (for reviews see Peyron et al., 2000, Davis, 2000a. Psychophysical evidence demonstrates that attention modulates the perceived intensity of a painful stimulus (Willer et al., 1979). Attention also modulates the intensity of pain Bushnell et al., 1985, Hodes et al., 1990, Good et al., 1999, thermal pain-evoked potentials (Siedenberg and Treede, 1996), and the intensity of S1 activation in neuroimaging studies (Bushnell et al., 1999).

While a growing number of studies have examined the effects of attention on pain and its neural correlates, few studies have addressed the converse issue of how pain engages attention and its neural correlates. Lesion and neuroimaging studies employing nonpainful stimuli have linked attentional control to a network of parietal, frontal, and cingulate cortical regions Corbetta et al., 1991, Melzack, 1993, Pardo et al., 1991, Nobre et al., 1997, Kastner and Ungerleider, 2000, thought to contribute to sensory, motor, and motivational aspects of attentional control, respectively Mesulam, 1981, Mesulam, 1999. Some studies have found similar patterns of cortical activation for both voluntary and involuntary attentional control mechanisms Gitelman et al., 1999, Kim et al., 1999. However, converging evidence from neuroimaging studies Corbetta et al., 1991, Melzack, 1993, Koski et al., 1999, Rosen et al., 1999, Hopfinger et al., 2000, Coull et al., 2000, lesion studies Alivisatos and Milner, 1989, Koski et al., 1998, and single-unit recordings (Schultz et al., 1995a) indicates that voluntary, sustained attention to nonpainful stimuli may rely on sustained activity in the prefrontal cortex and basal ganglia. This evidence raises the question of whether these areas also support the sustained salience of painful stimuli.

We have previously identified a right-lateralized frontal–parietal–cingulate–insular set of regions responding transiently to changes in visual, auditory, or tactile stimuli (Downar et al., 2000). The amplitude of the response is sensitive to factors influencing stimulus salience, such as behavioral relevance (Downar et al., 2001) and novelty (Downar et al., 2002). These regions have been proposed to play a general role in representing stimulus salience Knight et al., 1995, Downar et al., 2002. If so, they should show transient responses to the onset or offset of a nonpainful stimulus (which has transient salience during changes). We have previously observed such responses in these areas for tactile stimuli undergoing sudden changes (Downar et al., 2000). We also hypothesize that these areas should show sustained responses throughout the duration of a painful stimulus, due to the sustained motivational (i.e., aversive) salience of painful stimuli throughout their presence. To test these hypotheses, we used functional magnetic resonance imaging (fMRI) to examine the cortical response of the previously identified salience-sensitive regions to sustained painful and nonpainful transcutaneous electrical nerve stimulation (TENS). The overall approach to salience used here is analogous to a similar approach used successfully elsewhere to provide evidence for the representation of visual salience in parietal regions, in terms of response to stimulus onset, novelty, and behavioral/motivational relevance to a task or reward Gottlieb et al., 1998, Kusunoki et al., 2000. In conjunction with previous studies Downar et al., 2000, Frith, 2001, Ploner et al., 2002, the present study aims to show that the response of a frontal–parietal–cingulate–insular network is sensitive to a variety of influences, including stimulus change, novelty, task relevance, and (in the present study) motivational value and that the diverse functions of these regions may best be characterized overall as representing stimulus salience.