Elsevier

NeuroImage

Volume 32, Issue 4, 1 October 2006, Pages 1804-1814
NeuroImage

Predictability modulates the affective and sensory-discriminative neural processing of pain

https://doi.org/10.1016/j.neuroimage.2006.05.027Get rights and content

Abstract

Knowing what is going to happen next, that is, the capacity to predict upcoming events, modulates the extent to which aversive stimuli induce stress and anxiety. We explored this issue by manipulating the temporal predictability of aversive events by means of a visual cue, which was either correlated or uncorrelated with pain stimuli (electric shocks). Subjects reported lower levels of anxiety, negative valence and pain intensity when shocks were predictable. In addition to attenuate focus on danger, predictability allows for correct temporal estimation of, and selective attention to, the sensory input. With functional magnetic resonance imaging, we found that predictability was related to enhanced activity in relevant sensory-discriminative processing areas, such as the primary and secondary sensory cortex and posterior insula. In contrast, the unpredictable more aversive context was correlated to brain activity in the anterior insula and the orbitofrontal cortex, areas associated with affective pain processing. This context also prompted increased activity in the posterior parietal cortex and lateral prefrontal cortex that we attribute to enhanced alertness and sustained attention during unpredictability.

Introduction

A sensation of pain typically causes emotional distress. Pain signals a threat to bodily integrity and can be analyzed in framework of homeostatic mechanisms. It reflects one of several afferent modes (e.g., itch, tickle, sensual touch, vasomotor flush, hunger, thirst) that convey information about the state of the body to the brain, and this information might provide important input to the neural systems supporting emotion, feelings, and self-awareness (Craig, 2002). In the present study, we explore the interplay between pain, affect, and attentional processes in the context of pain stimuli that are predictable or unpredictable.

The nature of pain is reflected in both sensory-discriminative and affective components (Melzack and Casey, 1968). The sensory-discriminative system includes thalamic nuclei that mediate afferent information to the primary and secondary sensory cortices (SI and SII, respectively) and the posterior insula, which provides an interoceptive context as well as an interface to the affective system (Craig, 2002). The latter includes the anterior insula as a central structure and in addition the anterior cingulate and the orbitofrontal cortices (ACC and OFC, respectively) (Craig, 2002, Rainville, 2002). Singer et al. (2004) reported an interesting dissociation between these systems. Whereas both the sensory-discriminative and the affective systems were activated in a participant receiving actual pain stimulation, the affective, but not the sensory-discriminative, system was empathically activated in the participant observing a signal indicating that an emotionally close partner was receiving the pain stimulus.

Several lines of evidence suggest that interoceptive systems and the anterior insula play a central role in the affective system. First, the anterior insula activity is more closely correlated with subjective magnitude ratings of temperature than with physical increases in temperature (Craig, 2002). Second, visceral stimulation through inflation of a balloon in the esophagus results in insular activation (Aziz et al., 2000). This insular activity, as well as reported discomfort, is enhanced in a negative emotional context (Phillips et al., 2003b). Third, the accuracy in judgment of heartbeat timing is positively correlated with insular activity as well as with indices of negative emotion, providing support for its role in interoceptive awareness (Critchley et al., 2004). Fourth, perception and mapping of bodily states have since James (1894) and Lange (1922) been regarded as central for emotional experience (Mesulam and Mufson, 1985, Damasio et al., 2000, Craig, 2002). Accordingly, the anterior insula has consistently been implicated in studies with emotional manipulations (e.g., Phillips et al., 2003a, Carlsson et al., 2004) and specifically associated with the emotional component of pain (Singer et al., 2004).

The orbitofrontal cortex (OFC) has also been given a central position in affective pain processing. This region has a wide range of functions such as representing the value of reinforcing and punishing stimuli (Rolls, 2004). The OFC also responds to breaches in cue–target associations (Nobre et al., 1999), to the variability in response-reinforcement contingencies (Elliott et al., 2000) as well as to aversion prediction error signaling (Seymour et al., 2005). The ventral striatum, which is anatomically and functionally linked to the OFC (Selemon and Goldman-Rakic, 1985), has been suggested to support prediction-error processing, that is, violations of reward expectations, as well as attention to temporal structure (Schultz, 2002, Coull et al., 2004).

The context in which the pain is experienced has previously been shown to influence pain processing (Rainville et al., 1997, Petrovic et al., 2002). For example, subjects anticipating painful stimuli of high intensity report higher anxiety levels in anticipation of the stimulus, and higher pain intensity ratings at stimulus impact, compared to when they anticipate a stimulus of low intensity, even if the intensity in fact was constant (Ploghaus et al., 2001). Similarly, the threat of an electric stimulus increases the anxiety levels as well as the pain reactivity. In contrast to the threat of an electric shock, the exposure to three consecutive electric stimuli has the opposite effect: a decrease in pain reactivity is observed (Rhudy and Meagher, 2000). This reaction to immediate aversive stimuli can be characterized as a fear response, which mobilizes the organism to prepare and, if possible, take rapid action to remove the pain (Öhman, 2000b). Indeed, fear may be viewed as a homeostatic “error signal” conveying a threat to bodily integrity that motivates restoring the homeostatic balance by escaping from or inhibiting the pain stimulus. Thus, fear mobilizes the organism to take action here and now, and it has been suggested that fear and pain can be regarded as mutually inhibiting states, of which fear has priority when it comes to promoting fight and flight (Bolles and Fanselow, 1980). Anxiety, on the other hand, can be characterized as a future-oriented aversive state in the face of an uncertain and potentially uncontrollable threat (Öhman, 2000a). In terms of the homeostatic model, anxiety will prevail when pain reflects a long-term threat to homeostatic balance, probably reflected in salient interoceptive activation. Therefore, anxiety will be the dominating emotion when pain stimulation is uncontrollable and protracted. In such situations, it may be more appropriate to increase vigilance, environmental scanning and arousal, typical reactions observed in conditions of anxiety, with a consequent increase in pain reactivity (Bolles and Fanselow, 1980).

Behavioral control is a primary modulator of the impact of nociceptive input because it provides an avenue for impacting on the expected consequences of the aversive input (Maier and Watkins, 1998). Accordingly, numerous studies have demonstrated that control reduces the subjective experience of stress as well as the corresponding ratings of pain (Miller, 1980). Whereas control typically refers to the capability to do something in order to change a situation, the ability to predict implies knowledge about the relationship between events in a given environment1. Similar to control, prediction can modulate the extent to which aversive stimuli induce stress and anxiety (Miller, 1981). Predictability offers the possibility to develop a model of an aversive situation that specifies what can (and cannot) be expected. As a consequence there is potentially less surprise and focus on danger in an aversive situation, which in turn reduces anxiety as well as anticipatory and impact arousal (Berlyne et al., 1960).

Predictable pain stimuli entail a Pavlovian fear conditioning contingency in which the predictive cue (whatever its nature) serves as a conditioned stimulus and the pain stimulus as an unconditioned stimulus. As a result of this contingency, conditioned fear to the cue should be expected, because it is regularly followed by the painful unconditioned stimulus. In a functional perspective (Domjan, 2005) this conditioned fear response will serve to modulate the impact of the pain stimulus, through active avoidance if possible, or though the inhibitory effect of fear on pain (cf., Bolles and Fanselow, 1980) if the situation precludes active responses. Furthermore, the flip side of providing information about when the pain is due, is that the absence of the predictive cue denotes safety (Seligman, 1968). In an unpredictable situation, on the other hand, there are no designated safe periods, if the temporal distribution of pain stimuli is programmed to be random. With such a procedure, conditioning to the context, making the whole situation aversive, should be expected. In such an aversive situation, which lacks specific fear-inducing cues and means to influence the pain stimulus, the resulting emotional state is better described as anxiety than fear (e.g., Öhman, 2000a). Moreover, the emotional states generated in the predictable and unpredictable situations might have different brain substrates.

Predictable and unpredictable situations engender different psychological processes in other respects than conditioning. For example, the conditions for attention deployment are different with predictable and unpredictable stimuli. In the predictable case, selective attention can be precisely directed to the external stimuli, which is likely to result in increased processing in the brain regions that process the specific sensory input to which attention is directed (Rees et al., 1997, Carlsson et al., 2000). In the unpredictable conditions, on the other hand, stimulation cannot be anticipated, and hence attention cannot be selectively timed to the stimuli to the same degree as when they are predictable. However, as generating anxiety, unpredictable conditions are likely to induce a bias for detecting threat (e.g., Mogg and Bradley, 1998), which might activate brain circuits for sustained attention (Pardo et al., 1991).

Predictability can be differentiated in at least two types. The first involves knowing the conditions under which the event will occur (contingency predictability), and the second knowing what the event will be like (what-kind-of-event-predictability) (Miller, 1981). In the present study, the subjects were well acquainted with the nature and location of the stimuli through a procedure of choosing appropriate intensity levels. We designed experiments in which we varied the intensity and the temporal predictability of brief electric shocks. In a pilot experiment, the electrical shocks were presented with a regular rhythm or randomly. In the main experiment of the present study, shocks were preceded by a visual cue in the predictable conditions, and in the unpredictable conditions the cue and the shocks were randomly related. In other words, we manipulated the degree of contingency predictability of a somatosensory stimulus of high or low intensity.

The subjects received the same amount of nociceptive input in the predictable and unpredictable conditions, with the difference that they either could or could not predict when to expect the shocks. The primary aim of this functional magnetic resonance imaging (fMRI) study was to delineate functional neuroanatomical systems related to sensory-discriminative and affective aspects of pain as modulated by predictability and related contexts of anxiety, fear conditioning and attention. For the majority of the subjects, we expected the unpredictability of aversive events to result in an increase in subjective anxiety experience and negative valence with corresponding activity in affective neural processing areas such as the anterior insula and the OFC. We hypothesized predictability, on the other hand, to allow for direction of selective attention and to correlate with increased activity in areas supporting sensory-discriminatory processes, such as the SI, SII and the posterior insula.

Section snippets

Participants

Ten healthy subjects (3 F/7 M, right handed, non-smoking, mean age 24 years, range 21–34 years) participated in the pilot study. In the main experiment, thirteen healthy subjects participated. Four of these subjects made ratings that were inconsistent between series and were excluded from further analysis. The remaining nine subjects (5 F/4 M, right handed, non-smoking, mean age 25 years, range 21–32 years) were included in the statistical analysis of the behavioral and functional imaging data.

Behavioral results

In the pilot experiment, where predictability was based on rhythmicity, the ten subjects rated anxiety, F(1,9) = 24.84, P < 0.001, negative valence, F(1,9) = 22.19, P < 0.01 and pain intensity, F(1,9) = 9.88, P < 0.05 higher for the unpredictable, randomly presented shocks. The interactions between predictability (RHY/RA) and level (HI/LO) in anxiety F(1,9) = 49.43, P < 0.0001, negative valence F(1,9) = 28.79, P < 0.001 and intensity F(1,9) = 5.70 P < 0.05 indicated significantly higher effects

Discussion

This study was designed to examine cerebral mechanisms underlying the experience of aversive stimuli as altered by predictability. Compared to a condition in which pain stimuli were predictable by means of a visual cue, unpredictable pain stimuli resulted in increased ratings of anxiety and negative valence. The corresponding comparison of the fMRI data showed activations of structures belonging to the affective pain network, including the anterior insula and the lateral OFC. The unpredictable

Conclusions

We manipulated the stimulus presentation context in terms of predictability of aversive somatosensory events while keeping the input constant. The unpredictable situation induced more anxiety and correlated increases in the anterior insula, a region which has been suggested to provide an interface between interoceptive states and the representation of these states as feelings. This condition also increased the activity in the right fronto-parietal network, which we attribute to enhanced

Acknowledgment

This study was supported by grants from The Swedish Research Council (2003-5810), The family Hedlund Foundation and Karolinska Institutet. The project was finished in the context of Stockholm Brain Institute.

References (70)

  • K. Mogg et al.

    A cognitive-motivational analysis of anxiety

    Behav. Res. Ther.

    (1998)
  • J.S. Morris et al.

    Parallel neural responses in amygdala subregions and sensory cortex during implicit fear conditioning

    Neuroimage

    (2001)
  • M.L. Phillips et al.

    Neurobiology of emotion perception I: The neural basis of normal emotion perception

    Biol. Psychiatry

    (2003)
  • P. Rainville

    Brain mechanisms of pain affect and pain modulation

    Curr. Opin. Neurobiol.

    (2002)
  • J.L. Rhudy et al.

    Fear and anxiety: Divergent effects on human pain thresholds

    Pain

    (2000)
  • E.T. Rolls

    The functions of the orbitofrontal cortex

    Brain Cogn.

    (2004)
  • W. Schultz

    Getting formal with dopamine and reward

    Neuron

    (2002)
  • D.L. Walker et al.

    The role of the bed nucleus of the stria terminalis versus the amygdala in fear, stress and anxiety

    Eur. J. Pharmacol.

    (2003)
  • J. Ashburner et al.

    Nonlinear spatial normalization using basis functions

    Hum. Brain Mapp.

    (1999)
  • Q. Aziz et al.

    Functional neuroimaging of visceral sensation

    J. Clin. Neurophysiol.

    (2000)
  • D.E. Berlyne

    Conflict, Arousal and Curiosity

    (1960)
  • G.S. Berns et al.

    Brain regions responsive to novelty in the absence of awareness

    Science

    (1997)
  • R.C. Bolles et al.

    A perceptual-defensive-recuperative model of fear and pain

    Behav. Brain Sci.

    (1980)
  • C. Buchel et al.

    Dissociable neural responses related to pain intensity, stimulus intensity, and stimulus awareness within the anterior cingulate cortex: a parametric single-trial laser functional magnetic resonance imaging study

    J. Neurosci.

    (2002)
  • K. Carlsson et al.

    Tickling expectations: neural processing in anticipation of a sensory stimulus

    J. Cogn. Neurosci.

    (2000)
  • K. Carlsson et al.

    Fear and the Amygdala: Manipulation of Awareness Generates Differential Cerebral Responses to Phobic and Fear-Relevant (but Nonfeared) Stimuli

    Emotion

    (2004)
  • J.T. Coull et al.

    Functional anatomy of the attentional modulation of time estimation

    Science

    (2004)
  • A.D. Craig

    How do you feel? Interoception: the sense of the physiological condition of the body

    Nat. Rev., Neurosci.

    (2002)
  • A.D. Craig

    A new view of pain as a homeostatic emotion

    Trends Neurosci.

    (2003)
  • H.D. Critchley et al.

    Neural systems supporting interoceptive awareness.[see comment]

    Nat. Neurosci.

    (2004)
  • A.R. Damasio et al.

    Subcortical and cortical brain activity during the feeling of self-generated emotions

    Nat. Neurosci.

    (2000)
  • M. Devor et al.

    Patophysiology of damaged nerves in relation to chronic pain

  • M. Domjan

    Pavlovian conditioning: a functional perspective

    Annu. Rev. Psychol.

    (2005)
  • R. Elliott et al.

    Dissociable functions in the medial and lateral orbitofrontal cortex: evidence from human neuroimaging studies

    Cereb. Cortex

    (2000)
  • M.S. Fanselow et al.

    Conditional analgesia, defensive freezing, and benzodiazepines

    Behav. Neurosci.

    (1988)
  • Cited by (0)

    View full text