Learning pain-related fear: neural mechanisms mediating rapid differential conditioning, extinction and reinstatement processes in human visceral pain.

BACKGROUND AND AIMS

There exists converging evidence to support a role of pain-related fear in the pathophysiology and treatment of chronic pain conditions. Pain-related fear is shaped by associative learning and memory processes, which remain poorly characterized especially in the context of abdominal pain such as in irritable bowel syndrome (IBS). Therefore, using event-related functional magnetic resonance imaging (fMRI), we assessed the neural mechanisms mediating the formation, extinction and reinstatement of abdominal pain-related fear in healthy humans. Employing painful rectal distensions as clinically-relevant unconditioned stimuli (US), in this fear conditioning study we tested if differential excitatory and inhibitory learning is evocable after very few CS-US learning trials ("rapid conditioning"), and explored the underlying neural substrates of these learning and memory processes.

METHODS

In N=24 healthy men and women, "rapid" fear acquisition was accomplished by pairing visual conditioned stimuli (CS(+)) with painful rectal distensions as unconditioned stimuli (US), while different visual stimuli (CS(-)) were presented without US (differential delay conditioning with five CS(+) and five CS(-) presentations and a 80% reinforcement ratio). During extinction, all CS were presented without US. Subsequently, a reinstatement procedure was implemented, defined as the retrieval of an extinguished memory after unexpected and unpaired exposure to the US, followed by CS presentations. For each phase, changes in perceived CS-US contingency and CS unpleasantness were assessed with visual analogue scales and compared with analyses of variance. fMRI data were analyzed using whole-brain analyses (at p<.001 uncorrected) and in regions-of-interest analyses with familywise error correction of alpha (pFWE<.05). Differential neural activation in response to the CS during each experimental phase (i.e., CS(+)>CS(-); CS(+)<CS(-)) was analyzed without and subsequently also with a linear parametric modulation including trial number as a regressor.

RESULTS

A significant valence change (i.e. increased CS(+) unpleasantness) was observed following acquisition, indicating successful differential aversive learning. On the other hand, CS-US contingency awareness was not fully established. These behavioral results were paralleled by differential activation of the putamen (pFWE<.05), insula (pFWE<.05) and secondary somatosensory cortex (S2, p<.001 uncorrected) in response to the CS(+) during acquisition. The same analysis with a linear parametric modulation confirmed but also strengthened the resulting activations, which were all highly significant in ROI analyses at pFWE<.05. Extinction and reinstatement involved differential activation in response to the CS(-), involving the cingulate cortex and primary motor cortex (M1) during extinction and the posterior cingulate cortex (PCC) during reinstatement (all p<.001 uncorrected), without obvious effects upon linear parametric modulation analysis.

CONCLUSIONS

Abdominal pain stimuli are effective US that elicit conditioned pain-related fear even after very few learning experiences without full contingency awareness. These findings extend similar evidence of "rapid learning" in response to interoceptive US (e.g., conditioned taste aversion, conditioned nausea), and have implications for the pathophysiology and treatment of chronic abdominal pain such as in IBS.