Key points are not available for this paper at this time.
The assessment of pain and analgesic efficacy sometimes relies on the retrospective evaluation of pain felt in the immediate, recent or distant past, yet we have a very limited understanding of the processes involved in the encoding, maintenance and intentional retrieval of pain. We examine the properties of the short-term memory of thermal and pain sensation intensity with a delayed-discrimination task using pairs of heat pain, warm and cool stimulation in healthy volunteers. Performance decreased as a function of the inter-stimulus interval (ISI), indicating a robust deterioration of sensory information over the test period of 4-14 s. As expected, performance also decreased with smaller temperature differences (Delta-T) and shorter stimulus durations (6-2 s). The relation between performance and Delta-T was adequately described by a power function, the exponent of which increased linearly with longer ISI. Importantly, performance declined steadily with increasing ISI (from 6 to 14 s)--but only for pairs of heat pain stimuli that were relatively difficult to discriminate (Delta-T < or = 1.0 degree C; perceptual difference < or = 32/100 pain rating units) while no deterioration in performance was observed for the largest temperature difference tested (Delta T = 1.5 degrees C; perceptual difference of 50 units). These results are consistent with the possibility that short-term memory for pain and temperature sensation intensity relies on a transient analog representation that is quickly degraded and transformed into a more resistant but less precise categorical format. This implies that retrospective pain ratings obtained even after very short delays may be rather inaccurate but relatively reliable.
Rainville et al. (Wed,) studied this question.