The effect of stimulus number on the stability of responses for an extensive heat pain test.

Abstract

Background: The use of relatively lower stimulus presentation numbers in quantitative sensory testing may influence the computation accuracy of participants' discriminability. The minimum trial number for obtaining a stabilized participant discrimination ability was determined.

Materials and methods: Twelve participants' ability to discriminate between noxious heat stimuli pairs (45°C/46°C, 46°C/47°C, and 47°C/48°C) was assessed using a six-category confidence rating scale. Heat stimuli were administered to the forearm. Two conditions with presentation numbers of 17 trials per stimulus (representing the median number of trials in previous studies) and 40 trials per stimulus (used in a previous study with a similar protocol) were used.

Results and discussion: Participants' discriminability stabilized at approximately the 20th trial based on the lowest frequency of indeterminate and non-model conforming results under both conditions. A simple linear regression model showed a statistically significant positive relationship between discriminability for the two conditions (slope = 0.65, p < 0.001; constant = 0.33, p = 0.02; r(2 )= 0.51). As a rule of thumb, approximately 20 trials per stimulus intensity could be used to obtain a stabilized discriminability outcome.