Variance of sensory threshold measurements: Discrimination of feigners from trustworthy performers

Abstract

Sensory threshold measurements are criticized as subjective and therefore not to be relied upon in clinical diagnostic practice, particularly when deliberate deception by the patient is suspected. In an attempt to devise a method which permits dependable sensory threshold interpretation, individual variability of thresholds was examined in normal and neuropathic subjects. Normals were also instructed to feign sensory impairment resulting from hypothetical injury. For each subject, a number of threshold readings were averaged, yielding individual means and variances. Feigning normal subjects evidenced a larger variance compared to trustworthy normal and neuropathic subjects. Thus, alertness to variance reinforces the psychophysical analysis: small variance values suggest trustworthy normal or pathological results, whereas large variance calls the interpreter's attention to feigned results or inattentive test performance.