Short-term retention of temporal auditory information.

Abstract

To determine the short-term retention characteristics of temporal information when subjects experienced time under a conscious cognitive strategy for time estimation, i.e., subjects were instructed to refrain from employing time-aiding techniques, auditory durations of 1, 2, 4, 8, 16, and 32 sec. were estimated by 30 subjects by reproduction. Five levels of retention interval were used, viz., immediate reproduction, 30-sec. rest, 30-sec. interpolated activity with a response uncertainty of 0 bit per response, 30-sec. interpolated activity with a response uncertainty of 1 bit per response, and 30-sec. interpolated activity with a response uncertainty of 2 bits per response. Variable error evaluated effects of forgetting. When subjects hold durations of 8, 16, and 32 sec. in memory for a 30-sec. rest or 30-sec. interpolated activity, they become more variable than if they recall the item immediately. The variabilities between the 30-sec. rest and the 30-sec. interpolated activity retention intervals were similar. Two explanations for the presence of an interaction between duration and retention interval under the variable error were suggested. The constant error was used as an index of bias. Immediate reproduction of the retention interval had less mean constant error than 30-sec. rest retention under the 8-, 16-, and 32-sec. durations. That particular result was explained in perceptual terms.