Effect of signal-temporal uncertainty during childhood: Detection of a tonal signal in a random-frequency, two-tone masker.

Abstract

The purpose of this study was to determine if listeners benefit from a cue indicating when in time a signal occurs. In order to examine age-related differences in performance, three groups of children (5–7, 8–10, and 11–13 year-olds) and adults were tested. Thresholds were measured for a 120-ms, 1000-Hz signal presented in a continuous random-frequency, two-tone masker presented at 50 dB SPL. Each 120-ms masker burst was composed of a pair of tones, one selected from a uniform distribution 300–920 Hz and the other from 1080–3000 Hz. This masker was selected because it is expected to produce primarily informational masking. Listeners completed two temporal conditions: (1) temporally-defined, with the signal embedded in a 600-ms light cue and (2) temporally-uncertain, with no light cue. Two phases of testing were completed for each temporal condition. For phase 1, a single-interval, adaptive track procedure determined the midpoint of the psychometric function. In phase 2, signal levels around the midpoint were tested using a single-interval method of constant stimuli procedure. For both temporal conditions, listeners completed four runs of 40 trials, with an equal signal-nonsignal probability. Results indicate a substantial release from masking with reduced signal-temporal uncertainty for both children and adults.