Abstract
Three observers tried to detect a 500‐Hz target in white noise in one of two temporal intervals. A distractor tone was added to the target interval on half of the trials and to the nontarget interval on the remainder. When the distractor was within about 40 Hz of 500 Hz, the 80% correct threshold for the target was elevated or depressed depending on whether the distractor was added to the nontarget or target interval. Thus, the critical band around 500 Hz is estimated to be about 80 Hz wide. It is possible to compare the relative sizes of the two threshold shifts against the predictions of two models of detection: the classical, integrated‐energy model, and a model emphasizing contrast among the energy of the components within the band. To make these predictions, simulated observers were run under the identical procedure. Within the limits of experimental error, the simulation matched the data well only under the integrated‐energy model. Under that model, the observers are performing about 2. to 3 dB worse than what the simulation predicts, which provides an estimate of the internal noise as nearly equal to the external noise.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call