Abstract
A brief 1000‐Hz sinusoid served as signal in a two‐interval forced‐choice detection task. The noise waveform was digitized and in each of the two intervals either independent or identical waveforms were presented. Assume detection is based on waveform energy with internal Gaussian noise added to each sample. Detection performance is determined by the signal‐to‐noise ratio and the level of internal noise. We can therefore calculate estimates of noise levels using decision rules appropriate for each test condition. For narrow‐band noise (WT = 1) or wide‐band noise (WT > 100) these estimates are similar for either independent or identical noise samples. In the narrow‐band case the signal‐to‐noise ratio needed for a constant detection performance is smaller than the wide‐band case by almost 10 dB. Despite this change the internal noise estimates remain the same. [This work was supported by an NIH grant and post‐doctoral fellowship.]
Sign-in/Register to access full text options 
Free) 
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
