Abstract:

Discriminability of pairs of 100-ms bursts of noise was measured using a same--different psychophysical method. In the first condition, the correlation between bursts in a pair was fixed at 1.0 during each same trial. On each different trial, bursts were identical except for (tau) ms of independent noise located at either the beginning, the middle, or the end of the pairs. The correlation between bursts presented on each different trial was increased from 0.0 by decreasing the duration of (tau). The second condition was the converse of the first; the correlation between bursts presented on each different trial was fixed at 0.0 and the correlation between bursts presented on each same trial was decreased from 1.0. As the (tau) ms of either independent (condition 1) or the perfectly correlated (condition 2) noise is moved from the beginning to the end of bursts, discriminability increases in the first condition and decreases in the second condition. These results suggest that a ``difference'' is a more salient cue in the discrimination task than a ``sameness'' and are discussed in terms of a correlational model. [Work supported by NIH and AFOSR.]