Failure of selective attention in simple recognition tasks: Critical band effects

In the second in a series of studies examining the ability of human observers to perform two independent dichotic frequency difference limen tasks, the effect of separating the base frequencies was examined. We reported earlier [W.A. Ahroon, R.E. Pastore, and J.P. Wolz, J. Acoust. Soc. Am. 56, 545 (1974)] that trained observers performed no better during dichotic stimulation when attending to one frequency difference limen task than when they attempted to attend to both channels, when the base frequencies were either within the same critical band or separated by only two or three critical bands. The present study examined the effect of widely separating (by nine to ten critical bands) the base frequencies in the two-channel frequency difference limen task. Observers again showed a decrement in performance in the selective attention condition relative to monaural performance. However, this decrement did not seem as great as the decrement for the binaural (shared attention) condition. Selective attention continued to be impaired although not as greatly as when the two tasks involve frequency discriminations in the same critical band. [Work supported by NINDS.]