Failure to unlearn the precedence effect.

Studies of the precedence effect using two binaural clicks have shown that listeners' ability to discriminate changes in the interaural time difference (ITD) of the lagging click is much poorer than that for the leading click [e.g., Zurek, J. Acoust. Soc. Am. 67, 952-964 (1980)]. This difference is thought to reflect an auditory process that suppresses directional information from the lagging sound and attributes greater perceptual weight to directional information contained in the leading one. A report by Saberi and Perrott [J. Acoust. Soc. Am. 87, 1732-1737 (1990)] suggested that listeners can "unlearn" this suppression of the lag's directional information after training with an adaptive psychophysical procedure involving 100 reversals and extremely small step sizes. Here, an attempt was made to find a similar effect using psychophysical procedures that are more common to precedence studies. Eight subjects were rigorously trained on the precedence task using either a blocked procedure or an adaptive procedure to vary ITD. Listeners showed no sign of unlearning. After 9-31 h of participating in the task, all subjects maintained high lag just-noticeable differences (jnd's) and low single source jnd's. This failure to train away the precedence effect (as manifested in discrimination suppression) suggests that directional information contained in the lagging source is not easily accessed. Several possible explanations for the discrepancies between the present study and Saberi and Perrott's finding are discussed.