Across-channel interference in intensity discrimination: the role of practice and listening strategy.

Pure tone intensity discrimination thresholds can be elevated by the introduction of remote maskers with roved level. This effect is on the order of 10 dB [10 log(DeltaII)] in some conditions and can be demonstrated under conditions of little or no energetic masking. The current study examined the effect of practice and observer strategy on this phenomenon. Experiment 1 included observers who had no formal experience with intensity discrimination and provided training over 6 h on a single masked intensity discrimination task to assess learning effects. Thresholds fell with practice for most observers, with significant improvements in six out of eight cases. Despite these improvements significant masking remained in all cases. The second experiment assessed trial-by-trial effects of roved masker level. Conditional probability of a "signal-present" response as a function of the rove value assigned to each of the two masker tones indicates fundamental differences among observers' processing strategies, even after 6 h of practice. The variability in error patterns across practiced listeners suggests that observers approach the task differently, though this variability does not appear to be related to sensitivity.