A signal detection procedure eliminates priming biases in dichotic listening.

Dichotic digit names were presented to normal right-handed adults in two experiments, the first of which used a standard free report procedure and the second of which used a signal detection procedure. Prior to each test, a priming bias was generated by having subjects monitor either the left or right ear for 30 selective listening trials. The expected priming effect was found in Experiment 1: ear asymmetry and order of report were altered according to the ear previously monitored. In contrast, no priming effect was observed with the signal detection procedure used in Experiment 2. There was a right-ear advantage in sensitivity irrespective of the ear previously monitored. The findings suggest that priming biases act by altering the order in which multiple stimuli are processed and reported.