Retention of trained performance in consistent mapping search after extended delay.

We examined the retention of performance in memory, visual, and hybrid memory/visual search after subjects received extensive consistent mapping (CM) practice. In the first two experiments we examined retention of detection performance in memory scanning (Experiment 1) and in visual search (Experiment 2) approximately one month after training. In the third experiment we examined performance 1, 30, 90, 180, and 365 days following training on pure visual and hybrid memory/visual semantic-category search tasks. We examined retention at intervals up to one year after practice as a function of amount of CM practice. In addition, we assessed the degree of category activation (performance on untrained words from the trained categories) at each retention interval. Across experiments, we found no decay in CM trained memory search and minimal decay in pure visual search. Significant decline in CM performance was largely attributable to performance in the hybrid memory/visual search conditions. These data reconcile previously equivocal retention results for CM trained search performance.