| Three experiments measured the temporal dynamics of two study-test memory tasks. Comparisons of temporal dynamics were made for both recognition memory and memory for modality. Reaction time data were compared to electrophysiological measures and to retrieval functions derived from the response-signal methodology. Electrophysiological measures are typically expressed as event-related potentials generated by subjects as they perform cognitive tasks. The latency of the P300 component served as the electrophysiological measure for these studies. Using these three measures, reaction times, retrieval functions and the P300, the temporal dynamics of each task could be compared.; The results indicated that recognition-task information precedes modality task information. This finding supports the conclusions of Hintzman and Curran (1994) that familiarity information, the main mediator of recognition decisions, becomes available earlier than specific-item recall, suggested to mediate memory for modality. Dissociations between the response-signal method and the other two measures confirmed that these measures provide information about different parts of the retrieval episode. These results confirm that the response-signal procedure measures early aspects of the retrieval episode while the RT and P300 latency measures provide information about the end of the retrieval episode. Although P300 latency and RT were similarly affected by a number of variables, RT was dramatically affected by a speed vs. accuracy manipulation but P300 latency was not. This result can be accommodated by the predominant model of P300 latency only if subjects did not accelerate task computations in response to the speed manipulation, but rather accelerated motor responses. Several tests of this surprising prediction are proposed. |
