RECOGNITION MEMORY AND SEARCH FOR ATTENDED LETTERS - AN EVENT-RELATED POTENTIAL ANALYSIS

The purpose of this experiment was to evaluate whether the amplitude reductions in the P3 (or P300, P3b) component of the event-related brain potential (ERP) with increasing memory set size are due to overlap with a negative wave (Okita, Wijers, Mulder, & Mulder, 1985) or to equivocation, a factor that modulates P3 amplitude (Johnson, 1986). Twelve young adults performed a memory search task (memory set size 1, 2, or 4 consonants) and gave a speeded response to attended test items which matched one of the memory set letters. On half of the trials, the test items were presented in spatial locations to which the subjects were to attend, while on the other half of the trials the stimuli were presented in locations that were to be ignored. Our results for attended stimuli showed the usual inverse relation between P3 amplitude, maximal over parietal scalp, and memory set size, together with on increase in P3 latency, reaction time (RT), and error rate. This pattern of ERP and performance results is consistent with the previously well established relation between equivocation (due to weaker memory traces) and P3 and behavioral measures. In addition, we found a negative wave, present simultaneously with the P3, that was maximal over the left frontal scalp, whose amplitude was directly related to memory set size. In an analysis of scalp topographic information, we found that the effects of memory set size had significantly different scalp distributions at the Frontal and parietal electrode sites, signifying that these effects of memory set size were associated with different parts of the neural generator activity. Hence, it appears unlikely that the P3 amplitude reduction that occurs with increases in memory set size was due to overlap with any concomitant negative wave.