Studies using the lateralized readiness potential (LRP; an index of hand-specific motor preparation), suggest that the motor system can be activated on the basis of partial information, providing support for a parallel view of the human information processing system. However, data to the contrary have been obtained in memory-search paradigms, leading Heil et al. [Psychological Research 62 (1999) 289] to propose that hand-specific motor preparation cannot occur in parallel with memory scanning. To further investigate these findings we recorded the LRP in a fast-paced memory search task, in which the case of target letter and memory set size were manipulated and a choice-go/no-go procedure was used. The case of the test letter determined the response hand, whereas membership of the test stimulus in the memory set determined whether a response was required. We found different types of LRP effects as a function of time on task. Analyses based on the data from the first half of the experiment replicated those obtained by Heil et al. [Psychological Research 62 (1999) 289]. However, unlike Heil et al. [Psychological Research 62 (1999) 289], the second half of the data showed significant LRP activity on no-go trials and similar LRP onset latencies for go and no-go trials, as well as for different memory set size conditions in the stimulus-locked LRP waveforms. The response-related waveforms showed an earlier LRP onset in the set-size 4 than in the set-size 2 condition. These data suggest that motor preparation can occur simultaneously with memory search when participants are sufficiently practiced. Time-on-task and strategy differences between the two studies are discussed.