Electrophysiological correlates of word repetition spacing: ERP and induced band power old/new effects with massed and spaced repetitions.

The EEG was recorded in 21 participants who were required to make 'old' vs. 'new' decisions during a continuous recognition memory task with massed word repetitions (i.e., immediate repetitions) and spaced word repetitions (i.e., with six intervening words). After completion of the recognition task, they were given an unexpected free recall task. Behavioral measures revealed better recognition but worse free recall with massed repetitions, and worse recognition but better free recall with spaced repetitions. The ERP data revealed that for the 350- to 450-ms time window (representing the N400) the repetition effect was widely distributed across central and parietal scalp sites and was larger with massed than with spaced word repetitions. Peak latencies of the late positive complex (LPC) at Pz were shorter for massed than for spaced repetitions. Mean area around peak measures at Pz indicated that the LPC repetition effect was also larger with massed than with spaced repetitions. In the frequency domain, we found a theta (4-6 Hz) induced band power (IBP) repetition effect for massed repetitions, with theta power in the 250- to 625-ms time window being larger to repeated words than to new words. For spaced repetitions, we found an upper alpha (10-12 Hz) IBP repetition effect with the upper alpha power in the 625- to 1000-ms time window being smaller to repeated words than to new words. With spaced repetitions, we further found that larger LPC and theta IBP repetition effects were associated with smaller recall performance enhancements. The results mainly support a deficient processing account of the spacing effect.