Neural Transition From Short- to Long-Term Memory and the Medial Temporal Lobe: A Human Evoked-Potential Study Clara James, 1 Ste ´phanie Morand, 1 Sandra Barcellona-Lehmann, 1 Christoph M. Michel, 2 and Armin Schnider 1 * ABSTRACT: Recent studies indicated that the human medial temporal lobe (MTL) may not only be important for long-term memory consolida- tion but also for certain forms of short-term memory. In this study, we explored the interplay between short- and long-term memory using high-density event-related potentials. We found that pictures immedi- ately repeated after an unfilled interval were better recognized than pic- tures repeated after intervening items. After 30 min, however, the im- mediately repeated pictures were significantly less well recognized than pictures repeated after intervening items. This processing advantage at immediate repetition but disadvantage for long-term storage had an electrophysiological correlate: spatiotemporal analysis showed that im- mediate repetition induced a strikingly different electrocortical response after 200–300 ms, with inversed polarity, than new stimuli and delayed repetitions. Inverse solutions indicated that this difference reflected transient activity in the MTL. The findings demonstrate behavioral and electrophysiological dissociation between recognition during active maintenance and recognition after intervening items. Processing of novel information seems to immediately initiate a consolidation process, which remains vulnerable during active maintenance and increases its effectiveness during off-line processing. V V C 2008 Wiley-Liss, Inc. KEY WORDS: consolidation; evoked potentials; immediate recognition; long-term memory; medial temporal lobe; short-term memory INTRODUCTION Although the critical role of the human medial temporal lobe (MTL) for long-term memory is generally accepted, its role for short-term memory—the ability to temporarily retain information for processing (Fuster, 1995; Baddeley, 2003; Jonides et al., 2008)—is more controver- sial. In contradiction to earlier studies (Cave and Squire, 1992), recent studies found that amnesic subjects with proven or suspected MTL dam- age also displayed difficulty in short-term memory tasks involving the retention of spatial relations between items (Hannula et al., 2006; Olson et al., 2006; Hartley et al., 2007). Imaging studies showed MTL activation in such a task (Hannula and Ranganath, 2008) and during active maintenance of faces (Ranganath and D’Esposito, 2001). These studies suggested that MTL activation— which is undoubtedly essential for long-term-mem- ory—may be beneficial for short-term memory, too, and be involved in the processing of information dur- ing active maintenance. But it is unclear whether these MTL contributions to memory would be independent from each other, additive, or competitive. In the pres- ent study, we explored the interplay between short- and long-term memory for simple visual stimuli. Sub- jects performed a continuous recognition task while brain activity was recorded using high-density evoked potentials, allowing spatiotemporal analysis with very high temporal—albeit limited spatial—resolution (Michel et al., 2004). We compared the behavioral effects and electrocortical correlates of immediate stimulus repetition as opposed to delayed stimulus repetition and tested the effect of these manipulations on delayed recognition after 30 min. MATERIALS AND METHODS Participants Fifteen healthy, paid subjects (5 men, 10 women; 28.3 6 4.3 years) gave written informed consent to par- ticipate in the study, which was approved by the Ethical committee of the University Hospital of Geneva. Learning Task Subjects performed a continuous recognition task composed of 120 concrete black on white line drawings (Snodgrass and Vanderwart, 1980), all of which were repeated once, either immediately following a two-sec- onds stimulation-free interval after the initial presenta- tion (one-back items, N 5 60) or after nine intervening items (ten-back items, N 5 60). Stimuli were presented on a 17 in. monitor for 1000 ms, at the size of 88 of visual angle, with an interstimulus interval of 2000 ms filled with a fixation cross. Subjects had to indicate new pictures by pressing one button, picture recurrences by pressing another button with the right hand. 1 Division of Neurorehabilitation, Department of Clinical Neurosciences, University Hospital and University of Geneva, Geneva, Switzerland; 2 Functional Brain Mapping Laboratory, Department of Basic Neuro- sciences, University of Geneva, Geneva, Switzerland C.J. and S. M contributed equally to this work. Grant sponsor: Swiss National Science Foundation; Grant number: 32000- 113436. *Correspondence to: Armin Schnider, Service de neurore ´e ´ducation, Ho ˆpi- taux Universitaires de Gene `ve, 26, av. de Beau-Se ´jour, CH-1211 Geneva 14/Switzerland. E-mail: armin.schnider@hcuge.ch Accepted for publication 19 September 2008 DOI 10.1002/hipo.20526 Published online 20 November 2008 in Wiley InterScience (www. interscience.wiley.com). HIPPOCAMPUS 19:371–378 (2009) V V C 2008 WILEY-LISS, INC.