Research report Face repetition effects in direct and indirect tasks: an event-related brain potentials study Maja U. Trenner a,b , Stefan R. Schweinberger a, * , Ines Jentzsch a , Werner Sommer b a Department of Psychology, University of Glasgow, Glasgow, Scotland G12 8QB, UK b Department of Psychology, Humboldt-University at Berlin, Germany Accepted 17 June 2004 Available online 8 September 2004 Abstract We investigated immediate repetition effects on event-related potentials (ERPs) during direct and indirect tasks for sequentially presented face pairs. The first face (F1) was presented masked or unmasked, and at different stimulus onset asynchronies (SOAs, 67 vs. 1000 ms) preceding the second face (F2). Experiment I (indirect task) required a semantic classification of F2, with F1 identity being irrelevant. Experiment II (direct task) used the same stimulus sequence but required a physical identity matching of F1 and F2. Whereas no masked repetition effects in behaviour or ERPs were seen, such effects were clearly shown for unmasked F1 faces. For short SOAs , an early-onset (~100 ms) occipital repetition effect, an inferior temporal N250r (200–300 ms) and a central-parietal N400 modulation (300–500 ms) were seen in both tasks, whereas a parietal P600 effect (500–800 ms) was only present in the indirect task. For long SOAs , the early occipital effect disappeared, suggesting that it reflects a fast decaying iconic memory trace. Clear task differences were seen for N250r, N400, and P600 modulations: P600 was larger for the indirect task, and may be a correlate of semantic analysis required by this task. By contrast, N250r and N400 were larger for the direct task, suggesting that these components are sensitive to task relevance and/or attentional focus to F1, and thus do not reflect purely automatic facilitation in processing. This suggests an influence of strategic processing on the activation of both perceptual representations of faces and semantic representations of people. D 2004 Elsevier B.V. All rights resesrved. Theme: Neural basis of behaviour Topic: Cognition Keywords: Event-related potential; Face perception; Priming; Matching; Semantic processing 1. Introduction Stimulus repetitions have been shown to influence the processing of stimuli across a variety of tasks and across different time intervals between repetitions [55]. A great deal of research has been conducted in order to investigate the influence of experimental tasks on repetition effects. Repetition effects have been shown to depend on task complexity [3]. They also depend on the task relevance of the stimuli: An important distinction is between tasks in which successful performance requires processing of both the first and second stimulus, and tasks that require only processing of the second stimulus (although processing of the first stimulus is not necessarily excluded) [20,21,49]. With respect to stimulus repetitions, the former type of task is termed direct or explicit, whereas the latter is termed indirect or implicit. In the present context, we will use the terms direct versus indirect tasks. Event-related brain potentials (ERPs) have been reported in several studies in order to investigate differ- ences in brain activity related to the stimulus repetition, with the underlying goal to understand the brain processes involved in perception and memory. A sub- 0926-6410/$ - see front matter D 2004 Elsevier B.V. All rights resesrved. doi:10.1016/j.cogbrainres.2004.06.017 * Corresponding author. Tel.: +44 141 330 3947; fax: +44 141 330 4606. E-mail address: s.schweinberger@psy.gla.ac.uk (S.R. Schweinberger). Cognitive Brain Research 21 (2004) 388 – 400 www.elsevier.com/locate/cogbrainres