Cognitive Brain Research 10 (2000) 133–144 www.elsevier.com / locate / bres Research report Towards a functional neuroanatomy of self processing: effects of faces and words a,e, a a d * Tilo T.J. Kircher , Carl Senior , Mary L. Phillips , Philip J. Benson , b b c c Edward T. Bullmore , Mick Brammer , Andrew Simmons , Steven C.R. Williams , e a Mathias Bartels , Anthony S. David a Department of Psychological Medicine, Institute of Psychiatry and GKT School of Medicine, De Crespigny Park, London SE58AF, UK b Department of Biostatistics and Computing, Institute of Psychiatry and GKT School of Medicine, De Crespigny Park, London SE58AF, UK c Neuroimaging Unit, Institute of Psychiatry and GKT School of Medicine, De Crespigny Park, London SE58AF, UK d University Laboratory of Physiology, Parks Road, Oxford OX13PT, UK e ¨ Department of Psychiatry, University of Tuebingen, Osianderstr. 24, D-72076 Tubingen, Germany Accepted 23 May 2000 Abstract We studied the neural correlates of self vs. non-self judgements using functional magnetic resonance imaging (fMRI). Individually tailored faces and personality trait words were used as stimuli in three experiments (exp.). In the first two experiments, brain activation was measured while subjects viewed morphed versions of either their own (self face exp.) or their partner’s face (partner’s face exp.), alternating in blocks with presentation of an unknown face. In the self face exp. right limbic areas (hippocampal formation, insula, anterior cingulate), the right middle temporal lobe, left inferior parietal and left prefrontal regions showed signal changes. In the partner’s face exp., only the right insula was activated. In the third exp., subjects made decisions about psychological trait adjectives previously categorized as describing their own attributes. Activation was present in the precuneus, the left parietal lobe, left insula / inferior frontal gyrus and the left anterior cingulate. A reaction time advantage was present when subjects responded to self-relevant words. The main area with signal changes during self-reference processing, regardless of the type of stimulus, was the left fusiform gyrus. The self-relevant stimuli engaged to a differential extent long term and working memory, semantic and emotional processes. We suggest that regions activated by these stimuli are engaged in self-processing.  2000 Elsevier Science B.V. All rights reserved. Theme: Neural basis of behavior Topic: Cognition Keywords: Self; Self-recognition; Self-awareness; Familiar faces; Overlearned faces; Limbic system 1. Introduction the world and who we are, which we need for interaction with our environment. It is assumed that different brain Processing of self relevant information and self knowl- regions contribute to these various forms of memory edge is regarded as distinct from processing ‘objective’ processing [41,62]. In experiments on memory encoding information [30,56]. Self knowledge is stored in two main and retrieval it has been shown that relating information to memory systems, semantic and episodic memory. Episodic oneself (self-referent effect) enhances recall memory [61,68] deals with individual (‘narrative’) epi- [5,33,35,54,63]. This phenomenon is largely independent sodes that are definable with respect to time and place, of access to autobiographical narrative memory [34,37]. whereas semantic memory contains abstracted facts about There is evidence for the relative independence of seman- tic and episodic autobiographical memory systems, which can each be accessed independently [13,42,70]. The neural *Corresponding author. Tel.: 149-7071-298-2311; fax: 149-7071- systems involved in modality independent self-knowledge 294-141,. E-mail address: tilo.kircher@uni-tuebingen.de (T.T.J. Kircher). are unknown. 0926-6410 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0926-6410(00)00036-7