Research report Hemispheric dissociation of visual-pattern processing and visual rotation Boris Suchan a, *, Lidia Ya ´gu ¨ez b , Gilbert Wunderlich c , Anthony G.M. Canavan e , Hans Herzog d , Lutz Tellmann d , Volker Ho ¨ mberg e , Ru ¨ diger J. Seitz c a Department of Clinical Neuropsychology, Ruhr-University of Bochum, Universita ¨tsstr. 150, 44780 Bochum, Germany b Department of Psychology, Institute of Psychiatry, King’s College, University London, London, UK c Department of Neurology, Du ¨sseldorf Heinrich-Heine University, Du ¨sseldorf, Germany d Institute of Medicine, Research Centre Ju ¨lich, Ju ¨lich, Germany e Neurological Therapy Centre, Germany Received 30 April 2002; received in revised form 8 July 2002; accepted 8 July 2002 Abstract We aimed at investigating whether on-line and delayed visual pattern processing activated different areas in human prefrontal and parietal cortex. For this purpose we measured the regional cerebral blood flow (rCBF) during simultaneous and successive visual matrix processing in 10 right-handed subjects. Delayed matching to sample activated predominantly left hemispheric ventrolateral prefrontal cortex, Broca’s area and parts of the parietal cortex. In contrast, visuospatial matrix rotation showed activation of the right dorsolateral prefrontal cortex and parietal lobe. The present results suggest a hemispheric dissociation of fronto-parietal circuits with a left dominance for visual pattern processing like storage and a right dominance for visuospatial processing. # 2002 Elsevier Science B.V. All rights reserved. Keywords: PET; Visuospatial working memory; Mental rotation; Dorsolateral prefrontal cortex; Ventrolatral prefrontal cortex; Parietal cortex 1. Introduction Working memory as postulated by Baddeley and Hitch [4,5] mediates short-term storage of information and online manipulation of stored information. It is seen as a modular system, which includes separate, but integratively working subsystems. The Central Execu- tive system controls attention and information transfer to and from two modality specific short-term buffers, i.e. verbal (phonological loop) and visual (visuospatial sketchpad) subsystems. The visual subsystem can further be differentiated with respect to the type of the involved material into a visual and a spatial subcompo- nent. This distinction is supported by experimental [27] as well by neuropsychological findings [10,28]. Two opposite positions have emerged concerning the anatomical localisation and organisation. One position based on experimental work in non-human primates [14  /16] argues that working memory processes are organised according to the type of information being processed (i.e. the modality or domain). Thus, according to Goldman-Rakic, the dorsolateral frontal regions are mainly concerned with visuospatial material, whereas ventrolateral frontal areas are involved in working memory for non-spatial material. Levy and Goldman- Rakic [25] argue that in monkeys the homologue areas of the dorsolateral and ventrolateral prefrontal areas receive different input from posterior areas, that are themselves specialized in visuospatial processing and divided into spatial and features components. The authors also argue that in a lot of studies ‘‘spatial’’ tasks may be performed as ‘‘object’’ tasks and therefore * Corresponding author. Tel.:  /49-234-322-7575; fax:  /49-234- 321-4622 E-mail address: boris.suchan@ruhr-uni-bochum.de (B. Suchan). Behavioural Brain Research 136 (2002) 533 /544 www.elsevier.com/locate/bbr 0166-4328/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII:S0166-4328(02)00204-8