Selective Visual and Auditory Attention Toward Utterances—A PET Study Ryuta Kawashima,* , † Satoshi Imaizumi,‡ Koichi Mori,‡ Ken Okada,* Ryoi Goto,* Shigeru Kiritani,‡ Akira Ogawa,§ and Hiroshi Fukuda* , † *Department of Nuclear Medicine and Radiology, IDAC, Tohoku University, Sendai, 980–8575 Japan; †Aoba Brain Research Imaging Center, TAO, Sendai, Japan; ‡Department of Speech and Cognitive Sciences, Graduate School of Medicine, University of Tokyo, Tokyo, Japan; and §Department of Neurosurgery, Iwate Medical University, Morioka, Japan Received December 16, 1998 The purpose of this study was to reveal functional areas of the brain modulating processing of selective auditory or visual attention toward utterances. Re- gional cerebral blood flow was measured in six normal volunteers using positron emission tomography dur- ing two selective attention tasks and a control condi- tion. The auditory task activated the auditory, inferior parietal, prefrontal, and anterior cingulate cortices. The visual task activated the visual association, infe- rior parietal, and prefrontal cortices. Both conditions activated the same area in the superior temporal sulcus. During the visual task, deactivation was ob- served in the auditory cortex. These results indicate that there exists a modality-dependent selective atten- tion mechanism which activates or deactivates corti- cal areas in different ways.  1999 Academic Press INTRODUCTION During specific cognitive tasks, our attention is selec- tively directed toward the relevant sensory modality or submodality from which information necessary to per- form the task is obtained. Brain activity involved in the selective attentional processing within specific modal- ity has been investigated by many recent neuroimaging studies (Bench et al., 1993; Carter et al., 1995; Corbetta et al., 1991; Haxby et al., 1994; Le et al., 1998; Platel et al., 1997; Pugh et al., 1996; Tzourio et al., 1997). However, only a few studies (Roland, 1982; Woodruff et al., 1996; O’Leary et al., 1997) have attempted to study areas of the brain activated in relation to alternating attention between sensory modalities, and the brain mechanisms underlying this are still poorly under- stood. In the present study, therefore, we measured changes in regional cerebral blood flow (rCBF) using positron emission tomography (PET) during auditory- or visual-selective attention toward utterances. Sub- jects were simultaneously stimulated in both sensory modalities and were asked to direct their attention toward a specific modality and discriminate between the stimuli, but to ignore the irrelevant information from the other modality. METHODS Subjects Six right-handed male volunteers (aged 18–22 years) participated in the present study. Written informed consent was obtained from each subject on forms approved by the Iwate Medical University and the Declaration of Helsinki (1975). All the subjects were healthy, with no past history of psychiatric or neurologi- cal illness, and none was on any medication. MRI of the brain was performed on each subject using spoiled gradient echo sequence (TE = 12 ms, TR = 50 ms, flip angle = 45°) with a 0.5-TYokogawa Medical MRVectra scanner, on a separate occasion, consisting of 96 slices with a voxel size of 1  1  1.5 mm. Prior to the PET experiments, a catheter was placed in the right bra- chial vein of each subject for tracer administration. Positron Emission Tomography The rCBF was measured using a Headtome IV PET scanner (Shimadzu), after a bolus injection of H 2 15 O (35 mCi per scan). Each PET measurement was commenced immediately after radioactive counts were monitored on the PET camera and continued for a period of 60 s. Attenuation-corrected data were reconstructed into 14 image planes with a resultant resolution of 5.1  5.0  10.0 mm at full-width half-maximum. The room was darkened during the PET measurements (4.31 lux). Each subject wore a stereotaxic fixation helmet during the PET measurements. Video recordings were made for subse- quent analysis of each subject’s performance. An electro- oculogram (EOG) was also recorded for each subject during the PET measurements. Task Procedures Each subject performed the following three tasks during the PET measurements. The same set of audi- NeuroImage 10, 209–215 (1999) Article ID nimg.1999.0452, available online at http://www.idealibrary.com on 209 1053-8119/99 $30.00 Copyright  1999 by Academic Press All rights of reproduction in any form reserved.