The present study focuses on the question of whether inter-and intramodal forms of attention are reflected in activation of the same or different brain areas. ERPs were recorded while subjects were presented a random sequence of visual and auditory stimuli. They were instructed to attend to nonspatial attributes of either auditory or visual stimuli and to detect occasional target stimuli within the attended channel. An occipital selection negativity was found for intramodal attention to visual stimuli. Visual intermodal attention was also manifested in a similar negativity. A symmetrical dipole pair in the medial inferior occipital areas could account for the intramodal effects. Dipole pairs for the intermodal attention effect had a slightly more posterior location compared to the dipole pair for the intramodal effect. Auditory intermodal attention was manifested in an early enhanced negativity overlapping with the N1 and P2 components, which was localized using a symmetrical dipole pair in the lateral auditory cortex. The onset of the intramodal attention effect was somewhat later~around 200 ms!, and was reflected in a frontal processing negativity. The present results indicate that intra-and intermodal forms of attention were indeed similar for visual stimuli. Auditory data suggest the involvement of multiple brain areas.
Descriptors: Nonspatial attention, Intermodal attention, ERPs, Scalp topography, Source localizationSelective attention can be seen as a collection of hierarchical filters see Hansen & Hillyard, 1983;Heslenfeld, 1998! that provide various levels of selection. The first of these levels selects information among the sensory modalities, such as the visual, auditory, or tactile modality. At the dimensional level, a given stimulus dimension, such as spatial location, color, orientation~visual modality! pitch or intensity~auditory! is selected from the relevant modality, and at the feature level one particular feature, such as a given color~e.g., the color purple! is selected from the currently selected dimension. Based on this model, one could argue that selection of a relevant modality would precede the selection of relevant stimuli within the attended modality.Recent studies have questioned such a strict hierarchy, however. Is has been shown, for example, that the judgment of the location of auditory stimuli is improved by the presence of a structured visual field~Platt & Warren, 1972!, whereas discrepant visual information causes gross errors in auditory location~a phenomenon known as the "ventriloquist" effect; Pick, Warren, & Hay, 1969!. In addition, Eimer and Schröger~1998! have demonstrated the presence of location relevance for both visual and auditory stimuli, which not only existed when presented to the attended modality, but also when these stimuli were presented to a modality which was not attended~i.e., visual stimuli presented when attention was directed to auditory stimuli!.The present article focuses on the question of whether or not nonspatial forms of intermodal attention...