Visual Attention and Cognition – Synopsis

von ed. Wolfgang H Zangemeister, S.Stiehl und C.Freksa Elsevier, Amsterdam 1996

In this synopsis an introduction and overview is given with regard to the functional anatomical and clinical neurophysiological aspects of visual attention and cognition. Humans operate in a complex data rich world. Our visual system is constantly confronted with situations that require rapid processing and decision making. Yet, we are able to analyze almost effortlessly, information critical to the task at hand, while ignoring vast amounts of nonessential information. How is this done so effectively? Some have conjectured that top-down processes involving learning and semantic memory account for our ability to process information rapidly (Noton SL Stark 1971; Homa et al. 1976; Friedman 1979; Julesz 1991). These processes form higher level associations among the components in a scene. Hence, efficiency in processing is gained by reducing the need for an element—by—element encoding of each item in the scene. How are these associations created? The associations among the components of a scene may be formed, for example, by using their spatial relations (Ullmann 1985), when performing a visual search. A simple example is that a schematic of a face is more readily perceived when the components of the eyes, nose and mouth are normally arranged as opposed to being scrambled. Another means of consolidating information is by its contextual significance based on the relation among the objects in a scene. Context in the natural environment plays an important role. This is because in a natural scene objects have powerful and complex relations and the association of items in a scene provides the means for rapid and effective processing of visual information. The underlying association of this top-down contextual process with the bottom-up processes in early vision was recently investigated (Hung et al. 1995]. It was demonstrated that better performance for simultaneous over sequentially presented items was a general phenomenon, independent of the kind of icons used. They suggested that both a parallel buffer and a serial retrieval mechanism linked to short term memory were involved in the early stages of visual processing. At the connectivity level of visual processing, contextual effects involved higher level processes that incorporated the complex relations among objects. The fact that these relations could be formed rapidly for briefly presented objects indicated that cortical neural connectivity may already be in place to facilitate these interactions.