8 from Attention to Action in Frontal Cortex 8.1 Introduction

The purpose of the visual system is to transform light into action. For example, consider the visual search illustrated in figure 8.1, in which the observer must locate a T among many Ls. The observer makes a series of gaze shifts to inspect the elements in the array. Before each gaze shift, two selection processes have to take place (e.g., Allport 1987; Pashler 1991; Coles et al., 1995). The first process selects a stimulus to guide action, and the second selects the action. In this chapter , we will review our investigations into the processes by which visual stimuli are selected as targets for gaze shifts. Recognition is growing that overt eye movements and covert shifts of visual attention are guided by a common mechanism. Several experiments have shown a cost in perceptual reliability or saccade latency if attention is directed away from the target for a saccade (e.g., Hoffman and Subramaniam, 1995; Kowler et al., 1995; Deubel and Schneider , 1996). In addition, directing attention seems to influence the production of saccades (Sheliga et al., 1995; Kustov and Robinson, 1996). Further evidence is the common manner in which bottom-up factors influence visual selection for attention and saccades . Visual conspicuousness drives covert (e.g., Theeuwes , 1991) and overt (Theeuwes et al., 1998) selection. In fact, nontarget elements that resemble the target can be inadvertently selected covertly (e.g., Kim and Cave, 1995) or overtly (Findlay, 1997; Zelinsky and Sheinberg, 1997; Motter and Belky, 1998; Bichot and Schall, 1999b). Top-down factors also influence visual selection. Cognitive strategies can override both covert (e.g., Bacon and Egeth, 1994) and overt (e~g., Bichot et al., 1996; Nodine et al., 1996) selection of conspicuous target stimuli. In addition, target selection is influenced by implicit memory representations arising through short-term priming of location or stimulus features for covert (e.g., Maljkovic and Nakayama, 1994, 1996) and overt orienting (Bichot and Schall , 1999a; McPeek et al., 1999). Target selection is also influenced by longterm priming of target properties across sessions (Bichot and Schall , 1999b). Finally, an explicit memory representation is needed to identify the unique target during conjunction search (e.g., Treisman and Sato 1990; Bacon and Egeth, 1997). To explain all of these observations , most models of covert attention (e.g., Koch and Ullman, 1985; Treisman, 1988; Cave and Wolfe, 1990; Olshausen et al., 1993; Wolfe, 1994) and overt saccade generation (e.g., Findlay and Walker, 1999) postulate the existence of a map of salience derived from converging bottom-up and top-down influences. Peaks on the salience map that develop through winner-take-all competitive interactions represent locations that have been selected for further processing and can, but need not necessarily , lead to orienting saccadic eye movements . 8 From Attention to Action in Frontal Cortex