Copyright 2008 Psychonomic Society, Inc. 1148 Effective interaction with our visual environment re- quires that individual regions of interest within the visual field be selected for deeper processing. One way that peo- ple perform this selection is to deploy attention. Stemming from initial experiments into visual attention by Posner (1980), there are two traditionally accepted ways in which attention can be oriented in the visual field. Attention can be oriented through an exogenous or reflexive system, as has typically been found in experiments that have used uninformative peripheral cues (e.g., the abrupt onset of an object in the periphery that does not provide any in- formation regarding the location of the upcoming target; Posner, 1980). Or, attention can be oriented through an en- dogenous or volitional system, as has often been found in experiments using informative central cues (e.g., an arrow presented at fixation that indicates the most probable target location; Jonides, 1980, 1981). Endogenous shifts in atten- tion are not directly triggered by physical properties of the stimuli; instead, they are goal based. Recently, however, a considerable amount of research has been conducted on cues that are centrally presented and uninformative yet generate shifts of attention to peripheral locations. From this research has emerged the proposal that two types of uninformative central cues—arrows and gaze direction— may generate reflexive or exogenous shifts in attention (e.g., Friesen, Moore, & Kingstone, 2005; Hommel, Pratt, Colzato, & Godijn, 2001; Langton & Bruce, 1999). Many studies have now shown that the direction of gaze from a centrally presented face, be it a photograph of a real face (Driver et al., 1999) or a drawing of a schematic face (Friesen & Kingstone, 1998), can generate shifts of attention. The fact that such stimuli produce facilitation effects even when they are uninformative has led to the suggestion that they operate in a reflexive, or exogenous, manner. Indeed, facilitation effects toward the gazed-at location have also been found using counterpredictive central gaze cues (Friesen, Ristic, & Kingstone, 2004). Some evidence, however, suggests that gaze cues are not reflexive and, instead, produce endogenous shifts in atten- tion. Specifically, Vecera and Rizzo (2006) have described a frontal lobe patient, E.V.R., who did not show facilitated response times following gaze cues and endogenous cues. E.V.R. did, however, show the normal cuing effect for pe- ripheral exogenous cues. Because gaze cue orienting was damaged and peripheral orienting remained intact, it was suggested that gaze cues affect attentional processes in the same way that known endogenous cues do, and this type of orienting is disrupted by frontal lobe damage. There is also evidence that central arrow cues, which, when informative, have often been used to study endog- enous shifts in attention, may cause exogenous orienting. For example, Hommel et al. (2001) observed early facili- tation effects with central arrow cues to targets appearing in the cued location for uninformative, as well as counter- predictive, cues. Similar findings have been obtained by Ristic, Friesen, and Kingstone (2002), Tipples (2002), and Gibson and Bryant (2005). Ristic, Wright, and Kingstone (2007) also found reflexive attentional shifts using central arrow and gaze cues but noted that gaze cues may produce stronger reflexive orienting. In contrast, there has been evidence that arrow cues do not always produce facilita- tion. Jonides (1981) failed to find reflexive orienting to counterpredictive arrow cues. Gibson and Bryant (2005), reexamining arrow cues, found effects similar to those of Jonides (1981) when cue duration and stimulus onset asyn- chrony (SOA) were relatively brief; when these durations Testing whether gaze cues and arrow cues produce reflexive or volitional shifts of attention SARA A. STEVENS, GREG L. WEST, NASEEM AL-AIDROOS, ULRICH W. WEGER, AND JAY PRATT University of Toronto, Toronto, Ontario, Canada It has been suggested that two types of uninformative central cues produce reflexive orienting: gaze and arrow cues. Using the criterion that voluntary shifts of attention facilitate both response speed and perceptual accuracy, whereas reflexive shifts of attention facilitate only response speed (Prinzmetal, McCool, & Park, 2005), we tested whether these cues produce reflexive or volitional shifts of attention. A cued letter discrimination task was used with both gaze (Experiments 1A and 1B) and arrow (Experiments 2A and 2B) cues, in which participants responded to the identity of the target letter. In the response time (respond speed) tasks, participants were asked to respond as quickly as possible to the target; in the accuracy (perceptual quality) tasks, participants were asked to respond as accurately as possible. For both cue types, compatible cues were found to facilitate response speed but not perceptual accuracy, indicating that both gaze and arrow cues generate reflexive shifts in attention. Psychonomic Bulletin & Review 2008, 15 (6), 1148-1153 doi:10.3758/PBR.15.6.1148 S. A. Stevens, sara.stevens@utoronto.ca