When is object-based attention not based on objects?

Some studies using methods introduced by Egly, Driver, and Rafal (1994) to measure object-based attention have shown surprising effects of object orientation. Rectangles oriented horizontally produce evidence for object-based attention, whereas vertical rectangles do not. We explore these differences using a two-letter comparison task. Across all the experiments, responses are faster when the targets are arranged horizontally rather than vertically. The horizontal advantage persists when the rectangles are removed, demonstrating its independence from object-based attention. Furthermore, responses are faster for vertically configured targets when they are within the same rectangle. This effect only arises when the rectangle orientation is informative about the target configuration orientation. This same-object advantage would normally be attributed to object-based attention, but the same pattern of data emerged when the rectangles were replaced with a salient orientation cue. The rectangles can apparently serve as a cue, making the cuing effect appear to be an object-based attention effect. However, the cuing effect can also be triggered by a horizontal bar in the center of the display. Thus participants can adopt an attentional set for a horizontal orientation independently of stimulus location. Results from comparison tasks that might have been attributed to object-based attention could instead be due to a combination of a horizontal advantage and an orientation set cost. Although these results lead to a diminished role for object-based attention, they also call for a better understanding of how an attentional set can be adopted without selecting either locations or objects. (PsycINFO Database Record (c) 2019 APA, all rights reserved).