Academic journal article Perception and Psychophysics

Reinstating Object-Based Attention under Positional Certainty: The Importance of Subjective Parsing

Academic journal article Perception and Psychophysics

Reinstating Object-Based Attention under Positional Certainty: The Importance of Subjective Parsing

Article excerpt

Previous studies have shown that interference from flanking abstractors can be modulated by the object organization of the scene. The experiments reported here test for object-based attention under conditions of positional certainty, which allow a narrow focus of attention to the target. Prior research has suggested that object-based attention does not arise in these circumstances, but the experiments presented here show that object-based attention can still appear if previous experience with the stimuli leads participants to interpret the stimulus pattern as two separate objects. Two control experiments demonstrate that the appearance of object-based attention is not due simply to a widening of the focus of spatial attention. The presence of object-based attention in such a focused-attention task argues against Shomstein and Yantis's (2002) proposed explanation of object-based attention based on priority in the order of visual search.

A number of different experimental paradigms have demonstrated that the object organization of a scene can affect the allocation of attention, even when object organization is irrelevant to the task. One paradigm, which originated with EgIy, Driver, and Rafal (1994), comprised two simple rectangles and a spatial cue that appeared at the end of one of the rectangles. Responses to a test stimulus at the cued location were faster than responses to a test stimulus at the uncued end of the same rectangle, confirming earlier studies showing that attention weakens with distance from the cue (Downing & Pinker, 1985; LaBerge & Brown, 1989). However, test stimuli at the uncued end of the cued rectangle still elicited faster responses than stimuli on the other rectangle, which was completely uncued. Attention favored all parts of the cued object to some extent.

Other experiments using other techniques have demonstrated that this object-based attention extends to groups of objects that share a basic feature such as color. Harms and Bundesen (1983) used inhibition from flanking distractors to show that a distractor that was the same color as the target was to some extent selected with the target, even though the distractor was irrelevant to the task and the location of the target was known before it appeared. Kirn and Cave (2001 ) used spatial probes to show that the target-colored distractor also received more spatial attention than a distractor of a different color. Similar evidence has been provided by Kramer and Jacobson ( 1991 ).

However, as Shomstein and Yantis (2002) demonstrated, the effects of object organization do not always appear. In four of their five experiments, the location of the target letter to be reported was known in advance, allowing attention to be focused at that location in preparation for the stimulus. Flanking distractor letters were also present in the displays, sometimes within the same rectangle as the target and sometimes within a different rectangle. As expected, the flanking distractors interfered with the response when they were near the target (B. A. Eriksen & C. W. Eriksen, 1974), but the amount of interference was no greater when the distractors were within the same rectangle as the target than when they were in different rectangles. In the fifth experiment, the location of the target varied from trial to trial. This time, the object effect appeared, with more distractor interference when targets and distractors were within the same rectangle.

Shomstein and Yantis's (2002) results suggest that object effects appear only with spatial uncertainty. In their study, the object boundaries affected performance only when the target location was not known, and attention thus could not be focused at a single location. However, object-based effects do appear in the experiments by Harms and Bundesen (1983), Kirn and Cave (2001), and Kramer and Jacobson (1991), in all of which the target location was known. With this apparent conflict between the different results, it is not clear whether the deployment of object-based selection depends on spatial uncertainty or on other factors. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.