Multiple Constraints on Selection A Framework

Perhaps the most complete understanding of atten-tional selection is in the visuospatial domain, in which stimuli at specific locations are selected for processing. What are the critical parameters that determine those inputs that receive attention and those that do not? This is the question of attentional control.

There are different parameters or processes that can influence attentional control. Two general classes of control are top-down sources that arise from the current behavioral goals and bottom-up sources that arise from sensory stimuli present in a scene. These two sources can be illustrated by considering visual search, the act of looking for a visual target among distractors (e.g., finding a friend's face in a crowd). In a typical visual search task, observers are instructed to search for a particular target, such as a black vertical line, that appears in a field of distractors (Fig. 1). The target description can be conceptualized as a ''template'' temporarily stored in memory that can influence visual search in a top-down manner; as an observer, you would actively attempt to look at black and vertical items. The scene presented in a visual search task provides the bottom-up information that is searched; this information indicates where objects are located and which features are present at each location. Effective visual search would require finding a balance between the top-down information and the bottom-up information. An example of an effective search is searching for a single feature, such as a black vertical line among white vertical lines (Fig. 1a). In this example of a feature search, color uniquely defines the target, and the bottom-up information is consistent with the top-down information in constraining where an observer should search. A less efficient search would involve searching for a conjunction of features, such as

Figure 1 Sample visual search task: search for a black vertical line. (a) Feature search in which the target pops out from a homogeneous background. (b) Conjunction search in which the target does not pop out because the distractors share both black and vertical features with the target.

Figure 1 Sample visual search task: search for a black vertical line. (a) Feature search in which the target pops out from a homogeneous background. (b) Conjunction search in which the target does not pop out because the distractors share both black and vertical features with the target.

a black vertical line among black horizontal lines and white vertical lines (Fig. 1b). In this search, any single piece of bottom-up information is not unique to the target item, so the bottom-up constraints are weaker than in the feature search. Top-down constraints would be required to resolve the competition among the input items. In general, the control of spatial attention can be viewed as a ''biased competition'' model: Competition among bottom-up inputs is biased (i.e., some of the inputs are favored) by top-down inputs, such as a target template.

How is visual search controlled, particularly when search is inefficient and not determined by bottom-up information? Two possible control modes have been hypothesized for visual search. The serial search account proposes a sequential control of attention in which attention shifts from one item to the next until the target is found. The parallel search account proposes that attention is allocated to every item simultaneously, with less attention available for each item when a large number of items must be searched simultaneously. Much debate has surrounded the serial/parallel dichotomy, and current perspectives on the issue focus on the efficiency of search and not on an absolute dichotomy. Although there is evidence for a serial control process, this serial control process must be implemented in the brain's parallel hardware. Furthermore, if multiple attention systems exist, then some forms of attentional control may be more serial and others may be more parallel.

The control of spatial attention has also been examined by directing attention to a location before a target event occurs. In these ''spatial precuing'' tasks, observers are required to detect or identify a target item that appears at a peripheral location. Before the target appears, one location is precued by an arrow pointing to the location, a flash of light at the location, or some similar means. The subsequent target usually appears at the cued location (a ''valid'' trial), although it may occasionally appear at an uncued location (an ''invalid'' trial). Because the cue is usually valid, observers are motivated to attend to the cued location, and observers typically detect valid location targets more quickly and accurately than invalid location targets.

The type of precue used can bias attentional control to favor bottom-up factors or top-down factors. For example, if the precue is an abrupt luminance change (e.g., a flicker in the visual periphery), attention is automatically captured by the bottom-up input, irrespective of the observer's intentions. Such ''exogenous'' cues are extremely difficult to ignore and they are not interfered with by concurrent tasks such as a memory task. In contrast, if the precue is a symbol such as a centrally presented arrow that points to a location, attention will move to the cued location only if the observer wants to shift attention, and when attention does move it moves more slowly. These "endogenous" cues are not automatic: They can be ignored, and they are interfered with by concurrent tasks. Because endogenous cues are dependent on task-related goals and observers' expectancies, they involve top-down control processes. In many spatial precuing tasks, the control of attention involves a balance between bottom-up and top-down factors. Although bottom-up onset cues capture attention, they may be influenced by top-down attentional control settings (e.g., expectations of where the target will appear).

Was this article helpful?

0 0
Understanding And Treating Autism

Understanding And Treating Autism

Whenever a doctor informs the parents that their child is suffering with Autism, the first & foremost question that is thrown over him is - How did it happen? How did my child get this disease? Well, there is no definite answer to what are the exact causes of Autism.

Get My Free Ebook

Post a comment