Academic journal article Perception and Psychophysics

Searching for an O in an Array of Cs: Eye Movements Track Moment-to-Moment Processing in Visual Search

Academic journal article Perception and Psychophysics

Searching for an O in an Array of Cs: Eye Movements Track Moment-to-Moment Processing in Visual Search

Article excerpt

We examined how closely the underlying cognitive processing in a visual search task guides eye movements by comparing two different search tasks. In the extended search task, participants searched for an O in eight clusters of Landolt Cs with varying gap widths (four characters per cluster, arranged to look like words in text). In the single-cluster task, participants searched a single cluster (identical to the ones in the extended search). The key manipulation was gap size; although gap orientation for the distractors varied within a cluster, gap size was constant within a cluster but differed in size from cluster to cluster. The principal findings were that (1) gaze durations in the extended search were almost completely a function of the difficulty of the cluster (i.e., the gap size of the Cs) and (2) the effect of gap size on gaze durations in the extended search was very similar to its effect on response times in the single-cluster search. Thus, it appears that eye movements in the search task are determined almost exclusively by the ongoing cognitive processing on that cluster.

Searching for a target object is a task people do many times every day, and consistent with the importance of the task, it is one of the most widely studied areas hi cognitive psychology (Duncan & Humphreys, 1989; Schneider & Shifrrin, 1977; Treisman & Gelade, 1980; Wolfe, Cave, & Franzel, 1989; see Wolfe, 1998, for a review). However, although most visual searches that people perform outside the laboratory involve movements of the eyes (and head) that aid hi the search, a large majority of the studies of visual search do not record eye movements. Often, in these studies, the problem of not knowing where the eyes are fixating is dealt with by presenting an array that is to be searched briefly, so that it is seen within a single fixation of the eyes (e.g., Cave & Zimmerman, 1997; Kim & Cave, 1995). However, as a result, these studies involve quite small arrays that are different from those involved in most visual search tasks in the environment. Thus, we have an incomplete knowledge of what occurs hi the visual search task hi the environment when it involves eye movements and, perhaps, more global search strategies (Findlay, 1995; Findlay & Gilchrist, 1998).

In some recent studies, not only have eye movements been permitted, but also attempts have been made to incorporate eye movements into the study of visual search (e.g., Bertera & Rayner, 2000; Findlay, 1997; Findlay, Brown, & Gilchrist, 2001; Greene & Rayner, 200Ia, 2001b; Motter & Belky, 1998; Scialfa & Jofte, 1998; Zelinsky, 1996). One problem that arises with attempting to use eye movements as a tool is that the objects in typical search arrays are randomly distributed across the entire display. As a result, the experimenter often does not know what kinds of global strategies the person is employing (if any) and, thus, does not know what the participant intends to do. In addition, with the usual crowded search array, it is quite difficult to determine which element or elements the participant intended to fixate. Both of these problems make it difficult to relate the eye movement record to ongoing cognitive processing.

In a few studies, an attempt has been made to examine the relationship between the currently fixated information and the information in a subsequent fixation target, using post hoc analyses. Hooge and Erkelens (1999) found that fixation time on a particular item was highly influenced by the difficulty of the visual discrimination (i.e., how difficult the item was to discriminate from the target), with items more similar to the target being viewed longer than less similar items. They also found that people tended to direct eye movements to distractors that were more closely related to the target, indicating that targets of saccades were not random. In addition, the length of a fixation on an item directly influenced the selectivity of the next eye movement: the longer the fixation on the current item, the more similar to the target the next item fixated tended to be. …

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.