Academic journal article Memory & Cognition

Influence of Display Type and Cue Format on Task-Cuing Effects: Dissociating Switch Cost and Right-Left Prevalence Effects

Academic journal article Memory & Cognition

Influence of Display Type and Cue Format on Task-Cuing Effects: Dissociating Switch Cost and Right-Left Prevalence Effects

Article excerpt

In previous studies of task switching and of the right-left prevalence effect, researchers have used a procedure in which the stimulus on each trial occurs in one of four quadrants, and responses are made by pressing one of two diagonally arranged response keys. Across these studies, discrepant effects of cuing interval have been reported. These discrepancies need clarification because cue-based preparation effects are frequently interpreted as reflecting cognitive control processes. In Experiment 1, we compared performance with display formats used by Meiran (1996; Meiran, Chorev, & Sapir, 2000; small display, cues located at sides of quadrants and displayed until response) to study task switching and by Proctor and colleagues (Proctor, Koch, & Vu, 2006; large display, cues located at center of display and shown until target onset) to study right-left prevalence. We found a decrease in task-switch cost with increasing cuing interval with the Meiran display, but not with the Proctor display, but the right-left prevalence effect was of similar size for the two display formats and was relatively unaffected by cuing interval. To determine the basis of the discrepant task-switch results, we used small and large displays in Experiments 2 and 3, respectively, with cue type and cue offset varied. With the side cues, the task-switch cost decreased in all cases at the longer cuing interval, but with the centered cues, it decreased only when the display size was small. Thus, the effects of cuing interval on switch costs are sensitive to variations of display characteristics, whereas cuing interval and display characteristics have little influence on the right-left prevalence effect, suggesting that prevalence effect is due to processes that are independent from those producing the switch cost.

When stimuli and responses are arrayed in four quadrants of a real or imaginary square, each location can be represented on two dimensions: horizontal and vertical. Two-dimensional arrays have been used in order to investigate issues concerning switching between tasks for which the horizontal or vertical dimension is defined as relevant (e.g., Meiran, 1996), as well as issues of whether maintaining stimulus-response (S-R) compatibility (i.e., correspondence of stimuli and their assigned response locations) is more important along the vertical or horizontal dimension (e.g., Rubichi, Vu, Nicoletti, & Proctor, 2006). In both lines of research, participants are cued or instructed to respond along one dimension and ignore the other. Task-switching studies show robust switch costs, whereas compatibility studies typically show a right-left prevalence effect. However, despite the structural similarities of the experimental designs, and although some similar results have been obtained in the tasks used for investigating task switching and right-left prevalence, the results have not been entirely consistent.

In particular, manipulations of cuing intervals in taskswitching studies have shown that a long cuing interval (time between cue onset and onset of the target stimulus) leads to shorter reaction time (RT) and reduced switch costs than does a short cuing interval. These cuing interval effects have been interpreted as being a signature of advance preparation, which is an important aspect of cognitive control (see, e.g., Meiran, 1996; Monsell, 2003). In contrast, manipulations of cuing interval in compatibility studies have found no clear reduction of the right-left prevalence effect or of switch costs, and sometimes even longer RT at long cuing intervals than at short cuing intervals (Proctor, Koch, & Vu, 2006, Experiment 2). In the present study, we examine these discrepancies in effects of cuing interval.

Task switching was studied by Meiran (1996; Meiran, Chorev, & Sapir, 2000) with a display that consisted of a 2 × 2 grid, in which the target stimulus could appear in one of the four quadrants. The responses were arrayed diagonally along the positive ("1" and "9" keys) or negative ("7" and "3" keys) axis of the numeric keypad of a computer keyboard. …

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.