Identifying stimuli of different perceptual categories in mixed blocks of trials: Evidence for cost in switching between computational processes.

Abstract

Responding to stimuli of different perceptual categories is usually faster when the categories are presented isolated from each other, in pure blocks, than when they are presented intermixed, in mixed blocks. According to criterion models, these perceptual mixing costs result from the use of a less conservative response criterion in pure than in mixed blocks. According to alternate processing models, mixing costs result from time-consuming switching in mixed blocks between different computational processes called on by the different perceptual categories. In 5 experiments, participants had to identify number stimuli of different categories. The results showed clear mixing costs whenever these categories differed in their assumed computational processing requirements but not when they differed on features that seemed trivial from a computational viewpoint. The results favor the alternate processing conception.In the pure/mixed design, the levels of an independent variable are presented both in isolation, in pure blocks, and intermixed across trials, in mixed blocks. The common finding is that people respond slower to either level of the variable in mixed blocks than in pure blocks, a difference referred to as mixing costs. Mixing costs have been reported for a large number of variables, including stimulus intensity, stimulus quality, stimulus-response compatibility, foreperiod, and task set (see Los, 1996, for a review). In this article I focus on perceptual mixing costs, which are found when a perceptual variable is presented in the pure/mixed design. For instance, Van Duren and Sanders (1988) and Los (1994) had participants identify intact and degraded digits (i.e., two levels of the variable "stimulus quality") both in pure and mixed blocks of trials. In those studies, both intact and degraded digits showed mixing costs in that they were more slowly identified in mixed blocks than in pure blocks.Mixing costs have recently evoked interest among psychologists because they provide insights into human control of action. I discuss two issues: (a) the source of mixing costs and (b) the mechanism underlying mixing costs. This article is mainly concerned with the second issue, but for the sake of completeness I begin by addressing the first issue in some detail.

A critical test of the failure-to-engage theory of task switching

“…Rogers and Monsell also found that this switch cost decreased when the interval between one stimulus and the next (the response-to-stimulus interval, or RSI) was increased but that there was an irreducible switch cost that was not removed, no matter how long RSI became. Other researchers have since replicated and extended these findings, using the repeated runs design (e.g., De Jong, 2000;Gilbert & Shallice, 2002;Karayanidis, Coltheart, Michie, & Murphy, 2003;Lien, Schweickert, & Proctor, 2003;Los, 1999;Nieuwenhuis & Monsell, 2002;Sohn & Anderson, 2003;Yeung & Monsell, 2003).…”

“…Some theories use the concept of "task-set inertia" (TSI), which has garnered substantial empirical support (e.g., Allport, Styles, & Hsieh, 1994;Gilbert & Shallice, 2002;Los, 1999;Yeung & Monsell, 2003). Theories including TSI propose that each of the two tasks to be performed has an associated task set, loosely defined as a mental state that must be prepared in order to accomplish that task.…”

Age-Related Changes in Task-Switching Components: The Role of Task Uncertainty

“…For example, Mayr (2001) used 8 different stimuli in one experiment in comparison to the 62 stimuli used in the present study. If the number of stimulus repetitions is large, then uncertainty (or interference) to select among currently relevant task sets will be increased and the response criterion is set more conservatively for switching conditions (Los, 1999), so that latencies would increase. The current literature does not indicate whether this would have an impact on task switching performance.…”

“…Thus, general switch costs are derived at a block level of analysis, whereas specific switch costs are computed at the trial level (for a similar distinction, see Mayr, 2001). Other researchers define general switch costs, which they call ''mixing costs,'' as the difference between performance of trial repetitions in switch blocks and performance in single-task blocks (e.g., Los, 1999). Differences between these two approaches are discussed in more detail at the end of Results and Discussion.…”

Phonological encoding is not contingent on semantic feature retrieval: An electrophysiological study on object naming.

“…The basic semantic feature, animacy, was either repeatedly classified (blocked classification mode) or randomly alternated by an additional semantic classification (mixed classification mode). Task mixing has been shown to affect performance in various tasks (for a review see Los, 1996Los, , 1999aLos, , 1999b, and we expected it to slow down the animacy classification.…”