Modeling task switching without switching tasks: a short-term priming account of explicitly cued performance

Switch costs in task switching are commonly attributed to an executive control process of task-set reconfiguration, particularly in studies involving the explicit task-cuing procedure. The authors propose an alternative account of explicitly cued performance that is based on 2 mechanisms: priming of cue encoding from residual activation of cues in short-term memory and compound cue retrieval of responses from long-term memory. Their short-term priming account explains the repeated cue encoding benefit, switch cost, reduction in switch cost with preparation time, and other effects. The authors develop a mathematical model of their priming account and fit it to data from 3 experiments, demonstrating that a set of basic psychological processes can produce several effects--including putative switch costs--without switching tasks.     

Can the task-cuing paradigm measure an endogenous task-set reconfiguration process?

In 6 task-cuing experiments, with 2 cues per task, the authors varied cue-stimulus interval to investigate G. D. Logan and C. Bundesen's (2003) claim that when cue repetition is controlled for, task-switch cost and its reduction with preparation are largely eliminated and hence cannot index an endogenous control process. Experiment 1 replicates their result, but Experiments 2 and 3, with similar designs, demonstrate a substantial task-switch cost, reducing with increasing cue-stimulus interval. Experiments 4 to 6 show that the critical difference is the probability of a task change: If it is kept low enough to discourage reconfiguration of task set unless and until the cue signals a task change, robust evidence for anticipatory task-set reconfiguration is obtained, even in Experiment 6, modeled closely on Logan and Bundesen's.     

What it costs to implement a plan: Plan-level and task-level contributions to switch costs

Four experiments were conducted to identify the costs of implementing a plan in the task span procedure, which requires subjects to retrieve the task to perform on the current target from a list of planned tasks in memory. Experiment 1 compared switch costs in the task span procedure with switch costs in the explicit task-cuing procedure, which presented cues indicating the task to perform on each target. Switch costs were greater in the task span procedure. Experiments 2-4 were designed to identify the sources of this difference. Experiment 2 showed that the requirement of establishing a correspondence between the list of task names and the list of targets contributed to switch costs. Experiment 3 showed that retaining lists of similar task names produced greater switch costs than did retaining lists of dissimilar task names. Experiment 4 showed that memory load had no effect on switch costs. The results are discussed in terms of the interaction between plan-level and task-level processing in the implementation of plans.     

Decomposing task-switching costs with the diffusion model

In four experiments, task-switching processes were investigated with variants of the alternating runs paradigm and the explicit cueing paradigm. The classical diffusion model for binary decisions (Ratcliff, 1978) was used to dissociate different components of task-switching costs. Findings can be reconciled with the view that task-switching processes take place in successive phases as postulated by multiple-components models of task switching (e.g., Mayr & Kliegl, 2003; Ruthruff, Remington, & Johnston, 2001). At an earlier phase, task-set reconfiguration (Rogers & Monsell, 1995) or cue-encoding (Schneider & Logan, 2005) takes place, at a later phase, the response is selected in accord with constraints set in the first phase. Inertia effects (Allport, Styles, & Hsieh, 1994; Allport & Wylie, 2000) were shown to affect this later stage. Additionally, findings support the notion that response caution contributes to both global as well as to local switching costs when task switches are predictable.     

Task-set switching and long-term memory retrieval

The authors tested the hypothesis of a close relationship between the intentional component of task-set switching ("advance reconfiguration;" R. D. Rogers & S. Monsell, 1995) and long-term memory (LTM) retrieval. Consistent with this hypothesis, switch costs are reported to be larger when the switched-to task involves high retrieval demands (i.e., retrieval of episodic information) than when it involves low retrieval demands (i.e., retrieval of semantic information). In contrast, switch costs were not affected by a primary-task difficulty manipulation unrelated to intentional retrieval demands (Experiment 2). Also, the retrieval-demand effect on switch costs was eliminated when time for advanced preparation or task cues explicitly specifying the task rules were provided (Experiment 3). Overall, results were consistent with the hypothesis that the intentional switch-cost component reflects the time demands of retrieving appropriate task rules from LTM.     

Linguistic attention control: attention shifting governed by grammaticized elements of language

In 2 experiments, the authors investigated attention control for tasks involving the processing of grammaticized linguistic stimuli (function words) contextualized in sentence fragments. Attention control was operationalized as shift costs obtained with adult speakers of English in an alternating-runs experimental design (R. D. Rogers & S. Monsell, 1995). Experiment 1 yielded significant attention shift costs between tasks involving judgments about the meanings of grammatical function words. The authors used a 3-stage experimental design (G. Wylie & A. Allport, 2000), and the emerging pattern of results implicated task set reconfiguration and not task set inertia in these shift costs. Experiment 2 further demonstrated that shift costs were lower when the tasks involved shared attentional resources (processing the same grammatical dimension) versus unshared resources (different grammatical dimensions). The authors discuss the results from a cognitive linguistic perspective and for their implications for the view that language itself can serve a special attention-directing function.     

The nature of residual cost in regular switch response factors

Two experiments are presented that compare the residual cost found when switching from one task to another under predictable conditions. The aim of the study was to explore the roles played by the stimulus, the response, or both in the process of the mental set reconfiguration necessary to switch between two tasks. The experiments tested [Rogers, R. D., & Monsell, S. (1995). Cost of a predictable switch between simple cognitive tasks. Journal of Experimental Psychology: General, 124, 207-231] stimulus-cued-completion hypothesis and [Schuch, S., & Koch, I. (2003). The role of response selection for inhibition of task sets in task shifting. Journal of Experimental Psychology: Human Perception and Performance, 29, 92-105] hypothesis of response selection as the key factor in the nature of switch cost. In the first experiment, two conditions were created that varied in terms of a Go/No-Go signal: The Go trials were a replication of [Tornay, F. J., & Milán, E. G. (2001). A more complete task-set reconfiguration in random than in predictable task switch. The Quarterly Journal of Experimental Psychology, 54A, 785-803 Experiment 3]; The No-Go trials were identical to the first condition, except that participants did not execute a response in the trial n-1 (Schuch & Koch, 2003). In addition, the percentage of Go and No-Go trials was manipulated. The results showed that the cost was significant only in the high Go signal-frequency case (Experiment 2), with an abrupt offset in Go trials and a gradual offset in No-Go trials. Based on the results of these experiments, it was concluded that the crucial factor to complete a mental set reconfiguration is response-related and not stimulus-related.     

Task-switching performance with 1:1 and 2:1 cue-task mappings: not so different after all

When task-switching studies use the task-cuing procedure with a 1:1 cue-task mapping, task switching and cue switching are confounded, which is problematic for interpreting switch costs. The use of a 2:1 cue-task mapping is a potential solution to this problem, but it is possible that introducing more cues may also introduce marked changes in task-switching performance. In 5 experiments involving 160 subjects, the authors compared performance with 1:1 and 2:1 mappings across several methodological changes. Differences in switch costs between mappings were small and, in most analyses, nonsignificant. In all experiments, both mappings yielded significant reductions in switch cost across cue-target interval, and there were significant cue-switching effects with the 2:1 mapping. A model of cue encoding fit the data from both mappings about equally well. Overall, task-switching performance was more similar than it was different between mappings, leading the authors to suggest that the use of a 2:1 mapping is a viable solution to the problem associated with a 1:1 mapping.     

Is task switching nothing but cue priming? Evidence from ERPs

Recent findings suggesting that switch costs in the task-cuing paradigm are largely attributable to a change in the task-indicating cue have been interpreted in terms of a priming model of task-switch costs (Logan & Bundesen, 2003). According to this explanation, participants do not actually switch task sets, but merely use a cue-stimulus compound to disambiguate competing response tendencies associated with bivalent stimuli. Here, we report an event-related potential (ERP) experiment that provides evidence against this notion. In a paradigm with a 2:1 mapping between cues and tasks, we show that cue-switch and task-switch effects are dissociable on a neurophysiological level, indicating that task switching is more than a switch in the task-indicating cue. Moreover, a systematic analysis of the ERPs during the cue-stimulus interval suggests that updating processes can run in advance, before the stimulus is presented.     

任务切换理论争议新证据

任务切换是研究认知控制的主要范式之一。大量研究发现切换试次比重复试次的反应时更长,错误率更高,这种差异称为切换代价。任务切换时所产生切换代价的理论解释主要有惯性论、重构论和联结论。近十年来,这些理论均获得新的实验支持和发展,但其争议依旧,没有哪一理论能成功解释任务切换的所有效应。未来研究可以建立整合模型,以准确描述切换代价产生的认知机制。     

Clever homunculus: is there an endogenous act of control in the explicit task-cuing procedure?

Does the explicit task-cuing procedure require an endogenous a act of control? In 5 experiments, cues indicating which task to perform preceded targets by several stimulus onset asynchronies (SOAs). Two models were developed to account for changes in reaction time (RT) with SOA. Model 1 assumed an endogenous act of task switching for cue alterations but not for cue repetitions. Model 2 assumed no such act. In Experiments 1 and 2, the cue was masked or not masked. Masking interacted underadditively with repetition and alternation, consistent with Model 2 but not Model 1. In Experiments 3 and 4, 2 cues were used for each task. RT was slower for task repetition than for cue repetition and about the same as RT for task alternation, consistent with Model 2 but not Model 1. The results suggest that the explicit task-cuing procedure does not require an endogenous act of control.     

Task switching and the measurement of “switch costs”

The measurement of “switch costs” is held to be of interest because, as is widely believed, they may reflect the control processes that are engaged when subjects switch between two (or more) competing tasks. [In task-switching experiments, the reaction time (RT) switch cost is typically measured as the difference in RT between switch and non-switch (repeat) trials.] In this report we focus on the RT switch costs that remain even after the subject has had some time to prepare for the shift of task, when the switch cost may be approximately asymptotic (so-called residual switch costs). Three experiments are presented. All three experiments used Stroop colour/word, and neutral stimuli. Participants performed the two tasks of word-reading and colour-naming in a regular, double alternation, using the “alternating runs” paradigm (R. D. Rogers & S. Monsell, 1995). The experiments were designed to test the hypothesis that RT switch costs depend on a form of proactive interference (PI) arising from the performance of a prior, competing task. A. Allport, E. A. Styles and S. Hsieh (1994) suggested that these PI effects resulted from “task-set inertia”, that is, the persisting activation-suppression of competing task-sets, or competing task-processing pathways. The results confirmed the existence of long-lasting PI from the competing task as a major contributor to switch costs. Non-switch trials, used as the baseline in the measurement of switch costs, were also shown to be strongly affected by similar PI effects. However, task-set inertia was not sufficient to account for these results. The results appeared inconsistent also with all other previous models of task switching. A new hypothesis to explain these between-task interference effects was developed, based on the stimulus-triggered retrieval of competing stimulus-response (S-R) associations, acquired (or strengthened) in earlier trials. Consistent with this retrieval hypothesis, switch costs were shown to depend primarily on the S-R characteristics of the preceding task (the task that was switched from) rather than the upcoming task. Further, the effects of the other, competing task were found to persist over many successive switching trials, affecting switch costs long after the stimulus overlap (and hence the principal S-R competition) between the current tasks had been removed. Switch costs were also found to be affected by recent, item-specific experience with a given stimulus, in either the same or the competing task. Finally, the results showed that switch costs were massively affected by the ratio of the number of prior trials, in response to the same stimuli, that had implemented either the currently intended or the competing S-R mappings. None of these effects are predicted by current models of residual switch costs, which appeal to the differences in control processes assumed to be engaged in switch versus non-switch trials. Received: 31 March 1999 / Accepted: 23 July 1999     

Defining task-set reconfiguration: The case of reference point switching

Some researchers argue that task switching involves task-set reconfiguration-that is, changing the existing task set to perform a different task. Although this idea is intriguing, it is often unclear what is reconfigured and which processes are involved in reconfiguration. We addressed these issues by defining task sets, identifying differences between them, and obtaining evidence that was diagnostic of reconfiguration. In two experiments, subjects performed relative judgment tasks that involved comparing a target with a reference point that repeated or switched across trials. The task was the same on every trial, but the task set was not, because a task-set parameter-the reference point-had to be changed. Target-reference point distance effects that changed with the reference point provided diagnostic evidence that reconfiguration occurred, and this reconfiguration led to switch costs. We discuss how our approach can be generalized to define reconfiguration more clearly in other task-switching situations. nt|mis|This research was supported by Grants BCS 0133202 and BCS 0446806 from the National Science Foundation to G.D.L.     

Effects of associative learning on age differences in task-set switching

Costs of switching between tasks may disappear when subjects are able to learn associations between tasks, stimuli, and responses (cf. Rogers, R. D., & Monsell, S. (1995). Costs of a predictable switch between simple cognitive tasks. Journal of Experimental Psychology: General, 124, 207-231). The first aim of this study was to examine this possibility by manipulating stimulus-set size. We expected that costs of switching between tasks would be strongly reduced under conditions of small stimulus-set sizes (n=4) as compared to large stimulus-set sizes (n=96) with increasing time on task. The second aim was to determine whether younger as well as older adults were able to create associations between task components. As age differences in task switching are often found to be larger when response mappings are incompatible we also investigated interactions with response compatibility. Results of our study indicated that practice effects on switch costs were much more pronounced for small than large stimulus-set sizes, consistent with the view that the strength of associations between task components facilitates task switching. Furthermore, we found that practice benefits on task switching for small stimulus-set sizes were sensitive to age and response compatibility. In contrast to younger adults, who showed a reduction of switch costs for both response mapping conditions, older adults showed a reduction of switch costs only when response mappings were compatible. That is, older adults showed less associative learning when the currently irrelevant task feature had to be suppressed, supporting the view that older adults have primarily problems in separating overlapping task-set representations.     

The psychophysics of contingency assessment

The authors previously described a procedure that permits rapid, multiple within-participant evaluations of contingency assessment (the "streamed-trial" procedure, M. J. C. Crump, S. D. Hannah, L. G. Allan, & L. K. Hord, 2007). In the present experiments, they used the streamed-trial procedure, combined with the method of constant stimuli and a binary classification response, to assess the psychophysics of contingency assessment. This strategy provides a methodology for evaluating whether variations in contingency assessment reflect changes in the participant's sensitivity to the contingency or changes in the participant's response bias (or decision criterion). The sign of the contingency (positive or negative), outcome density, and imposition of an explicit payoff structure had little influence on sensitivity to contingencies but did influence the decision criterion. The authors discuss how a psychophysical analysis can provide a better understanding of findings in the literature such as mood and age effects on contingency assessment. They also discuss the relation between a psychophysical approach and an associative account of contingency assessment.     

Out with the old, in with the new: More valid measures of switch cost and retrieval time in the task span procedure

Two experiments provided new measures of switch cost and retrieval time in the task span procedure. In Experiment 1, subjects were given lists of six task names to remember, followed by six targets on which to perform the tasks named in the list. The lists contained alternations and repetitions, and switch costs were estimated by comparing reaction time (RT) on alternation and repetition trials. The experiment also included memory span and single task conditions, so switch costs could be estimated by subtracting the sum of the RT in those conditions from the task span RT, as in the original report (Logan, 2004). The data suggested that the original measure of switch cost was invalid and that the new measure was preferable. In Experiment 2, subjects performed each task on the list twice. Retrieval was required on the first but not on the second trial in each pair. Retrieval time was estimated by comparing the RT on trials that required retrieval with trials that did not require retrieval. This measure was more valid than the RT in the memory span condition of Experiment 1, which was used in the original report.     

准备时间和预先信息对任务切换的影响

两个实验采用任务线索范式,考察了准备时间和预先信息对任务切换的影响。实验1的准备时间为可预测,实验2的为不可预测,以考察准备时间的可预测性对任务切换的影响。以被试执行重复任务、切到任务和切离任务的正确反应时为主要指标,结果：（1）在准备时间可预测和不可预测两种条件下,准备时间和预先信息之间存在显著交互作用。在可预测条件中,重复和切到任务的反应时随着准备时间的增加而减少,但切到代价无显著变化,切离任务的反应时不受准备时间的影响,切离代价表现出反准备效应;不可预测条件下,三种任务的反应时都随着准备时间的增加而减少,但是切到和切离代价表现出反准备效应;（2）切离任务的反应时和代价显著高于切到任务的反应时和代价;（3）在这两种条件下,都没有出现切换代价的准备效应     

自主任务转换中的重构和干扰

任务转换是研究执行功能的常用范式。任务转换通常伴随着转换代价：执行转换任务比重复任务的反应时更长、错误率更高。转换代价可能反映了任务设置重构(重构理论), 也可能表明任务之间存在干扰(干扰理论)。与任务线索范式相比, 自主任务转换范式更具生态效度, 而且不仅能获得转换代价这个传统结果, 还引入了任务选择比例、任务转换率等新指标, 其结果倾向于支持重构理论。此外, 新近研究指出自主任务转换可能也包含干扰的作用。未来, 应通过进一步改进实验范式等方法, 实现两大理论的融合。     

What matters in the cued task-switching paradigm: Tasks or cues?

Schneider and Logan (2006) recently showed that cue-switch and task-switch costs are sensitive to the relative probability of cue switches versus task switches. From this they concluded that taskswitch costs reflect priming of cue-cue transitions rather than actual task-switching operations. However, because this design confounded probability of specific cue transitions with probability of task switches, the results could also reflect task-switch-level adjustments. The present experiment (N = 80) pits the critical prediction of the cue-priming account, namely that costs for high-probability cue-cue transitions are smaller than for low-probability cue-cue transitions, against the main prediction of the switch-probability account, namely that switch probability, irrespective of specific cue-cue transitions, determines switch costs. Whereas the cue-priming prediction was rejected, a specific version of the probability account—that subjects are sensitive to the probability of a task switch, given a cue switch—was fully confirmed. Thus, tasks are in fact the critical representational units that determine task-switch cost.     

The surface structure and the deep structure of sequential control: What can we learn from task span switch costs?

A large component of response time switch costs in the cued task-switching paradigm is linked to cue changes without task changes, suggesting costs might reflect passive priming rather than endogenous control. In contrast, the task span procedure requires subjects to guide task selection via sequences of memorized task cues and therefore may be better suited to reflect endogenous switch processes (Logan, 2004). The present experiments combined the task span procedure with a 2:1 mapping between cues and tasks, allowing separation of cue-switch costs from true task-switch costs. Replicating findings with the cued task-switching paradigm, results showed both substantial cue-switch costs and actual task-switch costs (Experiments 1 and 2) as well as sensitivity of cue-switch costs, but not of task-switch costs, to opportunity for preparation (Experiment 2). Apparently, simple action plans use “surface level” phonological or articulatory codes that contain no task information. These results suggest that the distinction between cue-related and task-related processes is critical no matter whether tasks are cued exogenously or endogenously.