ArticlePDF Available

Costs of a Predictable Switch Between Simple Cognitive Tasks

Authors:
Robert Rogers at School of Psychology
Robert Rogers
  • School of Psychology
Stephen Monsell at University of Exeter
Stephen Monsell
Download full-text PDF
Read full-text
Download citation

Abstract

In an investigation of task-set reconfiguration, participants switched between 2 tasks on every 2nd trial in 5 experiments and on every 4th trial in a final experiment. The tasks were to classify either the digit member of a pair of characters as even/odd or the letter member as consonant/vowel. As the response–stimulus interval increased up to 0.6 s, the substantial cost to performance of this predictable task-switch fell: Participants could partially reconfigure in advance of the stimulus. However, even with 1.2 s available for preparation, a large asymptotic reaction time (RT) cost remained, but only on the 1st trial of the new task. This is attributed to a component of reconfiguration triggered exogenously, i.e., only by a task-relevant stimulus. That stimuli evoke associated task-sets also explains why RT and switch costs increased when the stimulus included a character associated with the currently irrelevant task. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Content uploaded by Robert Rogers
Author content
All content in this area was uploaded by Robert Rogers on Sep 09, 2015
Content may be subject to copyright.
A preview of this full-text is provided by American Psychological Association.
Learn more
Content available from Journal of Experimental Psychology General
This content is subject to copyright. Terms and conditions apply.
... These findings have influenced cognitive modeling frameworks, including Brain-Inspired Cognitive Architectures (BICA) [6], [7], which attempt to replicate executive function processes in artificial systems. However, many existing models focus on task-switching in highly controlled experimental conditions [3], relying on cue-based paradigms or human subject trials [8], limiting their applicability to real-world multitasking scenarios. This tutorial introduces a computational approach for simulating task-switching behavior using biologically plausible spiking neural networks (SNNs), a thirdgeneration neural network model inspired by the brain's action potential dynamics [9]. ...
... Our work extends these studies by developing a two-layered SNN model that processes real-world stimuli, rather than relying on traditional cue-based switching paradigms. Inspired by task-switching cost experiments [8], we investigate how switching intervals impact neural adaptation and performance, providing insights into how networks reconfigure in response to new tasks. By leveraging SNNs with STDPbased learning, we demonstrate how biologically plausible mechanisms can encode dynamic task transitions, supporting the study of cognitive flexibility in more realistic, data-driven contexts. ...
... where R neuron (t) represents the global firing rate of excitatory neurons. Longer τ s values lead to smoother adaptation, as observed in cognitive task-switching studies [8], [28]. ...
Neural Models of Task Adaptation: A Tutorial on Spiking Networks for Executive Control
Preprint
Full-text available
  • Mar 2025
Understanding cognitive flexibility and task-switching mechanisms in neural systems requires biologically plausible computational models. This tutorial presents a step-by-step approach to constructing a spiking neural network (SNN) that simulates task-switching dynamics within the cognitive control network. The model incorporates biologically realistic features, including lateral inhibition, adaptive synaptic weights through unsupervised Spike Timing-Dependent Plasticity (STDP), and precise neuronal parameterization within physiologically relevant ranges. The SNN is implemented using Leaky Integrate-and-Fire (LIF) neurons, which represent excitatory (glutamatergic) and inhibitory (GABAergic) populations. We utilize two real-world datasets as tasks, demonstrating how the network learns and dynamically switches between them. Experimental design follows cognitive psychology paradigms to analyze neural adaptation, synaptic weight modifications, and emergent behaviors such as Long-Term Potentiation (LTP), Long-Term Depression (LTD), and Task-Set Reconfiguration (TSR). Through a series of structured experiments, this tutorial illustrates how variations in task-switching intervals affect performance and multitasking efficiency. The results align with empirically observed neuronal responses, offering insights into the computational underpinnings of executive function. By following this tutorial, researchers can develop and extend biologically inspired SNN models for studying cognitive processes and neural adaptation.
View
... Our observations imply that there was no residual switch cost after cue-only trials in these experiments, suggesting the possibility that the switch cost after cue-only trials may be qualitatively different from that after completed trials. Rogers and Monsell (1995) attributed the presence of the residual switch cost to the need for an exogenous reconfiguration process following target stimulus onset; it may be that there is no need for exogenous reconfiguration following cue-only trials which, by their nature, involve no target. Or it might be that the increased persistence of the switch cost after a completed trial reflects persisting inhibition of the competing task as a consequence of selecting a response to the target on these trials (Schuch & Koch, 2003). ...
... Alternatively, Schuch and Koch (2003) proposed that two task-sets can be active in parallel prior to response selection. These two task-sets may not interfere with each other directly (i.e., activation of the new task might not compete with activation of the old task as just described) but by being active concurrently they could potentially still create conflict between different responses, slowing response selection (cf. the responsecongruity effect often observed in cued task switching; e.g., Rogers & Monsell, 1995;Yamaguchi & Proctor, 2011). These possibilities could explain the short-lived nature of the switch cost after cue-only trials, as compared to the switch cost after completed trials, as observed in the current set of experiments. ...
Does preparation generate the cost of task switching? A recipe for a switch cost after cue-only trials
Article
Full-text available
  • Mar 2025
  • PSYCHOL RES-PSYCH FO
A switch cost can be observed in cued task-switching on trials that follow a cue-only trial, which presents a task cue indicating a task to be performed but does not present a target stimulus to be responded to. This finding has provided important implications as to the source of the performance cost that emerges when switching tasks. However, cue-only trials differ from completed trials (for which the target occurs and is responded to) in several task parameters, and there are a few untested assumptions about a task-switch cost after cue-only trials, which restricted the conditions under which cue-only trials have been used. The present study first examined whether a switch cost emerged after cue-only trials when cue-only trials were matched with completed trials in as many task parameters as possible, and found that an expected switch cost following cue-only trials was absent in response time. In the subsequent six experiments, we explored critical task parameters to obtain a switch cost after cue-only trials. The present results indicate that the use of a short preparation interval was an important factor and that the switch cost was more short-lived and dissipated more quickly after cue-only trials than after completed trials. These outcomes are consistent with the proposal that there are at least two sources of a task-switch cost, one that originates from processing a task cue and another that originates from performing a cued task. Early processes of task preparation (e.g., cue or task identification) may be sufficient to produce the switch cost after cue-only trials, but response-related processes might generate a more persistent switch cost.
View
... Fatigue induction. The fatigue induction in the experimental condition consisted of three different cognitive tasks (symbol counter task; SCT; Garavan et al., (2000), number letter task; NLT; Rogers & Monsell, (1995), math effort task; MET; Engle-Friedman et al., (2003)), adaptive to the participant's ability level. Each lasted for 17 minutes (total of 51 minutes on task). ...
... . The number-letter task (Rogers & Monsell, 1995) requires participants to either discriminate between even or odd numbers of consonants or vocals in pairs of letters (e.g., '4L') and numbers based on the font color of the pair (green font for number task and purple font for letter task). Participants were excluded if they made more errors than correct responses (below or at chance level response). ...
And yet They Tire: Cognitive Fatigue Facilitates Task Disengagement and Worsens Performance
Preprint
Full-text available
  • Mar 2025
Most scientists consider the ego depletion effect—worsened task performance after prior cognitive demand due to the exhaustion of a specific and limited resource—invalid. Yet, the negative effect of prior task engagement on self-control performance is a salient human experience. In the current study we aim to dissociate the effect of prior engagement on self-control performance from a limited resource account by applying recent advancements in fatigue literature to self-control empirically (the psychobiological model of endurance). The present study (N = 321) utilized understanding garnered from these recent advancements to overcome shortcomings in prior designs by (1) using an adaptive and longer (1h) fatigue induction and (2) the option to actively disengage from the main task in a dual-task design, allowing the theoretical influence prior task engagement has on motivation to influence performance through changes in persistence. Findings illustrated that participants that completed the fatigue induction protocol showed large post-induction differences in subjective fatigue (d = 0.83, BF10 > 1,000) and shorter persistence (d = -0.31, BF10 = 8.66), as well as lower performance (d = -0.31, BF10 = 9.21) in the main task, even when controlling for persistence differences, as compared to their control counterparts. Exploratory analysis further highlighted the predictive value of subjective post-induction fatigue and showed that performance is impaired, even when controlling for persistence. As suggested by recent findings in the fatigue literature, the present study supports the transition of self-control research away from a limited resource account and towards a perceived fatigue account.
View
... Masked stimuli, which are not consciously perceived, have been shown to trigger a wide range of cognitive control processes, including the activation of mental task representations (for reviews, see; Ansorge et al., 2014;Van Gaal & Lamme, 2012). The activation of such task representations, so-called task sets (Monsell, 2003;Rogers & Monsell, 1995), therefore did not appear to be bound to the conscious, intentional preparation for a task. Consciously initiated task sets are typically considered to be updated during the course of the implementation of a new task set, when one has to switch to a new task (called task set reconfiguration; Kiesel et al., 2010;Monsell, 2003;Vandierendonck et al., 2010). ...
Task set reconfiguration following masked and unmasked task cues
Article
Full-text available
  • Mar 2025
  • CONSCIOUS COGN
Numerous previous studies have shown that masked stimuli trigger cognitive control processes, including the activation of task sets, and thereby affect subsequent processing. However, it has not been directly tested whether unconsciously activated task sets also need to be reconfigured when switching to a new task, as has been shown for consciously triggered task sets. To test whether unconsciously activated task sets are subject to inhibitory processes, we measured n-2 repetition costs following masked cue presentation in a task switching design. We furthermore simultaneously assessed event-related potentials (ERPs) to gain additional insights into task set reconfiguration processes. Results showed that task sets were inhibited following the presentation of an unmasked task cue, as reflected by n-2 repetition costs. Furthermore, a cue-locked positivity ERP component indicated that task sets were reconfigured following both mere task preparation and task execution. In contrast, no evidence for a reconfiguration of unconsciously activated task sets was observed following masked cue presentation in either measure. Thus, task set reconfiguration, including the inhibition of a task set, is likely tied to conscious task set activation, suggesting that an unconscious process – once initiated – is not terminated by inhibitory processes.
View
... Box plots illustrate the interquartile ranges of the features along with the medians (white lines) and means (white points). ns: p > 0.05, *: p < 0.05 attention from one rule, feature, or task to another during task execution [75,76], which is crucial for adapting to changing demands, managing complex environments efficiently, and handling multitasking scenarios [77]. Previous studies have shown that, compared to control groups, individuals with ASD exhibit greater switching costs when processing complex, unpredictable (i.e., variable), and implicit (i.e., not guided by explicit rules) social-emotional transition tasks (emotional flexibility tasks) [78]. ...
Altered processing of consecutive changeable emotional voices in individuals with autistic traits: behavioral and ERP studies
Article
Full-text available
  • Mar 2025
Background Similar to individuals with autism spectrum disorder (ASD), individuals with autistic traits are expected to exhibit alterations in emotion recognition. However, many previous studies using single emotional stimuli did not observe these alterations in such individuals. Given that consecutive changeable emotional stimuli are more common in social interactions than single emotional stimuli, impaired mental processing of consecutive changeable emotions may be a key factor underlying the social interaction challenges faced by these individuals. Methods The present research aimed to investigate the behavioral and neural responses to consecutive changeable emotional voices in individuals with autistic traits through two studies (Study 1 and Study 2). Based on the autism-spectrum quotient (AQ) scores, participants were categorized into two groups: the High-AQ and the Low-AQ groups. In Study 1, both groups were asked to judge a single emotional voice (positive, negative, or neutral; S1) presented in each trial in Task 1, or the last presented emotional voice (S3) in a triplet of stimuli (S1-S2-S3, trains of three consecutive changeable emotional voices) in Task 2. In Study 2, both groups were instructed to passively listen to the stimulus triplet (S1-S2-S3), and event-related potential (ERP) technology was used to investigate their neural responses to each stimulus. Results No significant group difference was found in response to S1 voices in either Study 1 or Study 2. However, the High-AQ group elicited higher arousal levels (Study 1) and larger P2 amplitudes (Study 2) in response to S3 emotional voices (positive and negative) compared to the Low-AQ group. Conclusion These findings reveal that individuals with autistic traits may exhibit alterations in their processing of consecutive changeable emotions in the auditory modality.
View
... Responses involved four keys, corresponding to four different figures (circle, triangle, square, cross). In the number/letter task (adapted from Rogers & Monsell, 1995) with 256 experimental trials, participants saw pairs of a number and a letter, deciding based on the stimulus color whether the number was less/more than five (red) or if the letter was a vowel/consonant (green). Responses were made using two keys. ...
Trait characteristics of midfrontal theta connectivity as a neurocognitive measure of cognitive control and its relation to general cognitive abilities
Preprint
Full-text available
  • Mar 2025
Understanding the neurocognitive basis of cognitive control and its relationship with general cognitive ability is a key challenge in individual differences research. This study investigates midfrontal theta connectivity as a neurocognitive marker for individual differences in cognitive control. Using electroencephalography (EEG), we examined midfrontal global theta connectivity across three distinct cognitive control tasks in 148 participants. Our findings reveal that midfrontal theta connectivity can be modeled as a trait-like latent variable, indicating its consistency across tasks and stability over time. However, the reliability of the observed measures was found to be low to moderate, suggesting substantial measurement error. We also replicated previous results, finding a strong correlation (r = 0.64) between midfrontal theta connectivity and cognitive abilities, especially during higher-order stages of information processing. We disentangled the specific cognitive processes contributing to this relationship by employing a task-cueing paradigm with distinct cue and target intervals. The results indicated that only theta connectivity during response-related processes, not during cue-evoked task-set reconfiguration, correlated with cognitive abilities. These insights significantly advance theoretical models of intelligence, highlighting the critical role of specific aspects of cognitive control in cognitive abilities.
View
... In previous studies, performance-based experimental tasks have been prevalently employed as a means of assessing attention. These tasks include, for instance, the Sustained Attention to Response Task (Robertson et al., 1997), Metronome Response Task (Seli et al., 2013), Continuous Performance Test (Rosvold et al., 1956), A-X Continuous Performance Test (Wohlberg & Kornetsky, 1973), Gradual-Onset Continuous Performance Task (Esterman et al., 2013;Rosenberg et al., 2016), Attention Network Task (Fan et al., 2002), Number-Letter Task (Rogers & Monsell, 1995), Local-Global Task (Navon, 1977), Vigilance Task (Carter et al., 2013), and Multiple Object Tracking Task (Oksama & Hyönä, 2004). Furthermore, self-reported scales have also been utilized to measure attention, including the Adult ADHD Self-Report Scale (Kessler et al., 2005), Attention-Related Cognitive Errors Scale (Cheyne et al., 2006), Attentional Control Scale (Derryberry & Reed, 2002), Barratt Impulsiveness Scale-Attention Only (Patton et al., 1995), Mindful Attention Awareness Scale-Lapses Only (Carriere et al., 2008), Cognitive Failures Questionnaire-Attention Only (Zhou et al., 2016), and Academic Attention Problems Scale (Smits & Vorst, 1990). ...
The relationship between media multitasking and attention: a three-level meta-analysis
Article
Full-text available
  • Mar 2025
  • CURR PSYCHOL
Media multitasking, defined as the concurrent or rapid switching between multiple tasks, with at least one involving media use, has become increasingly prevalent in daily life. Research has suggested a possible correlation between media multitasking and attention. Nonetheless, the supporting evidence remains inconsistent. This study utilized a three-level meta-analysis, following PRISMA guidelines, to systematically explore this association and examine potential moderating factors. The analysis included 33 studies involving 36,861 participants and yielded 110 effect sizes. The results revealed a significant positive correlation between media multitasking and attention (r = 0.19, p < 0.001; a positive effect size indicates that greater levels of media multitasking are associated with poorer attention). Furthermore, this relationship was moderated by type of attention measurement, but was not by gender, culture, publication year, publication status, and type of media multitasking measurement. This study makes a theoretical contribution by offering a provisional conclusion regarding the relationship between media multitasking and attention, while also providing practical insights for the prevention and intervention of attention problems in the digital age.
View
... However, although previous researchers have discussed the difficulty of changing dfferent facets of a task from one trial to the next (Wulf & Lee, 1993;Sekiya, Magdl, Sidaway, & Anderson, 1994), the proactive interference or negative transfer between tasks has not yet been considered. Research on task switching (Rogers & Monsell, 1995) suggests further examination of this factor may provide useful insight into contextual interference effects. The fact that all the differences between practice conditions described above are found in outcome rather than form scores also deserves consideration. ...
Alternating versus Blocked Practice in Learning a Cartwheel
Article
Full-text available
This investigation examined the generalizability of contextual interference to learning the cartwheel in gymnastics. 32 participants ages 17 to 26 years completed five stages of practice of two versions of the skill, one in which the left hand led and one in which the right hand led, providing a total of 192 practice trials. Practice of these two versions of the task was completed in either a blocked or alternating order. Tests of retention and transfer were completed 20 min. and 1 wk. after acquisition. Outcome (errors) and form scores were derived from participants' performance. Alternating practice resulted in poorer acquisition, retention, and transfer performance as measured by outcome scores. The contextual interference effect did not generalize to this task, a finding which is attributed to task complexity and the interference associated with practicing the two tasks together.
View
View
Task Shifting in Dual-Task Settings
Article
Full-text available
When a participant is asked to perform two tasks in alternation, their mean reaction times were slower than when they performed the same tasks repeatedly. This “shift cost” has been hypothesized to reflect the time course of a single central executive that exerts control over thought and actions in task shifting. This study attempted to test this hypothesis using dual-task methodology. Participants were asked to carry out both a subtracting task and a rule-shifting task simultaneously. The main interest is to examine the effect of dual task on the magnitude of shift cost. The results showed that performing a concurrent subtracting task significantly interfered with the shifting operation resulting in over-additive time cost for shifting of task set. We further suggest that such interference may arise from the competition between activations of various rules.
View
The Cuing and Priming of Cognitive Operations
Article
Full-text available
  • Feb 1987
The effect on performance of advance information about the specific cognitive operations to be performed on a stimulus was investigated in two experiments using cues (information useful and necessary for performance) and primes (information useful but not necessary for performance). In the first experiment, a cue presented prior to a digit stimulus indicated whether the digit was to be classified as odd or even, or low (less than 6) or high (greater than 5). Results showed that performance improved with increasing time between cue and digit and with practice. A Stroop-like asymmetric interference of the low-high operation on the odd-even operation was also observed. In the second experiment, a prime that matched the cue, mismatched it, or was neutral was presented before the cue. Results showed facilitatory and inhibitory priming effects, as well as a distance effect based on the position of a digit relative to the boundary between 5 and 6. The results of both experiments were discussed in terms of a model based on relative processing speeds of the two relevant properties of the digits.
View
On the selection of signals
Article
Full-text available
  • Mar 1973
In a previous paper, it was argued that alertness, selectivity (set), and processing capacity (consciousness) could be identified and studied as separate components of attention. The current paper develops this theme by showing that alertness does not affect the buildup of information within the memory system but only the rate at which a later system responds to that information. Thus, in standard reaction-time tasks, increased alertness produces a reduction in reaction 'time but no decrease in errors. In contrast, providing a model of the signal the S is to process improved both speed and accuracy. The,. presence of a model of what the S is to process varies the vertex neural response to that specific signal as compared to a mismatching signal in the first 200-300 msec after its presentation. Three accounts of this effect are: speeded processing of a matching stimulus, habituation of the electrical response to a matching stimulus, and prolonged or enhanced processing of a mismatch. Evidence favors the first of these explanations, but the other two cannot be dismissed as possible contributors to this effect.
View
On the control of automatic processes: A parallel distributed processing model of the Stroop effect
Article
Full-text available
  • Jan 1988
Traditional views of automaticity are in need of revision. For example, automaticity often has been treated as an all-or-none phenomenon, and traditional theories have held that automatic processes are independent of attention. Yet recent empirical data suggest that automatic processes are continuous, and furthermore are subject to attentional control. A model of attention is presented to address these issues. Within a parallel distributed processing framework, it is proposed that the attributes of automaticity depend on the strength of a processing pathway and that strength increases with training. With the Stroop effect as an example, automatic processes are shown to be continuous and to emerge gradually with practice. Specifically, a computational model of the Stroop task simulates the time course of processing as well as the effects of learning. This was accomplished by combining the cascade mechanism described by McClelland (1979) with the backpropagation learning algorithm (Rumelhart, Hinton, & Williams, 1986). The model can simulate performance in the standard Stroop task, as well as aspects of performance in variants of this task that manipulate stimulus-onset asynchrony, response set, and degree of practice. The model presented is contrasted against other models, and its relation to many of the central issues in the literature on attention, automaticity, and interference is discussed.
View
View
Procedural learning: II. Intertrial repetition effects in speeded-choice tasks.
Article
  • Jan 1991
Ss' responses in a speeded-choice task are affected by the sequence of stimuli and responses. Generally, responses are faster when the same stimulus is repeated. The 1st 5 experiments demonstrate that the effect is stimulus specific: A succession of different stimuli that call for the same response produces little repetition benefit, unless the stimuli differ only in superficial ways (e.g., in color). Increasing the intertrial interval from 100 msec to 1,000 msec attenuates these repetition effects only slightly. These results would be consistent with a perceptual locus for the repetition effect, but Exp 6 shows that if the response mapping changes from trial to trial, the advantage for stimulus repetitions is abolished. Results indicate that locus of the repetition effect is at the stage of response selection. However, the stimulus specificity of the effects indicates that immediate repetitions produce or strengthen transient links that "shortcut" the response-selection stage. Results of H. Pasher and G. C. Baylis (see record 1991-17372-001 ) indicate that practice primarily strengthens response selection at the categorical level rather than shortcutting it.
View
View
View
Mental set and mental arithmetic
Article
  • Sep 1973
  • MEM COGNITION
State University of New York, Buffalo, New York 14226 Ss performed mental arithmetic problems in which they added, subtracted, or multiplied two one-digit numbers. The presentation order of the operator symbol and the digits was varied. With three possible operators, presentation of the operator prior to the digits (OD) led to faster RTs. With two possible operators, the opposite order (digits prior to operator, DO) led to faster RTs, because RTs in the OD condition were unaffected by the number of possible operators. These results are discussed in terms of the trade off between accessing active memory for a small number of items in the DO condition vs retrieving information from relatively large tables in long-term memory in the OD condition.
View
View
Mental Set and Mental Shift Revisited
Article
  • Dec 1976
In 1927, Jersild found that alternately subtracting 3 from a two-digit number and giving the common opposite to a word in a mixed list of numbers and words was faster than the average speed of subtracting 3s from a pure list of numbers and giving the opposites to a pure list of words. Experiment I replicated those findings: mixed lists were slightly, albeit nonsignificantly, faster than pure lists. Experiments II, III, and IV were designed to determine why changes of set did not slow performance on mixed lists: the results suggest that a shift of operations will take little or no time if the stimulus can serve as a retrieval cue for the operation to be performed on it. But changes of set will have a large effect when the selection of the appropriate operation requires that one keep track of previously performed operations.
View