Elsevier

Neuropsychologia

Volume 156, 18 June 2021, 107828
Neuropsychologia

Response variations can promote the efficiency of task switching: Electrophysiological evidence

https://doi.org/10.1016/j.neuropsychologia.2021.107828Get rights and content

Highlights

  • Smaller switch costs and higher accuracy for task switching in r-change context than in r-repeat context.

  • Greater frontal-central N2 in r-change context.

  • Larger stimulus-locked P3 for task-switch trials in r-change context.

  • Non-significant context effect on task switching during the response-locked P3 component.

  • Response variations during task repetition can promote the efficiency of followed task switching.

Abstract

Previous studies have investigated sequence effect on task switching and found that increased cognitive control in preceding trials would transfer to the current trial. However, it remains unclear whether response variations during task repetition can enhance cognitive control and promote task switching. In the present study, we designed two sequence contexts, the response-change (r-change) and response-repeat (r-repeat) contexts, by adopting a classical task-switching paradigm in which participants were asked to make an odd-even or large-small judgment of the presented digit. The only difference between the two sequence contexts was whether responses varied frequently during task repetition. Behavioral results showed that the r-change context induced smaller switch costs and higher accuracy for task switching than the r-repeat context. Event-related potential (ERP) results revealed (1) the effect of context on N2 amplitudes, with greater N2 in the r-change context than the r-repeat context at frontal-central regions; (2) the interaction between context and transition type during the stimulus-locked P3 component, with a marked context effect for the task-switch trials; (3) non-significant context effect on task switching during the response-locked P3 component. These findings suggest that response variations during a sequence of task-repeat trials can trigger the increase in cognitive control that promotes the efficiency of followed task switching.

Introduction

Cognitive control is the ability that guides people to focus on target behavior and inhibit distracting information. To investigate cognitive control, researchers often use conflict paradigms (e.g., Stroop task, Flanker task, Simon task), response inhibition paradigms (e.g., go/no-go task, stop-signal task), and task-switching paradigms (Allport et al., 1994; Baniqued et al., 2017; Braem et al., 2019; Jurado and Rosselli, 2007; Meiran et al., 2000; Monsell, 2003; Rogers and Monsell, 1995). It is widely accepted that prior conflict experience minimizes distraction on subsequent trials, which manifests as two types of sequence effects (for review see Chiu and Egner, 2019): trial-by-trial sequence effect like Gratton effect (Gratton et al., 1992), and proportion congruency effect (Bugg and Crump, 2012).

The trial-by-trial sequence effect is the finding that the cognitive control of the current conflict trial (N) is influenced by the preceding conflict trial (N−1). A number of studies on Gratton effect have demonstrated a smaller congruency effect (the time-costs of incongruent trials vs. congruent trials) on trials following an incongruent trial than a congruent trial (Braverman and Meiran, 2015; Brosowsky and Crump, 2018; Gratton et al., 1992). Botvinick et al. (2001) proposed a top-down conflict monitoring (or conflict adaptation) account to explain the trial-by-trial sequence effect. They indicated that, the cognitive processing in the ongoing conflict task is dynamically adjusted by previous conflict experience, so there is more attention to the task-relevant stimuli, which accelerates the congruency effect on trials after an incongruent trial.

The proportion congruency effect describes the finding that the cognitive control of the current conflict trial is affected by the frequency of congruent trials. Significant amount of evidence for the proportion congruency effect have shown that the congruency effect decreases in case of low proportion of congruent trials (Bugg and Crump, 2012; Crump et al., 2006; Egner, 2014; Jacoby et al., 2003; Logan and Zbrodoff, 1979). A parallel study on task switch frequency demonstrated that switch costs (the time-costs of task-switch trials vs. task-repeat trials) were smaller in case of high task switch frequency (Braem and Egner, 2018; Chiu and Egner, 2017; Monsell and Mizon, 2006; Nessler et al., 2012; Schneider and Logan, 2006). For example, Nessler et al. (2012) manipulated the switch probability (25%, 50%) and found that the switch costs decreased as the switch probability increased. The Event-related potentials (ERP) results showed that late parietal positivity (500–800 ms post-cue onset) was smaller in the frequent-switch condition than the infrequent-switch condition (Nessler et al., 2012). Moreover, a body of research has shown that conflict adaptation is a general control mechanism (Chiu and Egner, 2017; Dreisbach and Fischer, 2015; Fischer et al., 2014; Kan et al., 2013), which means that the control settings (e.g., a collection of task parameters, Vandierendonck et al., 2010) during preceding conflict trials or context can transfer to the current conflict trial or context.

Though numerous studies have been conducted to explore the sequence effect on task switching by focusing on stimulus set (i.e, bivalent effect, Grundy et al., 2013; Meier et al., 2009; Rogers and Monsell, 1995), response set (i.e, response effector, Philipp and Koch, 2011; Philipp et al., 2013; Tieges et al., 2007; West et al., 2009; Yeung and Monsell, 2003), or asymmetry (Kiesel et al., 2010; Li et al., 2019) and task-switch frequency (De Baene & Brass, 2013, 2014; Frober and Dreisbach, 2017; Frober et al., 2018; Monsell and Mizon, 2006; Nessler et al., 2012), the question of whether switch cost would be influenced by response variations in a sequence of task-repeat trials remains unaddressed. The purpose of the present study was to investigate whether the cognitive control would increase due to response variations during a sequence of task-repeat trials. The followed critical issue was whether the increased cognitive control during task repetition would transfer to the task-switch trials and improve the efficiency of task switching.

To address these two issues, we adopted a classical task-switching paradigm in which participants made a large-small or odd-even judgment of the presented number. After a sequence of task-repeat trials, participants should shift the task based on changes of the stimulus color. There were two contexts, response-change (r-change) and response-repeat (r-repeat) contexts, which only differed in whether the responses varied frequently before task switching (Fig. 1).

In the r-change context, cognitive control might increase to cope with response variations because response variations would induce more response conflict that should be detected and resolved by the increased cognitive control (Surrey et al., 2016; Tieges et al., 2007; Xie et al., 2020). The increased cognitive control in the task-repeat trials might transfer to the task-switch trials and accelerate subsequent task switching, resulting in smaller switch costs in the r-change context than the r-repeat context. Alternatively, in the task-repeat trials, increased cognitive control caused by response variations might not affect subsequent task switching. Instead, the effect of response variations might be limited to task-repeat trials. That is, in the r-change context, the reaction time (RT) of task-repeat trials would be longer due to response-switch cost (Tieges et al., 2007; West et al., 2009; Xie et al., 2020). In contrast, in the r-repeat context, the RT of task-repeat trials would be faster due to response-repeat benefit (Kleinsorge, 1999; Kleinsorge and Heuer, 1999; Kleinsorge et al., 2005; for a review see Altmann, 2011).

In brief, for task-repeat trials, there would be a difference in the RT between the r-change context and r-repeat context; the RT of task-repeat trials would be slower in the r-change context, and faster in the r-repeat context. However, according to the existing studies, it is unclear how the sequence context within task repetitions affects the RT of the subsequent task-switch trials. RT includes the processing time of multiple cognitive sub-processes when the participant completes a task. If we only analyze RT, we cannot examine which sub-processes are affected by context factors and which are not. Therefore, we used electroencephalogram (EEG), which has the merit of high temporal resolution (Luck, 2014), to investigate the dynamical features of sequence effect on task-switching.

It is known that cue-P3 is related with the endogenous task-set reconfiguration processing in the task-switch trials (Han et al., 2018; Karayanidis et al., 2010; Nicholson et al., 2005). The more effective the task switching, the smaller the cue-P3 amplitudes (Gajewski et al., 2018; Nicholson et al., 2005). When the cue-target interval was short enough (e.g., 0 ms), target-P3 was overlapped with cue-P3, the ERP waveform was similar to that of cue-P3 evoked in the experiment where the cue and target were separately presented (Jost et al., 2008; Han et al., 2019; Nicholson et al., 2005). In the present study, the cue-target interval was 0 ms, and the P3 component was expected to be similar to cue-P3 observed in previous studies. Therefore, we predicted that the amplitudes of the cue-P3 like component would be larger in the task-switch trials than in the task-repeat trials. If the response variations during task-repeat trials did not influence subsequent task switching, then the P3 component evoked by the task-switch trials (i.e., switch-P3) would not differ between the r-change and r-repeat contexts. Otherwise, if the switch-P3 amplitudes differed between the two contexts, then we could attribute it to sequence effect (conflict adaptation), i.e., response variations during task repetition. Furthermore, if the switch-P3 amplitudes were smaller in the r-change context than the r-repeat context, this might imply that increased cognitive control improves the efficiency of task switching. Moreover, if the increased cognitive control caused by response variations in the task-repeat trials only affected the cognitive processing related to the cue processing in the task-switch trials and did not affect the target processing, it might not affect the ERP components related to the target processing (such as P3 component locked to target); that is, the amplitudes of response-locked switch-P3 might not be different in the two contexts.

In addition, N2 component was related to cognitive control that involved the processing of conflict monitoring (Folstein and Van Petten, 2008; Van Veen and Carter, 2002; Yeung and Cohen, 2006), which was sensitive to conflict adaptation (Botvinick et al., 2001; Clayson and Larson, 2013; Whitehead2017). Studies on the trial-by-trial sequence effect and proportion congruency effect have found that N2 was adjusted by the degree of conflict in a given trial or context (Blais et al., 2016; Clayson and Larson, 2011; Forster et al., 2011; Gajewski et al., 2008; Riesel et al., 2017; Schreiter et al., 2018; Whitehead et al., 2017). Accordingly, in the r-change context of task repetition, due to frequent variations in response, more conflict exists, and more cognitive control might be recruited to deal with response conflicts and variations (Surrey et al., 2016; Tieges et al., 2007; West et al., 2009; Xie et al., 2020), which could be reflected in enhanced N2 amplitudes in the r-change context relative to the r-repeat context. Importantly, if the increased cognitive control would transfer to the subsequent task-switch trials, greater N2 was expected to be observed for task-switch trials in the r-change context than in the r-repeat context; otherwise, there would be no difference in switch-N2 between the two contexts.

Section snippets

Subjects

A total of 40 undergraduates (16 females, mean age 20) from Jiangxi Normal University (China) participated in this study. The data of six participants were not used due to lower accuracy or excessive artifacts. All participants were right-handed, were neurologically normal, and had normal or corrected-to-normal vision. After the experiment, they were paid ¥35 each. The experiment was approved by the Morals & Ethics Committee of the School of Psychology of Jiangxi Normal University (China), and

Behavioral results

For behavioral data analysis, error responses, responses after erroneous responses and RT outliers (±2 SD from the average RT of each condition) were removed when analyzing RT. About 11% of the total trials for each condition were eliminated. 2 × 2 repeated-measures ANOVA was conducted with transition type (repeat, switch) and context (r-change, r-repeat) as within-subject factors.

Mean accuracies and RTs for each condition are shown in Table 1. The ANOVA indicated that the mean accuracy was

Discussion

In order to investigate whether response variations during consecutive task-repeat trials can trigger the increase in cognitive control and promote task switching, we used a classical task-switching paradigm in which two sequence contexts were designed. In the r-change context, participants used various fingers to respond to the stimuli during a sequence of task-repeat trials, whereas in the r-repeat context, participants repeatedly used the same finger to respond to the stimuli during task

Credit author statement

Bingxin Zhuo: Conceptualization, Methodology, Software, Data curation, Validation, Formal analysis, Investigation, Visualization, Writing – original draft, Writing – review & editing. Y. Chen: Software, Data curation, Validation, Investigation. Yun Chen: Software, Data curation, Validation, Investigation. Mengqi Zhu: Software, Data curation, Validation, Formal analysis, Investigation, Visualization, Writing – original draft. Bihua Cao: Conceptualization, Methodology, Visualization, Writing –

Acknowledgment

Thanks to Junchen Li and Liufang Xie for their help in collecting data and valuable advice. This study was supported by the National Natural Science Foundation of China (31571118, 31860278, and 31760285).

References (90)

  • J.H. Han et al.

    More abstract, more difficult to shift: behavior and electrophysiological evidence

    Behav. Brain Res.

    (2019)
  • C.H. Hillman et al.

    A cross-sectional examination of age and physical activity on performance and event-related brain potentials in a task switching paradigm

    Int. J. Psychophysiol.: official journal of the International Organization of Psychophysiology

    (2006)
  • S. Hsieh et al.

    Electrophysiological correlates of task conflicts in task-switching

    Brain Res.

    (2008)
  • B.D. Isbel et al.

    Mental training affects electrophysiological markers of attention resource allocation in healthy older adults

    Neurosci. Lett.

    (2019)
  • I.P. Kan et al.

    To adapt or not to adapt: the question of domain-general cognitive control

    Cognition

    (2013)
  • R. Katz et al.

    Cognitive control in late-life depression: response inhibition deficits and dysfunction of the anterior cingulate cortex

    American Journal of Geriatric Psychiatry Official Journal of the American Association for Geriatric Psychiatry

    (2010)
  • T. Kleinsorge

    Response repetition benefits and costs

    Acta Psychol.

    (1999)
  • M.G. Lu et al.

    Event-related potentials and neural oscillations dissociate levels of cognitive control

    Behav. Brain Res.

    (2017)
  • N. Meiran et al.

    Component processes in task switching

    Cognit. Psychol.

    (2000)
  • S. Monsell

    Task switching

    Trends Cognit. Sci.

    (2003)
  • D. Nessler et al.

    A new account of the effect of probability on task switching: ERP evidence following the manipulation of switch probability, cue informativeness and predictability

    Biol. Psychol.

    (2012)
  • M.L. Schreiter et al.

    Neurophysiological processes and functional neuroanatomical structures underlying proactive effects of emotional conflicts

    Neuroimage

    (2018)
  • S. Schuch et al.

    Response-repetition effects in task switching with and without response execution

    Acta Psychol.

    (2010)
  • J.C. Seibold et al.

    Response repetitions in auditory task switching: the influence of spatial response distance and of the response-stimulus interval

    Acta Psychol.

    (2019)
  • R. Swainson et al.

    Using advance information in dynamic cognitive control: an ERP study of task-switching

    Brain Res.

    (2006)
  • K. Timmer et al.

    Earlier and more distributed neural networks for bilinguals than monolinguals during switching

    Neuropsychologia

    (2017)
  • K. Umebayashi et al.

    An ERP investigation of task switching using a flanker paradigm

    Brain Res.

    (2010)
  • V. Van Veen et al.

    The anterior cingulate as a conflict monitor: fMRI and ERP studies

    Physiol. Behav.

    (2002)
  • G.R. Wylie et al.

    Task switching: a high-density electrical mapping study

    Neuroimage

    (2003)
  • M. Zhu et al.

    Distinct brain activation in response to negative feedback at different stages in a variant of the Wisconsin Card Sorting Test

    Biological Psychological

    (2020)
  • A. Allport et al.

    Shifting intentional set: exploring the dynamic control of tasks

  • E.M. Altmann

    Testing probability matching and episodic retrieval accounts of response repetition effects in task switching

    J. Exp. Psychol. Learn. Mem. Cognit.

    (2011)
  • P.L. Baniqued et al.

    Shedding light on gray(ing) areas: connectivity and task switching dynamics in aging

    Psychophysiology

    (2017)
  • F. Barceló et al.

    An information theory account of late frontoparietal ERP positivities in cognitive control

    Psychophysiology

    (2018)
  • C. Blais et al.

    ERP evidence for implicit priming of top-down control of attention

    J. Cognit. Neurosci.

    (2016)
  • M.M. Botvinick et al.

    Conflict monitoring and cognitive control

    Psychol. Rev.

    (2001)
  • S. Braem et al.

    Getting a grip on cognitive flexibility

    Curr. Dir. Psychol. Sci.

    (2018)
  • A. Braverman et al.

    Conflict control in task conflict and response conflict

    Psychol. Res.

    (2015)
  • N.P. Brosowsky et al.

    Memory-guided selective attention: single experiences with conflict have long-lasting effects on cognitive control

    J. Exp. Psychol. Gen.

    (2018)
  • J.M. Bugg et al.

    In support of a distinction between voluntary and stimulus-driven control: a review of the literature on proportion congruent effects

    Front. Psychol.

    (2012)
  • Y.C. Chiu et al.

    Cueing cognitive flexibility: item-specific learning of switch readiness

    J. Exp. Psychol. Hum. Percept. Perform.

    (2017)
  • P.E. Clayson et al.

    Psychometric properties of conflict monitoring and conflict adaptation indices: response time and conflict N2 event-related potentials

    Psychophysiology

    (2013)
  • M.J.C. Crump et al.

    The context-specific proportion congruent Stroop effect: location as a contextual cue

    Psychon. Bull. Rev.

    (2006)
  • T. Cunillera et al.

    Brain oscillatory activity associated with task switching and feedback processing

    Cognit. Affect Behav. Neurosci.

    (2012)
  • E. Donchin et al.

    Is the P300 component a manifestation of context updating?

    Behav. Brain Sci.

    (1988)
  • Cited by (11)

    • Cognitive control is modulated by hierarchical complexity of task switching: An event-related potential study

      2022, Behavioural Brain Research
      Citation Excerpt :

      The task-switching paradigm is a common paradigm used to investigate cognitive control [2]. It has been shown that the response time(RT) for switch trials is longer and the error rate is higher than that of repeat trials, which is called the switch cost [3–10]. Switch cost can be modulated by cue features [11,12], cue-stimulus intervals [13–16], prior information [17–20], stimulus type [17,21], task context [10,22], and hierarchy levels of task-set [4,23–26].

    • Proactive and reactive control differ between task switching and response rule switching: Event-related potential evidence

      2022, Neuropsychologia
      Citation Excerpt :

      The human capacity to adopt appropriate actions in rapidly changing environments is a fundamental requisite for successful goal achievement. This ability is described as cognitive flexibility, which is usually explored by using the task-switching (T-switch) paradigm (Gratton et al., 2018; Diamond, 2013; Kiesel et al., 2010; Koch et al., 2018; Worringer et al., 2019; Verschooren et al., 2019; Zhuo et al., 2021). In task-switching studies, the response is slower and less accurate in switch trials than in repeat trials.

    • Trait anxiety modulates the temporal dynamics of Stroop task switching: An ERP study

      2021, Biological Psychology
      Citation Excerpt :

      A compatibility effect of N2 is often observed, as the amplitude of N2 is larger in incompatible trials than in compatible trials (Wei, Beuckelaer, & Zhou, 2021; Yeung et al., 2004), but sometimes N2 might not be sensitive to compatibility information (Rauchbauer, Lorenz, Lamm, & Pfabigan, 2021). In task-switching studies, the amplitude of N2 was larger in single trials than in non-switch trials and larger in switch trials than in non-switch trials (Gajewski, Ferdinand, Kray, & Jutta, 2018; Karayanidis, Whitson, Heathcote, & Michie, 2011), but sometimes the difference between switch trials and non-switch trials did not reach significance (Hillman, Kramer, Belopolsky, & Smith, 2006; Hsieh & Liu, 2008; Zhuo, Chen, Zhu, Cao, & Li, 2021). In related studies, anxiety was related to enhanced N2 amplitudes (Cavanagh, Meyer, & Hajcak, 2017; Wei et al., 2021), which was interpreted as greater response conflict in high-anxiety participants (compared to that in low-anxiety participants).

    View all citing articles on Scopus
    View full text