Elsevier

Neuropsychologia

Volume 72, June 2015, Pages 22-26
Neuropsychologia

Left anterior cingulate activity predicts intra-individual reaction time variability in healthy adults

https://doi.org/10.1016/j.neuropsychologia.2015.03.015Get rights and content

Highlights

  • Trial-to-trial RT variability was associated with activation in the left pregenual ACC.

  • Activation of the left pregenual ACC was associated with improved response stability.

  • Performance variability may serve as a marker for top down control of attention.

  • Future research should assess the relationship between RT variability and ACC subregions.

Abstract

Within-subject, or intra-individual, variability in reaction time (RT) is increasingly recognised as an important indicator of the efficiency of attentional control, yet there have been few investigations of the neural correlates of trial-to-trial RT variability in healthy adults. We sought to determine the neural correlates of intra-individual RT variability during a go/no-go response inhibition task in 27 healthy, male participants. We found that reduced trial-to-trial RT variability (i.e. greater response stability) was significantly associated with greater activation in the left pregenual anterior cingulate. These results support the role of the left anterior cingulate in the dynamic control of attention and efficient response selection. Greater understanding of intra-individual RT variability and top-down attentional control in healthy adults may help to inform disorders that impact executive/attentional control, such as attention deficit hyperactivity disorder and schizophrenia.

Introduction

The ability to maintain consistent attention to a task at hand is critical to all aspects of human learning and performance. Although fluctuations in attention can arise as a result of external distractors and fatigue, there is also inherent, internal fluctuation in our capacity to exert control over performance. Inherent fluctuations in neuronal activity occur both at rest and during cognitive tasks; this intrinsic variability in neuronal activity is a thought to be a primary driver of trial-to-trial behavioural variation (MacDonald et al., 2009b; Weissman et al., 2006). The variability in brain activity during cognitive tasks is thought to reflect the allocation of attentional resources at structural, functional, and neurochemical levels, which drives the subtle variation in an individual’s trial-to-trial behavioural response (MacDonald et al., 2009b).

Trial-to-trial variability in behavioural response, often referred to as intra-individual variability, is thought to be a hallmark of attention deficit hyperactivity disorder (ADHD; Bellgrove et al., 2005; Castellanos et al., 2006; Mullins et al., 2005) and other disorders that impact attention, such as schizophrenia and dementia (MacDonald et al., 2009b). Individuals with ADHD show greater intra-individual variability across a range of cognitive tasks, and this measure is more robust in differentiating ADHD from healthy control subjects compared to other indices, for instance, mean RT, directional errors, omission errors, or rates of inhibition on tasks such as the Continuous Performance Task and go–no go tasks (Castellanos et al., 2006, Klein et al., 2006, Mullins et al., 2005). In disorders such as ADHD, greater intra-individual variability may reflect inefficient response selection, greater frequency of lapses in attention or difficulty maintaining attention to the task, which may manifest as a greater proportion of trials with prolonged reaction times (RT; Castellanos et al., 2006). Intra-individual RT variability may therefore be an important indicator of efficient attentional control, rather than simply random noise.

Despite intra-individual RT variability being recognised as a hallmark of impaired cognitive control, there has been relatively little attention given to investigating its neural correlates. Research that has examined RT variability has largely focused on contrasting performance between clinical and control groups (Bellgrove et al., 2005; Castellanos et al., 2006; Klein et al., 2006; MacDonald et al., 2009a), however intra-individual variability within a healthy population is under-examined. Investigating fluctuations in RT in healthy individuals is essential to advancing both theoretical and clinical understandings of effective attentional control. It is well established that the timing of the haemodynamic response is correlated to an individual’s RT on a given task (Esterman et al., 2013, Fox et al., 2006, Yarkoni et al., 2009). However, it is necessary to decompose the relationship between RT variability and its neural correlates according to specific executive function contexts, such as response inhibition, response conflict, working memory, or motor planning, because intra-individual RT variability has been shown to vary across domains (Castellanos et al., 2006, Yarkoni et al., 2009). Studies have examined the functional neural correlates of RT variability in healthy adults across cognitive domains including response inhibition (Bellg rove et al., 2004, Connolly et al., 2005, Esterman et al., 2013, Esterman et al., 2014), global/local selective attention (Weissman et al., 2006), set shifting (MacDonald et al., 2009a), spatial attention (Hahn et al., 2007) or working memory and emotion processing tasks (Yarkoni et al., 2009). In assessing the relationship between neural activation and RT variability, the approach has been either to correlate global variability measures, such as coefficient of variability of RT, with brain activation (Bellg rove et al., 2004, Connolly et al., 2005), or alternatively, examine neural correlates of trial-to-trial RT fluctuations (Esterman et al., 2013, Weissman et al., 2006, Yarkoni et al., 2009). Of studies that have assessed intra-individual RT variability during response inhibition tasks, Connolly et al. (2005) used an anti-saccade task but only examined activation in the frontal and supplementary eye fields and intraparietal sulcus, rather than whole brain activation. Bellgrove et al. (2004) used a go/no-go task to examine the relationship between intra-individual RT variability and whole brain task-related activation in non-clinical adults, which revealed correlations between bilateral middle frontal areas and right inferior parietal and thalamic regions. Similar findings were reported in children (Simmonds et al., 2007). Previously our group has found that RT variability was inversely related to successful response inhibition, indicating that RT variability may be an important indicator of effective attentional control (Bellgrove et al., 2004). However this approach only informs us about global variability, rather than trial-to-trial fluctuations in response. Esterman et al. (2013) assessed whole brain correlates of trial-to-trial fluctuations in RT using a sustained attention task and found that low RT variability was associated with greater activation of the anterior cingulate cortex (ACC) and posterior cingulate cortex (PCC), both part of the default mode network (DMN), whereas higher RT variability was associated with greater FEF activation. Rapid, dynamic control of attention over trials is important to efficient cognitive control. Therefore to further elucidate brain regions that may play an important role in the dynamic top down control of attention in healthy adults, we sought identify brain regions whose activity showed a relationship with trial-to-trial RT variability during a go/no-go task using whole brain fMRI analysis.

Section snippets

Participants and task design

Detailed methods describing participants, task and procedures have been previously reported (Hester et al., 2012), therefore a summary of the study’s methodology is presented here. All participants were recruited in accordance with the principles of the Declaration of Helsinki and ethical guidelines of the University of Queensland and the Wesley Hospital (Brisbane, QLD, Australia).

Twenty-seven right-handed, male subjects (mean age, 22 years; range, 18–35 years) were recruited for this study via

Reaction time

As reported previously, mean RT on all valid go-trials was 428.5 ms (SD: 18.2; Hester et al., 2012); the average standard deviation of all valid go-trial RTs was 122.1 ms (SD: 8.33). An example of group-averaged fluctuations in absolute trial-to-trial RT ([RT-mean RT]) for one task block modelling go-trials (Block 3) is presented in Fig. 2. Mean RTs did not differ across blocks.

fMRI correlates of intra-individual reaction time variability

To assess correlates of intra-individual RT variability, we performed whole-brain multiple regression for each subject

Discussion

Increased intra-individual variability in RT has often been contrasted in clinical and control groups, and is increasingly recognised as a hallmark of a number of cognitive disorders, such as ADHD, schizophrenia and dementia. However, little attention has been given to the brain regions that underpin RT fluctuations and their relationship to cognitive performance in healthy controls. In the present study, we investigated the BOLD correlates of intra-individual RT variability during a go/no-go

Conclusion

In summary, we have identified that greater activation of the pgACC is associated with less intra-individual RT variability during a response inhibition task. Our findings extend previous evidence for the ACC in conflict monitoring and action selection in the presence of conflicting responses, demonstrating while selecting a behavioural goal in the presence of interfering rule sets or information, greater activity in the ACC is coupled with greater performance consistency. Future examination of

Acknowledgements

This work was supported by Australian Research Council Grant DP0770337 (Mark A. Bellgrove), the Australian National Health and Medical Research Council (Mark A. Bellgrove & Robert Hester, Grant 519730).

References (27)

Cited by (0)

View full text