Elsevier

Neuropsychologia

Volume 49, Issue 5, April 2011, Pages 1146-1153
Neuropsychologia

The neural correlates of impaired inhibitory control in anxiety

https://doi.org/10.1016/j.neuropsychologia.2011.01.019Get rights and content

Abstract

According to Attentional Control Theory (Eysenck et al., 2007) anxiety impairs the inhibition function of working memory by increasing the influence of stimulus-driven processes over efficient top-down control. We investigated the neural correlates of impaired inhibitory control in anxiety using an antisaccade task. Low- and high-anxious participants performed anti- and prosaccade tasks and electrophysiological activity was recorded. Consistent with previous research high-anxious individuals had longer antisaccade latencies in response to the to-be-inhibited target, compared with low-anxious individuals. Central to our predictions, high-anxious individuals showed lower ERP activity, at frontocentral and central recording sites, than low anxious individuals, in the period immediately prior to onset of the to-be-inhibited target on correct antisaccade trials. Our findings indicate that anxiety interferes with the efficient recruitment of top-down mechanisms required for the suppression of prepotent responses. Implications are discussed within current models of attentional control in anxiety (Bishop, 2009, Eysenck et al., 2007).

Research highlights

▸ High-anxious slower than the low-anxious in generating correct antisaccade. ▸ Lower frontocentral negativity prior to correct antisaccades in high-anxious. ▸ No anxiety-related group differences in ERP activity on prosaccade trials. ▸ No anxiety-related group differences in saccade error rates.

Introduction

A wealth of evidence associates anxiety with increased distractibility (see Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg, & van IJzendoorn, 2007; and Cisler & Koster, 2010 for reviews) and poor prepotent response inhibition (Ansari and Derakshan, 2010, Derakshan et al., 2009, Hardin et al., 2009, Jazbec et al., 2005, Sehlmeyer et al., 2010, Wieser et al., 2009). The Attentional Control Theory (ACT; Eysenck, Derakshan, Santos, & Calvo, 2007) was put forward to account for the effects of anxiety on central executive functions such as inhibition and shifting. According to ACT anxiety increases the influence of stimulus-driven processing over goal-directed regulatory processes resulting in poor executive attentional control which in turn impairs central executive functions such as inhibition. Inhibition is an important regulatory function that uses attentional control to restrain attentional resources from allocation to task-irrelevant salient stimuli or prepotent responses. It is believed to play a key role in “orchestrating” cognitive performance (Kok, 1999) in a number of domains including selective attention, memory, perception, problem solving and motor control (for comprehensive reviews, see Dempster and Corkhill, 1999, Kok, 1999, MacLeod et al., 2003). According to Friedman and Miyake (2004) the inhibition function of the central executive involves at least two components: prepotent response inhibition and resistance to interference from distracters. The notion that anxiety impairs inhibitory control has received support mainly from research assessing distractibility in the presence of emotional material (for reviews, see Cisler & Koster, 2010). However, recent research shows that anxiety can influence inhibitory control even in the absence of threat (see Derakshan and Eysenck, 2009, Bishop, 2009). Furthermore, research assessing the relationship between trait anxiety and prepotent response inhibition is scarce (but see Ettinger et al., 2005, on the relationship between antisaccade performance and broader constructs of negative affect including neuroticism).

There is accumulating evidence in support of predictions of ACT. For example Derakshan et al. (2009) used the ‘antisaccade’ task (Hallet, 1978) which requires top-down control of attention to inhibit the natural tendency to saccade toward an abrupt peripheral object and instead to generate a volitional saccade toward the opposite direction. Antisaccade performance is compared with ‘prosaccade’ performance where participants are simply instructed to make a saccade toward the abrupt peripheral object. Derakshan et al. (2009) found that high-anxious individuals compared with the low-anxious had significantly longer antisaccade latencies but the two groups did not differ in error rates. There were no group differences on the prosaccade task. This was taken to suggest that anxiety impaired inhibition of prepotent responses resulting in poor antisaccade latencies. Ansari and Derakshan (2010) extended these findings and confirmed that slower antisaccade latencies in high-anxious individuals was not attributable to impaired volitional action generation but more likely reflected impaired inhibitory control. ACT makes a distinction between ‘effectiveness’, and ‘efficiency’. Effectiveness refers to the overall ability to do the task and is often measured by error rates, while efficiency relates to the manner and amount of resources invested to achieve performance outcomes, and is often indicated by reaction times. ACT predicts that anxiety is more likely to impair efficiency but not effectiveness, a finding that has received accumulating support (Ansari and Derakshan, 2010, Ansari and Derakshan, in press, Ansari et al., 2008; see also Derakshan & Eysenck, 2009, for a review). It is therefore important to understand how anxiety impairs efficient inhibitory control. This can be done by assessing the more fine-grained neural mechanisms that underlie impaired inhibitory control in anxiety.

Neurocognitive models of executive attentional control implicate lateral PFC regions (DLPFC and VLPFC) and anterior cingulate cortex (ACC) in executive attentional control and particularly in the allocation of resources during conflict or distracting situations (e.g. Botvinick et al., 2001, Duncan and Owen, 2000, Miller and Cohen, 2001). Recent neuroimaging research suggests that trait anxiety interferes with the recruitment of prefrontal mechanisms required for attentional control. In a design that manipulated perceptual load, Bishop (2009) showed that in the presence of distracters anxiety interfered with target identification and was associated with lower activity in the DLPFC under low perceptual load. It was concluded that anxiety is associated with reduced recruitment of prefrontal attentional mechanisms required to restrain attention from distracter processing.

This study highlights an important point, one that seems to be largely overlooked in mainstream anxiety research, that anxiety could influence top-down attentional control in the absence of threat. Indeed ACT argues that anxiety influences attentional control even in the absence of threat resulting in poor attention regulation and impaired central executive functions. The current study aimed to evaluate this prediction by examining anxiety-modulated neural correlates of impaired inhibitory control using even-related potentials (ERPs) in combination with antisaccade performance.

Control of saccadic eye movements provides a highly reliable and precise assessment of executive control processes (e.g. McDowell et al., 2008, Munoz and Everling, 2004, Ridderinkhof et al., 2004). Studies examining the functional neuroanatomy of saccadic eye-movements indicate involvement of a network of frontoparietal brain areas with greater activation observed during more complex volitional eye-movements such as antisaccades (for recent reviews see Hutton and Ettinger, 2006, Johnston and Everling, 2008, McDowell et al., 2008). These include supplementary eye fields (SEF), frontal eye fields (FEF), dorsolateral prefrontal cortex (DLPFC), anterior cingulate cortex (ACC), areas of the parietal cortex (e.g. intra parietal sulcus; IPS) and superior colliculus (SC). Greater activation of these regions in antisaccade performance is thought to correspond to greater attentional control required for inhibition of a prepotent response and goal-directed generation of an antisaccade. This concurs with cognitive accounts of antisaccade performance as a complex cognitive task that requires recruitment of brain areas typically involved in executive control of behaviour and working memory (e.g. Kane et al., 2001, Mitchell et al., 2002, Roberts et al., 1994, Unsworth et al., 2004; for recent reviews, see Hutton, 2008, Hutton and Ettinger, 2006).

Numerous studies have assigned great importance to the role of ‘preparatory’ brain activity in the period immediately prior to stimulus onset in antisaccade performance (for reviews see Hutton, 2008, Hutton and Ettinger, 2006, McDowell et al., 2008). According to single-cell recording studies (for reviews see Johnston and Everling, 2008, Munoz and Everling, 2004) correct antisaccade performance is determined by preparatory activity of saccade neurons relative to the baseline activity of fixation neurons in the FEF and SC prior to stimulus onset. A reduction in saccade neuron activity in this critical time period will ensure that incoming responses to visual stimuli are insufficient to trigger a reflexive saccade. The relative increase in activity of fixation neurons on antisaccade relative to prosaccade trials in FEF and SC is believed to be a mechanism to account for this top-down influence (Johnston & Everling, 2008). SEF neurons project directly to the FEF and SC and as such can stimulate increased inhibition of saccade neurons in these areas. Everling, Spantekow, Krappmann, and Flohr (1998) observed greater pre-target negative ERP activity on correct versus incorrect antisaccade trials, maximal at frontocentral recording sites. This differential negativity began ∼60 ms before target onset reaching its maximal at ∼5 ms before the target appeared. Everling et al. (1998) concluded that this negativity in dorsomedial frontal cortex reflects “an activation of supplementary motor cortex or probably more specifically an activation of the SEF” (p. 32). These findings are thought to support the proposed role of the SEF in the generation of correct and suppression of incorrect antisaccades (e.g. Amador, Schlag-Rey, & Schlag, 2004).

Other structures that have been implicated as sources of top-down influence in correct antisaccade performance are the DLPFC and the ACC. Support for this comes from ERP and event-related fMRI studies. These studies have typically assessed preparation-related activity during a temporal ‘gap’ after central fixation offset and before peripheral target onset.1 For example, Ford, Goltz, Brown, and Everling (2005) used instructional cues at the beginning of every trial in a gap antisaccade task and compared brain activity in the preparatory period on correct versus incorrect antisaccade trials in a number of brain areas. In line with an important study by Connolly, Goodale, Goltz, and Munoz (2005) who found that greater preparatory activity in the FEF predicted antisaccade accuracy and latency, Ford et al. (2005) reported greater activation in the FEF as well as DLPFC and ACC on antisaccade compared with prosaccade trials at the end of the preparatory period (200 ms before stimulus onset). Similar results were reported by Brown, Vilis, and Everling (2007) supporting the role of DLPFC and ACC in correct antisaccade performance. Greater activity in these prefrontal regions before stimulus presentation is argued to reflect the involvement of these areas in ‘presetting’ or in other words biasing the occulomotor system for task-appropriate behaviour in the antisaccade task. These observations are in line with Miller and Cohen's (2001) influential model that considers these frontopareital regions as important sources of executive control that bias (or prepare) other brain areas for task-relevant processing.

Everling, Matthews, and Flohr (2001) compared ERPs preceding correct versus incorrect antisaccades in a distractor gap antisaccade task designed to increase the demand on inhibitory control. They found that an enhanced negative potential at frontocentral and parietal electrode sites immediately prior to stimulus presentation predicted correct antisaccade performance suggesting that greater negative activity immediately prior to stimulus presentation reflects a cortical inhibition mechanism that suppresses reflexive saccades when the visual fixation is removed and results in successful antisaccade performance.

As discussed above the prefrontal cortical areas that support antisaccade performance are considered to play a substantial role in the top-down regulation of attention in anxiety. The common involvement of these prefrontal structures in regulating executive control as measured by the antisaccade task makes this task ideal for evaluating predictions of attentional control deficits in anxiety (Eysenck et al., 2007). The current study measured ERP activity immediately prior to target onset in the antisaccade task to assess whether prolonged antisaccade latencies previously observed in high-anxious individuals is related to poor neural preparation (or recruitment) required for antisaccade performance. It was hypothesised that anxiety would modulate cortical activity at frontocentral sites in the period preceding correct antisaccade performance. Specifically high- compared with low-anxious individuals will exhibit slower antisaccade latencies and lower frontocentral activity at the end of the gap period. Furthermore, it was expected that anxiety-related group differences in cortical activity would only occur on correct antisaccade but not prosaccade trials. Also in line with our previous observations (Ansari et al., 2008, Ansari and Derakshan, 2010, Derakshan et al., 2009) we did not expect anxiety-related group differences in the measure of performance effectiveness indicated by comparable saccade error rates between high- and low-anxious individuals.

Section snippets

Participants

Thirty-four participants (mean age = 24.17 ± 4.16, min = 19, max = 34) were selected from a larger University of London subject panel based on their trait anxiety scores measured by the State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983). Participants were classified as low-anxious (LA) if they scored 34 or below and high-anxious (HA) if they scored 44 and above on the trait measure of STAI. Participants had normal or corrected-to-normal vision. Each individual was

Participants

Data for 24 participants were analysed.2 There were 12 low-anxious individuals (LA; mean trait anxiety score = 29.25 ± 2.99, min = 22, max = 33) and 12 high-anxious individuals (HA; mean trait

Discussion

Behavioural measures of antisaccade performance in the current study replicated our previous observations (Ansari and Derakshan, 2010, Derakshan et al., 2009) in that high-anxious compared with low-anxious participants took longer to perform correct antisaccades, while there were no group differences on prosaccade latencies. A central aim of the current study was to examine the underlying neural correlates of impaired antisaccade performance in an attempt to better understand how anxiety

Acknowledgements

The current research was supported by an ESRC PhD studentship awarded to Tahereh L. Ansari and carried out under the supervision of Nazanin Derakshan at Birkbeck, University of London.

References (47)

  • J.E. McDowell et al.

    Neurophysiology and neuroanatomy of reflexive and volitional saccades: Evidence from studies of humans

    Brain and Cognition

    (2008)
  • K.R. Ridderinkhof et al.

    Neurocognitive mechanisms of cognitive control: The role of prefrontal cortex in action selection, response inhibition, performance monitoring, and reward-based learning

    Brain and Cognition

    (2004)
  • C. Sehlmeyer et al.

    ERP indices for response inhibition are related to anxiety-related personality traits

    Neuropsychologia

    (2010)
  • N. Amador et al.

    Primate antisaccade. II. Supplementary eye field neuronal activity predicts correct performance

    Journal of Neurophysiology

    (2004)
  • T.L. Ansari et al.

    Anxiety impairs inhibitory control but not volitional action control

    Cognitive & Emotion

    (2010)
  • Ansari, T. L, & Derakshan, N. (in press). The neural correlates of cognitive effort in anxiety: Effects on processing...
  • T.L. Ansari et al.

    Effects of anxiety on task switching: Evidence from the mixed antisaccade task

    Cognitive, Affective & Behavioral Neuroscience

    (2008)
  • Y. Bar-Haim et al.

    Threat-related attentional bias in anxious and non-anxious individuals: A meta-analytic study

    Psychological Bulletin

    (2007)
  • S.J. Bishop

    Trait anxiety and impoverished prefrontal control of attention

    Nature Neuroscience

    (2009)
  • M. Botvinick et al.

    Conflict monitoring and cognitive control

    Psychological Review

    (2001)
  • M.R.G. Brown et al.

    Frontoparietal activation with preparation for antisaccades

    Journal of Neurophysiology

    (2007)
  • J.D. Connolly et al.

    FMRI activation in the human frontal eye field is correlated with saccadic reaction time

    Journal of Neurophysiology

    (2005)
  • F.N. Dempster et al.

    Interference and inhibition in cognition and behaviour: Unifying themes for educational psychology

    Educational Psychology Review

    (1999)
  • Cited by (114)

    • Social anxiety does not impair attention inhibition: An emotion anti-saccade task

      2022, Journal of Behavior Therapy and Experimental Psychiatry
    View all citing articles on Scopus
    View full text