Sustained neural alterations in anxious youth performing an attentional bias task: a pupilometry study

Abstract

Background: Biased attention patterns have been observed at early (16-500 ms poststimulus onset) and intermediate (1,500-4,000 ms post-onset) time points in anxious youth, but it is unclear whether a more sustained form of neural attentional bias, persisting well beyond the time frame of stimulus presentation and behavioral response, is also apparent. We investigated early, intermediate, and sustained forms of bias using behavioral measures and pupillary reactivity, an index of cognitive and affective load, to gain insight into potential neurocognitive targets for early intervention.

Method: Twenty nonanxious youth and 74 youth with generalized anxiety disorder (GAD), separation anxiety disorder (SAD), and/or social phobia (SP) completed a dot-probe task, which requires participants to respond to a dot replacing either a neutral or fearful face. Emotional faces were presented for short/early (200 ms) or intermediate (2 s) intervals and followed by a sustained (up to 10.5 s) poststimulus interval. Pupil dilation, gaze direction, and reaction times (RTs) were measured during task completion.

Results: Early and intermediate vigilance patterns in RTs and an avoidant pattern in gaze direction were observed in all participants irrespective of anxiety. Sustained pupil dilation in anxious youth was observed on trials in which the dot replaced fearful faces, along with an inflexible pattern of pupillary responding in comparison to controls.

Conclusion: Sustained cognitive-affective load following emotional face viewing is altered and inflexible in anxious youth. These prolonged alterations extend well beyond the time frame of behavioral attentional bias and may indicate inflexible and insufficient sustained cognitive control. Early interventions targeting these alterations could improve long-term mental health trajectories.

Figure 1

A) Representative long duration, congruent trial showing avoidant fixation pattern during face presentation. Color gradient depicts the passage of time, moving from darker (black) to lighter (nude) color as the trial progresses. Subject begins with a fixation at the screen center, shows an initial fixation to the neutral face, then fixates on the fearful face, and finally returns to the neutral face. B) Bias scores for eyetracking indices. Bias calculated for each individual as a difference score: % of trials (for initial fixation) or % of time during a given window (for fixation bias) with fixations to fearful face – % of trials/time with fixations to neutral face. Positive scores indicate vigilance, negative scores indicate avoidance. Error bars represent SD.

Figure 2

A) Differences in pupillary response to dot-probe trials as a function of group. Pupil diameter represents change from baseline. Initial constrictions are related to pupillary light reflex. Regions of statistically significant differences are highlighted along the X axis (red=p<.05, yellow=p<.1). Underlined area along the X-axis indicates that the length of statistically significant contiguous t-tests exceeded the contiguity threshold needed for multiple comparisons correction. Vertical red lines represent onset of faces (first line) and offset of faces/onset of dot (second line). B) Mean pupil diameter as a function of trial duration, dot-location, and group during the “early” and “late” time periods defined on the basis of pupil principal components analysis (PCA; see Supplement). The early factor (app. .5 to 1.5 seconds after stimulus onset on short trials and .5 to 3 seconds after stimulus onset on long trials) relates largely to pupillary light reflex in response to face presentation, while the late factor (app. 2 to 10 seconds after stimulus onset on short trials and app. 3 to 11 seconds after stimulus onset in long trials) represents sustained pupil dilation. Diagram highlights the simple effects within each group of dot-location, which are significant in the non-anxious group only.

Figure 2

A) Differences in pupillary response to dot-probe trials as a function of group. Pupil diameter represents change from baseline. Initial constrictions are related to pupillary light reflex. Regions of statistically significant differences are highlighted along the X axis (red=p<.05, yellow=p<.1). Underlined area along the X-axis indicates that the length of statistically significant contiguous t-tests exceeded the contiguity threshold needed for multiple comparisons correction. Vertical red lines represent onset of faces (first line) and offset of faces/onset of dot (second line). B) Mean pupil diameter as a function of trial duration, dot-location, and group during the “early” and “late” time periods defined on the basis of pupil principal components analysis (PCA; see Supplement). The early factor (app. .5 to 1.5 seconds after stimulus onset on short trials and .5 to 3 seconds after stimulus onset on long trials) relates largely to pupillary light reflex in response to face presentation, while the late factor (app. 2 to 10 seconds after stimulus onset on short trials and app. 3 to 11 seconds after stimulus onset in long trials) represents sustained pupil dilation. Diagram highlights the simple effects within each group of dot-location, which are significant in the non-anxious group only.