A neural oscillatory signature of sustained anxiety.

BACKGROUND: Anxiety is a sustained response to uncertain threats; yet few studies have explored sustained neurobiological activities underlying anxious states, particularly spontaneous neural oscillations. To address this gap, we reanalysed magnetoencephalographic (MEG) data recorded during induced anxiety to identify differences in sustained oscillatory activity between high- and low-anxiety states. METHODS: We combined data from three previous MEG studies in which healthy adults (total N = 51) were exposed to alternating periods of threat of unpredictable shock and safety while performing a range of cognitive tasks (passive oddball, mixed-saccade or stop-signal tasks). Spontaneous, band-limited, oscillatory activity was extracted from middle and late intervals of the threat and safe periods, and regional power distributions were reconstructed with adaptive beamforming. Conjunction analyses were used to identify regions showing overlapping spectral power differences between threat and safe periods across the three task paradigms. RESULTS: MEG source analyses revealed a robust and widespread reduction in beta (14-30 Hz) power during threat periods in bilateral sensorimotor cortices extending into right prefrontal regions. Alpha (8-13 Hz) power reductions during threat were more circumscribed, with notable peaks in left intraparietal sulcus and thalamus. CONCLUSIONS: Threat-induced anxiety is underpinned by a sustained reduction in spontaneous beta- and alpha-band activity in sensorimotor and parietal cortical regions. This general oscillatory pattern likely reflects a state of heightened action readiness and vigilance to cope with uncertain threats. Our findings provide a critical reference for which to identify abnormalities in cortical oscillatory activities in clinically anxious patients as well as evaluating the efficacy of anxiolytic treatments.

Anxious arousal alters prefrontal cortical control of stopping.

Anxiety heightens vigilance and stimulus-driven attention to the environment, which may in turn disrupt cognitive control processes such as response inhibition. How this unfolds at the neural level is unclear. Previous evidence implicates the right inferior frontal gyrus (IFG) as an important cortical node in both stimulus-driven attention and inhibitory control. Here we used magnetoencephalography (MEG) to investigate the neural mechanisms involved in the relationship between threat-induced anxiety and stopping during a stop-signal task, where a visual go signal was occasionally followed by an auditory stop signal. Healthy individuals (N = 18) performed the task during the threat of unpredictable shocks and safety to modulate anxious arousal. Behaviorally, we observed that stopping was impaired during threat (i.e. slower estimated stop-signal reaction times), indicating that anxious arousal weakens inhibitory control. MEG source analyses revealed that bilateral IFG and right dorsal prefrontal cortex showed increased beta-band activity (14-30 Hz) to the stop signal that varied as a function of successful stopping during nonanxious (safe) conditions only. Moreover, peak beta-band responses from right IFG were inversely correlated with stopping efficiency during nonanxious conditions. These findings support theoretical claims that beta oscillations function to maintain the current sensorimotor state, and that the lack of differential beta-band activity in prefrontal cortices underlies anxiety-related deficits in inhibitory control. We specifically argue that altered right IFG functioning might directly link impaired cognitive control to heightened stimulus-driven responding in anxiety states.

Negative urgency is related to impaired response inhibition during threatening conditions.

While it has been argued that impulsivity and inhibition are unrelated, previous evidence suggests that the relationship between the two can only be seen when their characteristics are closely matched. The negative urgency subscale of the UPPS-P describes impulsive action during negative affect. This was predicted to correlate more strongly with stop-signal reaction-time (SSRT) during threatening conditions than non-threatening conditions. Healthy participants (N = 68) completed the stop-signal task in threatening (induced by threat-of-shock) and non-threatening conditions after completing the UPPS-P and Spielberg State Trait Anxiety Inventory (STAI) scales. Negative urgency correlated with the difference in SSRT (threat - safe) after controlling for other variables. Conversely, similar correlations were not observed for positive urgency, suggesting threat increases the poorer inhibition seen in those high on negative urgency but not for those high on positive urgency. Additionally, sensation seeking correlated with the difference in SSRT (threat - safe) in the opposite direction, suggesting sensation seeking was related to a reduction in the effect of threat. The findings suggest the relationship between negative urgency and inhibition is facilitated by threatening conditions and that high sensation seekers experience threatening stimuli differently.

Threat-induced anxiety weakens inhibitory control.

Growing evidence indicates that anxiety impairs cognitive control processes, including inhibitory functioning. However, there are reports of anxiety state-related improvements in response inhibition performance in a go/nogo (GNG) task. Here we employed the stop-signal task (SST) to examine in complementary fashion the link between anticipatory anxiety and inhibitory control. Participants (N = 45) completed the SST under threat of unpredictable shocks and safe conditions while physiological activity (skin conductance and heart rate) was monitored. In addition to increased physiological activity, we found that stop-signal reaction time (SSRT), a robust measure of stopping efficiency, was prolonged during threat compared to safe without any difference in choice reaction times to go stimuli. This finding supports the claim of impaired inhibitory control in anxiety, and by consideration of differences between the SST and GNG tasks, can be reconciled with evidence of improved response inhibition on the latter under similar threat conditions.