Characteristics As Well As Cognitive And Neural Mechanisms Of Attentional Bias And Attentional Control In Trait Anxiety

Objective: It is well known that trait anxiety easily leads to a bias in attention towards emotionally threat-related information,and attentional bias to threat is thought to play a central role in the onset and sustainment of anxiety and anxiety disorder.However,due to the different experimental paradigms adopted and the different trait anxious individuals recruited,the experimental results by different investigators are considerably quite different.Up to date,the components of attentional bias and its characteristics over time in trait anxiety still remain unclear.Additionally,the relationship between trait anxiety and attentional bias may be regulated by attentional control.Therefore,attentional bias and attentional control of trait anxiety are key issues in our study.In this study,nonclinical low trait anxious(LTA)participants,high trait anxiety(HTA)participants,and clinical patients with generalized anxiety disorders(GAD)were recruited to clarify suitable ranges of the Neurocognitive Model of Anxiety and the Attentional Control Theory,and to explore characteristics as well as cognitive and neural mechanisms of attentional bias and attentional control in different levels of trait anxious individuals.Methods:(1)Emotional spatial-cuing experiment was adopted and inter-stimulus interval was systematically varied between 50 ms and 800 ms to clarify the components of attentional bias and its characteristics over time among different levels of trait anxious individuals.Additionally,the underlying mechanism was also explored by analysing frontal alpha asymmetry.(2)Emotional and nonemotional Flanker experiments were conducted to study the distracter inhibition and conflict adaptation effect among LTA,HTA and GAD groups,and to test the Neurocognitive Model of Anxiety and the Attentional Control Theory.(3)Single-task and dual-task emotional Flanker experiments were conducted to explore the shifting function and its underlying mechanism among different levels of trait anxious individuals.Results:(1)In the emotional spatial-cuing experiment,it was found that reaction times(RTs)of GAD group(473.74 ± 15.50)were significantly longer than those of the LTA group(383.75 ± 14.64,p < 0.001)and HTA group(381.24 ± 12.92,p < 0.001),while there were no significant difference between other two groups(p = 0.898).Furthermore,groups did not differ significantly with regard to the attentional bias at 50 ms inter-stimulus interval(ISI).However,at 800 ms ISI,we found impaired disengagement from threatening information in the HTA group and negative attentional avoidance in the GAD group.Resting-state EEG results showed that trait anxiety could significantly predict frontal alpha asymmetry after controlling for depression.It was also found that GAD individuals had lower left frontal cortical activity than other two groups.(2)In the emotional Flanker experiment,it was found that RTs of the GAD group(930.38 ± 22.87)were significantly longer than those of the LTA group(632.98 ± 19.81,p < 0.001)and HTA group(655.11 ± 17.71,p < 0.001).The error rates of the GAD group(2.51 ± 0.40)were greater than those of the LTA group(1.40 ± 0.31,p = 0.038)and HTA group(1.37 ± 0.31,p = 0.027).Compared to the other two groups,the GAD patients showed prolonged N2 latency and decreased N2 amplitude.HTA and GAD individuals had significantly smaller P3 amplitude than the LTA group.Compared to LTA,HTA and GAD individuals had higher levels of response conflict,as indexed by both behaviors and the N2 component.(3)In the nonemotional Flanker experiment,it showed that RTs of the GAD group(619.69 ± 129.85)were significantly longer than those of the LTA group(480.52 ± 48.18,p < 0.001)and HTA group(479.37 ± 57.71,p < 0.001),while there was no significant difference between the LTA group and HTA group(p = 0.945).The interference effects of GAD group were significantly higher than those of the LTA group(p = 0.001)and the HTA group(p < 0.001),while there was no significant difference between the LTA group and the HTA group.On the error rate index,the conflict adaptation level of the GAD group was significantly lower than those of the LTA group(p = 0.016).The latency of the N2 component in the GAD group(265.89 ± 5.45)was longer than those in the LTA(243.70 ± 5.45,p = 0.005)and HTA groups(252.54 ± 4.78,p = 0.070).The P3 amplitude in the GAD group(4.94 ± 0.86)was significantly lower than those in the LTA group(9.10 ± 0.86,p = 0.001)and the HTA group(7.87 ± 0.75,p = 0.013).There was no significant difference between LTA and HTA groups(p = 0.287).(4)In the dual-task emotional Flanker experiment,RTs of the GAD group(1052.39 ± 209.40)were significantly longer than those of LTA(719.88 ± 127.87,p < 0.001)and HTA groups(736.76 ± 106.72,p < 0.001).The error rates of GAD subjects(5.38 ± 0.63)were significantly larger than those of LTA(3.05 ± 0.52,p < 0.001)and HTA groups(2.95 ± 0.46,p < 0.001).On the error rate index,the shifting cost of the GAD group(6.10 ± 5.74,p < 0.001)was significantly larger than those of the other two groups(3.33 ± 2.47,p = 0.012;3.11 ± 2.27,p = 0.005,respectively).These three groups showed no significant difference in ERP components.(5)In the emotion spatial-cuing experiment,response time variabilities(RTCVs,3.52 ± 0.64)of clinical GAD patients were significantly higher than those of LTA(3.07 ± 0.62,p = 0.012)and HTA groups(2.96 ± 0.63,p = 0.001),while there was no significant difference between the LTA group and the HTA group(p = 0.487).Similar results were observed in the emotional,nonemotional,and dual-task of Flanker experiments.Conclusions:(1)At late attentional processing stage,compared to the LTA group,the HTA group was slower to disengage attention from threat stimuli,while GAD patients showed attention avoidance towards negative stimuli.Relatively lower resting left frontal cortical brain activation may be the neural basis of attentional avoidance in patients with GAD.(2)GAD group showed conflict maladaptation,poor inhibition function and shifting function in the contexts of emotional and unemotional stimuli,which indicated that GAD group had a malfunction in top-down attentional control.HTA group had conflict maladaptation only in the context of emotional stimuli,while it showed intact inhibition function and shifting function in the contexts of emotional and unemotional stimuli.Therefore,we proposed high trait anxiety and low attentional control may be the critical causes of clinical anxiety disorders.The Neurocognitive Model of Anxiety could explain the characteristics of attention in persons with GAD,while the Attentional Control Theory was more suitable for HTA individuals.(3)Reaction time variability is a hallmark of attentional control,which can distinguish individuals with GAD from normal persons.It is stable across experimental paradigms and experimental materials.