 |
 |
 |
 |
Within the affective cognitive neuroscience literature, there is an ongoing debate as to whether a fast subcortical thalamoamygdala pathway enables processing of threat-related stimuli that is independent of attentional resources. A number of studies have examined whether the amygdala response to threat is modulated by attention but with mixed results. My aim within this field has been to highlight the importance of considering both individual differences in anxiety and the role of top-down control mechanisms in regulating attention to threat distractors.
Cognitive studies of anxiety have shown that attentional biases towards threat-related stimuli vary dramatically between high anxious and low anxious individuals, yet this is often overlooked in neuroimaging studies. These between-participant differences could reflect up-regulation by anxiety of threat detection mechanisms and/or reduced recruitment of attentional control mechanisms. Drawing on mainstream models of selective attention we have attempted to elucidate the stages of processing modulated by anxiety across a series of neuroimaging studies. The first of these examined whether anxiety influences attentional modulation of the amygdala response to threat [Bishop et al. 2004 J. Neurosci.] We reported that only high anxious, and not low anxious, individuals showed an amygdala response to threat distractors while both high and low anxious individuals showed an amygdala response to attended threat stimuli (see Figure I). This finding does not however resolve whether the augmented amygdala response to threat distractors in high anxious individuals is pre- or post- attentive, there potentially being attentional ‘spillover’ to the nominally unattended items in the visual display.
Figure I. Anxiety influences attentional modulation of amygdala activity to fearful versus neutral faces. (a) Amygdala activity to attended fearful faces (AF) vs attended neutral faces (AN) relative to unattended fearful faces (UF) vs unattended neutral faces (UN) against state anxiety (STAI). Activation plotted is mean signal change associated with this contrast for the peak voxel from the left amygdala ROI, x, y, z = -18 -10 20, Z = 2.80, P-corrected < 0.02. A trend towards a similar relationship was observed within the right amygdala ROI, x, y, z = 26 -12 -18, Z = 2.05, P-corrected =.10 (b) Amygdala activity to fearful versus neutral faces by attentional condition and anxiety level. Participants were divided into ‘low’ and ‘high’ anxious groups using a median split on STAI scores. Amygdala activity is mean signal change for the peak voxel from (a). (Taken from Bishop et al. 2004. J Neurosci.),
This lead us to also investigate the modulatory influence of anxiety upon the recruitment of ‘top-down’ attentional mechanisms involved in regulating the processing of threat distractors. Here, we were interested in questioning the proposed ‘special status’ of threat stimuli, and examining whether mainstream models of attention can successfully account for findings with threat distractors. Specifically, Cohen and colleagues have attempted to dissociate mechanisms involved in detecting processing conflict from those involved in implementing attentional control. We tested whether the proposed role for the lateral prefrontal cortex in augmenting attentional control over expected salient distractors and that for the anterior cingulate cortex in monitoring processing conflict from unexpected (infrequent) salient distractors (Botvinick et al. 2001) also applies in the case of threat distractors. Our findings confirmed that this was indeed the case and that high anxious individuals showed deficits in recruiting both mechanisms [Bishop et al. 2004 Nat. Neurosci.]
More recently, we have been investigating the relevance of models of selective attention under load (Lavie 2001, 2005) to the processing of threat distractors, in particular testing the proposition that the effects of anxiety take place after an initial stage of perceptual competition but prior to the generation of behavioral responses. Our findings here indicate that the amygdala response to threat distractors observed in high anxious volunteers is abolished under high perceptual load [Bishop et al. 2007 Cerebral Cortex]. Within this framework, we are currently investigating whether state and trait anxiety have distinguishable effects upon the mechanisms involved in threat-detection versus cognitive control. Preliminary evidence suggests that high trait anxiety is linked to poor attentional control while state anxiety impacts primarily on mechanisms involved in the detection of threat-related stimuli (see Figures II and III).
Figure II. Two versions of a neurocognitive model of anxiety-related biased in selective attention. (a) Allocation of attentional resources to threat distractors is influenced both by the strength of a threat-detection signal from the amygdala and the strength of a top-down control signal supporting task-related processing. The latter is thought to
emanate from the lateral prefrontal cortex (LPFC), with a rostral anterior cingulate cortical (ACC) region signaling the presence of attentional competition from threat distractors [Bishop et al. 2004 Nat Neurosci.] Anxiety is held to modulate the magnitude of both the amygdala and prefrontal signals, being associated with amygdala hyper-responsivity and frontal hyporesponsivity [Bishop et al. 2004 J. Neuro; Bishop et al. 2004 Nat Neurosci.] (b) An extended version of the model in (a) incorporates recent findings suggesting that: (i) attentional competition involves both early competition for perceptual resources and later competition for further processing, with anxiety modulating processing subsequent to the initial stage of perceptual competition; and (ii) amygdala responsivity is primarily modulated by state anxiety whereas prefrontal recruitment is primarily influenced by individual differences in trait anxiety [Bishop et al. 2007 Cerebral Cortex].
Figure III. Amygdala and ventrolateral prefrontal cortical (VLPFC) activity to fearful versus neutral face distractors (F-N) under low perceptual load relative to fearful versus neutral face distractors (F-N) under high perceptual load as a function of state and trait anxiety scores, respectively (adapted from Bishop et al. 2007 Cerebral Cortex).
I am also interested in examining the neural mechanisms underlying associative threat processing and threat-related interpretative biases and investigating differences in the recruitment of these mechanisms in anxious and non anxious volunteer groups. Some of the literature of relevance to this is reviewed in [Bishop, 2007, Trends In Cognitive Science].
