Download files
Abstract
The selective attention towards emotionally significant stimuli, namely, ‘emotional attention’ is mediated by an interaction between top-down (e.g., attentional control) and bottom-up (e.g., sensory) mechanisms. Evidence suggests that viewing emotional facial expressions (EFEs) potentiates low-level perceptual mechanisms underlying stimulus visibility. This is thought to enhance attention towards an emotional item. Emotional attention processes are affected by emotional traits and states. Neuro-cognitive accounts posit that anxiety-related information-processing biases implicate the dissociative influence of state and trait anxiety on bottom-up and top-down mechanisms, respectively.
Domain-specific accounts hypothesise that emotional modulation of vision is based on level of threat and mediated by a ‘fear module’ that pre-attentively evaluates EFEs via an analysis of coarse information (low spatial frequencies; LSF). Anxiety is held to augment the responsivity of this pre-attentive threat detector. Domain-general accounts hypothesise that emotion modulation of visual processing is based upon more general dimensions of significance (e.g., ambiguity) and a ‘flexible use’ of coarse and fine (high spatial frequency; HSF) information, it does not occur independently of attention, and is context dependent.
This thesis explored the emotional modulation of perception and attention, and how different types of anxiety differentially modulated these effects. It employed emotional priming paradigms devised by Phelps, Ling and Carrasco (2006) and Bocanegra and Zeelenberg (2009), which measure the effects of a briefly presented EFE cue on probe contrast sensitivity. Across seven experiments and one post-hoc analysis it aimed to differentiate the underexplored points of contention between domain-specific and domain-general accounts.
Results indicated that emotional effects on perception are not robust without taking into account individual differences in anxiety. State and social anxiety (SA) had the greatest bottom-up effects on emotional priming. Trait anxiety and SA had some smaller top-down effects on emotional priming. In general, the results support the domain-general view that anxiety is a disorder of vigilance, representing continued attempts to acquire additional information about the possibility of threat based on even mild and ambiguous threat or positive EFE cues. Anxiety-related emotional priming appeared to be driven by a ‘flexible use’ of HSFs and LSFs. Notably, anxiety-related emotional priming by ambiguous EFEs cues was context dependent.