Timing the fearful brain: unspecific hypervigilance and spatial attention in early visual perception

doi:10.1093/scan/nst044 SCAN (2014) 9, 723^729 Timing the fearful brain: unspecific hypervigilance and spatial attention in early visual perception Mathias Weymar,1,2 Andreas Keil,2 and Alfons O. Hamm1 1 Department of Biological and Clinical Psychology, University of Greifswald, 17487 Greifswald, Germany and 2Center for the Study of Emotion and Attention (CSEA), University of Florida, Gainesville, FL 32611, USA Numerous studies suggest that anxious individuals are more hypervigilant to threat in their environment than nonanxious individuals. In the present event-related potential (ERP) study, we sought to investigate the extent to which afferent cortical processes, as indexed by the earliest visual component C1, are biased in observers high in fear of specific objects. In a visual search paradigm, ERPs were measured while spider-fearful participants and controls searched for discrepant objects (e.g. spiders, butterflies, flowers) in visual arrays. Results showed enhanced C1 amplitudes in response to spatially directed target stimuli in spider-fearful participants only. Furthermore, enhanced C1 amplitudes were observed in response to all discrepant targets and distractors in spider-fearful compared with non-anxious participants, irrespective of fearful and non-fearful target contents. This pattern of results is in line with theoretical notions of heightened sensory sensitivity (hypervigilance) to external stimuli in high-fearful individuals. Specifically, the findings suggest that fear facilitates afferent cortical processing in the human visual cortex in a non-specific and temporally sustained fashion, when observers search for potential threat cues. Keywords: emotion; hypervigilance; spatial attention; C1; event-related potentials (ERPs) INTRODUCTION Voluntary attention to a specific location in the environment results in faster detection and enhanced discrimination for stimuli presented at that location than stimuli at unattended locations. This well-known effect of spatial attention has been demonstrated in striate and extrastriate visual cortical areas (V1–V4; Kastner et al., 1999; Martinez et al., 1999), resulting in increased electrophysiological and hemodynamic activity for attended compared with unattended locations. Recently, Kelly et al. (2008) showed that spatial attention may modulate the initial passage of visual input to primary visual cortex (V1). Using event-related potentials (ERPs), the authors observed enhanced amplitudes of the C1 component in response to spatially cued patterns 50 ms after presenting the visual stimulus. The C1 component has been described for many decades as the earliest cortical component in the visual evoked potential (Regan, 1989). Extensive evidence points to the striate cortex as the neural generator of the C1 (Jeffreys and Axford, 1972, Clark et al., 1995), consistent with its peak latencies 50–100 ms. The C1 topography evinces pronounced retinotopy, with a voltage reversal over posterior scalp sites reliably related to the location of the stimulus in the upper versus lower visual field. Although best established with systematic manipulation of visual field location, pattern onset at foveal and symmetric locations has been shown to elicit C1 in a robust fashion, given sufficient signal-to-noise ratio (Regan, 1989). Under these conditions, the C1 tends to be negative and widely distributed in topography (Baseler and Sutter, 1997). Besides spatial attention and perceptual learning (Bao et al., 2010), the C1 has also been described as sensitive to contextual features such as fearful faces and fear-conditioned stimuli (Pourtois et al., 2004, Stolarova et al., 2006), indicating superior evaluation of biologically relevant stimuli at the earliest stage of cortical processing. If biological significance is capable of biasing sensory neurons, then observers high Received 2 July 2012; Accepted 28 March 2013 Advance Access publication 1 April 2013 We are grateful to Janine Wirkner for her assistance in data collection. This research was supported in part by a post-doctoral stipend from the German Research Society (Deutsche Forschungsgemeinschaft, DFG) to Mathias Weymar (Forschungsstipendium, WE 4801/1-1). Correspondence should be addressed to Mathias Weymar, Department of Biological and Clinical Psychology, University of Greifswald, Franz-Mehring-Strasse 47, 17487 Greifswald, Germany. E-mail: in specific object fear should show sensory facilitation when expecting to confront the feared object. Presently, however, it is unclear how sensory processing varies with inter-individual fear status. A substantial amount of studies have found that anxious individuals are more vigilant to environmental threat than non-anxious individuals (see for review Bar-Haim et al., 2007; Cisler and Koster, 2010; Yiend, 2010; Miskovic and Schmidt, 2012). For instance, participants reporting fear of spiders (snakes) detect spider (snake) stimuli more rapidly than non-fearful participants, in visual search paradigms (Öhman et al., 2001). Participants high in specific fear also display enhanced N2pc amplitudes in the ERP, indicating enhanced attention capture by these stimuli (Weymar et al., 2013). In the same vein, hightrait anxious individuals detect angry faces faster than low-trait anxious individuals (Byrne and Eysenck, 1995), a difference not observed for happy faces. Similar results were found for socially anxious individuals who show a preferential processing of angry facial expression relative to happy expressions and also shorter responses for angry relative to disgust expressions (Gilboa-Schechtman et al., 1999), pointing to a greater attention bias in anxious individuals. In addition to rapid initial attention capture by threat-relevant stimuli, other studies have shown that high anxious individuals have difficulties disengaging attention from fear-relevant stimuli once they are detected (e.g. Fox et al., 2002; Gerdes et al., 2008). Furthermore, others have proposed that anxious individuals tend to avoid the threat stimulus immediately after detection (Mogg and Bradley, 1998; Pflugshaupt et al., 2005). Unspecific hypervigilance or heightened alertness even prior to detecting a threat stimulus has previously been described as characteristic of anxious individuals (Beck et al., 1985; Eysenck, 1992). In this perspective, anxious individuals are constantly looking out for signs of threat or harm in their environment and selectively attend to stimuli signaling possible danger. Specifically, Eysenck proposes a broadening of attention (general hypervigilance) during excessive environmental scanning for threat cues followed by a narrowing of attention when a stimulus is being processed (enhanced selective attention). General sensory hypervigilance has also been framed in the defense cascade model based on work in the animal model of defensive behavior (Fanselow, 1994; Lang et al., 1997). According to this model, defensive behavior is characterize (...truncated)