Biological Psychology 88 (2011) 188–195 Contents lists available at ScienceDirect Biological Psychology jo ur nal homep age : www.elsevier.com/locate/biopsycho Neural activation during the processing of ambiguous fearful facial expressions: An ERP study in anxious and nonanxious individuals Tahl I. Frenkel ∗ , Yair Bar-Haim The Adler Center for Research in Child Developmental and Psychopathology, Department of Psychology, Tel Aviv University, Tel Aviv 69978, Israel a r t i c l e i n f o Article history: Received 27 December 2010 Accepted 1 August 2011 Available online 16 August 2011 Keywords: Anxiety ERP Threat-processing Attention bias Face-processing a b s t r a c t Event-related brain potentials (ERPs) were recorded from anxious and nonanxious participants during performance on a fear detection task. Sequential presentation of gradually increasing fear cues from neutral to fearful allowed an examination of anxiety-related differences in the neural activation patterns corresponding to participants’ overt detection of fear in ambiguous stimuli as well as the activation pat- terns corresponding to stages of fear processing preceding overt fear detection. While centro-parietal Late Positive Potential (LPP) amplitude of nonanxious participants was significantly modulated by increases in stimulus fear intensity preceding overt fear detection, no such LPP sensitivity was detected in anx- ious participants. Additionally, anxiety group differences as well as emotion related modulation were found for earlier ERP components (P1, P2 and EPN). These findings reveal an anxiety-related dissociation between the early and late processing stages of threat processing. Implications are discussed in light of existing theories of cognitive biases in anxiety. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Behavioral research has documented anxiety-related differ- ences in the classification of ambiguous threat-related facial expressions (e.g., Richards et al., 2002). Based on such findings, biased processing of ambiguously threatening facial cues has been implicated as an etiologic and maintaining factor in anxiety (e.g., Mathews and Mackintosh, 2000; Mathews and MacLeod, 2002; Mathews et al., 2007; Wilson et al., 2006). Here we examine the neural correlates of anxiety-related dif- ferences in the fear detection process. Specifically, event-related potentials (ERPs) were recorded during a fear detection task in which interpolated face stimuli were gradually unfolding from neutral, through mildly fearful, and up to moderately fearful expressions which eventually elicited an overt detection of fear. As such, the study design allowed neural activation to be examined both at the time of overt fear detection, and during the processing of fears’ precursors in slightly less fearful stimuli not yet eliciting overt detection of fear. Previous studies have employed a variety of emotional facial expressions as threat-related stimuli, for instance angry (e.g., Bar- Haim et al., 2005; Fox et al., 2002; Horley et al., 2003, 2004; Kolassa ∗ Corresponding author. Tel.: +972 3 6405465; fax: +972 3 6409547. E-mail address: tahlfren@post.tau.ac.il (T.I. Frenkel). et al., 2009; Mogg et al., 2004; Mueller et al., 2009; Richards et al., 2002; Wilson and MacLeod, 2003), fearful (Fox, 2002; Holmes et al., 2008; Richards et al., 2002), disgust (Gilboa-Schechtman et al., 1999; Mansell et al., 1999). While some facial expressions are thought to be highly relevant for specific types of anxiety such as disgusted faces for socially anxious individuals (Amir et al., 2010; Cobb et al., 2005), or to elicit specific threat cues (e.g., an angry face in a frontal presentation), fearful faces are thought to signal the presence of a diffuse/unknown danger in the immedi- ate environment thus, serving as a threatening stimulus that is generally effective across a wide range of contexts. Indeed, neu- roimaging research indicates that fearful faces elicit the strongest fear-circuitry response (e.g., Dolan and Vuilleumier, 2003). Thus, here we examine anxiety-related differences in the fear detection process. Threat detection in faces is comprised of several stages of infor- mation processing that can be grossly divided into early processes such as stimulus encoding and selective allocation of attention, and late strategic processes reflecting sustained motivated atten- tion and stimulus interpretation. In attempt to provide further information on the time course and neural correlates of these, the present study focused on the examination of both early and late ERP components that are thought to differentially reflect these dis- tinct cognitive processes. Specifically, early ERP components such as the attention-related P1 (e.g., Batty and Taylor, 2003; Eimer and Holmes, 2002, 2007; Holmes et al., 2003; Streit et al., 2003) and P2 0301-0511/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.biopsycho.2011.08.001