Biological Psychology 92 (2013) 342–352 Contents lists available at SciVerse ScienceDirect Biological Psychology j ourna l h o me pag e: www.elsevier.com/locate/biop sycho Deficient safety learning characterizes high trait anxious individuals Femke J. Gazendam ∗ , Jan H. Kamphuis, Merel Kindt Department of Clinical Psychology, University of Amsterdam, Weesperplein 4, 1018 XA Amsterdam, The Netherlands a r t i c l e i n f o Article history: Received 18 October 2011 Accepted 10 November 2012 Available online 19 November 2012 Keywords: Fear conditioning Startle response Fear extinction Trait anxiety Fear inhibition Fear generalization Anxiety disorders a b s t r a c t Trait anxiety is a well-established risk factor for developing anxiety disorders, but evidence for abnormal associative fear learning in high trait anxious (HTA) individuals is inconclusive. In part, this may due to limitations in the scope and measures used to assess fear learning. The current study therefore assessed fear learning across multiple response domains and multiple test phases in a two-day discriminative fear-conditioning paradigm. We tested whether trait anxiety is associated with deficient safety learning, by comparing HTA individuals (N = 20) and healthy Controls (N = 22). HTA participants showed stronger fear on the startle response and distress ratings to the safety (CS - ) but not to the threat stimulus (CS + ) during acquisition, along with impaired extinction and re-extinction. Trait anxiety did not affect skin conductance responses and effects on UCS-expectancy were limited. We conclude that high trait anxiety may be characterized by deficient safety learning which in turn may promote persistent and generalized fear responses. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Anxiety disorders are among the most prevalent and debilitat- ing psychiatric disorders (Kessler et al., 2005; Olatunji et al., 2007), underscoring the need for research into its core pathogenic factors. A well-established risk factor for the development of anxiety disor- ders is neuroticism or trait anxiety (Gershuny and Sher, 1998; Jorm et al., 2000; McNally et al., 2003). Trait anxiety refers to the gen- eral tendency to react negatively to stressful situations (Andrade et al., 2001; Bados et al., 2010; Clark et al., 1994; Spielberger et al., 1970). To advance our understanding of trait anxiety as a risk factor for anxiety disorders, research approaches range from prospective designs examining the predictive relation between personality characteristics and anxiety disorders (e.g., Gershuny and Sher, 1998) to experimental studies testing how distinct (e.g., neurocognitive or behavioral) systems engaged in anxiety devel- opment behave as a function of trait anxiety. A particularly suitable approach to elucidate the mechanisms by which trait anxiety promotes the development of anxiety disor- ders is fear conditioning. In the pathogenesis of anxiety disorders, associative fear learning is considered a central mechanism (e.g., Mineka and Zinbarg, 2006), bearing many similarities with how anxiety develops in real life (Field, 2006; Merckelbach et al., 1996; Ollendick and King, 1991). In a typical discriminative fear ∗ Corresponding author at: University of Amsterdam, Faculty of Social and Behavioral Sciences, Department of Clinical Psychology, Weesperplein 4, 1018 XA Amsterdam, The Netherlands. Tel.: +31 20 525 8601; fax: +31 20 639 1369. E-mail address: F.J.Gazendam@UvA.nl (F.J. Gazendam). conditioning procedure, a neutral or ambiguous stimulus (i.e., con- ditioned stimulus, CS1 + ) (e.g., picture) acquires the capacity to elicit fear responses after pairing it with an intrinsically aversive event (i.e., unconditioned stimulus, UCS) (e.g., electric stimulus) (Pavlov, 1927). A second, nonreinforced conditioned stimulus (CS2 - ), which is never paired with the UCS, serves as a safety cue. Conditioning to the CS1 + elicits excitatory processes, and conditioning to the CS2 - elicits inhibitory processes (e.g., Lissek et al., 2005). The procedure to reduce conditioned fear responding is extinction: the repeated presentation of the feared stimulus without the aversive conse- quence (UCS). During extinction, inhibitory processes are thought to deactivate the (excitatory) threat (CS1 + /UCS) association (for a review, see Davis et al., 2000). Although the fear conditioning paradigm is a strong model for the learning and unlearning of fear, the basic conditioning proce- dures may fall short in modeling pathological fear development. Fear acquisition (CS1 + ) itself is an adaptive phenomenon, facili- tating detection of warning signals of threat (e.g., Beckers et al., 2012; Frijda, 1986). Therefore, anxiety disorders may only develop if associative fear learning progresses in a maladaptive manner (e.g., Mineka and Zinbarg, 1996, 2006). Several potential mechanisms have been proposed to explain elevated fear (Craske et al., 2009; Lissek et al., 2005). First, a long standing hypothesis to explain elevated fear expression to the threat stimulus (CS1 + ) is enhanced conditionability (e.g., Orr et al., 2000; Otto et al., 2007), referring to stronger excitatory fear mech- anisms. A second hypothesis is that elevated fear responses to the safety stimulus (CS2 - ) are related to a deficit in awareness of the contingencies (i.e., the general discrete knowledge on the (non)occurrence of the UCS) (Grillon, 2002). Third, some consider 0301-0511/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.biopsycho.2012.11.006