Context and renewal of conditioned fear: An experimental evaluation using 20% carbon dioxide-enriched air as an unconditioned stimulus Carlos G. Finlay *, John P. Forsyth University at Albany, State University of New York, United States Exposure-based interventions are at the core of virtually all efficacious treatments for anxiety disorders (see Richard & Lauterbauch, 2007, for a comprehensive edited volume). Such interventions involve prolonged in vivo or imaginal contact with feared stimuli both to attenuate fearful responding and to counteract avoidance or escape behaviors (Barlow, 2002; Bouton, 2002; Bouton & Nelson, 1998; Forsyth, Barrios, & Acheson, 2007; Forsyth & Eifert, 1996, 1998; Rachman, 1990, 1991; Wolpe, 1958). Though efficacy of exposure-based interventions is well estab- lished, increasing attention has focused on the durability of treatment gains and prevention of fear relapse. It is widely known that some patients experience a re- emergence of subjective and autonomic fearful responding following successful exposure-based treatments. This ‘‘fear renewal’’ is considered distinct from fears that are re-acquired via additional aversive learning (i.e., a second, unrelated traumatic event signaled by the same feared stimulus), the acquisition of new fears, or a full fear relapse in that reported elevations are not necessarily at clinically significant levels (Rachman, 1979, 1989). Fear renewal is thought to denote one of several precipitants for relapse in anxious persons. Recent accounts suggest fear renewal is best understood as the result of new learning applied to an established CS during exposure therapy, rather than unlearning of conditioned response (CR; Bouton, Mineka, & Barlow, 2001; Bouton & Swartzentruber, 1991). That is, extinction is viewed as a suppression of the CR by additional association of a previously established CS to the absence of the US. This CR suppression serves as a competing response elicited by the CS, and it is argued that other environmental stimuli (i.e., context) determine which response is likely elicited (Bouton, 2000, 2002; Bouton & Nelson, 1998; Holland & Bouton, 1999). Laboratory investigations with animals have shown renewal of classically conditioned responses following extinction when a subsequent test phase is conducted in a novel context or following return to the context in which the conditioned response was established (Bouton & Bolles; 1979; Bouton & Ricker, 1994). Additional work has also demonstrated contextual control over mixed histories of reinforcement and nonreinforcement (Bouton & King, 1986), drug seeking behaviors (Crombag & Shaham, 2002), appetitive conditioning (Brooks & Bouton, 1994), competing conditioned fear inhibitors and excitors (Thomas & Ayres, 2004), and taste aversion preparations (Rosas & Bouton, 1997). Overall, Journal of Anxiety Disorders 23 (2009) 737–745 ARTICLE INFO Article history: Received 2 October 2008 Received in revised form 24 February 2009 Accepted 26 February 2009 Keywords: Anxiety Fear renewal Context Fear conditioning Exposure therapy ABSTRACT This analogue experiment used fear conditioning and extinction procedures to establish and reduce fearful responding and then test for fear renewal following a context change. Healthy undergraduates (N = 61) underwent a differential fear conditioning procedure using geometric shapes as conditioned stimuli (CS) and inhalations of 20% CO 2 -enriched air as an aversive unconditioned stimulus (US). Across phases (i.e., habituation, acquisition, extinction, and test), red and green ambient lighting served as contextual stimuli, and electrodermal and evaluative ratings were assessed as indices of conditioned fear. The control group underwent extinction and test phases under identical conditions (i.e., no context changes). The experimental groups either underwent acquisition and test phases in one context and the intervening extinction phase in a different context (A-B-A), or underwent acquisition and extinction phases in one context and the test phase in a novel context (A-A-B). Consistent with expectation, fear renewal was observed when the test context matched the acquisition context. This effect was modest for electrodermal responses, but reasonably robust for evaluative responses. The role of context in the subsequent renewal of fear following exposure-based treatment is discussed. Published by Elsevier Ltd. * Corresponding author at: National Center for PTSD, VA Boston Healthcare System, 116-B, 150 South Huntington Avenue, Boston, MA 02130, United States. Tel.: +1 857 364 6322; fax: +1 857 364 4501. E-mail address: carlos.finlay@va.gov (C.G. Finlay). Contents lists available at ScienceDirect Journal of Anxiety Disorders 0887-6185/$ – see front matter . Published by Elsevier Ltd. doi:10.1016/j.janxdis.2009.02.014