Background and objectives: research shows that after exposure treatment, re-exposure to a previously feared stimulus outside the treatment context can result in renewal of fear. This study investigates whether conducting exposure treatment in multiple real-life context can attenuate renewal of fear.

Introduction

Following successful exposure based treatment of specific phobias, there’s high risk of relapse of anxiety symptoms. Conditioning research has given evidence that the renewal effect is an underlying mechanisms responsible for the return of fear. Research concludes that a renewal of fear may happen when a feared stimulus is encountered outside of the treatment context. So its required to establish methods to enhance generalizability of exposure treatment across contexts.

Lab-based research with humans has shown that extinction treatment in multiple contexts does and does not attenuate renewal – methodological differences? Clinical analogue studies, however, have consistently found that conducting exposure treatment in multiple contexts attenuates renewal of fear when follow up is conducted in novel contexts. Support was found that exposure treatment using multiple stimuli (different spiders) enhances generalizability of exposure treatment.

Vasteenwegen et al. (2007) exposed spider anxious students to videos of a spider in different filmed contexts. Follow up testing in a novel context found a significant renewal of fear indicated by self-report and skin conductance for the group of the spider filmed in one context. Renewal of fear was attenuated for the group exposed to the video of a spider in multiple contexts.

Recently, Shiban et al. (2013) attenuated renewal of fear in 40 spider phobic individuals using virtual spider and multiple virtual contexts that differed by background colour. Self-reported fear and skin-conductance responses showed significant renewal of fear to a virtual spider for those that received exposure treatment in only one virtual context. Those receiving treatment in multiple virtual contexts, renewal was attenuated.

This study aims to extend the findings of these previous reports. Neither study used real-life contextual changes which could’ve limited the applicability of those studies to real-life situations where contexts may: a) vary by multiple sensory cues, b) present unique challenges, c) vary on the informative value of the present cues. Both experience with a task and the informative value of contextual cues have been found to moderate attention to contextual cues and consequently affect the context dependence of learning.

Shiban et al. conducted a behavioural avoidance test (BAT) using a real-life contextual change and spider to examine the generalizability of their virtual reality treatment. But during the test group differences were limited to behavioural avoidance and participants weren’t told to touch the spider, potentially resulted in ceiling effects on fear renewal. But single extinction context group found to be more avoidant of the real spider than the multiple extinction context group.

This study examines whether conducting exposure treatment in multiple real-life context with a real spider enhances the generalizability of exposure treatment to novel contexts and attenuates fear renewal. This study allowed participants to complete any step they were willing at each stage, enhancing likelihood of observing avoidance.

Participants were randomly allocated to either:

• Control group (BBB): B) treatment in one context, B&B) following up in the same context.
• Single extinction context group (BEF): B) treatment in one context, E&F) follow up in novel context.
• Multiple exposure context group (BCDEF): B, C, and D) treatment in three different contexts, E&F) each follow up in a different context.

Follow up testing was conducted one week and four weeks after treatment for all groups. It was hypothesized that there would be renewal of fear for the BEF group. Also hypothesized that renewal of fear would be attenuated for the BCDEF group.

Method

Participants

46 moderate to extremely fearful participants scoring between 17-26 on the Spider Phobia Questionnaire (SPQ) participated for treatment benefits and/or in exchange for course credit. Recruitment was via website advertisement or mass testing sessions using SPQ during university classes. Participants were randomly assigned to the three conditions, and group membership was independent of gender.

Therapist

To ensure consistency in treatment adherence and pace of treatment across participants, the same exposure hierarchy was used for all participants, a treatment manual was devised and used at each session, and the researchers frequently discussed adherence to the treatment manual.

Apparatus

Non-harmful spider was used – nephila plumipes. Same spider was used throughout the experiment. The experimental contexts were locations in the university campus and were counterbalanced across groups and phases of the experiment.

Procedure

The procedure consisted of three sessions lasting a maximum of 4 hours. Initial session was 35 minutes of pre- and post-exposure assessment, and 1.45-hour exposure treatment. Follow up testing sessions one (FU1) and four weeks (FU2) after exposure were 35 minutes each. Participants were randomly assigned to the treatment groups.

Spider was placed 3 meters opposite the participant and the participant had to perform a self-chosen step on the hierarchy and report their fear level at that distance. They rated their maximum fear at the current step and anticipated fear of the next step of the hierarchy. If a participants fear during a step dropped to 1- or below on the 100 point SUDS (subjective unit of distress) scale, they had to perform the step again without the therapist being present.

Treatment was completed when all steps of the hierarchy were completed. They were given a 5-minute resting period without the spider to provide a post-treatment baseline. Next they were asked to complete the last step of the hierarchy again without the therapist present, providing a post-treatment SUDS.

Results

There was a significant increase in SUDS, HR, and avoidance measures for all groups between post-treatment and follow up sessions. This increase is probably due to the expected spontaneous recovery effect caused by the delay from post-treatment to follow up. The BEF group showed an overall pattern of larger fear than the BBB and BCDEF groups. This indicates that renewal of fear occurred and that it was attenuated by conducting exposure treatment in multiple contexts.

Discussion

Study showed successful treatment of moderate to high spider fear using one session exposure treatment. Results showed that renewal of fear happened when exposure was conducted in one context when subsequent fear tests were conducted in novel contexts, confirming the first hypothesis.

Strength of this study is that renewal was found with a triad of measures including, verbal, physiological, and behavioural measures.

Other methods to attenuate fear renewal have been identified. Future research could combine extended exposure or context similarity with exposure in multiple contexts to examine if this maximizes treatment benefits and reduces likelihood of renewal.

The effects of extinction treatment in multiple contexts on renewal of fear can be explained using the memory model of learning. During fear acquisition learning CS-US associations (spider-pain) are learnt and stored in memory. During extinction treatment, CS-noUS associations (spider-no pain) are learnt and stored in memory alongside CS-US making the relationship between CS and US ambiguous. When the feared object (CS) is encountered following treatment, the ambiguity between the CS and US is resolved by memory retrieval cues from the environment. If the cues in the follow up context overlap with the treatment context more than the acquisition context, the CS-noUS association will be retrieved and renewal of fear attenuated. If not CS-US association is retrieved and renewal of fear will occur.

So the more cues present form the exposure treatment context at follow up, the more likely that the CS-noUS association will be retrieved and renewal will be attenuated. For the BCDEF group, conducting exposure treatment in multiple contexts theoretically created more overlapping cues between the exposure treatment and follow up contexts when compared to the initial fear acquisition context and novel follow up contexts, so their renewal of fear was attenuated. For the BEF group (single exposure context) it seems the overlap in contextual cues between treatment and follow up weren’t enough to avoid the CS-US association retrieval, leading to renewal of fear.

Some limitations: it used moderate to high spider fearful participants rather than people with spider phobia. Secondly, it used the same therapist throughout the experiment meaning they weren’t blind to the therapist – could result in some bias in the results. This was fixed by having the therapist be absent during post-treatment and follow up.

In conclusion, the study observed verbal, physiological, and behavioural evidence that renewal can be attenuated by conducting exposure treatment in multiple contexts. This provides clinicians with a better understanding of the role of context in relapse and how the likelihood of relapse can be reduced by making small but important adjustments to already empirically validated behavioural treatment protocols.