Effects of increased attention allocation to threat and safety stimuli on fear extinction and its recall

Highlights

• A subtle attentional manipulation was implemented during extinction.

• The manipulation resulted in increased attention to stimuli during this phase.

• Increased attention during extinction influenced fear-related physiological response.

Abstract

Background and objectives

Attention plays an important role in the treatment of anxiety. Increased attention to threat has been shown to yield improved treatment outcomes in anxious patients following exposure-based therapy. This study examined whether increasing attention to learned stimuli during fear extinction, an experimental analogue for exposure-based treatments, could improve extinction learning and its maintenance.

Methods

Sixty-five healthy adults were randomized into experimental or control conditions. All completed a differential fear conditioning task. During extinction, a subtle attentional manipulation was implemented in the experimental group, designed to increase participants’ attention to both threat and safety cues. Three days later, an extinction recall test was conducted using the original cues and two perceptually similar morphs.

Results

Fear conditioning was achieved in both behavioral and psychophysiological measures. In addition, between-group differences emerged during extinction. The experimental group exhibited increased attention to stimuli and lower fear responses in physiological measure than the control group. Similarly, during extinction recall, the experimental group exhibited lower startle responses than the control group. Last, across groups, attending to the safety cue during extinction was associated with lower self-reported risk of the two generalization morphs displayed during extinction recall.

Limitations

Skin conductance response (SCR) was not measured during extinction recall. Future research should include both SCR and additional generalization morphs so as to allow for the examination of more subtle individual differences.

Conclusions

Results indicate that the attentional manipulation increased attention allocation to stimuli during extinction; this, in turn, affected fear-related physiological response.

Introduction

Anxiety disorders are a common form of psychopathology associated with significant impairment (Kessler et al., 2005). While exposure-based cognitive behavioral therapy (CBT) is the gold-standard for treatment, approximately half of patients do not respond (for a review, see Loerinc et al., 2015). Investigations using an experimental analogue of exposure-based therapy, called fear extinction, have yielded new insights into the mechanisms by which these treatments work (Craske, Hermans, & Vervliet, 2018). For example, the way anxious individuals attend to or avoid fear-related stimuli may influence fear extinction and, thus, treatment response (for a review, see Barry, Vervliet, & Hermans, 2015; also Barry, Vervliet, & Hermans, 2016, 2017; O'Malley & Waters, 2018). To further assess the influence of attention on fear extinction processes, we examined whether fear extinction could be enhanced if healthy participants were trained to direct their attention towards stimuli during extinction learning.

Reductions in fear during and after exposure-based CBT are often explained within a classical fear conditioning and extinction framework (Waters, LeBeau, & Craske, 2017). In these models, fear is acquired when a neutral stimulus (conditioned stimulus; CS) co-occurs with an aversive stimulus (unconditioned stimulus; US), creating an association between them. The formerly neutral stimulus comes to represent a danger cue (CSP), resulting in a conditioned fear response in anticipation of the US (i.e., fear conditioning). Fear is extinguished when the CSP is repeatedly presented in the absence of the US, such that it no longer serves as a danger cue (i.e., fear extinction). In differential conditioning paradigms, an additional CS is presented during both fear conditioning and extinction that is not paired with an aversive stimulus, thereby serving as a safety cue (CSM). This allows for the assessment of threat-safety discrimination and fear generalization. Notably, in a clinical context, during exposure therapy, anxious clients are systematically and repeatedly exposed to the objects and objects related to their fears until they learn they no longer signal danger. Impaired extinction has been observed in lab-based studies among anxious adults (Duits et al., 2015) and youth (Dvir, Horovitz, Aderka, & Shechner, 2019) compared to their healthy counterparts. Arguably, then, enhancing extinction learning is clinically relevant and could lead to improvements in exposure-based therapy.

Relatedly, the maintenance and generalization of extinction learning is also of clinical relevance. In a laboratory setting, extinction recall tasks are typically used to assess these factors, whereby the extinguished danger cue is presented again following extinction. The generalization of extinction learning can then be examined through the use of generalization stimuli (GS). Often, this is done by presenting participants with a series of GSs ranging in perceptual similarity from the CSM to the CSP (e.g., Michalska et al., 2019; Shechner et al., 2018). Plotted fear responses to each morph can then be examined. Although the majority of laboratory work examining fear generalization has been conducted in healthy adults (for review, see Dymond, Dunsmoor, Vervliet, Roche, & Hermans, 2015), anxious participants have been shown to generalize fear onto stimuli that share perceptual characteristics to the CSP as compared to healthy controls (e.g., GAD: Lissek et al., 2014; panic disorder: Lissek et al., 2010). Further, original danger cues are often not readily accessible during exposure-based treatments, and so participants are exposed to fear-relevant GSs. Therefore, understanding individuals’ responses to GSs during extinction recall is imperative in order to understand the persistence and generalization of extinction learning.

Recently there has been increased interest in understanding the precise mechanisms impacting the fear reduction observed following extinction. One such mechanism is that of attention, which plays an important role in fear learning. Stimuli are prioritized and processed through either voluntary, top-down, goal-directed processes or automatic, bottom-up attentional capture (Weierich, Treat, & Hollingworth, 2008). Within these two dimensions, attention can be divided into overt (e.g., eye movements) or covert (e.g., internal attention) processes, which together, form a gateway to information processing. A growing body of research suggests biases in visual attention allocation may be especially important during extinction learning. For example, a recent study of anxious and non-anxious youth found a tendency to attend to, as opposed to avoid, threat resulted in better extinction in behavioral and psychophysiological measures (Waters & Kershaw, 2015). Further, several treatment studies found clinically anxious individuals who selectively attended to threat had better recovery rates following exposure-based therapy than individuals who displayed threat avoidance (Barry, Sewart, Arch, & Craske, 2015; Price, Tone, & Anderson, 2011; Waters, Mogg, & Bradley, 2012) or did not display a specific threat-related attentional allocation pattern (Niles, Mesri, Burklund, Lieberman, & Craske, 2013). Other studies found selective attention away from threat was associated with CBT success among anxious youth (Legerstee et al., 2009, 2010). Importantly, these equivocal results highlight that much is still unclear regarding the precise relationship between attention and fear reduction. Further, notably in all of the above studies differences in attentional processes were measured before extinction or exposure treatment, making it difficult to ascertain the degree to which attention allocation during extinction/exposure directly impacted learning or treatment success.

To the best of our knowledge, only two studies have attempted to manipulate attention allocation during extinction (Barry, Vervliet, & Hermans, 2017; O'Malley & Waters, 2018). O'Malley and Waters (2018) used a differential conditioning task. Participants were instructed to either actively avoid or attend to the CSP during extinction; a control group received no instructions. Participants in the avoidant group displayed poorer extinction and a greater return of fear to the CSP than the attending and control groups. Barry et al. (2017) also used a differential conditioning task; in this case, stimulus features were modified between the acquisition, extinction, and extinction-test phases. During extinction, participants were instructed to examine either unique or common features of the CSP used in extinction, which shared perceptual characteristics with the CSP used in acquisition. The group instructed to look at unique features showed a greater return of fear, as measured by skin conductance, in the test phase when a third, fear-related stimulus was shown. Both studies used explicit instructions to increase participants' attention during extinction, which did not allow for a direct examination of whether or not participants heeded their received instructions. Further, all phases of the experiments were administered on a single day, in which only the immediate effect of attention allocation can be tested. That said, together the two aforementioned studies suggest that attending to specific stimuli during extinction may enhance extinction learning and its maintenance.

The current study expanded on the above findings by increasing attention to both threat and safety stimuli presented during extinction through use of a subtle attentional manipulation, with an overall aim of enhancing extinction learning and further elucidating the relationship between attention allocation and fear reduction. This was implemented through eye tracking measurements, which allowed for a direct examination of whether or not individuals did in fact allocate increased visual attention to stimuli during extinction. In addition, extinction recall was conducted three days following fear conditioning and extinction in order to examine the long-term effect of attention allocation on extinction learning. Last, during extinction recall, the influence of visual attention allocation on fear generalization gradient ranging in perceptual similarity from the CSP to the CSM was tested. We hypothesized that participants in the experimental condition would exhibit enhanced extinction learning compared to the control group, and these differences would be maintained in a subsequent extinction recall task.