Seeing the Worst or Saying the Words? A Multilevel Comparison of Occasional Reinforcement and Affect Labeling as Strategies to Augment an Imagery-Based Exposure Intervention

Abstract

Research has shown that some individuals do not benefit from exposure therapy and for those who do, fear can return. Occasional reinforcement, which involves intermittently reinforcing the feared outcome during exposure, and affect labeling, in which individuals describe their current affective state during exposure, are two inhibitory retrieval strategies that have shown promise for augmenting exposure interventions. Yet, it remains unclear whether these strategies differ in their efficacy for attenuating return of fear across multiple levels of analysis. Accordingly, the present treatment-analogue experiment examined the effects of a multi-session imagery-based exposure manipulation that included reminders of the feared outcome or prompts to label one’s affective experience in a phobic sample on threat expectancy, behavioral avoidance, and attentional bias for threat. Community adults (N = 136) who met diagnostic criteria for snake phobia were randomized to a single-cue video exposure alone condition (S), a multiple-cue video exposure condition that occasionally reinforced the feared outcome (i.e., snake bite; FO), or a multiple-cue video exposure condition that instructed participants to label their affective response (AL). Results revealed significant reductions in threat expectancy and behavioral avoidance, but not attentional bias for threat, for all three groups. Although there were no significant group differences in threat expectancy and attentional bias for threat at a 1-week follow up, those in the FO condition completed significantly more BAT steps than the AL group and this group difference was partially mediated by distress variability and within-session fear reduction. The implications of these findings for the inhibitory retrieval theory are discussed.

Keywords

exposure inhibitory retrieval fear extinction phobia

Approximately one third of the population is affected by an anxiety disorder at some point in their lifetime, and many anxious individuals suffer for years or decades after the onset of the disorder (Bandelow & Michaelis, 2015). Translational approaches to understanding how anxiety disorders are developed, maintained, and treated have been directly informed by fear conditioning models. Learned, conditioned associations between a neutral stimulus (CS; e.g., snake) and an unconditioned stimulus (US; e.g., bite) promote excessive fear and avoidance, and are extinguished as the CS is repeatedly presented in the absence of the US (e.g., Lonsdorf et al., 2017). Exposure therapy, which leverages these principles of fear extinction, is considered the key ingredient to successful treatment for anxiety disorders (Olatunji et al., 2010). In fact, meta-analytic research has shown that exposure-based treatment for specific phobias produced large effects sizes relative to no treatment, and outperformed placebo conditions and alternative active psychotherapeutic approaches (Wolitzky-Taylor et al., 2008).

The emotional processing theory (EPT) posits that exposure therapy is effective as individuals acquire corrective information that is incompatible with their original threat beliefs (e.g., I held the snake and it didn’t bite me; Foa & Kozak, 1986). This theory has generated novel hypotheses about the psychopathology of anxiety and related disorders that have directly informed the development and refinement of specific treatment protocols (Foa & McLean, 2016). Craske and colleagues (2014, 2022) have since built upon the EPT and suggested that learning occurs as an individual identifies a mismatch between their expected and observed exposure outcome (i.e., expectancy violation; Rescorla & Wagner, 1972). The inhibitory retrieval theory (Craske et al., 2008) postulates that these expectancy violations yield new inhibitory (i.e., safety) learning that competes with the existing fear association in memory. Given that the original fear association is more easily retrieved across contexts and time, fear often returns after exposure therapy (Craske & Mystkowski, 2006). Accordingly, Craske and colleagues (2014, 2022) have identified inhibitory retrieval strategies that may enhance expectancy violation and attenuate return of fear.

One proposed inhibitory retrieval strategy that may attenuate the return of fear is occasional reinforcement, which involves intermittently reinforcing the fear association throughout exposure therapy. When the aversive outcome is intermittently reinforced, the original fear association may be especially salient, thus facilitating a greater expectancy violation when the feared stimulus is subsequently presented in the absence of the aversive event (Craske et al., 2014; Pearce & Hall, 1980). In a human fear conditioning paradigm that paired neutral faces (CS) with a shock (US), extinction procedures that involved partial reinforcement (i.e., occasional shocks during extinction) resulted in elevated physiological fear responding, but also protected against subjective and physiological return of fear through tests of spontaneous recovery and rapid reacquisition, respectively (Culver et al., 2018). Similarly, Thompson et al. (2018) showed that both paired and unpaired presentations of the unconditioned stimulus (a shock) eliminated spontaneous recovery of differential skin conductance responses in an extinction paradigm. Given that extant human fear conditioning research has shown promising effects of occasional reinforcement, this strategy may be an effective approach for maximizing the effectiveness of exposure therapy. However, experimental approaches that allow for repeated exposure to one’s feared outcome in vivo, especially in instances of specific phobias or trauma (Craske et al., 2014), have been difficult to implement.

One approach that may allow for examining the effects of occasional reinforcement in exposure therapy is to leverage imagery-based (e.g., video) exposure paradigms in translational research. Treatment-analogue experiments have used this approach to show that the use of multiple contexts during exposure might improve its generalizability (Vansteenwegen et al., 2007). In a recent treatment-analogue experiment examining occasional reinforcement, Jessup and Olatunji (2022) randomized snake-fearful individuals to a video exposure manipulation that was intermittently interrupted by a different video of a snake biting someone. Compared to a multiple-cue, video-based exposure alone condition (i.e., videos of different individuals handling different snakes), the multiple-cue exposure condition that also involved reminders of the feared outcome resulted in lower self-reported negative expectancy ratings and higher behavioral approach 1 week later (Jessup & Olatunji, 2022). Such findings have since been extended using multi-session exposure manipulation to show that a multiple-cue + fear outcome exposure condition resulted in higher behavioral approach compared to both multiple-cue and single-cue exposure alone conditions, which did not significantly differ from one another (Jessup et al., in press). Together these findings suggest that occasional reinforcement, especially in tandem with exposure to multiple cues, may facilitate greater learning and lower return of fear. However, it remains unclear whether these findings generalize to clinical samples and how well occasional reinforcement with multiple cues performs in comparison to other inhibitory retrieval strategies.

Affect labeling, or putting one’s feelings into words, is another inhibitory retrieval strategy through which exposure may facilitate greater learning and attenuate return of fear (Craske et al., 2014). Affect labeling has been shown to activate executive functioning areas of the cortex that reduce amygdala activity and attenuate anxious responding (Lieberman et al., 2007). In a within-subjects design, repeated exposure to fear-relevant pictures plus affective labels led to greater attenuation of physiological fear renewal in spider-fearful participants compared to exposure alone (Tabibnia et al., 2008). Results during a re-test with novel stimuli were strongest when affect labeling included negatively labeled words (e.g., war), compared to neutral labels (Tabibnia et al., 2008), suggesting that labeling negative words may be especially beneficial in promoting generalization of learning. Kircanski and colleagues (2012) expanded upon these findings with a between-subjects design examining affect labeling with fear-relevant, negative words among spider phobic individuals. The findings showed that compared to an exposure alone group, the affect labeling group demonstrated reduced skin conductance responses during a test of generalization. Niles and colleagues (2015) also examined whether affect labeling would enhance the effectiveness of exposure for public speaking anxiety and more explicitly examined this strategy in relation to feared outcomes. As part of speech exposure exercises, participants were instructed to complete the sentence “I feel _____, the audience will _____” with words and feared outcomes provided during the task. Compared to a control condition, participants in the affect labeling condition showed steeper decreases in heart rate and skin conductance responses from post-exposure to follow-up (Niles et al., 2015). Together, these findings suggest affect labeling may be an advantageous supplement to exposure therapy that weakens return of fear. However, the extent to which affect labeling changes negative expectancies, which are a core facet of inhibitory learning (Craske et al., 2008) and other relevant processes known to maintain anxiety disorders (e.g., behavioral avoidance, attention bias) remains unclear. It is also unclear how affect labeling when exposed to multiple exposure cues compares to other inhibitory retrieval strategies (e.g., multiple-cue + fear outcome exposure) and traditional (single-cue) exposure with regards to attenuating return of fear.

In addition to determining which inhibitory retrieval strategies most effectively attenuate fear renewal, it is also necessary to understand potential mechanisms of change. As Craske and colleagues (2008) have acknowledged, exposure process mechanisms proposed by the classic emotional processing theory (e.g., within-session fear reduction; Foa & Kozak, 1986) lack consistent empirical support. Instead, processes that promote increased tolerance of distress may help to facilitate greater learning (Craske et al., 2008). For instance, sustained levels of distress and greater fluctuations in distress responding during exposure have shown promising preliminary evidence in predicting better outcomes (Culver et al., 2012). In fact, Jessup and Olatunji (2022) identified variability in distress responding during exposure as a mechanism through which the effects of occasional reinforcement attenuate return of fear. Similarly, conditioning studies have shown occasional reinforcement sustains higher mean distress levels during exposure which then predicts better treatment outcomes (Culver et al., 2018). However, it remains unclear whether these process variables are specific mechanisms for occasional reinforcement or if they also help to explain the effects of other inhibitory retrieval strategies.

The present study sought to examine the comparative effects of confronting the feared outcome versus labeling one’s affective experience during a two-session imagery-based exposure manipulation for those with snake phobia. To address prior gaps in the literature, the present study incorporated subjective, behavioral, and attentional measures to examine exposure outcomes across multiple levels of analysis. The present study also obtained distress ratings throughout exposure to examine different process variables that may function as mechanisms of change. Consistent with the extant literature on the efficacy of inhibitory retrieval strategies (Jessup & Olatunji, 2022; Kircanski et al., 2012), it was hypothesized that multiple-cue exposure conditions that incorporated affect labeling or reminders of the feared outcome would experience significantly weakened return of fear at a 1-week follow-up compared to the single-cue exposure condition. Furthermore, it was predicted that snake phobic participants randomly assigned to affect labeling or occasional reinforcement would not significantly differ in return of fear at follow-up. Exploratory analyses were also conducted to examine potential group differences on exposure process variables (i.e., fear reduction, mean distress, and distress variability) that may be related to treatment outcome.

Methods

Participants

Adults were recruited from across the United States using ResearchMatch, a national health volunteer registry supported by the U.S. National Institutes of Health, and from the local community using a university-sponsored mass distribution email listserv. Interested individuals were contacted via email and completed a phone screen to assess for eligibility. During the phone screen, the Anxiety Disorders Interview Schedule for DSM-5 (ADIS; Brown & Barlow, 2014) was administered to assess for clinical diagnosis of snake phobia. Participants were considered eligible if they met DSM-5 criteria for snake phobia, had a working computer with high-speed internet, and did not endorse current psychotic disorder, pervasive developmental disorder, bipolar disorder, or blindness. All of these criteria were assessed during the phone screen. Although not considered an inclusionary criterion, 100% of participants endorsed a feared outcome of a snake bite on the ADIS. The sample (see CONSORT flow diagram) consisted of participants from 19 states, the majority of whom were from Tennessee (45.1%; n = 73), Illinois (12.3%; n = 20), or Texas (11.1%; n = 18). The majority of the sample was female (n = 95; 69.3%) and White (n = 72, 52.6%), and the mean age of participants was 36.63 (SD = 12.85; range = 19–66).

Target enrollment was determined with a-priori power analyses (G*Power 3.1; Faul et al., 2009). For repeated measures ANOVA tests, it was determined that a sample size of 90 was necessary to achieve 80% power to detect a small- to medium-sized effect at a = .05. Additional participants were recruited to account for possible attrition. A 5.13% attrition rate yielded a final sample of 137 snake-phobic participants.

Participants were randomly assigned to one of three experimental conditions: a multiple-cue + fear-outcome exposure group [FO], n = 46, a multiple-cue + affect labeling exposure group [AL], n = 38, or a single-cue exposure group [S], n = 53. Randomization was completed by an online random number generator with a range from 1 to 3 (where 1 = FO condition, 2 = AL condition, 3 = S condition). The three conditions were compared across a three-session, 11-day period: pre-exposure (session 1, day 1), post-exposure (session 2, day 4), and 1-week follow-up (session 3, day 11).

Measures

Fear of Snakes Questionnaire (FSQ; Milosevic & Radomsky, 2008; Olatunji et al., 2017)

The FSQ is an 18-item measure assessing snake phobia. Participants rated items on a Likert scale from 0 (totally disagree) to 7 (totally agree), and scores range from 0 to 126. The FSQ demonstrated good internal consistency in the present study (α = .89).

Beck Anxiety Inventory (BAI; Beck & Steer, 1990)

The BAI is a 21-item measure assessing common symptoms of anxiety. Items are rated on a Likert scale ranging from 0 (not at all) to 3 (severely). The BAI demonstrated excellent internal consistency in the present sample (α = .95).

Exposure Task

Prior to completing the assigned exposure task, participants were given the following brief form of psychoeducation: “while it’s often scary to confront the things that scare us, research suggests repeated exposure is an effective strategy in reducing fear because we have the opportunity to learn that our feared outcomes aren’t as likely or as severe as we think they might be.” The exposure task consisted of four 5-min videos of snakes with 1-minute inter-trial intervals. Four run orders were determined using a random number generator to counterbalance the sequence of the exposure videos within each condition. Participants in the S condition viewed the same five-minute snake video four times. Participants in the FO condition viewed four novel 5-min snake videos (i.e., 4 different snakes) that were each interrupted halfway through with a 10-s clip of a different snake biting someone. Participants in the AL condition watched the same 4, 5-min snake videos as the FO condition, but each video was paused three times for 10 s. During the paused video and consistent with procedures from Niles et al. (2015), participants were shown a still image of the snake and asked to complete the sentence: “I feel ________. I think the snake will ________.” Participants were given examples for each response on the video screen and were told they could pick a word off the list or generate a new word that best captured their emotion and feared outcome in the moment (see Supplemental Figure A). For emotions, examples provided included distressed, scared, afraid, anxious, disgusted, and nervous. For the feared outcome, examples provided included lunge, bite, slither, pounce, jump, and move. All three exposure conditions received exactly 20 min of exposure per session.

After each exposure task, participants were asked to rate how emotionally intense they found each video on a scale of 0 (not intense at all) to 100 (the most intensity I have ever felt). The videos were consistent with stimuli presented during exposure, such that the S group rated the intensity of the one snake video they saw, the AL group rated the intensity of the four snake videos they saw, and the FO group rated the intensity of the four snake videos plus the bite video. For each group, ratings were averaged to yield a single emotional intensity score for each session of exposure.

Primary Outcome Measures

Threat Expectancy Ratings

Prior to each exposure video, participants were presented with a still image of the video and asked, “to what extent do you expect this snake to bite someone?” Participants rated their threat expectancy using a visual analogue scale that ranged from 0 (definitely not going to happen) to 100 (definitely going to happen). The same threat expectancy question was asked immediately following each video and at follow-up to measure changes in threat expectancy. At each stage of exposure (session 1 pre-/post-exposure, session 2 pre-/post-exposure, and session 3 follow-up), the four threat expectancy ratings for the FO and AL groups were averaged to yield a single expectancy score. Since the S group watched the same video four times, only the first threat expectancy rating at each time point was used.

Behavioral Avoidance Task (BAT)

A computer-delivered BAT modeled after prior BATs (Jessup & Olatunji, 2022; Meng et al., 2004) was administered to measure visual avoidance of snakes. The BAT consisted of 25 images and videos presented in a hierarchical manner, such that each stimulus was more threatening than the previous. The BAT task was pilot tested in a smaller group of participants to determine the hierarchical order of the steps (stimuli sorted from lowest average anxiety reported on the pilot task to highest average anxiety reported). To successfully complete a step on the BAT, participants were asked to view the stimulus for 10 consecutive seconds¹. The experimenter recorded BAT scores as the highest step completed, which ranged from 0 to 25. Higher scores indicated greater behavioral approach.

Attention Bias Task (MouseView.js; Anwyl-Irvine et al., 2021)

Attention bias was assessed by comparing the time spent attending to fear-relevant (snake) images versus pleasant images using MouseView.js over time. Mouseview.js is a program designed to mimic the visual system’s peripheral blur and foveal clarity by enabling participants to move a high-fidelity aperture around an obscured field with their computer mouse. MouseView.js tracks and records the movement of a participant’s cursor and cursor placement on each image over time. These data provide a measure analogous to fixation duration and dwell time presented in eye tracking research (Anwyl-Irvine et al., 2021). Anwyl-Irvine and colleagues (2021) have shown that MouseView.js is a highly reliable and valid alternative to eye-tracking in preferential-looking experiments. In the present study, MouseView.js was customized with an aperture size that was 5% of the screen with a Gaussian blurred edge and a blurred overlay applied to the full screen. The MouseView.js task for the present study was created in Gorilla Experiment Builder (Anwyl-Irvine et al., 2020) and consisted of 20 images (5 snake, 5 pleasant, and 10 neutral) from the International Affective Picture System (Lang & Bradley, 2007). The affective images were randomly paired with a neutral image for a total of 10 unique trials. During each trial, two images were presented side-by-side for 10 s. The task consisted of 40 total trials, such that each presentation was repeated four times and yielded a total of 20 pleasant-neutral trials and 20 snake-neutral trials. The program tracks and records the duration of time participants spent attending to images to capture the proportion of time spent dwelling on each image type. This MouseView.js task has been previously validated in a clinical sample and showed that snake phobic participants exhibit significantly shorter dwell time on threat images than non-phobic individuals (Woronko et al., 2023).

Exposure Process Measures

Consistent with prior research (Culver et al., 2012; Jessup & Olatunji, 2022), Subjective Units of Distress Scores (SUDS) were collected every 60 s throughout each exposure video. At each time point, participants were asked to verbally report their current distress level on a scale of 0 (no distress) to 100 (most distressed I’ve ever felt). These ratings yielded a total of 24 SUDS during each exposure session, which were used to calculate exposure process measures. In line with previous research (Culver et al., 2012; Jessup & Olatunji, 2022), exposure process measures were computed as follows: (1) within-session fear reduction (WSFR): highest SUD level during the exposure videos minus the final SUD level during the fourth exposure video, (2) between-session fear reduction (BSFR): initial SUD level during the first exposure video in session 1 minus initial SUD level during the first exposure video in session 2, (3) mean distress: average SUD level across all exposure videos, and (4) distress variability: standard deviation of SUD ratings across all exposure videos. WSFR, mean distress, and distress variability were averaged across sessions 1 and 2 to yield a single score for each process measure.

Procedure

Vanderbilt University’s Institutional Review Board approved all aspects of the study. The study consisted of two sessions held on a secure Zoom call with a trained researcher. After providing informed consent, participants completed the FSQ and BAI via REDCap (Harris et al., 2009). The researcher then guided them through the BAT to assess baseline snake avoidance. Next, participants were sent a personalized link to complete the pre-exposure MouseView.js task on their computer. After completing the MouseView.js task, participants received the brief form of psychoeducation and an overview of the exposure task, including familiarization with the 0–100 SUD and negative expectancy scales. Participants then completed their randomly assigned exposure manipulation.

Participants returned 3 days later to complete the same exposure manipulation. Like session one, the exposure task consisted of four exposure videos of snakes that lasted 5 min each. Participants were asked to provide threat expectancy ratings before and after each video and reported their SUD ratings every 60 s. After the exposure task, participants rated the emotional intensity of each video on a scale of 0 (not intense at all) to 100 (the most intensity I’ve ever experienced). Participants then completed the same MouseView.js task and BAT to measure changes in visual attention and avoidance of snakes, respectively. One week later, participants returned to complete the same threat expectancy ratings, MouseView.js, and behavioral avoidance tasks. At the end of session three, participants were debriefed and compensated $40.

Results

Descriptive Statistics

Chi square and Univariate Analysis of Variance (ANOVA) tests were conducted to examine baseline differences between the experimental groups on demographic variables and baseline self-report questionnaires. There were no significant differences between the groups on any demographic variable (see Table 1).

Table 1.

Descriptive Characteristics of Participants within the Single-Context, Multiple-Context + Fear-Outcome, and Multiple-Context + Affect Labeling Exposure Conditions

Intervention condition	Single-context exposure group (n = 53)	Multiple-context + fear outcome exposure group (n = 46)	Multiple-context + affect labeling exposure group (n = 38)	F/χ²	p
Age	37.85 (12.51)	37.93 (13.37)	31.95 (13.34)	2.85	.06
% Female	69.81 (n = 37)	69.57 (n = 32)	68.42 (n = 26)	3.67	.72
% White	64.15 (n = 34)	54.35 (n = 25)	34.21 (n = 13)	18.11	.11
FSQ	102.24 (21.33)	106.72 (12.60)	98.39 (22.89)	1.94	.15
BAI	24.85 (18.49)	19.80 (16.52)	27.46 (18.71)	2.01	.14
Session 1 EI	37.79 (25.20)	56.03 (20.98)	54.09 (22.73)	9.81	<.001
Session 2 EI	30.25 (27.80)	48.05 (25.84)	44.95 (24.33)	6.64	.002

Note. FSQ = Fear of Snakes Questionnaire. BAI = Beck Anxiety Inventory. EI = Emotional Intensity ratings.

Differences in Emotional Intensity

One-way ANOVAs were conducted to compare group differences in mean emotional intensity ratings following each exposure session. There were significant group differences in emotional intensity ratings following the first, F(2,143) = 9.81, p < .001, and second exposure sessions, F(2,136) = 6.64, p = .002, such that the FO and AL groups reported the videos were significantly more intense than the S group (ps < .01). The FO and AL groups did not differ from one another in emotional intensity ratings following either exposure session (ps > .10).

Changes in Threat Expectancy

A 3 (Condition: FO, AL, S) × 5 (Time: session 1 pre-exposure, session 1 post-exposure, session 2 pre-exposure, session 2 post-exposure, follow-up) repeated measures ANOVA was conducted to examine changes in threat expectancy. As depicted in Figure 1, there was a significant main effect of time, F = 50.70, p < .001, ηp² = .28 suggesting an overall decrease in negative threat expectancies, and a significant time by condition interaction, F = 3.40, p = .004, ηp² = .05. LSD post-hoc tests revealed that the S group reported marginally significantly lower negative expectancy ratings at session 2 pre-exposure (p = .06), session 2 post-exposure (p = .08) and session 3 follow-up (p = .06) than the AL group. There were no other significant differences among groups (ps > .10). Additional post-hoc analyses were conducted to examine group differences in change across two consecutive time points. Results revealed a greater reduction in threat expectancy ratings from session 1 pre- to post-exposure for the S group compared to the AL group (p = .004). Results also revealed a larger increase in threat expectancy ratings from session 2 post-exposure to session 3 for the FO group compared to the S group (p = .02) but not the AL group (p > .10). There were no other significant group differences, including comparisons of change from session 1 post-exposure to session 2 pre-exposure or from session 2 pre-exposure to session 2 post-exposure (ps > .10).

Figure 1.

Self-Reported Threat Expectancy Ratings for the Three Exposure Conditions During Session 1 Pre-Exposure, Session 1 Post-Exposure, Session 2 Pre-exposure, Session 2 Post-Exposure, and the 1-Week Follow-Up Behavioral Approach Tasks. Error Bars: ±1 Standard Error

Changes in Behavioral Avoidance

A 3 (Condition: FO, AL, S) × 3 (Time: pre-exposure, post-exposure, follow-up) repeated measures ANOVA was conducted to examine changes in the number of BAT steps completed (See Figure 2). Results revealed a significant time by condition interaction, F = 3.24, p = .01, ηp² = .05. Main effect of time was examined separately for the three conditions and was not significant for the FO or S conditions (ps > .10), but was significant for the AL condition, F = 7.81, p = .002, ηp² = .18, suggesting increased behavioral avoidance over time. LSD post-hoc tests of multiple comparisons showed that individuals in the FO and S conditions completed significantly more BAT steps at session 2 post-exposure than individuals in the AL condition (ps = .01 and .02, respectively). Further, individuals in the FO condition completed significantly more BAT steps at follow-up compared to the AL condition (p = .036), but the S and AL conditions did not differ from each other (p = .16).

Figure 2.

Number of Behavioral Approach Test (BAT) Steps Completed for the Three Exposure Conditions at Pre-Exposure, Post-Exposure, and the One-Week Follow-Up. Error Bars: ±1 Standard Error

Attentional Bias Outcomes

Differences in Pleasant and Snake Dwell Time Proportions

A paired-samples t-test showed that participants spent significantly less time viewing snake images (M = .42, SD = .18) than pleasant images (M = .55, SD = .07) during pre-exposure, t = −6.78, p < .001, d = −.62. These differences in dwell time proportion are consistent with an attentional avoidance of threatening stimuli.

Changes in Visual Attention

A 3 (Condition: FO, AL, S) × 3 (Time: pre-exposure, post-exposure, follow-up) repeated measures ANOVA was conducted to examine changes in the proportion of time spent attending to snake images. Results revealed no significant main effect of time (p = .45) and the predicted time by condition interaction was not significant (p = .15). Figure 3 depicts visual approach and avoidance patterns to snake versus pleasant images at pre-exposure, post-exposure, and follow-up.

Figure 3.

Attentional Patterns During the MouseView.js Task at Pre-Exposure (Top Row), Post-Exposure (Middle Row), and Follow-Up (Bottom Row) for the Full Sample. Dwell Time (Y Axis) is Shown as Percentage Difference for Pleasant-Neutral and Threat-Neutral Trials. Positive Values Indicate a Higher Dwell Time on the Affective Image (i.e., Approach) and Negative Values Indicate a Higher Dwell Time on the Neutral Image (i.e., Avoidance). The Solid Line Represents the Average Dwell Time for Each Stimulus Presentation and the Shaded Region Represents a 95% Confidence Interval

Group Differences in Exposure Process Measures

Figure 4 depicts the subjective distress ratings for each group during exposure that were used to examine exposure process variables. A series of one-way ANOVAs revealed significant group differences in within-session habituation, F = 4.75, p = .01, distress variability, F = 7.84, p < .001, and mean distress, F = 7.45, p < .001, and differences in between-session fear reduction approached but did not reach significance, F = 2.77, p = .07. Post-hoc tests revealed that the FO group reported significantly greater within-session fear reduction than the AL group (p = .006) and the S group (p = .013), and there was not a significant difference between the AL and S groups (p = .62). The S group reported significantly greater between-session fear reduction than the AL group (p = .02), but no other group differences were significant (ps > .10). The FO group reported significantly greater distress variability than the AL group (p = .001) and the S group (p < .001), but the AL and S groups did not significantly differ from one another (p = .90). Finally, the FO and AL groups reported significantly greater mean distress than the S group (ps = .002 and .001, respectively), but did not significantly differ from one another (p = .75).

Figure 4.

Subjective units of distress Ratings for Each Group During Session One Exposure (Top) and Session Two Exposure (Bottom). FO = Fear-Outcome Exposure Group; AL = Affect Labeling Exposure Group; S = Single-Context Exposure Group

Associations Between Exposure Process Variables and Outcome Measures

Within-session fear reduction across the three conditions was not significantly correlated with any outcome measures (ps > .10). Between-session fear reduction was significantly correlated with only follow-up expectancy ratings (r = −.244, p = .004). Distress variability was significantly associated with only visual attention (i.e., dwell proportion) at follow-up (r = −.22, p = .01). Lastly, mean distress was significantly correlated with follow-up expectancy ratings (r = .69, p < .001) and follow-up BAT steps completed (r = −.24, p = .004).

Exploring the Mediating Role of Exposure Process Mechanisms

Prior research suggests that variability in distress responding during exposure is a central mechanism through which occasional reinforcement attenuates the return of fear (Jessup & Olatunji, 2022) and may be a better predictor of outcome than within-session fear reduction (Culver et al., 2012). Given significant group differences between the FO and AL groups on distress variability, within-session fear reduction, and avoidance at follow-up, an exploratory mediational model was conducted to determine the extent to which distress variability and within-session fear reduction mediate the relationship between exposure condition and number of BAT steps completed at follow-up. A 95% bootstrap confidence interval revealed the true indirect effect for distress variability was estimated to lie between .59 and 7.72 (Effect = 3.63, SE = 1.84), and that the true indirect effect for within-session fear reduction was estimated to lie between −6.98 and −.76 (Effect = −3.14, SE = 1.60). Because these 95% confidence intervals do not contain zero, it can be concluded that distress variability and within-session fear reduction significantly mediate the relationship between exposure condition (FO vs. AL) and BAT steps completed at follow-up (see Figure 5).

Discussion

The primary aim of the present study was to examine the comparative efficacy of two inhibitory retrieval strategies, occasional reinforcement and affect labeling (AL), in augmenting learning during an exposure manipulation. Results of this treatment-analogue experiment indicated that participants in the three exposure conditions experienced declines in threat expectancy ratings over the course of exposure, suggesting that exposure was effective in reducing the perceived likelihood of a snake biting someone for participants with snake phobia.

Participants in the occasional reinforcement (FO) group and the S group did not significantly differ in threat expectancy ratings over the course of the exposure intervention or 1 week later, suggesting that an exposure intervention that explicitly incorporates confronting the feared outcome does not necessarily increase overall expectancies of that outcome compared to an exposure alone intervention. However, participants in the FO condition did demonstrate a greater increase in average threat expectancy ratings from session 2 post-exposure to follow-up compared to the S group but not the AL group. It is worth noting that the FO group provided expectancy ratings for four different snakes whereas the S group viewed and provided threat expectancy ratings for just one snake. Thus, it is unclear if the expectancy ratings for the two groups are capturing a similar process. Additional research including follow up tests to novel stimuli for both the S and FO conditions is needed before definitive inferences about return of fear can be made.

Figure 5.

Mediational Model of the Association Between Exposure Condition (Fear Outcome vs. Affect Labeling Groups), Distress Variability (i.e., Standard Deviation of Subjective Units of Distress), Within-Session Fear Reduction (i.e., Peak Distress Minus Distress at End of Video 4), and Number of Behavioral Approach Task Steps Completed at the 1-Week Follow-Up. ** = p ≤ .01,* = p < .05

The present study did observe a general pattern where snake phobic participants in the AL exposure condition reported marginally higher threat expectancy ratings than the S exposure group after the exposure intervention and at the 1-week follow-up, though this difference did not reach statistical significance. When considering differences between inhibitory retrieval strategies, this finding raises the possibility that in contrast to an exposure intervention with occasional reinforcement, AL may increase overall expectancies of feared outcomes compared to an exposure alone intervention. To the extent to which AL during exposure therapy may result in more desirable outcomes, the present findings suggest that it is unlikely that those effects are achieved by decreasing threat expectancy outcomes. This is perhaps not surprising given that research has shown that individuals do not expect AL to be helpful for downregulating negative emotions that are often evoked by heightened threat expectancies (Lieberman et al., 2011). In fact, some predict AL will make their distress worse, not better (Lieberman et al., 2011).

Examination of behavioral outcomes in the present study showed that individuals in the FO condition completed significantly more BAT steps at session 2 post-exposure and the 1-week follow-up than individuals in the AL condition, but did not significantly differ from the S condition. The finding that snake phobic participants in the occasional reinforcement condition did not outperform those in the exposure alone condition is contrary to predictions and also contradicts the recent findings by Jessup and Olatunji (2022), which suggested that introducing reminders of the feared outcome into exposure leads to significant increases in behavioral approach. However, given that snake phobic participants in the FO condition completed 21 out of 25 steps at pre-exposure and did not demonstrate significant improvement on the BAT over time, the present results may be partially constrained by baseline ceiling effects on the BAT. Thus, future studies should investigate whether in-vivo BATs can potentially reduce pre-treatment ceiling effects that will allow for more sensitivity to detect predicted group differences. Nonetheless, it is important to note that although reminders of the feared outcome did not significantly augment exposure, they were also not detrimental to learning. That is, the presentation of a snake bite during exposure did not exacerbate avoidance behaviors or increase negative expectancies. Fellow clinicians (Lokers, 2020) and media outlets (Slater, 2003) alike have raised ethical concerns that exposure therapy may exacerbate symptoms, despite an abundance of evidence contradicting these claims (Olatunji et al., 2009).

The present study did show that those in the AL condition completed significantly fewer BAT steps than the FO and S conditions at session 2 post-exposure, and significantly fewer steps than the FO, but not the S condition at follow-up. These results suggest that imagery-based exposure + AL may be less effective than occasional reinforcement with regards to attenuating the return of fear. Of course, this group difference may be an artifact of the study design rather than a difference between the two interventions per se. For example, one explanation for the group difference is that individuals labeled their affect every 2 min in the present study, which could have detracted from a participant’s ability to process disconfirmatory information during the exposure manipulation. Although the exposure videos were paused when the AL prompt appeared to preemptively address this concern, it may be the case that AL individuals were distracted during exposure (i.e., between pauses) by considering what to say next rather than focusing on the exposure stimuli itself. Research has shown that distraction is a pernicious behavior that can interfere with exposure interventions, particularly when it demands additional cognitive or attentional resources and prevents the processing of corrective information (Craske et al., 2022; Parrish et al., 2008). Consistent with this view, AL has been termed an “incidental emotion regulation strategy” and is believed to share mechanisms with established emotion regulation strategies including distraction (Lieberman et al., 2011).

The finding that snake phobic participants in the AL condition were more avoidant than those in the FO and S conditions at either post-exposure or follow-up may be a product of the intensity of the treatment-analogue design. Recent research has also shown that although AL reduces distress in high-intensity aversive conditions, it increases distress in low-intensity conditions (Levy-Gigi & Shamay-Tsoory, 2022). The treatment-analogue approach in the present study that capitalizes on video exposure may represent a low intensity condition compared to actual exposure therapy in vivo that will be of higher intensity. That being said, it is worth noting that Kircanski and colleagues (2012) found that an AL condition did not significantly differ from exposure alone and distraction conditions in BAT steps completed at follow-up. Similarly, Plaisted and colleagues (2022) more recently found that AL failed to enhance exposure as assessed with self-rated anxiety, heart rate, and observer ratings of expressed anxiety for adolescents with public speaking anxiety from pre-test to 1-week follow-up. The present findings converge with this prior research that suggests that AL may not increment exposure alone and also suggests that AL may be inferior to occasional reinforcement in attenuating the return of fear.

An important contribution of the present study is the examination of exposure outcomes across multiple levels, including attentional bias. Although attentional biases are characteristic of many anxiety disorders including specific phobias (Bar-Haim et al., 2007; Cisler & Koster, 2010), the extent to which exposure interventions reduce such biases remains unclear. Prior to the exposure intervention, participants spent significantly less time viewing snake images than pleasant images, which is consistent with prior research showing that individuals with specific phobias demonstrate sustained attentional bias away from threat (Armstrong et al., 2013). Despite robust changes in threat expectancies in the present study, attentional biases did not significantly change over the course of the exposure paradigm. Furthermore, there were no significant differences in attentional biases between the three exposure conditions. The absence of an effect of exposure on attentional bias for threat is consistent with existing findings. For example, Kampmann and colleagues (2018) found that pre- to post-treatment changes in attention bias for those with social anxiety disorder did not significantly differ between an exposure condition and waitlist control condition. This lack of change in attentional bias for threat may partially explain why fear re-emerges with the passage of time, a change in context, or following an aversive event. Although the therapeutic benefit of attention bias modification is rather small for anxiety (Mogoaşe et al., 2014), a novel protocol that leverages principles of operant conditioning has shown promise in reducing attention to threat and increasing attention control (Azriel et al., 2024). However, no studies to date have examined how such protocols augment exposure-based interventions. Future research that targets attentional bias for threat during exposure therapy from an inhibitory learning framework may be especially promising for attenuating the return of fear.

Another important aim of the present study was to examine potential group differences in exposure process variables that may explain differences in outcomes. The FO condition experienced significantly greater within-session fear reduction and distress variability than the AL and S conditions. The FO condition also experienced significantly greater mean distress than the S, but not AL, condition. Given that within-session fear reduction is believed to reflect within-session performance (Craske et al., 2008) and distress variability is thought to enhance learning (Craske et al., 2014), these group differences in process measures may help to explain why the AL condition underperformed in the present study. This is despite the fact that the FO and AL groups perceived their manipulation as more emotionally intense than the S group but did not significantly differ from each other in emotional intensity ratings. In fact, when comparing the FO and AL groups, both distress variability and within-session fear reduction partially mediated the relationship between exposure intervention condition and number of BAT steps completed at follow-up. This finding suggests that distress variability and within-session fear reduction may be mechanisms that differentiate how the effects of occasional reinforcement differ from that of AL in modifying behavioral outcomes. However, it is important to note that computation of within-session fear reduction and distress variability both rely on the peak in distress. Accordingly, additional research is necessary to determine the distinctiveness of the process variables and to further clarify the extent to which they reliably predict and/or explain when and how the various inhibitory retrieval strategies produce differences in exposure outcomes.

The present findings provide preliminary data on the comparative efficacy of two inhibitory retrieval strategies. The study has several important strengths, including the use of a clinical sample and multi-session nature, which allowed us to assess both between and within-session fear reduction. In addition, the virtual nature of the study yields more generalizable results, given that participants were recruited from across the United States, including rural communities. Despite the geographic diversity, the study participants were largely White and female, limiting our ability to generalize study findings to other demographic groups. The present study is not without additional limitations, which should be carefully considered when interpreting results. For instance, it was previously noted that the virtual nature of the study required that the BAT was administered on the computer, which may have contributed to potential ceiling effects. Future research should employ an exposure intervention in a laboratory setting where in-vivo behavioral approach can be measured. Relatedly, the BAT hierarchy was designed based on pilot data and thus, ascending threat potential could not be established for each individual participant. Additionally, although the present study did assess outcomes across multiple levels of analysis, collecting data in a laboratory setting would also allow for the measurement of psychophysiological outcomes (e.g., skin conductance and heart rate), which may yield different results than the subjective, behavioral, and attentional outcomes collected in the present study. Another important study limitation is the use of a single-cue exposure alone condition as a control group. Although this condition allowed inferences to be made regarding how the multiple-cue + fear outcome and affect labeling groups compare to a more traditional exposure approach, the current study design makes it difficult to delineate whether the effects were due to a single or multiple-cue exposure, or the group manipulations. While prior research has shown that multiple-cue + fear outcome exposure outperforms both multiple-cue and single-cue exposure alone conditions (and the multiple and single cue groups do not differ from one another; Jessup et al., in press), future research may benefit from comparisons of multiple-cue + affect labeling and multiple-cue exposure alone conditions to further delineate effects. Further, the present study only assessed spontaneous recovery (i.e., fear that returns with the passage of time) as the measure of return of fear. Given that fear can also return with a change in context (context renewal) or re-exposure to the US (reinstatement), additional tests with exposure therapy in vivo in a more diverse sample may provide further insight into the differential efficacy of inhibitory retrieval strategies in augmenting exposure.

Supplemental Material

Supplemental Material - Seeing the Worst or Saying the Words? A Multilevel Comparison of Occasional Reinforcement and Affect Labeling as Strategies to Augment an Imagery-Based Exposure Intervention

Supplemental Material for Seeing the Worst or Saying the Words? A Multilevel Comparison of Occasional Reinforcement and Affect Labeling as Strategies to Augment an Imagery-Based Exposure Intervention by Sarah C. Jessup, Thomas Armstrong, Kavi S. Jakes, Edwin S. Dalmaijer, and Bunmi O. Olatunji in Journal of Experimental Psychopathology

Footnotes

ORCID iDs

Sarah C. Jessup

Edwin S. Dalmaijer

Ethical Statement

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: S. Jessup was supported by a predoctoral fellowship training grant awarded by the National Institute of Mental Health (T32-MH018921-29) and received financial support from the Vanderbilt’s Institute for Clinical and Translational Research. The funding agency had no role in data collection, data analysis, manuscript writing, or in deciding to submit the article for publication.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data will be made available upon request.*

Supplemental Material

Supplemental material for this article is available online.

Note

Author Biographies

Sarah C. Jessup is a doctoral candidate at Vanderbilt University and a current pre-doctoral intern at McLean Hospital/Harvard Medical School.

Thomas Armstrong is an Associate Professor of Psychology at Whitman College.

Kavi S. Jakes is a research assistant at the University of Louisville.

Edwin S. Dalmaijer is a cognitive neuroscientist at the University of Bristol.

Bunmi O. Olatunji is a Gertrude Conaway Vanderbilt Professor of Psychology, Associate Professor of Psychiatry, and the Director of Clinical Training at Vanderbilt University.

References

Anwyl-Irvine

A. L.

Armstrong

Dalmaijer

E. S.

(2021). MouseView.js: Reliable and valid attention tracking in web-based experiments using a cursor-directed aperture. Behavior Research Methods, 54(4), 1663–1687. https://doi.org/10.3758/s13428-021-01703-5

Anwyl-Irvine

A. L.

Massonnié

Flitton

Kirkham

Evershed

J. K.

(2020). Gorilla in our midst: An online behavioral experiment builder. Behavior Research Methods, 52(1), 388–407. https://doi.org/10.3758/s13428-019-01237-x

Armstrong

Hemminger

Olatunji

B. O.

(2013). Attentional bias in injection phobia: Overt components, time course, and relation to behavior. Behaviour Research and Therapy, 51(6), 266–273. https://doi.org/10.1016/j.brat.2013.02.008

Azriel

Arad

Pine

D. S.

Lazarov

Bar-Haim

(2024). Attention bias vs. attention control modification for social anxiety disorder: A randomized controlled trial. Journal of Anxiety Disorders, 101, 102800. https://doi.org/10.1016/j.janxdis.2023.102800

Bandelow

Michaelis

(2015). Epidemiology of anxiety disorders in the 21st century. Dialogues in Clinical Neuroscience, 17(3), 327–335. https://doi.org/10.31887/DCNS.2015.17.3/bbandelow

Bar-Haim

Lamy

Pergamin

Bakermans-Kranenburg

M. J.

van IJzendoorn

M. H.

(2007). Threat-related attentional bias in anxious and nonanxious individuals: A meta-analytic study. Psychological Bulletin, 133(1), 1–24. https://doi.org/10.1037/0033-2909.133.1.1

Beck

A. T.

Steer

R. A.

(1990). Manual for the Beck anxiety inventory. The Psychological Corporation.

Brown

T. A.

Barlow

D. H.

(2014). Anxiety and Related Disorders Interview Schedule for DSM-5 (ADIS-5) - adult and lifetime version: Clinician manual. Oxford University Press.

Cisler

J. M.

Koster

E. H. W.

(2010). Mechanisms of attentional biases towards threat in the anxiety disorders: An integrative Review. Clinical Psychology Review, 30(2), 203–216. https://doi.org/10.1016/j.cpr.2009.11.003

10.

Craske

M. G.

Kircanski

Zelikowsky

Mystkowski

Chowdhury

Baker

(2008). Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy, 46(1), 5–27. https://doi.org/10.1016/j.brat.2007.10.003

11.

Craske

M. G.

Mystkowski

J. L.

(2006). Exposure therapy and extinction: Clinical studies. In Fear and learning: From basic processes to clinical implications (pp. 217–233). American Psychological Association. https://doi.org/10.1037/11474-011

12.

Craske

M. G.

Treanor

Conway

Zbozinek

Vervliet

(2014). Maximizing exposure therapy: An inhibitory learning approach. Behaviour Research and Therapy, 58, 10–23. https://doi.org/10.1016/j.brat.2014.04.006

13.

Craske

M. G.

Treanor

Zbozinek

T. D.

Vervliet

(2022). Optimizing exposure therapy with an inhibitory retrieval approach and the OptEx Nexus. Behaviour Research and Therapy, 152, 104069. https://doi.org/10.1016/j.brat.2022.104069

14.

Culver

N. C.

Stevens

Fanselow

M. S.

Craske

M. G.

(2018). Building physiological toughness: Some aversive events during extinction may attenuate return of fear. Journal of Behavior Therapy and Experimental Psychiatry, 58, 18–28. https://doi.org/10.1016/j.jbtep.2017.07.003

15.

Culver

N. C.

Stoyanova

Craske

M. G.

(2012). Emotional variability and sustained arousal during exposure. Journal of Behavior Therapy and Experimental Psychiatry, 43(2), 787–793. https://doi.org/10.1016/j.jbtep.2011.10.009

16.

Faul

Erdfelder

Buchner

Lang

A. G.

(2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160. https://doi.org/10.3758/BRM.41.4.1149

17.

Foa

E. B.

Kozak

M. J.

(1986). Emotional processing of fear: Exposure to corrective information. Psychological Bulletin, 99(1), 20–35. https://doi.org/10.1037/0033-2909.99.1.20

18.

Foa

E. B.

McLean

C. P.

(2016). The efficacy of exposure therapy for anxiety-related disorders and its underlying mechanisms: The case of OCD and PTSD. Annual Review of Clinical Psychology, 12(1), 1–28. https://doi.org/10.1146/annurev-clinpsy-021815-093533

19.

Harris

P. A.

Taylor

Thielke

Payne

Gonzalez

Conde

J. G.

(2009). Research Electronic Data Capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics, 42(2), 377–381. https://doi.org/10.1016/j.jbi.2008.08.010

20.

Jessup

S. C.

Armstrong

Rast

C. E.

Woronko

S. E.

Jackson

Anwyl-Irvine

A. L.

Dalmaijer

E. S.

Olatunji

B. O.

(in press). Benefits of the “worst-case scenario”: A multi-level examination of the effects of confronting the feared outcome during imagery-based exposure. Cognitive Behaviour Therapy. https://doi.org/10.1080/16506073.2025.2456479

21.

Jessup

S. C.

Olatunji

B. O.

(2022). Effects of confronting the feared outcome during exposure therapy on the return of fear: An analogue study. Journal of Behavior Therapy and Experimental Psychiatry, 76, 101747. https://doi.org/10.1016/j.jbtep.2022.101747

22.

Kampmann

I. L.

Emmelkamp

P. M. G.

Morina

(2018). Does exposure therapy lead to changes in attention Bias and approach-avoidance bias in patients with social anxiety disorder? Cognitive Therapy and Research, 42(6), 856–866. https://doi.org/10.1007/s10608-018-9934-5

23.

Kircanski

Lieberman

M. D.

Craske

M. G.

(2012). Feelings into words: Contributions of language to exposure therapy. Psychological Science, 23(10), 1086–1091. https://doi.org/10.1177/0956797612443830

24.

Lang

Bradley

M. M.

(2007). The International Affective Picture System (IAPS) in the study of emotion and attention. Handbook of Emotion Elicitation and Assessment, 29, 70–73.

25.

Levy-Gigi

Shamay-Tsoory

(2022). Affect labeling: The role of timing and intensity. Plos One, 17(12), Article e0279303. https://doi.org/10.1371/journal.pone.0279303

26.

Lieberman

M. D.

Eisenberger

N. I.

Crockett

M. J.

Tom

S. M.

Pfeifer

J. H.

Way

B. M.

(2007). Putting feelings into words. Psychological Science, 18(5), 421–428. https://doi.org/10.1111/j.1467-9280.2007.01916.x

27.

Lieberman

M. D.

Inagaki

T. K.

Tabibnia

Crockett

M. J.

(2011). Subjective responses to emotional stimuli during labeling, reappraisal, and distraction. Emotion, 11(3), 468–480. https://doi.org/10.1037/a0023503

28.

Lokers

L. M.

(2020). Fear over facts: When fellow clinicians become a barrier to dissemination of exposure and response prevention for OCD—a clinical response to licensing board investigation of ERP. Journal of Cognitive Psychotherapy.

29.

Lonsdorf

T. B.

Menz

M. M.

Andreatta

Fullana

M. A.

Golkar

Haaker

Heitland

Hermann

Kuhn

Kruse

Drexler

S. M.

Meulders

Nees

Pittig

Richter

Römer

Shiban

Schmitz

Straube

Merz

C. J.

(2017). Don’t fear ‘fear conditioning’: Methodological considerations for the design and analysis of studies on human fear acquistion, extinction, and return of fear. Neuroscience and Biobehavioral Reviews, 77, 247–285. https://doi.org/10.1016/j.neubiorev.2017.02.026

30.

Meng

C. T. T.

Kirkby

K. C.

Martin

Gilroy

L. J.

Daniels

B. A.

(2004). Computer-delivered behavioural avoidance tests for spider phobia. Behaviour Change, 21(3), 173–185. https://doi.org/10.1375/bech.21.3.173.55994

31.

Milosevic

Radomsky

A. S.

(2008). Safety behaviour does not necessarily interfere with exposure therapy. Behaviour Research and Therapy, 46(10), 1111–1118. https://doi.org/10.1016/j.brat.2008.05.011

32.

Mogoaşe

David

Koster

E. H.

(2014). Clinical efficacy of attentional bias modification procedures: An updated meta‐analysis. Journal of Clinical Psychology, 70(12), 1133–1157. https://doi.org/10.1002/jclp.22081

33.

Niles

A. N.

Craske

M. G.

Lieberman

M. D.

Hur

(2015). Affect labeling enhances exposure effectiveness for public speaking anxiety. Behaviour Research and Therapy, 68, 27–36. https://doi.org/10.1016/j.brat.2015.03.004

34.

Olatunji

B. O.

Cisler

J. M.

Deacon

B. J.

(2010). Efficacy of cognitive behavioral therapy for anxiety disorders: A Review of meta-analytic findings. The Psychiatric Clinics of North America, 33(3), 557–577. https://doi.org/10.1016/j.psc.2010.04.002

35.

Olatunji

B. O.

Deacon

B. J.

Abramowitz

J. S.

(2009). The cruelest cure? Ethical issues in the implementation of exposure-based treatments. Cognitive and Behavioral Practice, 16(2), 172–180. https://doi.org/10.1016/j.cbpra.2008.07.003

36.

Olatunji

B. O.

Tomarken

Wentworth

Fritzsche

(2017). Effects of exposure in single and multiple contexts on fear renewal: The moderating role of threat-specific and nonspecific emotionality. Journal of Behavior Therapy and Experimental Psychiatry, 54, 270–277. https://doi.org/10.1016/j.jbtep.2016.09.004

37.

Parrish

C. L.

Radomsky

A. S.

Dugas

M. J.

(2008). Anxiety-control strategies: Is there room for neutralization in successful exposure treatment? Clinical Psychology Review, 28(8), 1400–1412. https://doi.org/10.1016/j.cpr.2008.07.007

38.

Pearce

J. M.

Hall

(1980). A model for Pavlovian learning: Variations in the effectiveness of conditioned but not of unconditioned stimuli.

39.

Plaisted

Waite

Creswell

(2022). Optimising exposure for adolescents with public speaking anxiety: Affect labelling or positive coping statements? Behaviour Research and Therapy, 148, 103997. https://doi.org/10.1016/j.brat.2021.103997

40.

Rescorla

R. A.

Wagner

(1972). A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and non-reinforcement. Classical Conditioning, Current Research and Theory, 2, 64–99.

41.

Slater

(2003). The cruelest cure. New York Times, November 3.

42.

Tabibnia

Lieberman

M. D.

Craske

M. G.

(2008). The lasting effect of words on feelings: Words may facilitate exposure effects to threatening images. Emotion, 8(3), 307–317. https://doi.org/10.1037/1528-3542.8.3.307

43.

Thompson

McEvoy

P. M.

Lipp

O. V.

(2018). Enhancing extinction learning: Occasional presentations of the unconditioned stimulus during extinction eliminate spontaneous recovery, but not necessarily reacquisition of fear. Behaviour Research and Therapy, 108, 29–39. https://doi.org/10.1016/j.brat.2018.07.001

44.

Vansteenwegen

Vervliet

Iberico

Baeyens

Van den Bergh

Hermans

(2007). The repeated confrontation with videotapes of spiders in multiple contexts attenuates renewal of fear in spider-anxious students. Behaviour Research and Therapy, 45(6), 1169–1179. https://doi.org/10.1016/j.brat.2006.08.023

45.

Wolitzky-Taylor

K. B.

Horowitz

J. D.

Powers

M. B.

Telch

M. J.

(2008). Psychological approaches in the treatment of specific phobias: A meta-analysis. Clinical Psychology Review, 28(6), 1021–1037. https://doi.org/10.1016/j.cpr.2008.02.007

46.

Woronko

S. E.

Jessup

S. C.

Armstrong

Anwyl-Irvine

A. L.

Dalmaijer

E. S.

Olatunji

B. O.

(2023). A novel probe of attentional bias for threat in specific phobia: Application of the “MouseView. js” approach. Journal of Anxiety Disorders, 96, 102700. https://doi.org/10.1016/j.janxdis.2023.102700

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.72 MB