Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Mar 1.
Published in final edited form as: Behav Res Ther. 2022 Jan 7;150:104031. doi: 10.1016/j.brat.2022.104031

Modulation of threat extinction by working memory load: an event-related potential study

Yuhan Cheng 1, T Bryan Jackson 1, Annmarie MacNamara 1
PMCID: PMC8844280  NIHMSID: NIHMS1771472  PMID: 35032699

Abstract

Distraction is typically discouraged during exposure therapy for anxiety, because it is thought to interfere with extinction learning by diverting attention away from anxiety-provoking stimuli. Working memory load is one form of distraction that might interfere with extinction learning. Alternatively, working memory load might reduce threat responding and benefit extinction learning by engaging prefrontal brain regions that have a reciprocal relationship with brain circuits involved in threat detection and processing. Prior work examining the effect of working memory load on threat extinction has been limited and has found mixed results. Here, we used the late positive potential (LPP), an event-related potential that is larger for threatening compared to non-threatening stimuli to assess the effect of working memory load on threat extinction. After acquisition, 38 participants performed three blocks of an extinction task interspersed with low and high working memory load trials. Results showed that overall, the LPP was reduced under high compared to low working memory load, and that working memory load slowed extinction learning. Results provide empirical evidence in support of limiting distraction during exposure therapy in order to optimize extinction learning efficiency.

Keywords: working memory load, threat extinction, distraction, exposure therapy, late positive potential (LPP), event-related potential (ERP)

Exposure therapy is an effective means of treating specific anxieties and fear disorders that involves repeated contact with a fearful/threatening stimulus (Craske et al., 2008, 2014). A key mechanism thought to underlie exposure therapy is extinction learning, in which newly acquired, non-threatening memories come to inhibit previously learned threat-stimulus associations (Bouton, 2004; Myers & Davis, 2007). Increased understanding of the factors that affect extinction learning may help improve exposure therapy success. One factor thought to be critical to exposure therapy success is that the patient allocates sufficient attention to the fear-eliciting stimulus. For this reason, distraction and avoidance are typically discouraged in exposure therapy (Craske et al., 2008, 2014). Nonetheless, the effect of distraction on threat extinction is poorly understood, and some work has even suggested that distraction might benefit extinction learning, at least for certain phenotypes (e.g., 5-HTT knockout rats; Nonkes et al., 2012).

Attention is a limited resource that balances top-down control with stimulus-driven capture, which is allocated to salient stimuli even when irrelevant to task goals (Hickey et al., 2006; Van der Burg et al., 2007). This is especially true for emotional stimuli (Smith et al., 2004), such as threat-conditioned stimuli (Pittig et al., 2014). The late positive potential (LPP), a centro-parietal event-related potential (ERP) that begins approximately 400 ms following stimulus onset and persists until the end of stimulus presentation, is sensitive to the emotional salience of stimuli (Cuthbert et al., 2000; Hajcak & Olvet, 2008; Pastor et al., 2007) and is larger for threat-conditioned stimuli (CS+) compared to stimuli that have not been paired with threat (CS-; MacNamara & Barley, 2018). While the LPP is sensitive to emotional salience, it can also be modulated by top-down effects. For example, it is smaller when participants are asked to reduce their emotional responses to stimuli (Hajcak et al., 2010; Parvaz et al., 2012). Moreover, the stimulus-locked LPP is also sensitive to less willful modulations of attention. For example, when participants must devote attention to a more demanding cognitive task (i.e., holding six versus two letters in memory), the LPP to task-irrelevant negative and neutral pictures is smaller, indicating reduced salience of these stimuli (MacNamara et al., 2011, 2019; MacNamara & Proudfit, 2014). As such, the LPP provides a sensitive measure of the intrinsic significance of stimuli, but is also expected to be reduced when processing resources are consumed by other, ongoing tasks, such as a working memory task. However, beyond this expected main effect of working memory load, threat-potentiation of the LPP in an extinction task might be increased under working memory load, which would indicate that threat-conditioned stimuli retain their threat value more when participants’ attention is diverted to a working memory task.

Though there has been limited prior work examining the effect of working memory load on extinction learning, some studies have performed more overt manipulations of attention to conditioned stimuli during extinction. For example, O’Malley and Waters (2018) instructed participants to look at or away from conditioned stimuli during extinction. Results showed that participants who were instructed to avoid looking at the CS during extinction had larger skin conductance responses (SCRs) to the CS+ and CS- during extinction, as well as larger SCRs to the CS+ during a subsequent test of extinction retention. Similarly, Klein and colleagues (2021) showed that, when the eyes of conditioned faces turned blue as participants looked at them, participants fixated more on these stimuli during extinction and showed reduced fear potentiated startle to both the CS+ and the CS- during extinction and later during extinction recall (see also Barry et al., 2017). Therefore, the literature on overt manipulations of attention away from or towards conditioned stimuli is relatively cohesive, indicating that relative to when stimuli are the focus of attention, extinction learning is compromised when attention is diverted from conditioned stimuli.

Studies that have examined the effect of working memory load on extinction learning have been less consistent. For example, Raes and colleagues (2009) asked participants to hold two- or six-number strings in working memory during exposure to conditioned stimuli, and reported that working memory load negatively affected extinction learning, as measured by SCR. Therefore, working memory load may hinder extinction learning by competing with extinction learning for attentional resources. On the other hand, de Voogd and Phelps (2020) found that when conditioned stimuli were followed by an n-back working memory task, working memory load facilitated extinction learning and the retention of extinction memories 24 hours later, as measured using SCR.

One reason that working memory load could benefit extinction is that it may reorganize the allocation of cognitive resources between the central-executive and salience networks, thereby facilitating the suppression of fear-related responses (Carter et al., 2003; de Voogd, Kanen, et al., 2018; Vytal et al., 2012). In support of this view, another study (de Voogd, Kanen, et al., 2018) found that goal-directed eye movements during extinction increased activation in the dorsal frontoparietal network while improving extinction learning and reducing activity in the amygdala, with larger reductions associated with less threat recovery the next day. Moreover, cognitively demanding tasks might engage domain-general prefrontal brain regions, such as the dorsolateral prefrontal cortex (dlPFC; Smith et al., 1998), which is involved in extinction learning (Picó-Pérez et al., 2019) and might inhibit the amygdala (de Voogd, Kanen, et al., 2018; Erk et al., 2007; Kanske et al., 2011; McRae et al., 2010; Van Dillen et al., 2009). Therefore, performing an unrelated task that exerts cognitive load, such as a working memory task, could conceivably facilitate rather than hinder extinction, via activation of brain regions that inhibit those involved in threat responding (Banks et al., 2007). Indeed, there is evidence to suggest cognitive load is parametrically and inversely related to amygdala inhibition, such that higher levels of cognitive load are associated with lower levels of amygdala threat responding (for a review, see de Voogd, Hermans, et al., 2018; see also de Voogd & Phelps, 2020).

The LPP provides a measure of central autonomic system activity, and therefore provides insight into extinction learning at a different level of psychophysiological response than SCR. Moreover, the high temporal resolution afforded by ERPs has the potential to inform understanding of the time-course of working memory load’s effects on attention to conditioned stimuli. Here, we used the LPP to help resolve inconsistencies regarding the effect of working memory load on threat extinction. First, participants performed a threat acquisition task in which two stimuli were paired with shock (CS+s) and two stimuli were never paired with shock (CS-s). Next, participants performed an extinction learning task interspersed with a working memory task used in our prior work (MacNamara et al., 2011, 2019; MacNamara & Proudfit, 2014). On each trial, participants viewed two or six letters, prior to the presentation of a CS+ that had previously been paired with shock or a CS- that had not been paired with shock; at the end of each trial, participants typed in the letters they had seen at the beginning of the trial. One of the CS+s was always presented on High-load working memory trials, whereas the other CS+ was always presented on Low-load working memory trials; similarly, one of the CS-s was always presented on High-load working memory trials, whereas the other CS- was always presented on Low-load working memory trials. By measuring the LPP elicited by each of the CS+s and CS-s, we could assess differences in extinction learning under high compared to low working memory load.

During acquisition, we expected to observe larger LPPs and larger stimulus preceding negativities (SPNs - a negative-going component that provides a measure of anticipation of salient stimuli) for the CS+ versus the CS-, as well as ratings of increased shock expectancy for the CS+ versus the CS-. However, our hypotheses of interest concerned extinction, where we expected to observe a main effect of working memory load (Low-load > High-load) and a main effect of stimulus (CS+ > CS-) on the LPP. We also expected to observe a larger effect of stimulus (CS+ > CS-) on High-load versus Low-load trials for the LPP, which would indicate that working memory load impaired threat extinction. We expected that by the end of extinction, participants would no longer rate the CS+ as more likely to be paired with shock.

Method

Participants

Participants were 38 unselected undergraduates who completed the experiment for course credit; sample size was determined by the a priori decision to run the study for one semester. Twenty-eight participants (17 females, 11 males; M age = 20.21, SD = 6.14) were included in the final sample of acquisition data (four participants were excluded because they did not meet our criteria for inclusion in analyses of extinction task data; four additional participants were excluded for EEG artifacts in excess of 50% of trials, and two additional participants were excluded because of technical errors with data recording). Thirty-four participants (21 females, 13 males; M age = 19.94 years, SD = 5.59) were included in the final sample for analyses of data from the working memory and extinction task (three participants were excluded because they did not learn to associate the unconditioned stimulus with the conditioned stimuli during acquisition and one participant was excluded because of EEG artifacts in excess of 50% of trials). Therefore, all participants who performed the extinction task were included in analyses of acquisition data, however six additional participants were excluded from analyses of acquisition data because of problems specific to data collection in that task.

Post-hoc sensitivity analyses using G*Power (Faul et al., 2009) suggested that with a final sample size of n = 34, power 1- β = .80, and α = .05, extinction analyses were adequately powered to detect a small-to-medium effect size of f = .20, which converts to ηp2 = .04. For acquisition data, with a final sample size of n = 28, power 1- β = .80, and α = .05, was analyses were powered to detect a medium effect size of dz = .55.

Stimulus Materials

During the acquisition and extinction tasks, participants viewed four shapes - a blue circle, yellow triangle, red square, or green diamond. Visual stimuli were presented on a black background on a 21.75 in (55 cm) monitor and participants were seated approximately 70 cm from the screen. Shapes subtended approximately 10.4° of visual angle horizontally and 6.6° of visual angle vertically.

The unconditioned stimulus used in the acquisition task was a mild electric shock, delivered to the wrist. To control for individual differences in shock sensitivity, we set the shock level individually for each participant (Bradford et al., 2014; MacNamara & Barley, 2018). In brief, participants rated a series of gradually increasing shocks, using a scale from 0 (“Can’t feel shock”) to 100 (“Highest you can tolerate”). When participants indicated that the shock level had reached the highest level they could tolerate, no further shocks were administered, and the shock level selected by the participant was used for the acquisition task.

Letter strings were created using a random number generator (Reed, 2002) and were the same as in the original version of the task that used negative and neutral pictures (MacNamara et al., 2011). Vowels were not included in the strings; there were 60 two-consonant strings and 60 six-consonant strings (Ashcraft & Kirk, 2001).

Procedure

After giving their consent to participate in the experiment, participants completed a demographics questionnaire. Next, they performed an acquisition task, followed by an extinction task interspersed with a working memory task. Stimuli were presented using Presentation software (Neurobehavioral Systems Inc., Berkeley, CA). Study procedures were in compliance with the Helsinki Declaration of 1975 (as revised in 1983) and were approved by the Texas A&M institutional review board.

Acquisition

In the first part of the experiment, participants underwent a Pavlovian conditioning task, in order to learn conditioned shape associations. On each trial, participants viewed one of four shapes (a blue circle, yellow triangle, red square, or green diamond) for 4000 ms. Two of these shapes (CS+s) co-terminated with a 200 ms shock to the wrist 50% of the time; the other two shapes (CS-s) were never paired with shock. Shape assignment to condition was counterbalanced across participants. During the intertrial interval, participants viewed a white fixation presented on a black background for 3000 – 4500 ms. To encourage learning of shape-contingencies, the first three presentations of each CS+ in the task were paired with shock. Trials were evenly distributed between three blocks such that 12 trials of each CS were viewed in each block, for a total of 144 trials (36 of each trial type). Participants received a self-paced break between each block. At the end of each block, participants were asked to rate their shock expectancy for each CS on a 5-point Likert-type scale (“How likely were you to receive a shock when this shape appeared?” 1 = “Sure Shock”, 3 = “Not Sure”, 5 = “Sure No Shock”). Acquisition trials were not interspersed with working memory trials; acquisition was performed only to threat condition stimuli for use in the working memory and extinction task.

Working Memory and Extinction Task

Next, participants performed a working memory load task similar to that used in our prior work (MacNamara et al., 2011, 2019; MacNamara & Proudfit, 2014). Each trial began with the presentation of a two-letter (“Low-load”) or six-letter (“High-load”) string for 5000 ms, followed by a white fixation cross presented on a black background for 500 ms. Next, one of four shapes used in the previous task (i.e., blue circle, yellow triangle, red square, or green diamond) was presented for 4000 ms. Load-shape assignment was fixed for each participant, such that each shape was only presented under low or high working memory load, throughout the task. For example, one of the shapes previously paired with shock (e.g., green diamond CS+) was always presented under High-load and the other shape previously paired with shock (e.g., blue circle CS+) was always presented under Low-load. Similarly, one shape that had not previously been paired with shock (e.g., red square CS-) was always presented under High-load and the other shape that had not been paired with shock (e.g., yellow triangle CS-) was always presented under low working memory load. CS assignment to working-memory load condition was counterbalanced across participants. Participants were asked to remember the letters presented at the beginning of the trial and to enter them at the end of each trial. As in our prior work, letter entry was self-paced – i.e., when participants finished entering the letters, they pressed the enter key to proceed to the next trial.

Participants were told that the shapes were irrelevant to the task, but that they should keep their eyes on the screen the entire time. During the intertrial interval, a white fixation cross was displayed on a black background for 3000 – 4500 ms. Trial types were evenly distributed across 120 trials, such that four conditions were created: 30 CS+ Low-load trials; 30 CS+ High-load trials; 30 CS- Low-load trials and 30 CS- High-load trials. In total, the task consisted of three blocks, with an even distribution of trial types in each (i.e., 10 of each trial type). We used more trials in the acquisition compared to the extinction task, because: a) we wanted a high likelihood that participants would learn the meaning of each shape; b) trials in the working memory task were longer (because of time needed for the presentation of letters and letter recall); and c) our prior work using this working memory task had 30 trials per condition, as in the current extinction task (MacNamara et al., 2011, 2019; MacNamara & Proudfit, 2014). Trials were presented in a random order in each block and participants received a self-paced break between blocks. Throughout the task, a shock electrode was connected to participants’ wrists, but no shocks were delivered. At the end of the task, participants were asked to rate their shock expectancy for each of the four shapes, using the same Likert-type scale used in acquisition.

EEG Recording and Data Reduction

For both the acquisition and the working memory and extinction tasks, continuous EEG was recorded using an ActiCap and the ActiCHamp amplifier system (Brain Products, Munich, Germany). Thirty-two electrode sites were used based on the 10/20 system. The electrooculogram (EOG) was recorded from four facial electrodes: two that were placed approximately 1 cm above and below the right eye, forming a bipolar channel to measure vertical eye movement and blinks and two that were placed approximately 1 cm beyond the outer edges of each eye, forming a bipolar channel to measure horizontal eye movements. The EEG data were digitized at 24-bit resolution and a sampling rate of 1000 Hz.

EEG data were processed offline using Brain Vision Analyzer 2 software (Brain Products, v2.1.2). Data were processed separately for the acquisition and working memory extinction tasks, though processing parameters were the same across tasks. For both tasks, data were segmented for each trial beginning 200 ms prior to CS onset and continuing for 4000 ms (i.e., until CS offset); baseline correction for each trial was performed using the 200 ms prior to CS onset.

The signal from each electrode was re-referenced to the average of the left and right mastoids (TP9/10) and band-pass filtered with high-pass and low-pass filters of 0.01 and 30 Hz, respectively. Eye blink and ocular corrections used the method developed by (Miller et al., 1988). Artifact analysis was used to identify a voltage step of more than 50.0 μV between sample points, a voltage difference of 300.0 μV within a trial, and a maximum voltage difference of less than 0.50 μV within 100 ms intervals. Trials were also inspected visually for any remaining artifacts, and data from individual channels containing artifacts were rejected on a trial-to-trial basis.

Acquisition

ERPs were time-locked to each CS. The LPP was scored by averaging amplitudes at a parieto-occipital pooling of Pz, PO3 and PO4 between 400–800 ms following CS onset (Bauer et al., 2020; Kujawa et al., 2015; MacNamara et al., 2016). The SPN was scored by averaging amplitudes at Fz, from 2000–3800 ms following CS onset (MacNamara & Barley, 2018). The time window and electrodes used to score the ERPs were selected using a collapsed localizer approach (S. J. Luck & Gaspelin, 2017).

Working Memory and Extinction Task

ERPs time-locked to each CS were created separately for each block. Based on functional differences between early and late portions of the ERP (Hajcak et al., 2010; MacNamara et al., 2009; Schupp et al., 2006), the LPP was scored by averaging amplitudes at a centro-parietal pooling of Pz, CP1, and CP2, separately for 400–2000 ms and for 2000–4000 ms following CS onset (Bauer et al., 2020; MacNamara et al., 2011; MacNamara & Proudfit, 2014), and separately for each block. The time window and electrodes used to score the LPP were selected using a collapsed localizer approach (S. J. Luck & Gaspelin, 2017).

Working Memory Performance

Responses during the working memory and extinction task were considered correct if the letters entered matched those presented at the beginning of the trial and were entered in the same order. The percentage of correct responses was calculated separately for each condition and block.

Data Analyses

Analyses were performed using SPSS statistical software, version 25 (IBM, Armonk, NY).

Acquisition

Though our primary interest and hypotheses of interest concerned extinction data, we include analyses of acquisition data here for completeness and to demonstrate successful acquisition. Paired sample t-tests were used to assess the effect of stimulus (CS+ versus CS-) on the LPP, SPN and shock likelihood ratings across all blocks. In addition, we divided Block 1 into an early and a late portion, each consisting of 24 trials. Paired samples t-tests were used to compare the ERPs elicited on CS+ and CS- trials between early and late Block 1, in order to assess the time course of threat conditioning during Block 1. Additional analyses including block as a factor are presented in the Supplementary Material section.

Working Memory and Extinction Task

For the working memory and extinction task, a 2 (stimulus: CS-, CS+) x 2 (working memory load: Low-load, High-load) repeated measures analysis of variance (ANOVA) was performed for the LPP and for working memory performance, separately for each block. We used a paired samples t-test to compare participant ratings of shock likelihood for CS+ and CS- trials made at the end of extinction (ratings were only made at the end of the task, not after each block). Analyses were performed separately for each block to provide a more fine-grained assessment of the effect of working memory load on extinction learning at various time points throughout the task. Nonetheless, for completeness and for future hypotheses, we also present analyses including the factor block in the Supplementary Material section.

Results

Acquisition

LPP

Fig. 1 depicts grand-averaged waveforms for the LPP and headmaps depicting the spatial distribution of voltage for the LPP elicited on CS+ trials and CS- trials. There was a significant main effect of stimulus, t(27) = 5.05, p < .001, such that CS+ trials elicited larger LPPs than CS- trials (CS+: M = 7.09 μV, SD = 4.29 μV; CS-: M = 5.06 μV, SD = 4.13 μV). Moreover, the LPP elicited on CS+ trials was larger in late Block 1 compared to early Block 1, t(27) = 3.99, p < .001 (early Block 1 CS+: M = 6.25 μV, SD = 6.99 μV; late Block 1 CS+: M = 9.23 μV, SD = 6.83 μV). The LPP elicited by the CS- did not differ between early and late Block 1 (p = .16; early Block 1 CS-: M = 4.36 μV, SD = 6.93 μV; late Block 1 CS-: M = 6.15 μV, SD = 5.06 μV).

Fig. 1.

Fig. 1

Open in a new tab

Acquisition: Grand-averaged waveforms at the parieto-occipital pooling where the LPP was scored and headmaps depicting the spatial distribution of voltage for the CS+ and CS- from 400–800 ms after cue onset. Waveforms were filtered with a high-pass filter of 12 Hz for illustrative purposes only (not analyses).

SPN

Fig. 2 depicts grand-averaged waveforms for the SPN and headmaps depicting the spatial distribution of voltage for the SPN for each stimulus. There was a significant main effect of stimulus, t(27) = 4.57, p < .001, such that CS+ trials elicited larger (more negative) SPNs compared to CS- trials (CS+: M = −2.80 μV, SD = 7.19 μV; CS-: M = 1.42 μV, SD = 6.95 μV). The SPN elicited by the CS+ did not differ between early and late Block 1 (p = .84; early Block 1 CS+: M = −2.80 μV, SD = 15.42 μV; late Block 1 CS+: M = −2.17 μV, SD = 9.82 μV). Similarly, the SPN elicited by the CS- did not differ between early and late Block 1 (p = .22; early Block 1 CS-: M = −1.80 μV, SD = 13.78 μV; late Block 1 CS-: M = 2.07 μV, SD = 11.90 μV) .

Fig. 2.

Fig. 2

Open in a new tab

Acquisition: Grand-averaged waveforms at Fz where the SPN was scored and headmaps depicting the spatial distribution of voltage for the CS+ and CS- from 2000–3800 ms after cue onset. Waveforms were filtered with a high-pass filter of 12 Hz for illustrative purposes only (not analyses).

Participant Perception of Shock Likelihood

There was a significant effect of stimulus, t(27) = 27.40, p < .001 (CS+: M = 2.25, SD = .53; CS-: M = 4.98, SD = .13), such that participants thought they were more likely to be shocked on CS+ compared to CS- trials.

Working Memory and Extinction Task

Table 1 presents means and standard deviations for the LPP and working memory performance, shown separately for each of CS+ and CS-, level of working memory load and block.

Table 1.

Means (and standard deviations) for the LPP and working memory performance in the extinction task, shown separately for each condition and level of working memory load, and separately for each block.

Block 1 Block 2 Block 3
CS- Low- load CS- Highload CS+ Low- load CS+ High- load CS- Low- load CS- High- load CS+ Low- load CS+ High- load CS- Low- load CS- High- load CS+ Low- load CS+ High- load
 LPP 400–2000 ms (μv) .48 (4.36) −1.43 (4.58) 1.12 (4.17) .19 (4.80) 1.29 (4.63) −1.06 (4.62) .11 (3.54) .68 (5.57) 1.33 (4.20) −.20 (4.28) 1.16 (3.62) .11 (4.23)
 LPP 2000–4000 ms (μv) 0.03 (6.24) −3.52 (7.09) 1.01 (5.35) −1.00 (7.52) .62 (7.58) −2.99 (7.23) −1.21 (5.10) −1.09 (7.14) 2.09 (7.17) −2.14 (6.02) .95 (5.18) −1.59 (6.30)
 Working memory performance (%
 correct)		 100.00 (.00) 72.94 (26.80) 97.94 (4.79) 70.59 (21.59) 97.06 (6.29) 75.29 (19.27) 97.35 (5.67) 70.29 (21.81) 97.65 (6.54) 71.18 (22.26) 98.24 (4.59) 63.53 (31.90)
Open in a new tab

LPP

Fig. 3 depicts grand-averaged waveforms and the spatial distribution of voltages for the LPP in A) Block 1; B) Block 2 and C) Block 3.

Fig. 3.

Fig. 3

Open in a new tab

Working memory and extinction: Grand-averaged waveforms at the centro-parietal pooling where the LPP was scored (left) and headmaps depicting the spatial distribution of voltage for each condition, level of working memory load and LPP time window, shown separately for A) Block 1, B) Block 2 and C) Block 3. Waveforms were filtered with a high-pass filter of 12 Hz for illustrative purposes only (not analyses).

Block 1.
400–2000 ms.

There was a main effect of stimulus, F(1, 33) = 4.22, p = .05, ηp2 = .11, such that CS+ trials elicited larger LPPs compared to CS- trials. The effect of working memory load and the interaction between stimulus X working memory load did not reach significance (ps > .11).

2000–4000 ms.

There was a main effect of stimulus, F(1, 33) = 6.03, p = .02, ηp2 = .16, such that CS+ trials elicited larger LPPs compared to CS- trials. There was also a main effect of working memory load, F(1, 33) = 5.94, p = .02, ηp2 = .15, such that the LPP was larger for CSs that were presented on Low-load compared to High-load trials. The interaction between stimulus X working memory load did not reach significance (p = .34).

Block 2 .
400–2000 ms.

The effects of stimulus and working memory load did not reach significance (ps > .37). There was a significant interaction between stimulus X working memory load, F(1,33) = 4.42, p = .04, ηp2 = .12, such that the effect of stimulus was larger on High- compared to Low-load trials. Follow-up tests indicated that this was driven by larger LPPs to the CS+ versus the CS- on High-load, t(33) = 2.13, p = .04, but not Low-load trials, p = .26.

2000–4000 ms.

The effects of working memory load and stimulus did not reach significance (ps > .22). The interaction between stimulus X working memory load did not reach significance (p = .06), but was in the same direction as in the 400–2000 ms time window.

Block 3.
400–2000 ms.

The effects of working memory load, stimulus and their interaction did not reach significance (ps > .10).

2000–4000 ms.

There was a significant main effect of working memory load, F(1, 33) = 8.57, p = .01, ηp2 = .21, such that the LPP was larger on Low-load compared to High-load trials. The effect of stimulus and the interaction between stimulus X working memory load did not reach significance (ps > .31).

Working Memory Performance

Block 1.

A main effect of working memory load, F(1, 33) = 50.30, p < .001, ηp2 = .60 indicated that participants recalled more letters on Low-load compared to High-load trials. The effect of stimulus and the interaction between stimulus X working memory load did not reach significance (ps > .22).

Block 2.

The main effect of stimulus did not reach significance (p = .09). A main effect of working memory load, F(1, 33) = 53.54, p < .001, ηp2 = .62 indicated that participants recalled more letters on Low-load compared to High-load trials. This was qualified by an interaction between stimulus X working memory load, F(1, 33) = 4.11, p = .05, ηp2 = .11, such that the stimulus (CS+ versus CS-) had a more deleterious effect on working memory performance on High-load compared to Low-load trials.

Block 3.

The main effect of stimulus did not reach significance (p = .11). A main effect of working memory load, F(1, 33) = 54.01, p < .001, ηp2 = .62 indicated that participants recalled more letters on Low-load compared to High-load trials. This was qualified by an interaction between stimulus X working memory load, F(1, 33) = 3.99, p = .05, ηp2 = .11, such that the stimulus (CS+ versus CS-) had a more deleterious effect on working memory performance on High-load compared to Low-load trials.

Participant Perception of Shock Likelihood

At the end of extinction, participants still thought that it was somewhat more likely that they would be shocked on CS+ trials (M = 4.38, SD = 1.07) compared to CS- trials (M = 4.85, SD = .50), t(33) = 2.77, p = .011. Nonetheless, participants thought it was less likely that they would be shocked on CS+ trials at the end of extinction (M = 4.43, SD = 1.03) compared to at the end of acquisition (M = 2.32, SD = .70), t(27) = 10.00, p < .001. By contrast, shock expectancy for the CS- did not differ between the end of extinction (M = 4.89, SD = .42) and the end of acquisition (M = 5.00, SD = .00), t(27) = 1.36, p = .182.

Discussion

Exposure therapy for anxiety is based on the principles of extinction learning. Therefore, examination of the factors that modulate extinction learning may provide a useful means of testing assumptions underlying recommended best practice for exposure therapy and in improving or updating these recommendations as needed (Craske et al., 2018). Here, we tested the effect of working memory load on extinction learning, using the LPP as a measure of the motivational salience of conditioned stimuli. Following acquisition, the LPP was larger for the CS+ compared to the CS- and participants thought it was more likely they would be shocked on CS+ compared to CS- trials. During extinction, we observed a main effect of working memory load (Blocks 1 and 3), such that it reduced the LPP elicited by both the CS+ and the CS-. Nonetheless, working memory load also slowed extinction learning. Specifically, we observed larger LPPs to the CS+ versus the CS- presented under high compared to low working memory load, during the middle portion of extinction learning (i.e. Block 2). Therefore, while the LPP did not differ for CS+ versus CS- trials by the end of extinction, more trials were necessary to extinguish conditioned responding under high working memory load.

Working memory load can be thought of as a form of distraction, which may interfere with extinction by detracting attention from the conditioned stimulus, and by impeding new learning regarding the absence of the US (Weisman & Rodebaugh, 2018). While it is often assumed that distraction impedes exposure therapy, research has been less conclusive, with distraction found to be beneficial (Johnstone & Page, 2004; Oliver & Page, 2008; Penfold & Page, 1999), harmful (Dethier et al., 2015; Kamphuis & Telch, 2000; Mohlman & Zinbarg, 2000), and inconsequential (Antony et al., 2001; Telch et al., 2004) to exposure therapy success. Few studies to-date have investigated how distraction might affect extinction learning in a nontherapeutic context (i.e., in the lab), and those that did have also yielded inconsistent results. For example, Raes and colleagues (2009) found that participants who underwent extinction under high working memory load showed larger differences in CS+ > CS- SCR as compared to those under low working memory load. By contrast, other work has suggested that working memory load improves extinction learning, as assessed using SCR (de Voogd & Phelps, 2020). Importantly, however, participants in the latter study memorized and recalled digits in an n-back task presented after each extinction trial, such that attention to conditioned stimuli was not compromised during extinction learning in this design.

Though there are differences between the current study and that of Raes and colleagues (2009), including working memory trials during both acquisition and extinction in the latter, both studies used working memory tasks that were executed concurrently with extinction learning. Therefore, the deleterious effect of working memory load on threat extinction that was found in both studies might be explained in part by diversion of processing resources away from the threat-conditioned stimuli, in line with prior work that has used more overt manipulations of attention away from conditioned stimuli during extinction (Barry et al., 2017; Klein et al., 2021; O’Malley & Waters, 2018). Extinction learning is thought to occur via ventromedial prefrontal cortex (vmPFC) inhibition of the amygdala, which leads to a reduction in the expression of threat responses (Milad & Quirk, 2012). Working memory load and other goal-directed cognitive tasks, on the other hand, are believed to mainly engage the dorsolateral prefrontal cortex dlPFC (D’Esposito et al., 2000; Miller & Cohen, 2001; Smith et al., 1998). Though anatomical connectivity studies have failed to find direct connections between the amygdala and the dlPFC (Barbas, 2000; McDonald et al., 1996; Stefanacci & Amaral, 2002), fMRI work has found that dlPFC engagement in the context of high cognitive demands inhibits amygdala activation (Erk et al., 2007; Kanske et al., 2011; McRae et al., 2010; Van Dillen et al., 2009), in line with the main effect of working memory load observed on the LPP in the current study. The vmPFC has also been shown to mediate dlPFC inhibition of the amygdala (i.e., during the reappraisal of threat-conditioned stimuli; Delgado et al., 2008). Therefore, recruitment of the dlPFC by a working memory task during extinction learning might have diminished its capacity to contribute to the downregulation of CS+ responding via the vmPFC, thereby slowing extinction learning.

Importantly, the meaning behind larger LPPs to the CS+ versus the CS- on High-load trials in Block 2 must not be confused with evidence of smaller LPPs to both the CS+ and the CS- on High-load trials (i.e., main effect of working memory load) in Blocks 1 and 3. Whereas the interaction in Block 2 indicates that the threat salience of the CS+ was preserved to a better extent on High-load compared to Low-load trials, the main effect of working memory load indicates that fewer processing resources were allocated to the CSs overall under High-load (i.e., irrespective of threat value). That is, high levels of cognitive load likely led to the availability of fewer processing resources for vmPFC-mediated inhibition of CS+ threat salience as well as an overall reduction in the availability of processing resources for brain regions involved in the generation of the LPP (e.g., in the visual cortex, medial and lateral prefrontal cortex; Liu et al., 2012; MacNamara et al., 2018). In line with the notion that the threat salience of the CS+ was preserved on High-load trials, we also observed that working memory performance was reduced for CS+ High-load trials (in Block 2), which fits with prior evidence showing that threatening stimuli can interfere with working memory performance (MacNamara et al., 2011). Of note, future research using functional brain imaging would be necessary to test the hypotheses outlined above and to increase understanding of the effect of working memory load on the neurocircuitry of threat extinction (beyond CS+ threat value output/LPP).

Main effects of working memory load (in Blocks 1 and 3) were only observed during the later portion of picture presentation (i.e., 2000–4000 ms after CS onset). In our prior work, we found that working memory load reduced the LPP elicited by negative and neutral scenes both during both early and later portions of picture presentation (MacNamara et al., 2011). Therefore, differences in stimulus complexity might explain our failure to observe an early effect of working memory load in the current study. For example, because the conditioned stimuli used here were less visually complex than pictorial scenes, participants may have been able to allocate comparable levels of attentional resources to these stimuli under both high and low levels of working memory load, at least early on during stimulus presentation. Nonetheless, condition means indicated that although non-significant, the effect of working memory load was in the expected direction during the early time window (Low-load > High-load), suggesting that it might have reached significance with more trials and/or participants.

The LPP elicited by the CS+ and CS- no longer differed by the end of extinction (i.e. Block 3), and participants showed a reduction in shock likelihood ratings of the CS+ from the beginning to the end of extinction. Nonetheless, participants still rated the CS+ as more likely to elicit shock than the CS- at the end of extinction. While this result was unexpected, it aligns with findings in prior studies (Klein et al., 2021; Lipp & Edwards, 2002; C. C. Luck & Lipp, 2015; O’Malley & Waters, 2018; Raes et al., 2009; Vansteenwegen et al., 2005), and with the broader finding that measures of threat responding do not always coincide across subjective, behavioral and psychophysiological levels (Beckers et al., 2013; MacNamara et al., 2013). Subjective measures may involve higher cognitive processes that may be more resistant to extinction than more immediate physiological measures (C. C. Luck & Lipp, 2016).

In sum, results indicate that working memory load modulates the electrocortical processing of conditioned threat stimuli and that concurrent working memory load slows (but does not ultimately prevent) extinction learning. Conclusions are limited by our use of a one-day paradigm, which precludes examination of how working memory load during extinction learning might affect extinction retention the next day or several days later. Nonetheless, in a therapeutic context, treatment duration matters, with fewer sessions reducing burden on therapist and patient – and may be facilitated by keeping distractions (including the patient’s internal or self-generated ones) to a minimum. In this context, our results offer empirical support for the practice of maximizing attention to threatening stimuli during extinction learning/exposure therapy, and suggest that by reducing distraction/cognitive load, exposure therapy may be more efficient.

Supplementary Material

1

Highlights.

  • Knowing how working memory load affects extinction is relevant for exposure therapy

  • An extinction task was interspersed with low & high working memory load during EEG

  • Overall, working memory load reduced processing of conditioned stimuli

  • Extinction learning was slower under high versus low working memory load

  • There is empirical support for limiting distraction to optimize extinction learning

Acknowledgements

AM was supported by NIMH K23 MH105553 during the design, data collection and data analysis of this project and by NIMH R01 MH125083 during the writing and editing of this manuscript. Thank you to the MAClab research assistants.

Footnotes

Author CRediT statement

Yuhan Cheng: statistical analyses; writing - original draft preparation. T. Bryan Jackson: writing - reviewing and editing. Annmarie MacNamara: conceptualization; methodology; writing – reviewing and editing; supervision.

Declarations of interest: none

1

A t-test was used instead of a 2 X 2 repeated measures ANOVA, because ratings were identical for the CS- presented on high versus low working memory load trials and for the CS+ presented on high versus low working memory load trials.

2

M and SD for shock expectancy ratings made at the end of extinction are descriptively different depending on whether analyses were restricted to individuals with useable data for both acquisition and extinction (i.e., when comparing ratings made at the end of acquisition with those made at the end of extinction) or just extinction (i.e., when comparing CS+ and CS- ratings made at the end of extinction).

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Antony MM, McCabe RE, Leeuw I, Sano N, & Swinson RP (2001). Effect of distraction and coping style on in vivo exposure for specific phobia of spiders. Behaviour Research and Therapy, 39(10), 1137–1150. 10.1016/S0005-7967(00)00089-9 [DOI] [PubMed] [Google Scholar]
  2. Ashcraft MH, & Kirk EP (2001). The relationships among working memory, math anxiety, and performance. Journal of Experimental Psychology: General, 130(2), 224–237. 10.1037/0096-3445.130.2.224 [DOI] [PubMed] [Google Scholar]
  3. Banks SJ, Eddy KT, Angstadt M, Nathan PJ, & Phan KL (2007). Amygdala–frontal connectivity during emotion regulation. Social Cognitive and Affective Neuroscience, 2(4), 303–312. 10.1093/scan/nsm029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barbas H. (2000). Connections underlying the synthesis of cognition, memory, and emotion in primate prefrontal cortices. Brain Research Bulletin, 52(5), 319–330. 10.1016/S0361-9230(99)00245-2 [DOI] [PubMed] [Google Scholar]
  5. Barry TJ, Vervliet B, & Hermans D. (2017). Feature Specific Attention and Return of Fear after Extinction. Journal of Experimental Psychopathology, 8(1), 76–87. 10.5127/jep.051115 [DOI] [Google Scholar]
  6. Bauer EA, MacNamara A, Sandre A, Lonsdorf TB, Weinberg A, Morriss J, & Reekum C. M.van (2020). Intolerance of uncertainty and threat generalization: A replication and extension. Psychophysiology, 57(5), e13546. 10.1111/psyp.13546 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Beckers T, Krypotos A-M, Boddez Y, Effting M, & Kindt M. (2013). What’s wrong with fear conditioning? Biological Psychology, 92(1), 90–96. 10.1016/j.biopsycho.2011.12.015 [DOI] [PubMed] [Google Scholar]
  8. Bouton ME (2004). Context and behavioral processes in extinction. Learning & Memory (Cold Spring Harbor, N.Y.), 11(5), 485–494. 10.1101/lm.78804 [DOI] [PubMed] [Google Scholar]
  9. Bradford DE, Magruder KP, Korhumel RA, & Curtin JJ (2014). Using the Threat Probability Task to Assess Anxiety and Fear During Uncertain and Certain Threat. Journal of Visualized Experiments, 91, 51905. 10.3791/51905 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Carter RM, Hofstötter C, Tsuchiya N, & Koch C. (2003). Working memory and fear conditioning. Proceedings of the National Academy of Sciences, 100(3), 1399–1404. 10.1073/pnas.0334049100 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Craske MG, Hermans D, & Vervliet B. (2018). State-of-the-art and future directions for extinction as a translational model for fear and anxiety. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1742). 10.1098/rstb.2017.0025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Craske MG, Kircanski K, Zelikowsky M, Mystkowski J, Chowdhury N, & Baker A. (2008). Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy, 46(1), 5–27. 10.1016/j.brat.2007.10.003 [DOI] [PubMed] [Google Scholar]
  13. Craske MG, Treanor M, Conway CC, Zbozinek T, & Vervliet B. (2014). Maximizing exposure therapy: An inhibitory learning approach. Behaviour Research and Therapy, 58, 10–23. 10.1016/j.brat.2014.04.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Cuthbert BN, Schupp HT, Bradley MM, Birbaumer N, & Lang PJ (2000). Brain potentials in affective picture processing: Covariation with autonomic arousal and affective report. Biological Psychology, 52(2), 95–111. 10.1016/S0301-0511(99)00044-7 [DOI] [PubMed] [Google Scholar]
  15. de Voogd LD, Hermans EJ, & Phelps EA (2018). Regulating defensive survival circuits through cognitive demand via large-scale network reorganization. Current Opinion in Behavioral Sciences, 24, 124–129. 10.1016/j.cobeha.2018.08.009 [DOI] [Google Scholar]
  16. de Voogd LD, Kanen JW, Neville DA, Roelofs K, Fernández G, & Hermans EJ (2018). Eye-Movement Intervention Enhances Extinction via Amygdala Deactivation. The Journal of Neuroscience, 38(40), 8694–8706. 10.1523/JNEUROSCI.0703-18.2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. de Voogd LD, & Phelps EA (2020). A cognitively demanding working-memory intervention enhances extinction. Scientific Reports, 10(1), 7020. 10.1038/s41598-020-63811-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Delgado MR, Nearing KI, LeDoux JE, & Phelps EA (2008). Neural Circuitry Underlying the Regulation of Conditioned Fear and Its Relation to Extinction. Neuron, 59(5), 829–838. 10.1016/j.neuron.2008.06.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. D’Esposito M, Postle BR, & Rypma B. (2000). Prefrontal cortical contributions to working memory: Evidence from event-related fMRI studies. Experimental Brain Research, 133(1), 3–11. 10.1007/s002210000395 [DOI] [PubMed] [Google Scholar]
  20. Dethier V, Bruneau N, & Philippot P. (2015). Attentional focus during exposure in spider phobia: The role of schematic versus non-schematic imagery. Behaviour Research and Therapy, 65, 86–92. 10.1016/j.brat.2014.12.016 [DOI] [PubMed] [Google Scholar]
  21. Erk S, Kleczar A, & Walter H. (2007). Valence-specific regulation effects in a working memory task with emotional context. NeuroImage, 37(2), 623–632. 10.1016/j.neuroimage.2007.05.006 [DOI] [PubMed] [Google Scholar]
  22. Faul F, Erdfelder E, Buchner A, & 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. 10.3758/BRM.41.4.1149 [DOI] [PubMed] [Google Scholar]
  23. Hajcak G, MacNamara A, & Olvet DM (2010). Event-Related Potentials, Emotion, and Emotion Regulation: An Integrative Review. Developmental Neuropsychology, 35(2), 129–155. 10.1080/87565640903526504 [DOI] [PubMed] [Google Scholar]
  24. Hajcak G, & Olvet DM (2008). The Persistence of Attention to Emotion: Brain Potentials During and After Picture Presentation. Emotion (Washington, D.C.), 8(2), 250–255. 10.1037/1528-3542.8.2.250 [DOI] [PubMed] [Google Scholar]
  25. Hickey C, McDonald JJ, & Theeuwes J. (2006). Electrophysiological Evidence of the Capture of Visual Attention. Journal of Cognitive Neuroscience, 18(4), 604–613. 10.1162/jocn.2006.18.4.604 [DOI] [PubMed] [Google Scholar]
  26. Johnstone KA, & Page AC (2004). Attention to phobic stimuli during exposure: The effect of distraction on anxiety reduction, self-efficacy and perceived control. Behaviour Research and Therapy, 42(3), 249–275. 10.1016/S0005-7967(03)00137-2 [DOI] [PubMed] [Google Scholar]
  27. Kamphuis JH, & Telch MJ (2000). Effects of distraction and guided threat reappraisal on fear reduction during exposure-based treatments for specific fears. Behaviour Research and Therapy, 38(12), 1163–1181. 10.1016/S0005-7967(99)00147-3 [DOI] [PubMed] [Google Scholar]
  28. Kanske P, Heissler J, Schönfelder S, Bongers A, & Wessa M. (2011). How to Regulate Emotion? Neural Networks for Reappraisal and Distraction. Cerebral Cortex, 21(6), 1379–1388. 10.1093/cercor/bhq216 [DOI] [PubMed] [Google Scholar]
  29. Klein Z, Ginat-Frolich R, Barry TJ, & Shechner T. (2021). Effects of increased attention allocation to threat and safety stimuli on fear extinction and its recall. Journal of Behavior Therapy and Experimental Psychiatry, 72, 101640. 10.1016/j.jbtep.2021.101640 [DOI] [PubMed] [Google Scholar]
  30. Kujawa A, Macnamara A, Fitzgerald KD, Monk CS, & Phan KL (2015). Enhanced Neural Reactivity to Threatening Faces in Anxious Youth: Evidence from Event-Related Potentials. Journal of Abnormal Child Psychology, 43(8), 1493–1501. 10.1007/s10802-015-0029-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lipp OV, & Edwards MS (2002). Effect of instructed extinction on verbal and autonomic indices of Pavlovian learning with fear-relevant and fear-irrelevant conditional stimuli. Journal of Psychophysiology, 16(3), 176–186. 10.1027//0269-8803.16.3.176 [DOI] [Google Scholar]
  32. Liu Y, Huang H, McGinnis-Deweese M, Keil A, & Ding M. (2012). Neural Substrate of the Late Positive Potential in Emotional Processing. Journal of Neuroscience, 32(42), 14563–14572. 10.1523/JNEUROSCI.3109-12.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Luck CC, & Lipp OV (2015). A potential pathway to the relapse of fear? Conditioned negative stimulus evaluation (but not physiological responding) resists instructed extinction. Behaviour Research and Therapy, 66, 18–31. 10.1016/j.brat.2015.01.001 [DOI] [PubMed] [Google Scholar]
  34. Luck CC, & Lipp OV (2016). Instructed extinction in human fear conditioning: History, recent developments, and future directions. Australian Journal of Psychology, 68(3), 209–227. 10.1111/ajpy.12135 [DOI] [Google Scholar]
  35. Luck SJ, & Gaspelin N. (2017). How to get statistically significant effects in any ERP experiment (and why you shouldn’t): How to get significant effects. Psychophysiology, 54(1), 146–157. 10.1111/psyp.12639 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. MacNamara A, & Barley B. (2018). Event- related potentials to threat of predictable and unpredictable shock. Psychophysiology, 55(10), e13206–n/a. 10.1111/psyp.13206 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. MacNamara A, Ferri J, & Hajcak G. (2011). Working memory load reduces the late positive potential and this effect is attenuated with increasing anxiety. Cognitive, Affective, & Behavioral Neuroscience, 11(3), 321–331. 10.3758/s13415-011-0036-z [DOI] [PubMed] [Google Scholar]
  38. MacNamara A, Foti D, & Hajcak G. (2009). Tell me about it: Neural activity elicited by emotional pictures and preceding descriptions. Emotion, 9(4), 531–543. 10.1037/a0016251 [DOI] [PubMed] [Google Scholar]
  39. MacNamara A, Jackson TB, Fitzgerald JM, Hajcak G, & Phan KL (2019). Working Memory Load and Negative Picture Processing: Neural and Behavioral Associations With Panic, Social Anxiety, and Positive Affect. Biological Psychiatry : Cognitive Neuroscience and Neuroimaging, 4(2), 151–159. 10.1016/j.bpsc.2018.04.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. MacNamara A, Kappenman ES, Black SR, Bress JN, & Hajcak G. (2013). Integrating behavioral and electrocortical measures of attentional bias toward threat. In Handbook of self-regulatory processes in development: New directions and international perspectives (pp. 215–243). Psychology Press. 10.4324/9780203080719.ch11 [DOI] [Google Scholar]
  41. MacNamara A, & Proudfit GH (2014). Cognitive Load and Emotional Processing in Generalized Anxiety Disorder: Electrocortical Evidence for Increased Distractibility. Journal of Abnormal Psychology (1965), 123(3), 557–565. 10.1037/a0036997 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. MacNamara A, Rabinak CA, Kennedy AE, & Phan KL (2018). Convergence of fMRI and ERP measures of emotional face processing in combat- exposed U. S. military veterans. Psychophysiology, 55(2), e12988–n/a. 10.1111/psyp.12988 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. MacNamara A, Vergés A, Kujawa A, Fitzgerald KD, Monk CS, & Phan KL (2016). Age-related changes in emotional face processing across childhood and into young adulthood: Evidence from event-related potentials. Developmental Psychobiology, 58(1), 27–38. 10.1002/dev.21341 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. McDonald AJ, Mascagni F, & Guo L. (1996). Projections of the medial and lateral prefrontal cortices to the amygdala: A Phaseolus vulgaris leucoagglutinin study in the rat. Neuroscience, 71(1), 55–75. 10.1016/0306-4522(95)00417-3 [DOI] [PubMed] [Google Scholar]
  45. McRae K, Hughes B, Chopra S, Gabrieli JDE, Gross JJ, & Ochsner KN (2010). The Neural Bases of Distraction and Reappraisal. Journal of Cognitive Neuroscience, 22(2), 248–262. 10.1162/jocn.2009.21243 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Milad MR, & Quirk GJ (2012). Fear Extinction as a Model for Translational Neuroscience: Ten Years of Progress. Annual Review of Psychology, 63(1), 129–151. 10.1146/annurev.psych.121208.131631 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Miller EK, & Cohen JD (2001). An integrative theory of prefrontal cortex function. Annual Review of Neuroscience, 24, 167–202. [DOI] [PubMed] [Google Scholar]
  48. Miller GA, Gration G, & Yee CM (1988). Generalized Implementation of an Eye Movement Correction Procedure. Psychophysiology, 25(2), 241–243. 10.1111/j.1469-8986.1988.tb00999.x [DOI] [Google Scholar]
  49. Mohlman J, & Zinbarg RE (2000). What kind of attention is necessary for fear reduction? An empirical test of the emotional processing model. Behavior Therapy, 31(1), 113–133. 10.1016/S0005-7894(00)80007-6 [DOI] [Google Scholar]
  50. Myers KM, & Davis M. (2007). Mechanisms of fear extinction. Molecular Psychiatry, 12(2), 120–150. 10.1038/sj.mp.4001939 [DOI] [PubMed] [Google Scholar]
  51. Nonkes LJP, de Pooter M, & Homberg JR (2012). Behavioural therapy based on distraction alleviates impaired fear extinction in male serotonin transporter knockout rats. Journal of Psychiatry & Neuroscience : JPN, 37(4), 224–230. 10.1503/jpn.110116 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Oliver NS, & Page AC (2008). Effects of internal and external distraction and focus during exposure to blood-injury-injection stimuli. Journal of Anxiety Disorders, 22(2), 283–291. 10.1016/j.janxdis.2007.01.006 [DOI] [PubMed] [Google Scholar]
  53. O’Malley KR, & Waters AM (2018). Attention avoidance of the threat conditioned stimulus during extinction increases physiological arousal generalisation and retention. Behaviour Research and Therapy, 104, 51–61. 10.1016/j.brat.2018.03.001 [DOI] [PubMed] [Google Scholar]
  54. Parvaz MA, MacNamara A, Goldstein RZ, & Hajcak G. (2012). Event-related induced frontal alpha as a marker of lateral prefrontal cortex activation during cognitive reappraisal. Cognitive, Affective, & Behavioral Neuroscience, 12(4), 730–740. 10.3758/s13415-012-0107-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Pastor MC, Bradley MM, Löw A, Versace F, Moltó J, & Lang PJ (2007). Affective picture perception: Emotion, context, and the late positive potential. Brain Research, 1189, 145–151. 10.1016/j.brainres.2007.10.072 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Penfold K, & Page AC (1999). The effect of distraction on within-session anxiety reduction during brief in vivo exposure for mild blood-injection fears. Behavior Therapy, 30(4), 607–621. 10.1016/S0005-7894(99)80028-8 [DOI] [Google Scholar]
  57. Picó-Pérez M, Alemany-Navarro M, Dunsmoor JE, Radua J, Albajes-Eizagirre A, Vervliet B, Cardoner N, Benet O, Harrison BJ, Soriano-Mas C, & Fullana MA (2019). Common and distinct neural correlates of fear extinction and cognitive reappraisal: A meta-analysis of fMRI studies. Neuroscience and Biobehavioral Reviews, 104, 102–115. 10.1016/j.neubiorev.2019.06.029 [DOI] [PubMed] [Google Scholar]
  58. Pittig A, Schulz AR, Craske MG, & Alpers GW (2014). Acquisition of Behavioral Avoidance: Task-Irrelevant Conditioned Stimuli Trigger Costly Decisions. Journal of Abnormal Psychology (1965), 123(2), 314–329. 10.1037/a0036136 [DOI] [PubMed] [Google Scholar]
  59. Raes AK, De Raedt R, Verschuere B, & De Houwer J. (2009). Failure to loose fear: The impact of cognitive load and trait anxiety on extinction. Behaviour Research and Therapy, 47(12), 1096–1101. 10.1016/j.brat.2009.08.002 [DOI] [PubMed] [Google Scholar]
  60. Reed D. (2002). Random letter sequence generator. Http://www.davereed.com/Nifty/randSeq.html—GoogleSearch. (n.d.). Retrieved July 14, 2021, from https://www.google.com/search?q=Reed%2C+D.+(2002).+Random+letter+sequence+generator.+http%3A%2F%2Fwww.davereed.com%2FNifty%2FrandSeq.html&rlz=1C1GCEA_enUS924US924&oq=Reed%2C+D.+(2002).+Random+letter+sequence+generator.+http%3A%2F%2Fwww.davereed.com%2FNifty%2FrandSeq.html&aqs=chrome..69i57.311j0j7&sourceid=chrome&ie=UTF-8
  61. Schupp HT, Flaisch T, Stockburger J, & Junghöfer M. (2006). Emotion and attention: Event-related brain potential studies. In Anders S, Ende G, Junghofer M, Kissler J, & Wildgruber D. (Eds.), Progress in Brain Research (Vol. 156, pp. 31–51). Elsevier. 10.1016/S0079-6123(06)56002-9 [DOI] [PubMed] [Google Scholar]
  62. Smith APR, Dolan RJ, & Rugg MD (2004). Event-Related Potential Correlates of the Retrieval of Emotional and Nonemotional Context. Journal of Cognitive Neuroscience, 16(5), 760–775. 10.1162/089892904970816 [DOI] [PubMed] [Google Scholar]
  63. Smith EE, Jonides J, Marshuetz C, & Koeppe RA (1998). Components of verbal working memory: Evidence from neuroimaging. Proceedings of the National Academy of Sciences of the United States of America, 95(3), 876–882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Stefanacci L, & Amaral DG (2002). Some observations on cortical inputs to the macaque monkey amygdala: An anterograde tracing study. Journal of Comparative Neurology, 451(4), 301–323. 10.1002/cne.10339 [DOI] [PubMed] [Google Scholar]
  65. Telch MJ, Valentiner DP, Ilai D, Young PR, Powers MB, & Smits JAJ (2004). Fear activation and distraction during the emotional processing of claustrophobic fear. Journal of Behavior Therapy and Experimental Psychiatry, 35(3), 219–232. 10.1016/j.jbtep.2004.03.004 [DOI] [PubMed] [Google Scholar]
  66. Van der Burg E, Olivers CN, Bronkhorst AW, Koelewijn T, & Theeuwes J. (2007). The absence of an auditory-visual attentional blink is not due to echoic memory. Perception & Psychophysics, 69(7), 1230–1241. 10.3758/bf03193958 [DOI] [PubMed] [Google Scholar]
  67. Van Dillen LF, Heslenfeld DJ, & Koole SL (2009). Tuning down the emotional brain: An fMRI study of the effects of cognitive load on the processing of affective images. NeuroImage, 45(4), 1212–1219. 10.1016/j.neuroimage.2009.01.016 [DOI] [PubMed] [Google Scholar]
  68. Vansteenwegen D, Hermans D, Vervliet B, Francken G, Beckers T, Baeyens F, & Eelen P. (2005). Return of fear in a human differential conditioning paradigm caused by a return to the original acquistion context. Behaviour Research and Therapy, 43(3), 323–336. 10.1016/j.brat.2004.01.001 [DOI] [PubMed] [Google Scholar]
  69. Vytal K, Cornwell B, Arkin N, & Grillon C. (2012). Describing the interplay between anxiety and cognition: From impaired performance under low cognitive load to reduced anxiety under high load. Psychophysiology, 49(6), 842–852. 10.1111/j.1469-8986.2012.01358.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Weisman JS, & Rodebaugh TL (2018). Exposure therapy augmentation: A review and extension of techniques informed by an inhibitory learning approach. Clinical Psychology Review, 59, 41–51. 10.1016/j.cpr.2017.10.010 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1771472-supplement-1.pdf (61.2KB, pdf)

RESOURCES