- Email: [email protected]
d -Cycloserine for the augmentation of an attentional training intervention for trait anxiety
Journal of Anxiety Disorders 24 (2010) 440–445
Contents lists available at ScienceDirect
Journal of Anxiety Disorders
d-Cycloserine for the augmentation of an attentional training intervention for trait anxiety Evelyn Behar a,∗ , R. Kathryn McHugh b , Andrew Peckham b , Michael W. Otto b,1 a b
University of Illinois at Chicago, 1007 W. Harrison St. (M/C 285), Chicago, IL 60607, United States Boston University, United States
a r t i c l e
i n f o
Article history: Received 23 November 2009 Received in revised form 3 February 2010 Accepted 15 February 2010 Keywords: d-Cycloserine Attention training Cognitive Attentional bias Trait anxiety
a b s t r a c t The present study investigates the combination of two novel strategies for the treatment of anxiety that resulted from translational research. We examined whether the putative memory enhancer, dcycloserine (DCS), offered benefit to procedures designed to train attention away from threat. Participants were 44 adults selected on the basis of high trait anxiety. In this randomized study, DCS or placebo was administered 1 h prior to attentional training away from threat using the dot probe task. On the following day, the effectiveness of this training was assessed along with emotional reactivity following two stressful tasks. We found that the addition of DCS resulted in significantly stronger reduction in attentional bias toward threat relative to placebo, but found no additive effects for the DCS condition on subsequent emotional reactivity. These results provide initial support for the efficacy of DCS for augmenting attentional training tasks; potential strategies for enhancing these results are discussed. © 2010 Elsevier Ltd. All rights reserved.
1. Introduction Despite the successes of cognitive-behavioral and pharmacologic interventions for the treatment of anxiety disorders, many patients continue to need additional or alternative treatment (Hofmann & Smits, 2008; Pollack et al., 2008). Two recent translational research successes – attentional training interventions and the use of d-cycloserine (DCS) to augment exposure-based cognitive-behavioral therapy (CBT) – have shown promise as innovative strategies for the treatment of anxiety disorders. Attentional training interventions emerged from basic research demonstrating elevated vigilance toward threatening cues (attentional bias) among individuals with anxiety disorders or sub-syndromal anxiety symptoms (for a review, see Bar-Haim, Pergamin, Bakermans-Kranenberg, & van Ijzendoorn, 2007). In a seminal study, MacLeod, Rutherford, Campbell, Ebsworthy, and Holker (2002) found that training individuals to attend to threatening stimuli led to enhanced emotional reactivity during a subsequent stressful task. This study both provided evidence for a causal effect of vigilance toward threat on emotional reactivity, and support for the possibility of using a training procedure to modify attentional bias. Attentional training consists of a computerized
∗ Corresponding author. E-mail address: [email protected] (E. Behar). 1 Served as a consultant for Organon (Schering-Plough) and Jazz Pharmaceuticals, has received royalties for use of the SIGH-A, and receives research support from Schering-Plough. The other authors have no disclosures to report. 0887-6185/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.janxdis.2010.02.009
task that provides repeated presentations of stimulus pairs (one threatening and one neutral) with one type of stimulus followed more frequently by a target stimulus, thus training early attentional allocation either toward or away from the threatening information. Early applications of this strategy demonstrated that the training procedure modified both attentional bias and emotional reactivity to stress (MacLeod et al., 2002; See, MacLeod, & Bridle, 2009). Several investigators have recently applied this strategy to the treatment of various anxiety conditions by randomly assigning individuals to a condition in which they receive attentional training away from threat stimuli or to a placebo training condition. Results have consistently shown that attentional training is associated with reduced attentional bias and positive effects on diagnostic or behavioral indices of anxiety. For example, Amir, Beard, Burns, and Bomyea (2009) randomly assigned 29 treatmentseeking individuals with GAD to eight sessions of either attention modification or to a placebo training procedure. Results indicated a decrease in both attentional bias toward threat and disorderspecific symptoms for the attentional training group as compared to the placebo training group. Similarly, Hazen, Vasey, and Schmidt (2009) randomly assigned 24 severe worriers to receive five sessions of either attention retraining or to a sham training procedure. Results indicated that compared to the shame training condition, participants in the attention retraining condition evidenced significant reductions in both attentional bias to threat and symptoms of anxiety and depression. Similar results have been reported for the use of attentional training for social anxiety. Amir, Weber, Beard, Bomyea, and Taylor (2008) randomly assigned 94 socially anxious undergraduate stu-
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445
dents to either an attention modification or placebo training procedure that was completed in a single day. They reported that in comparison to participants in the placebo training condition, participants in the attention modification condition evidenced lower levels of anxiety in response to a public speaking challenge, and were rated as having given higher quality speeches. Similarly, Schmidt, Richey, Buckner, and Timpano (2009) randomly assigned 36 individuals with social anxiety disorder to an attention training or control condition, and found that compared to control participants, participants receiving attention training evidenced significantly greater reductions in social anxiety and trait anxiety, and that 72% of participants in the attention training condition no longer met diagnostic criteria for social anxiety disorder (as compared to 11% of participants in the control condition). Together, these results support the hypothesis that attention plays a causal role in anxiety conditions, and provide strong evidence in favor of attention training procedures for the reduction of anxiety symptoms. The mechanism of action of attentional training interventions is not well understood at this early stage of its evaluation. One hypothesis is that these procedures create change through repeated exposure to an attentional contingency that can ameliorate preexisting cognitive processing biases (MacLeod, Koster, & Fox, 2009). With a reduced attentional bias toward threat, the cascade of anxiogenic responses to threat cues is hypothesized to decrease, thereby interrupting the self-perpetuating cycles that characterize the anxiety disorders (see Amir et al., 2009). In a separate advance in translational research, d-cyloserine (DCS), a partial agonist at the NMDA receptor in the amygdala, has been shown to augment extinction learning in animal and human paradigms (see Davis, Myers, Ressler, & Rothbaum, 2005; Richardson, Ledgerwood, & Cranney, 2004). A number of trials support the efficacy of DCS as a way to enhance the therapeutic learning provided by exposure-based CBT (Norberg, Krystal, & Tolin, 2008), with the use of individual doses of DCS taken prior to exposure sessions showing benefit relative to placebo for the treatment of acrophobia (Ressler et al., 2004), social phobia (Guastella et al., 2008; Hofmann et al., 2006), panic disorder (Otto et al., 2010), and obsessive-compulsive disorder (Kushner et al., 2007; Wilhelm et al., 2008).2 DCS appears to exert its beneficial effects by aiding the consolidation of extinction learning (Davis et al., 2005; Richardson et al., 2004), but the generality of these memory effects is not yet clear. DCS has performed poorly in applications of this singledose paradigm to non-emotional memory tasks that do not involve extinction in humans (Otto et al., 2009; Goff et al., 2008). However, animal models have suggested that DCS facilitates other types of memory in tasks that may involve limbic activation, such as spatial memory (Land & Riccio, 1999; Lelong, Dauphin, & Boulouard, 2001) and taste aversion acquisition (Davenport & Houpt, 2009; Nunnink, Davenport, Ortega, & Houpt, 2007). Similarly, in humans, a recent study showed that DCS can facilitate the retention of fear conditioning using mild electric shock as a stimulus (Kalisch et al., 2009). Attentional biases such as those manipulated in the attentional training tasks have been associated with both cortical and subcortical neural activation, including activation of the amygdala (e.g., van den Heuvel et al., 2005). Moreover, at the neuronal level, there is some evidence for a role of NMDA receptors in competitive attention (e.g., Deco & Rolls, 2005). Finally, it has been proposed that DCS may act preferentially on implicit extinction learning rather than on cognitively-mediated processes (Grillon, 2009); hence, the
2 Studies on the effects of DCS in the treatment of OCD are more equivocal than for other anxiety conditions. See, for example, Storch et al. (2007).
441
implicit conditioning of responses away from threat cues, as likely occurs in attentional training (see Van Damme, Crombez, Hermans, Koster, & Eccleston, 2006), may be a mechanism open to DCS modulation. The present study targets the confluence of two lines of translational research. First, the training of attentional bias away from threat has potential as a novel treatment of anxiety disorders and may operate through retraining of associative learning networks that are, at least in part, amygdala-based. Second, in repeated small-scale trials DCS augmentation has shown clinical benefit. It remains unclear whether DCS has therapeutic applications other than extinction learning, although there are encouraging findings for other emotional associative learning paradigms. Accordingly, this investigation has two primary aims: (1) replication of the findings from MacLeod and colleagues (2002) indicating a change in attentional bias and emotional reactivity subsequent to modification of attentional bias, and (2) evaluation of the efficacy of DCS relative to placebo for the augmentation of this intervention. Consistent with previous findings (e.g., Amir et al., 2008), we hypothesized that participants would exhibit reductions in attentional bias and emotional reactivity following the attentional training procedure. In addition, we hypothesized that participants receiving DCS would exhibit greater decreases in attentional bias following training relative to those receiving placebo, and that participants receiving DCS would exhibit greater reductions in emotional reactivity to stressful tasks relative to those receiving placebo. Although many current applications of attentional training utilize multiple sessions (e.g., Amir et al., 2009), we selected a single session in training to be consistent with MacLeod et al. (2002) and to be consistent with the limited number of exposure sessions used in trials of DCS augmentation (e.g., Ressler et al., 2004).
2. Methods 2.1. Participants Adults between the ages of 18–65 were recruited from the greater Boston area through newspaper advertisements, flyers, and internet postings. Potential participants were excluded if they reported being pregnant or nursing, had a history of memory or cognitive impairment, or if they were taking any prescription psychotropic medications or medications contraindicated with DCS. Criteria for inclusion in the study included a score of 50 or higher on the State-Trait Anxiety Inventory-Trait Version (STAIT; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983), which was administered as part of a telephone screen. A score of 50 on the STAI, which is 1.5 standard deviations above the mean for the general population, was selected to reflect an elevated level of trait anxiety and is consistent with recruitment of high trait anxious individuals in previous investigations (e.g., Wilson & MacLeod, 2003). No other evaluation of anxiety or Axis I conditions was conducted during the phone screen or physician evaluation. Eligible participants were evaluated by the prescribing physician prior to engaging in any laboratory procedures. Upon medical approval, participants were invited to the laboratory for two experimental sessions on two consecutive days. Participants were compensated $50 for the completion of all study procedures. Forty-four participants (23 women) provided written informed consent and were enrolled in the study. The mean age of study participants was 44.5 years (SD = 12.4, range = 18–65). Participants in the sample were 45.5% Caucasian, 41% African-American, 7% Hispanic, and 4.5% Asian, with 2% of participants electing to not indicate their race/ethnicity.
442
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445
2.2. Procedure Following provision of informed consent, participants were randomly assigned to receive 50 mg of d-cycloserine (DCS) or 50 mg of pill placebo (PBO); both the participant and the research staff were blind to study condition. Study medication was dispensed by onsite medical staff not otherwise involved in the study procedures. Participants then completed a battery of self-report questionnaires described below. Two stress-induction tasks (described below) were then completed in a randomized order, which was stratified by randomized condition. Prior to and following this stress induction task, participants completed state versions of the STAI (STAI-S) and the Positive and Negative Affect Schedule (PANASNow; Watson, Clark, & Tellegen, 1988) to evaluate current levels of affect. Participants were then administered DCS or pill placebo by the experimenter and were asked to wait for 50–60 min in the clinic waiting room before continuing with study procedures, allowing for sufficient metabolism to ensure peak levels of DCS during and immediately subsequent to the training procedure. Participants then completed the attentional bias assessment and training using procedures similar to the original MacLeod et al. (2002) study as described below. The second experimental session was conducted on the day immediately following the first experimental session at approximately the same time of day. Participants completed the STAI-S and PANAS-Now before and after repeating the two stress induction tasks. Additionally, a final set of attentional bias assessment trials to assess the durability of attentional training effects was conducted. 2.3. Materials 2.3.1. Measures The State-Trait Anxiety Inventory (STAI; Spielberger et al., 1983) was used in both its state (STAI-S) and trait (STAI-T) forms to assess both situational and baseline levels of anxiety. The STAI is a 20item questionnaire in which participants rate their general level of anxiety. Research indicates that the STAI possesses good internal consistency, retest reliability, and validity (Spielberger et al., 1983). The Positive and Negative Affect Schedule (PANAS; Watson et al., 1988) is a brief measure of affect and yields Positive Affect and Negative Affect factors. This self-report measure includes 20 items in which participants rate themselves on words describing general positive or negative affect. The measure was administered in both its “past 24 h” (PANAS-24) and situational (PANAS-Now) forms. The PANAS evidences excellent internal consistency as well as favorable convergent, discriminant, and predictive validities (Watson et al., 1988).
mean response time to threatening stimuli from the mean response time to neutral stimuli; thus, a positive score reflects quicker attentional allocation to threatening stimuli. On Day 1, participants first completed 12 practice trials (in which only neutral-neutral pairs were presented), followed by 96 assessment trials to measure baseline attentional bias. Participants then completed 672 training trials. In training trials, the neutral stimulus was replaced by the dot probe on 92% of trials, similar to the procedures used by MacLeod et al. (2002). On Day 2, an additional 96 assessment trials (consisting of novel word pairs) were administered to test attentional bias 24 h after the training phase. 2.3.3. Stress induction tasks All participants completed two stress induction tasks, presented in counterbalanced order, on both days of the experiment. In the Anagram Stress Task (originally used by Mogg, Mathews, Bird, & Macgregor-Morris, 1990), participants were given a list of 28 extremely difficult anagrams and told to complete as many as possible within 3 min. Upon completion of the task, regardless of the participant’s actual performance, the experimenter gave the participant negative feedback. Participants were then told that they could use up to an additional 4 min to continue working on the anagrams, but that they could stop at any point. The experimenter recorded the total amount of time each participant worked on the anagram task (3 min plus any additional time used) as a measure of tolerance to stress. Participants also completed the Computerized Mirror-Tracing Persistence Task (MTPT-C; Strong et al., 2003), a computerized tracing task in which all visual feedback is reversed (as in a mirror). In this task, participants use the computer mouse to move a red dot along a shape on the monitor. When a mistake is made, the computer emits a loud and noxious buzzing sound. There are three shapes of increasing difficulty, and on the last shape, participants are instructed to discontinue the task at any point by pressing a computer key (with a maximum of 5 min). To provide motivation for participants to engage in the tasks, participants were told that they would be entered in a lottery to win another $40 if they performed in the top 25% on these tasks, consistent with prior investigations (Brown, Lejuez, Kahler, & Strong, 2002). Immediately prior to and following the mirror tracing task, participants were asked to self report their levels of irritability, frustration, anxiety, and difficulty concentrating in order to provide an index of their emotional reactivity to this stressor. 3. Results 3.1. Data reduction
2.3.2. Dot probe paradigm The dot probe paradigm (MacLeod, Mathews, & Tata, 1986) is a method for assessing visual attentional bias that utilizes response time as an index of relative attentional allocation to stimuli. Recently, this procedure has been adapted for use also as an intervention to alter attentional biases (e.g., Amir et al., 2009; MacLeod et al., 2002). Dot probe assessment and training procedures, including stimuli, were adapted from those used by MacLeod and colleagues (2002). All dot probe trials involved two verbal stimuli – one threatening (e.g., assault) and one neutral (e.g., bottles) – presented one above the other. A fixation cross (+) cueing participants to allocate attention to the center of the computer screen was presented for 500 ms, followed by the stimuli for 500 ms. Following the disappearance of the stimuli, one was replaced by a small dot. Participants were asked to identify the location of the dot as replacing the top word or the bottom word using a keyboard. The inter-trial interval was 500 ms. Attentional bias scores were calculated by subtracting the
Consistent with past evaluations of attentional bias (e.g., Beevers, Gibb, McGeary, & Miller, 2007), any individual trial for which a participant responded incorrectly or responded with an exceptionally low (<150 ms) or high (>1500 ms) reaction time (thus likely reflecting a different stage of attention) were excluded from calculation of individual means. Any participant for whom >10% of trials were excluded based on these criteria was excluded from overall data analyses for the study (n = 6). This degree of invalid data was judged to be reflective of a likely failure of the participant to sufficiently engage in the training (i.e., failed manipulation), possibly resulting in a questionably valid measure of attentional bias. In addition, we excluded from all analyses one participant for whom a computer malfunction resulted in a failure to record dot probe data, and three participants who only completed Day 1 of the study procedures. Prior to analysis, the data were examined for fit between the distributions of the self-report questionnaires (STAI-T, STAI-S, PANAS),
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445
443
Table 1 Dependent variables at Day 1 and Day 2. Day 1
STAI-T total at baseline PANAS—positive affect in the last 24 h PANAS—negative affect in the last 24 h Attentional bias during dot probe procedure (milliseconds) Number of seconds to discontinue Anagram task Number of seconds to discontinue Mirror tracing task Dysphoria following mirror tracing task Change in state anxiety from pre- to post-stressors Change in Negative affect from pre- to post-stressors
Day 2
DCS
PBO
56.21 (7.44) 25.67 (8.60) 24.39 (8.51) 11.48 (18.35) 359.30 (63.01) 227.60 (162.90) 183.06 (100.21) +3.97 (7.20) +5.28 (6.41)
56.07 (7.49) 21.60 (9.35) 19.47 (6.12) 9.98 (33.44) 330.00 (112.95) 125.27 (160.96) 201.27 (131.52) +6.07 (7.08) +5.93(5.99)
DCS
PBO – – – −.37 (15.30) 320.15 (96.86) 171.63 (151.04) 184.60 (80.56) +2.68 (5.56) −4.94 (13.62)
– – – 2.64 (24.32) 301.00 (102.00) 147.33 (163.50) 139.53 (123.07) +5.15 (11.34) −4.14 (14.57)
DCS = d-cycloserine group, PBO = placebo group. No between-group differences were found at baseline for these measures.
and reaction time data and the assumptions of normality. Any value exceeding a z-score of 3.3 was considered to be a univariate outlier and was omitted from analyses. No outliers were found in the self-report data. Two participants (both in the PBO condition) evidenced outliers on the dot probe reaction time data and were thus excluded from all analyses. For the mirror tracing emotional reactivity data, composite dysphoria scores were created by summing participants’ 0–100 ratings of irritability, frustration, anxiety, and difficulty concentrating as measured immediate prior to and immediately following that task. Thus, dysphoria scores ranged from 0 to 400. This dysphoria scale has been utilized in prior studies of emotional reactivity during the mirror tracing task (e.g., Daughters et al., 2005). 3.2. Preliminary analyses Preliminary analyses also sought to ensure that levels of selfreported symptoms and affect did not differ between the two experimental conditions at baseline on Day 1 of the study. A series of one-way t-tests indicated that the DCS and PBO groups did not differ on baseline levels of trait anxiety (t[32] = 0.06 ns), positive affect (t[31] = 1.30 ns), or negative affect (t[31] = 1.87 ns) over the 24 h prior to the experimental session. Similarly, we sought to ensure that participants in the two groups did not differ in their reactivity to study procedures and stressors prior to the administration of the study pill. Indeed, participants in the two groups did not differ with respect to attentional bias during the first dot-probe procedure (t[32] = 0.17 ns), level of dysphoria in reaction to the mirror tracing task (t[30] = −0.44 ns), number of seconds to discontinue the mirror tracing task (t[30] = 1.78 ns), or number of seconds to discontinue the anagram task (t[30] = 0.91 ns). Levels of these variables on both days of the study are reported in Table 1. Finally, we conducted manipulation checks to ensure that emotional reactivity did in fact increase in reaction to the stressors employed at baseline (prior to any training taking place and prior to pill administration) and that this reactivity did not differ between DCS and PBO groups. We employed a 2 (Phase: Pre-stressor, Poststressor) × 2 (Drug Condition: DCS, PBO) mixed model ANOVA on self-reported state anxiety and negative affect scores prior to attentional training. For the state anxiety analysis, although there was no effect of Drug Condition, there was a main effect of Phase such that state anxiety was higher following the stressors (M = 56.29, SD = 9.01) than it was prior to the stressors (M = 51.40, SD = 10.19), F(1,32) = 16.55, p < .001. For the negative affect analysis, although there was likewise no effect of Drug Condition, there was a main effect of Phase such that NA was higher following the stressors (M = 23.73, SD = 8.73) than it was prior to the stressors (M = 18.15, SD = 8.43), F(1,31) = 26.55, p < .001. Thus, as expected, the stress tasks were successful in eliciting emotional reactivity in individuals at baseline, and random assignment to condition was successful in producing two groups
Fig. 1. Change in attentional bias by drug condition.
who did not differ in the degree of that reactivity prior to experimental manipulations being instituted. 3.3. Attentional bias We first examined the effectiveness of the attention training procedure in reducing attentional biases in the full sample of participants. A paired samples t-test revealed that attentional bias scores did in fact decrease across the two days of the experiment, t(31) = 2.00, p = .05. Next, to compare the effects of DCS and PBO on change in attentional bias, we conducted a one-way (Drug Condition) ANOVA on attentional bias change scores from Day 1 to Day 2 of the experiment. Results indicated a significant effect of Drug Condition, indicating that participants in the DCS group evidenced a significantly greater decrease in attentional bias towards threatening information relative to participants in the placebo condition, F(1,31) = 5.18, p < .05 (see Fig. 1). This Drug Condition effect (DCS vs. placebo) over time evidenced a large effect size (d = 0.82) according to Cohen’s (1988) standards. 3.4. Distress tolerance Similar analyses were conducted to examine across-day and between-group differences in the number of seconds participants engaged in the anagram and mirror-tracing stressors before discontinuing those tasks. For the anagram task, there was a significant
444
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445
decrease in the total amount of time (in seconds) study participants tolerated the task from Day 1 (M = 345.08, SD = 86.31) to Day 2 (M = 305.26, SD = 98.54) of the study, F(1,27) = 7.88, p < .01. However, there was no effect of Drug Condition on number of seconds to discontinue the task, F(1,27) = 0.13 ns. For the mirror-tracing task, results did not indicate a significant change in the amount of time study participants tolerated the task from Day 1 (M = 160.96, SD = 151.32) to Day 2 (M = 159.48, SD = 155.15) of the study, F(1,28) = 0.003 ns. Results also indicated no effect of Drug Condition on number of seconds to discontinue the task, F(1,28) = 0.87 ns. 3.5. Emotional reactivity We also examined across-day and between-group differences in emotional reactivity to the stress tasks. For the mirror tracing task, dysphoria scores did not change significantly from Day 1 (M = 189.90, SD = 117.48) to Day 2 (M = 162.07, SD = 104.74) of the study, t(30) = 1.84, p = .08. A one-way ANOVA further showed that there was no effect of Drug Condition on across-day change in dysphoria between the two conditions, F(1,30) = 3.40, p = .08. We also examined affective reactivity to the two stressors as measured by the change in state anxiety and negative affect from pre- to post-stressor tasks. For the state anxiety analyses, change in state anxiety from pre- to post-stressors did not change significantly from Day 1 (M = +4.64, SD = 7.19) to Day 2 (M = +3.69, SD = 8.32) of the study, t(31) = 0.74 ns. Additionally, a one-way ANOVA indicated no effect of Drug Condition on pre- to poststressors change from Day 1 to Day 2 of the study, F(1,30) = 3.40 ns. For the negative affect analyses, change in NA from pre- to poststressors significantly decreased from Day 1 (M = +5.69, SD = 6.20) to Day 2 (M = −4.59, SD = 13.81) of the study, t(31) = 3.58, p < .01. However, a one-way ANOVA indicated no effect of Drug Condition on pre- to post-stressors change from Day 1 to Day 2 of the study, F(1,30) = 3.40 ns. 4. Discussion This study provides preliminary evidence for a facilitating effect of DCS on an attentional training intervention for individuals with elevated trait anxiety. In our sample of individuals who were high in trait anxiety and who completed only a single session of attentional training, DCS administration led to significantly greater training efficacy. Although all participants evidenced a decrease in attentional bias from Day 1 to Day 2 of the experiment, this decrease was greater among individuals who took DCS relative to those who took a placebo, and this difference evidenced a large effect size. This differential effect on attentional bias, however, was not associated with differential tolerance for stressful tasks or differential reactivity on the emotional stress tasks, although the results indicated a reduction in emotional reactivity over time for some (negative affect) but not all (state anxiety) measures. The absence of effects for the emotional tasks may be attributable to the single session of the attentional training intervention utilized in this analogue study. Other studies examining the effects of attention training on affective reactivity have found affective shifts; however, these studies were characterized by the use of a no-training comparison (Amir et al., 2008) or utilized conditions comparing attentional training toward threat to attentional training away from threat (MacLeod et al., 2002), which may have increased the relative effect size of the intervention. In our study design we selected a single dose of attentional training to avoid potential ceiling effects from repeated training trials, as per the designs of early studies of exposure augmentation using DCS (e.g., Ressler et al., 2004). Furthermore, this investigation uti-
lized two nomothetic stressors that were selected for their ability to elicit frustration, anxiety, and irritability in participants, given that the study sample (individuals with high trait anxiety) is not characterized by the presence of specific feared stimuli. The questionable degree to which these tasks were able to elicit consistently strong emotional reactions among participants may have limited our ability to find results for state anxiety measures. We anticipate that both stronger attentional training results, and affective consequences of this training, may be achieved with additional sessions of training and/or with more ideographic stressors that are more likely to elicit strong emotional responses. One pragmatic question that our results raise is whether the reduction in attentional bias resulting from the training procedure is worth the added effort that is entailed in prescribing DCS. Indeed, several investigations have found enhanced effects of DCS in the treatment of various anxiety conditions (e.g., Wilhelm et al., 2008), but the provision of DCS in those studies and in our study requires the need for physiological and/or psychiatric evaluations, the involvement of a prescribing physician, and potentially the use of additional personnel. Although additional research certainly needs to be conducted before one can determine the relative costs and benefits of administering DCS in conjunction with attention training or any other cognitive-behavioral intervention, our results provide evidence in favor of DCS’s potential therapeutic applications to attention-based interventions such as this one. Limitations of our study include the deletion of data from several participants because of particularly poor quality of dot probe data, likely reflecting a failure to engage in study procedures. Also, the small sample size limited our ability to detect other than large effect sizes for the experimental manipulations. Replication in larger samples is an important area for future research. In summary, this study provides preliminary support for the combination of two novel intervention strategies for anxiety disorders, attentional training and DCS. This finding is of particular importance given the continued need for strategies to improve the rates of treatment response in anxiety disorders, and the need to investigate the limits of the DCS augmentation effects (e.g., Otto et al., 2009). Future research is needed to replicate these results and to extend them to clinical samples. References Amir, N., Beard, C., Burns, M., & Bomyea, J. (2009). Attention modification program in individuals with generalized anxiety disorder. Journal of Abnormal Psychology, 118, 28–33. Amir, N., Weber, G., Beard, C., Bomyea, J., & Taylor, C. T. (2008). The effect of a single-session attention modification program on response to a public-speaking challenge in socially anxious individuals. Journal of Abnormal Psychology, 117, 860–868. Bar-Haim, Y., Pergamin, L., 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, 1–24. Beevers, C. G., Gibb, B. E., McGeary, J. E., & Miller, I. W. (2007). Serotonin transporter genetic variation and biased attention for emotional word stimuli among psychiatric inpatients. Journal of Abnormal Psychology, 116, 208–212. Brown, R. A., Lejuez, C. W., Kahler, C. W., & Strong, D. R. (2002). Distress tolerance and duration of past smoking cessation attempts. Journal of Abnormal Psychology, 111, 180–185. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates. Daughters, S. B., Lejuez, C. W., Bornovalova, M. A., Kahler, C. W., Strong, D. R., & Brown, R. A. (2005). Distress tolerance as a predictor of early treatment dropout in a residential substance abuse treatment facility. Journal of Abnormal Psychology, 114, 729–734. Davenport, R. A., & Houpt, T. A. (2009). d-cycloserine enhances short-delay, but not long-delay, conditioned taste aversion learning in rats. Pharmacology Biochemistry and Behavior, 91, 596–603. Davis, M., Myers, K. M., Ressler, K. J., & Rothbaum, B. O. (2005). Facilitation of extinction of conditioned fear by d-cycloserine: Implications for psychotherapy. Current Directions in Psychological Science, 14, 214–219. Deco, G., & Rolls, E. T. (2005). Neurodynamics of biased competition and cooperation for attention: A model with spiking neurons. Journal of Neurophysiology, 94, 295–313.
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445 Goff, D. C., Cather, C., Gottlieb, J. D., Evins, A. E., Walsh, J., Raeke, L., et al. (2008). Once-weekly d-cycloserine effects on negative symptoms and cognition in schizophrenia: An exploratory study. Schizophrenia Research, 106, 320–327. Grillon, C. (2009). d-Cycloserine facilitation of fear extinction and exposure-based therapy might rely on lower-level, automatic mechanisms. Biological Psychiatry, 66, 1227–1234. Guastella, A. J., Richardson, R., Lovibond, P. F., Rapee, R. M., Gaston, J. E., Mitchell, P., et al. (2008). A randomized controlled trial of d-cycloserine enhancement of exposure therapy for social anxiety disorder. Biological Psychiatry, 63, 544–549. Hazen, R. A., Vasey, M. W., & Schmidt, N. B. (2009). Attentional retraining: A randomized clinical trial for pathological worry. Journal of Psychiatric Research, 43, 627–633. Hofmann, S. G., Meuret, A. E., Smits, J. A. J., Simon, N. M., Pollack, M. H., Eisenmenger, K., Shiekh, M., & Otto, M. W. (2006). Augmentation of exposure therapy with d-cycloserine for social anxiety disorder. Archives of General Psychiatry, 63, 298–304. Hofmann, S. G., & Smits, J. A. (2008). Cognitive-behavioral therapy for adult anxiety disorders: A meta-analysis of randomized placebo-controlled trials. Journal of Clinical Psychiatry, 69, 621–632. Kalisch, R., Holt, B., Petrovic, P., De Martino, B., Klöppel, S., Büchel, C., & Dolan, R. J. (2009). The NMDA agonist d-cycloserine facilitates fear memory consolidation in humans. Cerebral Cortex, 19, 187–196. Kushner, M. G., Kim, S. W., Donahue, C., Thuras, P., Adson, D., Kotlyar, M., et al. (2007). d-Cycloserine augmented exposure therapy for obsessive-compulsive disorder. Biological Psychiatry, 62, 835–838. Land, C., & Riccio, D. (1999). d-Cycloserine: Effects on long-term retention of a conditioned response and on memory for contextual attributes. Neurobiology Learning and Memory, 72, 158–168. Lelong, V., Dauphin, F., & Boulouard, M. (2001). RS 67333 and d-cycloserine accelerate learning acquisition in the rat. Neuropharmacology, 41, 517–522. MacLeod, C., Koster, E. H. W., & Fox, E. (2009). Whither cognitive bias modification research? Commentary on the special section articles. Journal of Abnormal Psychology, 118, 89–99. MacLeod, C., Mathews, A., & Tata, P. (1986). Attentional bias in emotional disorders. Journal of Abnormal Psychology, 95, 15–20. MacLeod, C., Rutherford, E., Campbell, L., Ebsworthy, G., & Holker, L. (2002). Selective attention and emotional vulnerability: Assessing the causal basis of their association through the experimental manipulation of attentional bias. Journal of Abnormal Psychology, 111, 107–123. Mogg, K., Mathews, A., Bird, C., & Macgregor-Morris, R. (1990). Effects of stress and anxiety on the processing of threat stimuli. Journal of Personality and Social Psychology, 59, 1230–1237. Norberg, M. M., Krystal, J. H., & Tolin, D. F. (2008). A meta-analysis of d-cycloserine and the facilitation of fear extinction and exposure therapy. Biological Psychiatry, 63, 1118–1126. Nunnink, M., Davenport, R. A., Ortega, B., & Houpt, T. A. (2007). d-Cycloserine enhances conditioned taste aversion learning in rats. Pharmacology Biochemistry and Behavior, 87, 321–330.
445
Otto, M. W., Basden, S. L., McHugh, R. K., Kantak, K. M., Deckersbach, T., Cather, C., et al. (2009). Effects of d-cycloserine administration on weekly nonemotional memory tasks in healthy participants. Psychotherapy and Psychosomatics, 78, 49–54. Otto, M. W., Tolin, D. F., Simon, N. M., Pearlson, G. D., Basden, S., Meunier, S. A., et al. (2010). Efficacy of D-Cycloserine for enhancing response to cognitive-behavior therapy for panic disorder. Biological Psychiatry, 67, 365–370. Pollack, M. H., Otto, M. W., Roy-Byrne, P. P., Coplan, J. D., Rothbaum, B. O., Simon, N. M., et al. (2008). Novel treatment approaches for refractory anxiety disorders. Depression and Anxiety, 25, 467–476. Ressler, K. J., Rothbaum, B. O., Tannenbaum, L., Anderson, P., Graap, K., Zimand, E., Hodges, L., & Davis, M. (2004). Cognitive enhancers as adjuncts to psychotherapy: Use of d-cycloserine in phobics to facilitate extinction of fear. Archives of General Psychiatry, 1136–1144. Richardson, R., Ledgerwood, L., & Cranney, J. (2004). Facilitation of fear extinction by d-cycloserine: Theoretical and clinical implications. Learning and Memory, 11, 510–516. See, J., MacLeod, C., & Bridle, R. (2009). The reduction of anxiety vulnerability through the modification of attentional bias: A real-world study using a home-based cognitive bias modification procedure. Journal of Abnormal Psychology, 118, 65–75. Schmidt, N. B., Richey, J. A., Buckner, J. D., & Timpano, K. R. (2009). Attention training for generalized social anxiety disorder. Journal of Abnormal Psychology, 118, 5–14. Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychologists Press. Storch, E. A., Merlo, L. J., Bengtson, M., Murphy, T. K., Lewis, M. H., Yang, M. C., Jacob, M. L., Larson, M., Hirsh, A., Fernandez, M., Geffken, G. T., & Goodman, W. K. (2007). d-Cycloserine does not enhance exposure-response prevention therapy in obsessive-compulsive disorder. International Clinical Psychopharmacology, 22, 230–237. Strong, D. R., Lejuez, C. W., Daughters, S. B., Marinello, M., Kahler, C. W., & Brown, R. A. (2003). The Computerized Mirror Tracing Task Version 1. Unpublished manuscript. Van Damme, S., Crombez, G., Hermans, D., Koster, E. H. W., & Eccleston, C. (2006). The role of extinction and reinstatement in attentional bias to threat: A conditioning approach. Behaviour Research and Therapy, 44, 1555–1563. van den Heuvel, O. A., Veltman, D. J., Groenewegen, H. J., Witter, M. P., Merkelbach, J., Cath, D. C., et al. (2005). Disorder-specific neuroanatomical correlates of attentional bias in obsessive-compulsive disorder, panic disorder, and hypochondriasis. Archives of General Psychiatry, 62, 922–933. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. Wilhelm, S., Buhlmann, U., Tolin, D. F., Meunier, S. A., Pearlson, G. D., Reese, H. E., et al. (2008). Augmentation of behavior therapy with d-cycloserine for obsessivecompulsive disorder. American Journal of Psychiatry, 165, 335–341. Wilson, E., & MacLeod, C. (2003). Contrasting two accounts of anxiety-linked attentional bias: Selective attention to varying levels of stimulus threat intensity. Journal of Abnormal Psychology, 112, 212–218.
Contents lists available at ScienceDirect
Journal of Anxiety Disorders
d-Cycloserine for the augmentation of an attentional training intervention for trait anxiety Evelyn Behar a,∗ , R. Kathryn McHugh b , Andrew Peckham b , Michael W. Otto b,1 a b
University of Illinois at Chicago, 1007 W. Harrison St. (M/C 285), Chicago, IL 60607, United States Boston University, United States
a r t i c l e
i n f o
Article history: Received 23 November 2009 Received in revised form 3 February 2010 Accepted 15 February 2010 Keywords: d-Cycloserine Attention training Cognitive Attentional bias Trait anxiety
a b s t r a c t The present study investigates the combination of two novel strategies for the treatment of anxiety that resulted from translational research. We examined whether the putative memory enhancer, dcycloserine (DCS), offered benefit to procedures designed to train attention away from threat. Participants were 44 adults selected on the basis of high trait anxiety. In this randomized study, DCS or placebo was administered 1 h prior to attentional training away from threat using the dot probe task. On the following day, the effectiveness of this training was assessed along with emotional reactivity following two stressful tasks. We found that the addition of DCS resulted in significantly stronger reduction in attentional bias toward threat relative to placebo, but found no additive effects for the DCS condition on subsequent emotional reactivity. These results provide initial support for the efficacy of DCS for augmenting attentional training tasks; potential strategies for enhancing these results are discussed. © 2010 Elsevier Ltd. All rights reserved.
1. Introduction Despite the successes of cognitive-behavioral and pharmacologic interventions for the treatment of anxiety disorders, many patients continue to need additional or alternative treatment (Hofmann & Smits, 2008; Pollack et al., 2008). Two recent translational research successes – attentional training interventions and the use of d-cycloserine (DCS) to augment exposure-based cognitive-behavioral therapy (CBT) – have shown promise as innovative strategies for the treatment of anxiety disorders. Attentional training interventions emerged from basic research demonstrating elevated vigilance toward threatening cues (attentional bias) among individuals with anxiety disorders or sub-syndromal anxiety symptoms (for a review, see Bar-Haim, Pergamin, Bakermans-Kranenberg, & van Ijzendoorn, 2007). In a seminal study, MacLeod, Rutherford, Campbell, Ebsworthy, and Holker (2002) found that training individuals to attend to threatening stimuli led to enhanced emotional reactivity during a subsequent stressful task. This study both provided evidence for a causal effect of vigilance toward threat on emotional reactivity, and support for the possibility of using a training procedure to modify attentional bias. Attentional training consists of a computerized
∗ Corresponding author. E-mail address: [email protected] (E. Behar). 1 Served as a consultant for Organon (Schering-Plough) and Jazz Pharmaceuticals, has received royalties for use of the SIGH-A, and receives research support from Schering-Plough. The other authors have no disclosures to report. 0887-6185/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.janxdis.2010.02.009
task that provides repeated presentations of stimulus pairs (one threatening and one neutral) with one type of stimulus followed more frequently by a target stimulus, thus training early attentional allocation either toward or away from the threatening information. Early applications of this strategy demonstrated that the training procedure modified both attentional bias and emotional reactivity to stress (MacLeod et al., 2002; See, MacLeod, & Bridle, 2009). Several investigators have recently applied this strategy to the treatment of various anxiety conditions by randomly assigning individuals to a condition in which they receive attentional training away from threat stimuli or to a placebo training condition. Results have consistently shown that attentional training is associated with reduced attentional bias and positive effects on diagnostic or behavioral indices of anxiety. For example, Amir, Beard, Burns, and Bomyea (2009) randomly assigned 29 treatmentseeking individuals with GAD to eight sessions of either attention modification or to a placebo training procedure. Results indicated a decrease in both attentional bias toward threat and disorderspecific symptoms for the attentional training group as compared to the placebo training group. Similarly, Hazen, Vasey, and Schmidt (2009) randomly assigned 24 severe worriers to receive five sessions of either attention retraining or to a sham training procedure. Results indicated that compared to the shame training condition, participants in the attention retraining condition evidenced significant reductions in both attentional bias to threat and symptoms of anxiety and depression. Similar results have been reported for the use of attentional training for social anxiety. Amir, Weber, Beard, Bomyea, and Taylor (2008) randomly assigned 94 socially anxious undergraduate stu-
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445
dents to either an attention modification or placebo training procedure that was completed in a single day. They reported that in comparison to participants in the placebo training condition, participants in the attention modification condition evidenced lower levels of anxiety in response to a public speaking challenge, and were rated as having given higher quality speeches. Similarly, Schmidt, Richey, Buckner, and Timpano (2009) randomly assigned 36 individuals with social anxiety disorder to an attention training or control condition, and found that compared to control participants, participants receiving attention training evidenced significantly greater reductions in social anxiety and trait anxiety, and that 72% of participants in the attention training condition no longer met diagnostic criteria for social anxiety disorder (as compared to 11% of participants in the control condition). Together, these results support the hypothesis that attention plays a causal role in anxiety conditions, and provide strong evidence in favor of attention training procedures for the reduction of anxiety symptoms. The mechanism of action of attentional training interventions is not well understood at this early stage of its evaluation. One hypothesis is that these procedures create change through repeated exposure to an attentional contingency that can ameliorate preexisting cognitive processing biases (MacLeod, Koster, & Fox, 2009). With a reduced attentional bias toward threat, the cascade of anxiogenic responses to threat cues is hypothesized to decrease, thereby interrupting the self-perpetuating cycles that characterize the anxiety disorders (see Amir et al., 2009). In a separate advance in translational research, d-cyloserine (DCS), a partial agonist at the NMDA receptor in the amygdala, has been shown to augment extinction learning in animal and human paradigms (see Davis, Myers, Ressler, & Rothbaum, 2005; Richardson, Ledgerwood, & Cranney, 2004). A number of trials support the efficacy of DCS as a way to enhance the therapeutic learning provided by exposure-based CBT (Norberg, Krystal, & Tolin, 2008), with the use of individual doses of DCS taken prior to exposure sessions showing benefit relative to placebo for the treatment of acrophobia (Ressler et al., 2004), social phobia (Guastella et al., 2008; Hofmann et al., 2006), panic disorder (Otto et al., 2010), and obsessive-compulsive disorder (Kushner et al., 2007; Wilhelm et al., 2008).2 DCS appears to exert its beneficial effects by aiding the consolidation of extinction learning (Davis et al., 2005; Richardson et al., 2004), but the generality of these memory effects is not yet clear. DCS has performed poorly in applications of this singledose paradigm to non-emotional memory tasks that do not involve extinction in humans (Otto et al., 2009; Goff et al., 2008). However, animal models have suggested that DCS facilitates other types of memory in tasks that may involve limbic activation, such as spatial memory (Land & Riccio, 1999; Lelong, Dauphin, & Boulouard, 2001) and taste aversion acquisition (Davenport & Houpt, 2009; Nunnink, Davenport, Ortega, & Houpt, 2007). Similarly, in humans, a recent study showed that DCS can facilitate the retention of fear conditioning using mild electric shock as a stimulus (Kalisch et al., 2009). Attentional biases such as those manipulated in the attentional training tasks have been associated with both cortical and subcortical neural activation, including activation of the amygdala (e.g., van den Heuvel et al., 2005). Moreover, at the neuronal level, there is some evidence for a role of NMDA receptors in competitive attention (e.g., Deco & Rolls, 2005). Finally, it has been proposed that DCS may act preferentially on implicit extinction learning rather than on cognitively-mediated processes (Grillon, 2009); hence, the
2 Studies on the effects of DCS in the treatment of OCD are more equivocal than for other anxiety conditions. See, for example, Storch et al. (2007).
441
implicit conditioning of responses away from threat cues, as likely occurs in attentional training (see Van Damme, Crombez, Hermans, Koster, & Eccleston, 2006), may be a mechanism open to DCS modulation. The present study targets the confluence of two lines of translational research. First, the training of attentional bias away from threat has potential as a novel treatment of anxiety disorders and may operate through retraining of associative learning networks that are, at least in part, amygdala-based. Second, in repeated small-scale trials DCS augmentation has shown clinical benefit. It remains unclear whether DCS has therapeutic applications other than extinction learning, although there are encouraging findings for other emotional associative learning paradigms. Accordingly, this investigation has two primary aims: (1) replication of the findings from MacLeod and colleagues (2002) indicating a change in attentional bias and emotional reactivity subsequent to modification of attentional bias, and (2) evaluation of the efficacy of DCS relative to placebo for the augmentation of this intervention. Consistent with previous findings (e.g., Amir et al., 2008), we hypothesized that participants would exhibit reductions in attentional bias and emotional reactivity following the attentional training procedure. In addition, we hypothesized that participants receiving DCS would exhibit greater decreases in attentional bias following training relative to those receiving placebo, and that participants receiving DCS would exhibit greater reductions in emotional reactivity to stressful tasks relative to those receiving placebo. Although many current applications of attentional training utilize multiple sessions (e.g., Amir et al., 2009), we selected a single session in training to be consistent with MacLeod et al. (2002) and to be consistent with the limited number of exposure sessions used in trials of DCS augmentation (e.g., Ressler et al., 2004).
2. Methods 2.1. Participants Adults between the ages of 18–65 were recruited from the greater Boston area through newspaper advertisements, flyers, and internet postings. Potential participants were excluded if they reported being pregnant or nursing, had a history of memory or cognitive impairment, or if they were taking any prescription psychotropic medications or medications contraindicated with DCS. Criteria for inclusion in the study included a score of 50 or higher on the State-Trait Anxiety Inventory-Trait Version (STAIT; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983), which was administered as part of a telephone screen. A score of 50 on the STAI, which is 1.5 standard deviations above the mean for the general population, was selected to reflect an elevated level of trait anxiety and is consistent with recruitment of high trait anxious individuals in previous investigations (e.g., Wilson & MacLeod, 2003). No other evaluation of anxiety or Axis I conditions was conducted during the phone screen or physician evaluation. Eligible participants were evaluated by the prescribing physician prior to engaging in any laboratory procedures. Upon medical approval, participants were invited to the laboratory for two experimental sessions on two consecutive days. Participants were compensated $50 for the completion of all study procedures. Forty-four participants (23 women) provided written informed consent and were enrolled in the study. The mean age of study participants was 44.5 years (SD = 12.4, range = 18–65). Participants in the sample were 45.5% Caucasian, 41% African-American, 7% Hispanic, and 4.5% Asian, with 2% of participants electing to not indicate their race/ethnicity.
442
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445
2.2. Procedure Following provision of informed consent, participants were randomly assigned to receive 50 mg of d-cycloserine (DCS) or 50 mg of pill placebo (PBO); both the participant and the research staff were blind to study condition. Study medication was dispensed by onsite medical staff not otherwise involved in the study procedures. Participants then completed a battery of self-report questionnaires described below. Two stress-induction tasks (described below) were then completed in a randomized order, which was stratified by randomized condition. Prior to and following this stress induction task, participants completed state versions of the STAI (STAI-S) and the Positive and Negative Affect Schedule (PANASNow; Watson, Clark, & Tellegen, 1988) to evaluate current levels of affect. Participants were then administered DCS or pill placebo by the experimenter and were asked to wait for 50–60 min in the clinic waiting room before continuing with study procedures, allowing for sufficient metabolism to ensure peak levels of DCS during and immediately subsequent to the training procedure. Participants then completed the attentional bias assessment and training using procedures similar to the original MacLeod et al. (2002) study as described below. The second experimental session was conducted on the day immediately following the first experimental session at approximately the same time of day. Participants completed the STAI-S and PANAS-Now before and after repeating the two stress induction tasks. Additionally, a final set of attentional bias assessment trials to assess the durability of attentional training effects was conducted. 2.3. Materials 2.3.1. Measures The State-Trait Anxiety Inventory (STAI; Spielberger et al., 1983) was used in both its state (STAI-S) and trait (STAI-T) forms to assess both situational and baseline levels of anxiety. The STAI is a 20item questionnaire in which participants rate their general level of anxiety. Research indicates that the STAI possesses good internal consistency, retest reliability, and validity (Spielberger et al., 1983). The Positive and Negative Affect Schedule (PANAS; Watson et al., 1988) is a brief measure of affect and yields Positive Affect and Negative Affect factors. This self-report measure includes 20 items in which participants rate themselves on words describing general positive or negative affect. The measure was administered in both its “past 24 h” (PANAS-24) and situational (PANAS-Now) forms. The PANAS evidences excellent internal consistency as well as favorable convergent, discriminant, and predictive validities (Watson et al., 1988).
mean response time to threatening stimuli from the mean response time to neutral stimuli; thus, a positive score reflects quicker attentional allocation to threatening stimuli. On Day 1, participants first completed 12 practice trials (in which only neutral-neutral pairs were presented), followed by 96 assessment trials to measure baseline attentional bias. Participants then completed 672 training trials. In training trials, the neutral stimulus was replaced by the dot probe on 92% of trials, similar to the procedures used by MacLeod et al. (2002). On Day 2, an additional 96 assessment trials (consisting of novel word pairs) were administered to test attentional bias 24 h after the training phase. 2.3.3. Stress induction tasks All participants completed two stress induction tasks, presented in counterbalanced order, on both days of the experiment. In the Anagram Stress Task (originally used by Mogg, Mathews, Bird, & Macgregor-Morris, 1990), participants were given a list of 28 extremely difficult anagrams and told to complete as many as possible within 3 min. Upon completion of the task, regardless of the participant’s actual performance, the experimenter gave the participant negative feedback. Participants were then told that they could use up to an additional 4 min to continue working on the anagrams, but that they could stop at any point. The experimenter recorded the total amount of time each participant worked on the anagram task (3 min plus any additional time used) as a measure of tolerance to stress. Participants also completed the Computerized Mirror-Tracing Persistence Task (MTPT-C; Strong et al., 2003), a computerized tracing task in which all visual feedback is reversed (as in a mirror). In this task, participants use the computer mouse to move a red dot along a shape on the monitor. When a mistake is made, the computer emits a loud and noxious buzzing sound. There are three shapes of increasing difficulty, and on the last shape, participants are instructed to discontinue the task at any point by pressing a computer key (with a maximum of 5 min). To provide motivation for participants to engage in the tasks, participants were told that they would be entered in a lottery to win another $40 if they performed in the top 25% on these tasks, consistent with prior investigations (Brown, Lejuez, Kahler, & Strong, 2002). Immediately prior to and following the mirror tracing task, participants were asked to self report their levels of irritability, frustration, anxiety, and difficulty concentrating in order to provide an index of their emotional reactivity to this stressor. 3. Results 3.1. Data reduction
2.3.2. Dot probe paradigm The dot probe paradigm (MacLeod, Mathews, & Tata, 1986) is a method for assessing visual attentional bias that utilizes response time as an index of relative attentional allocation to stimuli. Recently, this procedure has been adapted for use also as an intervention to alter attentional biases (e.g., Amir et al., 2009; MacLeod et al., 2002). Dot probe assessment and training procedures, including stimuli, were adapted from those used by MacLeod and colleagues (2002). All dot probe trials involved two verbal stimuli – one threatening (e.g., assault) and one neutral (e.g., bottles) – presented one above the other. A fixation cross (+) cueing participants to allocate attention to the center of the computer screen was presented for 500 ms, followed by the stimuli for 500 ms. Following the disappearance of the stimuli, one was replaced by a small dot. Participants were asked to identify the location of the dot as replacing the top word or the bottom word using a keyboard. The inter-trial interval was 500 ms. Attentional bias scores were calculated by subtracting the
Consistent with past evaluations of attentional bias (e.g., Beevers, Gibb, McGeary, & Miller, 2007), any individual trial for which a participant responded incorrectly or responded with an exceptionally low (<150 ms) or high (>1500 ms) reaction time (thus likely reflecting a different stage of attention) were excluded from calculation of individual means. Any participant for whom >10% of trials were excluded based on these criteria was excluded from overall data analyses for the study (n = 6). This degree of invalid data was judged to be reflective of a likely failure of the participant to sufficiently engage in the training (i.e., failed manipulation), possibly resulting in a questionably valid measure of attentional bias. In addition, we excluded from all analyses one participant for whom a computer malfunction resulted in a failure to record dot probe data, and three participants who only completed Day 1 of the study procedures. Prior to analysis, the data were examined for fit between the distributions of the self-report questionnaires (STAI-T, STAI-S, PANAS),
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445
443
Table 1 Dependent variables at Day 1 and Day 2. Day 1
STAI-T total at baseline PANAS—positive affect in the last 24 h PANAS—negative affect in the last 24 h Attentional bias during dot probe procedure (milliseconds) Number of seconds to discontinue Anagram task Number of seconds to discontinue Mirror tracing task Dysphoria following mirror tracing task Change in state anxiety from pre- to post-stressors Change in Negative affect from pre- to post-stressors
Day 2
DCS
PBO
56.21 (7.44) 25.67 (8.60) 24.39 (8.51) 11.48 (18.35) 359.30 (63.01) 227.60 (162.90) 183.06 (100.21) +3.97 (7.20) +5.28 (6.41)
56.07 (7.49) 21.60 (9.35) 19.47 (6.12) 9.98 (33.44) 330.00 (112.95) 125.27 (160.96) 201.27 (131.52) +6.07 (7.08) +5.93(5.99)
DCS
PBO – – – −.37 (15.30) 320.15 (96.86) 171.63 (151.04) 184.60 (80.56) +2.68 (5.56) −4.94 (13.62)
– – – 2.64 (24.32) 301.00 (102.00) 147.33 (163.50) 139.53 (123.07) +5.15 (11.34) −4.14 (14.57)
DCS = d-cycloserine group, PBO = placebo group. No between-group differences were found at baseline for these measures.
and reaction time data and the assumptions of normality. Any value exceeding a z-score of 3.3 was considered to be a univariate outlier and was omitted from analyses. No outliers were found in the self-report data. Two participants (both in the PBO condition) evidenced outliers on the dot probe reaction time data and were thus excluded from all analyses. For the mirror tracing emotional reactivity data, composite dysphoria scores were created by summing participants’ 0–100 ratings of irritability, frustration, anxiety, and difficulty concentrating as measured immediate prior to and immediately following that task. Thus, dysphoria scores ranged from 0 to 400. This dysphoria scale has been utilized in prior studies of emotional reactivity during the mirror tracing task (e.g., Daughters et al., 2005). 3.2. Preliminary analyses Preliminary analyses also sought to ensure that levels of selfreported symptoms and affect did not differ between the two experimental conditions at baseline on Day 1 of the study. A series of one-way t-tests indicated that the DCS and PBO groups did not differ on baseline levels of trait anxiety (t[32] = 0.06 ns), positive affect (t[31] = 1.30 ns), or negative affect (t[31] = 1.87 ns) over the 24 h prior to the experimental session. Similarly, we sought to ensure that participants in the two groups did not differ in their reactivity to study procedures and stressors prior to the administration of the study pill. Indeed, participants in the two groups did not differ with respect to attentional bias during the first dot-probe procedure (t[32] = 0.17 ns), level of dysphoria in reaction to the mirror tracing task (t[30] = −0.44 ns), number of seconds to discontinue the mirror tracing task (t[30] = 1.78 ns), or number of seconds to discontinue the anagram task (t[30] = 0.91 ns). Levels of these variables on both days of the study are reported in Table 1. Finally, we conducted manipulation checks to ensure that emotional reactivity did in fact increase in reaction to the stressors employed at baseline (prior to any training taking place and prior to pill administration) and that this reactivity did not differ between DCS and PBO groups. We employed a 2 (Phase: Pre-stressor, Poststressor) × 2 (Drug Condition: DCS, PBO) mixed model ANOVA on self-reported state anxiety and negative affect scores prior to attentional training. For the state anxiety analysis, although there was no effect of Drug Condition, there was a main effect of Phase such that state anxiety was higher following the stressors (M = 56.29, SD = 9.01) than it was prior to the stressors (M = 51.40, SD = 10.19), F(1,32) = 16.55, p < .001. For the negative affect analysis, although there was likewise no effect of Drug Condition, there was a main effect of Phase such that NA was higher following the stressors (M = 23.73, SD = 8.73) than it was prior to the stressors (M = 18.15, SD = 8.43), F(1,31) = 26.55, p < .001. Thus, as expected, the stress tasks were successful in eliciting emotional reactivity in individuals at baseline, and random assignment to condition was successful in producing two groups
Fig. 1. Change in attentional bias by drug condition.
who did not differ in the degree of that reactivity prior to experimental manipulations being instituted. 3.3. Attentional bias We first examined the effectiveness of the attention training procedure in reducing attentional biases in the full sample of participants. A paired samples t-test revealed that attentional bias scores did in fact decrease across the two days of the experiment, t(31) = 2.00, p = .05. Next, to compare the effects of DCS and PBO on change in attentional bias, we conducted a one-way (Drug Condition) ANOVA on attentional bias change scores from Day 1 to Day 2 of the experiment. Results indicated a significant effect of Drug Condition, indicating that participants in the DCS group evidenced a significantly greater decrease in attentional bias towards threatening information relative to participants in the placebo condition, F(1,31) = 5.18, p < .05 (see Fig. 1). This Drug Condition effect (DCS vs. placebo) over time evidenced a large effect size (d = 0.82) according to Cohen’s (1988) standards. 3.4. Distress tolerance Similar analyses were conducted to examine across-day and between-group differences in the number of seconds participants engaged in the anagram and mirror-tracing stressors before discontinuing those tasks. For the anagram task, there was a significant
444
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445
decrease in the total amount of time (in seconds) study participants tolerated the task from Day 1 (M = 345.08, SD = 86.31) to Day 2 (M = 305.26, SD = 98.54) of the study, F(1,27) = 7.88, p < .01. However, there was no effect of Drug Condition on number of seconds to discontinue the task, F(1,27) = 0.13 ns. For the mirror-tracing task, results did not indicate a significant change in the amount of time study participants tolerated the task from Day 1 (M = 160.96, SD = 151.32) to Day 2 (M = 159.48, SD = 155.15) of the study, F(1,28) = 0.003 ns. Results also indicated no effect of Drug Condition on number of seconds to discontinue the task, F(1,28) = 0.87 ns. 3.5. Emotional reactivity We also examined across-day and between-group differences in emotional reactivity to the stress tasks. For the mirror tracing task, dysphoria scores did not change significantly from Day 1 (M = 189.90, SD = 117.48) to Day 2 (M = 162.07, SD = 104.74) of the study, t(30) = 1.84, p = .08. A one-way ANOVA further showed that there was no effect of Drug Condition on across-day change in dysphoria between the two conditions, F(1,30) = 3.40, p = .08. We also examined affective reactivity to the two stressors as measured by the change in state anxiety and negative affect from pre- to post-stressor tasks. For the state anxiety analyses, change in state anxiety from pre- to post-stressors did not change significantly from Day 1 (M = +4.64, SD = 7.19) to Day 2 (M = +3.69, SD = 8.32) of the study, t(31) = 0.74 ns. Additionally, a one-way ANOVA indicated no effect of Drug Condition on pre- to poststressors change from Day 1 to Day 2 of the study, F(1,30) = 3.40 ns. For the negative affect analyses, change in NA from pre- to poststressors significantly decreased from Day 1 (M = +5.69, SD = 6.20) to Day 2 (M = −4.59, SD = 13.81) of the study, t(31) = 3.58, p < .01. However, a one-way ANOVA indicated no effect of Drug Condition on pre- to post-stressors change from Day 1 to Day 2 of the study, F(1,30) = 3.40 ns. 4. Discussion This study provides preliminary evidence for a facilitating effect of DCS on an attentional training intervention for individuals with elevated trait anxiety. In our sample of individuals who were high in trait anxiety and who completed only a single session of attentional training, DCS administration led to significantly greater training efficacy. Although all participants evidenced a decrease in attentional bias from Day 1 to Day 2 of the experiment, this decrease was greater among individuals who took DCS relative to those who took a placebo, and this difference evidenced a large effect size. This differential effect on attentional bias, however, was not associated with differential tolerance for stressful tasks or differential reactivity on the emotional stress tasks, although the results indicated a reduction in emotional reactivity over time for some (negative affect) but not all (state anxiety) measures. The absence of effects for the emotional tasks may be attributable to the single session of the attentional training intervention utilized in this analogue study. Other studies examining the effects of attention training on affective reactivity have found affective shifts; however, these studies were characterized by the use of a no-training comparison (Amir et al., 2008) or utilized conditions comparing attentional training toward threat to attentional training away from threat (MacLeod et al., 2002), which may have increased the relative effect size of the intervention. In our study design we selected a single dose of attentional training to avoid potential ceiling effects from repeated training trials, as per the designs of early studies of exposure augmentation using DCS (e.g., Ressler et al., 2004). Furthermore, this investigation uti-
lized two nomothetic stressors that were selected for their ability to elicit frustration, anxiety, and irritability in participants, given that the study sample (individuals with high trait anxiety) is not characterized by the presence of specific feared stimuli. The questionable degree to which these tasks were able to elicit consistently strong emotional reactions among participants may have limited our ability to find results for state anxiety measures. We anticipate that both stronger attentional training results, and affective consequences of this training, may be achieved with additional sessions of training and/or with more ideographic stressors that are more likely to elicit strong emotional responses. One pragmatic question that our results raise is whether the reduction in attentional bias resulting from the training procedure is worth the added effort that is entailed in prescribing DCS. Indeed, several investigations have found enhanced effects of DCS in the treatment of various anxiety conditions (e.g., Wilhelm et al., 2008), but the provision of DCS in those studies and in our study requires the need for physiological and/or psychiatric evaluations, the involvement of a prescribing physician, and potentially the use of additional personnel. Although additional research certainly needs to be conducted before one can determine the relative costs and benefits of administering DCS in conjunction with attention training or any other cognitive-behavioral intervention, our results provide evidence in favor of DCS’s potential therapeutic applications to attention-based interventions such as this one. Limitations of our study include the deletion of data from several participants because of particularly poor quality of dot probe data, likely reflecting a failure to engage in study procedures. Also, the small sample size limited our ability to detect other than large effect sizes for the experimental manipulations. Replication in larger samples is an important area for future research. In summary, this study provides preliminary support for the combination of two novel intervention strategies for anxiety disorders, attentional training and DCS. This finding is of particular importance given the continued need for strategies to improve the rates of treatment response in anxiety disorders, and the need to investigate the limits of the DCS augmentation effects (e.g., Otto et al., 2009). Future research is needed to replicate these results and to extend them to clinical samples. References Amir, N., Beard, C., Burns, M., & Bomyea, J. (2009). Attention modification program in individuals with generalized anxiety disorder. Journal of Abnormal Psychology, 118, 28–33. Amir, N., Weber, G., Beard, C., Bomyea, J., & Taylor, C. T. (2008). The effect of a single-session attention modification program on response to a public-speaking challenge in socially anxious individuals. Journal of Abnormal Psychology, 117, 860–868. Bar-Haim, Y., Pergamin, L., 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, 1–24. Beevers, C. G., Gibb, B. E., McGeary, J. E., & Miller, I. W. (2007). Serotonin transporter genetic variation and biased attention for emotional word stimuli among psychiatric inpatients. Journal of Abnormal Psychology, 116, 208–212. Brown, R. A., Lejuez, C. W., Kahler, C. W., & Strong, D. R. (2002). Distress tolerance and duration of past smoking cessation attempts. Journal of Abnormal Psychology, 111, 180–185. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates. Daughters, S. B., Lejuez, C. W., Bornovalova, M. A., Kahler, C. W., Strong, D. R., & Brown, R. A. (2005). Distress tolerance as a predictor of early treatment dropout in a residential substance abuse treatment facility. Journal of Abnormal Psychology, 114, 729–734. Davenport, R. A., & Houpt, T. A. (2009). d-cycloserine enhances short-delay, but not long-delay, conditioned taste aversion learning in rats. Pharmacology Biochemistry and Behavior, 91, 596–603. Davis, M., Myers, K. M., Ressler, K. J., & Rothbaum, B. O. (2005). Facilitation of extinction of conditioned fear by d-cycloserine: Implications for psychotherapy. Current Directions in Psychological Science, 14, 214–219. Deco, G., & Rolls, E. T. (2005). Neurodynamics of biased competition and cooperation for attention: A model with spiking neurons. Journal of Neurophysiology, 94, 295–313.
E. Behar et al. / Journal of Anxiety Disorders 24 (2010) 440–445 Goff, D. C., Cather, C., Gottlieb, J. D., Evins, A. E., Walsh, J., Raeke, L., et al. (2008). Once-weekly d-cycloserine effects on negative symptoms and cognition in schizophrenia: An exploratory study. Schizophrenia Research, 106, 320–327. Grillon, C. (2009). d-Cycloserine facilitation of fear extinction and exposure-based therapy might rely on lower-level, automatic mechanisms. Biological Psychiatry, 66, 1227–1234. Guastella, A. J., Richardson, R., Lovibond, P. F., Rapee, R. M., Gaston, J. E., Mitchell, P., et al. (2008). A randomized controlled trial of d-cycloserine enhancement of exposure therapy for social anxiety disorder. Biological Psychiatry, 63, 544–549. Hazen, R. A., Vasey, M. W., & Schmidt, N. B. (2009). Attentional retraining: A randomized clinical trial for pathological worry. Journal of Psychiatric Research, 43, 627–633. Hofmann, S. G., Meuret, A. E., Smits, J. A. J., Simon, N. M., Pollack, M. H., Eisenmenger, K., Shiekh, M., & Otto, M. W. (2006). Augmentation of exposure therapy with d-cycloserine for social anxiety disorder. Archives of General Psychiatry, 63, 298–304. Hofmann, S. G., & Smits, J. A. (2008). Cognitive-behavioral therapy for adult anxiety disorders: A meta-analysis of randomized placebo-controlled trials. Journal of Clinical Psychiatry, 69, 621–632. Kalisch, R., Holt, B., Petrovic, P., De Martino, B., Klöppel, S., Büchel, C., & Dolan, R. J. (2009). The NMDA agonist d-cycloserine facilitates fear memory consolidation in humans. Cerebral Cortex, 19, 187–196. Kushner, M. G., Kim, S. W., Donahue, C., Thuras, P., Adson, D., Kotlyar, M., et al. (2007). d-Cycloserine augmented exposure therapy for obsessive-compulsive disorder. Biological Psychiatry, 62, 835–838. Land, C., & Riccio, D. (1999). d-Cycloserine: Effects on long-term retention of a conditioned response and on memory for contextual attributes. Neurobiology Learning and Memory, 72, 158–168. Lelong, V., Dauphin, F., & Boulouard, M. (2001). RS 67333 and d-cycloserine accelerate learning acquisition in the rat. Neuropharmacology, 41, 517–522. MacLeod, C., Koster, E. H. W., & Fox, E. (2009). Whither cognitive bias modification research? Commentary on the special section articles. Journal of Abnormal Psychology, 118, 89–99. MacLeod, C., Mathews, A., & Tata, P. (1986). Attentional bias in emotional disorders. Journal of Abnormal Psychology, 95, 15–20. MacLeod, C., Rutherford, E., Campbell, L., Ebsworthy, G., & Holker, L. (2002). Selective attention and emotional vulnerability: Assessing the causal basis of their association through the experimental manipulation of attentional bias. Journal of Abnormal Psychology, 111, 107–123. Mogg, K., Mathews, A., Bird, C., & Macgregor-Morris, R. (1990). Effects of stress and anxiety on the processing of threat stimuli. Journal of Personality and Social Psychology, 59, 1230–1237. Norberg, M. M., Krystal, J. H., & Tolin, D. F. (2008). A meta-analysis of d-cycloserine and the facilitation of fear extinction and exposure therapy. Biological Psychiatry, 63, 1118–1126. Nunnink, M., Davenport, R. A., Ortega, B., & Houpt, T. A. (2007). d-Cycloserine enhances conditioned taste aversion learning in rats. Pharmacology Biochemistry and Behavior, 87, 321–330.
445
Otto, M. W., Basden, S. L., McHugh, R. K., Kantak, K. M., Deckersbach, T., Cather, C., et al. (2009). Effects of d-cycloserine administration on weekly nonemotional memory tasks in healthy participants. Psychotherapy and Psychosomatics, 78, 49–54. Otto, M. W., Tolin, D. F., Simon, N. M., Pearlson, G. D., Basden, S., Meunier, S. A., et al. (2010). Efficacy of D-Cycloserine for enhancing response to cognitive-behavior therapy for panic disorder. Biological Psychiatry, 67, 365–370. Pollack, M. H., Otto, M. W., Roy-Byrne, P. P., Coplan, J. D., Rothbaum, B. O., Simon, N. M., et al. (2008). Novel treatment approaches for refractory anxiety disorders. Depression and Anxiety, 25, 467–476. Ressler, K. J., Rothbaum, B. O., Tannenbaum, L., Anderson, P., Graap, K., Zimand, E., Hodges, L., & Davis, M. (2004). Cognitive enhancers as adjuncts to psychotherapy: Use of d-cycloserine in phobics to facilitate extinction of fear. Archives of General Psychiatry, 1136–1144. Richardson, R., Ledgerwood, L., & Cranney, J. (2004). Facilitation of fear extinction by d-cycloserine: Theoretical and clinical implications. Learning and Memory, 11, 510–516. See, J., MacLeod, C., & Bridle, R. (2009). The reduction of anxiety vulnerability through the modification of attentional bias: A real-world study using a home-based cognitive bias modification procedure. Journal of Abnormal Psychology, 118, 65–75. Schmidt, N. B., Richey, J. A., Buckner, J. D., & Timpano, K. R. (2009). Attention training for generalized social anxiety disorder. Journal of Abnormal Psychology, 118, 5–14. Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychologists Press. Storch, E. A., Merlo, L. J., Bengtson, M., Murphy, T. K., Lewis, M. H., Yang, M. C., Jacob, M. L., Larson, M., Hirsh, A., Fernandez, M., Geffken, G. T., & Goodman, W. K. (2007). d-Cycloserine does not enhance exposure-response prevention therapy in obsessive-compulsive disorder. International Clinical Psychopharmacology, 22, 230–237. Strong, D. R., Lejuez, C. W., Daughters, S. B., Marinello, M., Kahler, C. W., & Brown, R. A. (2003). The Computerized Mirror Tracing Task Version 1. Unpublished manuscript. Van Damme, S., Crombez, G., Hermans, D., Koster, E. H. W., & Eccleston, C. (2006). The role of extinction and reinstatement in attentional bias to threat: A conditioning approach. Behaviour Research and Therapy, 44, 1555–1563. van den Heuvel, O. A., Veltman, D. J., Groenewegen, H. J., Witter, M. P., Merkelbach, J., Cath, D. C., et al. (2005). Disorder-specific neuroanatomical correlates of attentional bias in obsessive-compulsive disorder, panic disorder, and hypochondriasis. Archives of General Psychiatry, 62, 922–933. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. Wilhelm, S., Buhlmann, U., Tolin, D. F., Meunier, S. A., Pearlson, G. D., Reese, H. E., et al. (2008). Augmentation of behavior therapy with d-cycloserine for obsessivecompulsive disorder. American Journal of Psychiatry, 165, 335–341. Wilson, E., & MacLeod, C. (2003). Contrasting two accounts of anxiety-linked attentional bias: Selective attention to varying levels of stimulus threat intensity. Journal of Abnormal Psychology, 112, 212–218.