To address shortcomings of purely reaction-time based attention bias modification (ABM) paradigms, a novel eye-tracking based ABM training (ET-ABM) was developed. This training targets the late disengagement from negative stimuli and the lack of attention for positive information, which are characteristics of depression. In the present study, 75 dysphoric students (BDI ≥ 9) were randomly assigned to either this positive training (PT), or a sham-training (ST) that did not train any valence- specific gaze pattern (positive and negative pictures had to be disengaged from and attended to equally often). Results showed that the PT induced a positive attentional bias (longer fixations of positive than negative pictures). Although the ST group showed an increase in positive attentional bias as well, this increase was not as strong as in the PT group. Compared to the ST, the PT specifically induced faster disengagement from negative pictures. No differential training effects were found on stress responses or state rumination. These results show that the ET-ABM successfully modifies attentional processes, specifically late disengagement from negative stimuli, in dysphoric students, and hence might be a promising alternative to existing ABM paradigms. Keywords Depression · Attention bias modification · Attentional disengagement · Eye-tracking Introduction A potential cognitive vulnerability factor which is thought to maintain depressive symptoms and predispose individu- With more than 300 million people suffering from depres- als to repeatedly develop new episodes, is the heightened sion worldwide, it is amongst the most prevalent mental attention to depression-relevant, negative information com- disorders and, according to the World Health Organization pared to positive information, commonly found in depressed (WHO 2017), has become the leading cause of disabili- individuals (e.g., Armstrong and Olatunji 2012). Research ties, with a major contribution to the overall global disease making use of eye-tracking technology suggests that in burden. Despite the variety of existing treatment options depression, this so-called negative attentional bias is speci- fi for depression, relapse rates are high (e.g., about 30% as cally characterized by difficulties with disengaging attention reported in Seemüller et al. 2014), with an increasing risk from negative stimuli once they have become the focus of of recurrent depression with each subsequent depressive epi- attention (Sanchez et al. 2013). At the same time, depressed sode (Steinert et al. 2014). This suggests that the underlying individuals show reduced maintained attention to positive mechanisms which maintain depression are not very well stimuli, compared to healthy individuals (e.g., Ellis et al. understood yet, and hence may not be targeted sufficiently 2010; Kellough et al. 2008; Sears et al. 2010). by current treatment programs. According to cognitive theories of depression, atten- tional biases play a causal role in both development and maintenance of this disorder (Beck 1976; Teasdale 1988). * Martin Möbius As a consequence, computerized training paradigms have firstname.lastname@example.org been developed, aiming at reducing depressive symptoms Behavioural Science Institute, Radboud University through altering maladaptive information processing ten- Nijmegen, PO Box 9104, 6500 HE Nijmegen, dencies; the so-called cognitive bias modification tech- The Netherlands niques (CBM; Mathews and MacLeod 2005). The most Pro Persona, Center for Mental Health Care, Nijmegen, frequently used paradigm for modifying (and measuring) The Netherlands Vol:.(1234567890) 1 3 Cognitive Therapy and Research (2018) 42:408–420 409 attentional processes is the dot-probe task (MacLeod et al. stimuli (for a more thorough discussion of the limitations 2002). On each trial of this task, two stimuli are presented of the dot-probe task, see Ferrari et al. 2016). simultaneously on the computer screen, usually a negative Based on the limitations of the dot-probe task and other picture and a positive (or neutral) picture (or word). After RT-based ABM paradigms, Ferrari et al. (2016) developed a short delay, both stimuli disappear and a target replaces a new ABM paradigm, incorporating eye-tracking technol- one of the two stimuli, which participants have to react to. ogy. This eye-tracking based attentional bias modification Faster reactions to targets replacing negative compared to (ET-ABM) paradigm allows for the continuous assessment positive stimuli indicate a negative attentional bias. In the of eye-movements, and hence for a potentially more reliable training version of this task, the targets replace mostly the assessment and training of the specific attentional compo- positive (or neutral) stimuli, such that participants’ atten- nents that are biased in depression. During this ET-ABM tion is selectively trained away from the negative stimuli. task, participants are trained to disengage their attention To date, the dot-probe task has mainly been used to from negative pictures and to keep their attention on posi- assess and modify attentional bias in anxious populations tive pictures. On each trial of the task, participants first have (for review see Cisler and Koster 2010). Only a few stud- to fixate a cross on a computer screen, after which two posi - ies have tried to modify attentional processes in depressed tively and two negatively valenced pictures are presented individuals, with inconsistent findings so far. While some in a 2 × 2 grid. Importantly, the trial only continues after studies managed to reduce the bias toward negative a sufficiently long fixation of a positive picture (1000 ms). stimuli and accordingly the depressive symptoms, many Hence, if a negative picture replaces the cross, the trial only other studies failed to modify an attentional bias and to continues when participants look away from the negative replicate the beneficial therapeutic effects on depressed picture and fixate one of the positive pictures. If a positive mood or symptoms, questioning the efficacy of attentional picture replaces the cross, the pictures disappear when par- bias modification (ABM) procedures for depression (for ticipants keep their attention on the fixated positive picture reviews see Cristea et al. 2015; Mogoaşe et al. 2014). In a or fixate the other positive picture. As soon as a positive paper reflecting on the increasing number of ABM failures picture has been fixated for 1000 ms, all pictures disappear Clarke et al. (2014), however, cautioned against taking the and a target stimulus is presented at the location of the last absence of evidence in some studies as evidence against fixated picture. The target has to be identified by pressing the theoretical basis of ABM in general. The authors drew a corresponding button. Different from previous RT-based attention to the fact that most studies that succeeded in ABM paradigms, the training trials in this task only continue modifying an attentional bias also induced emotional if participants show the required viewing pattern, tailoring change. By contrast, those studies that failed to modify the pace of the task to the individuals’ performance. an attentional bias also failed to find beneficial effects on In a first proof-of principle study (Ferrari et al. 2016) mood. This suggests that conventional ABM paradigms with an unselected student sample, this positive training was may not be optimal for reliably modifying and measuring compared to a negative training with the opposite training attentional bias, and that more research is needed into the contingencies. Thus, participants were trained to direct their task conditions under which ABM actually changes atten- gaze to negatively valenced pictures. Results of this first tional processes. According to this argument, more prom- study showed that this ET-ABM training is suited to alter ising clinical applications of ABM depend on the develop- attentional processes relevant in depression: The positive ment of more effective attention modification procedures. training induced a positive sustained attention bias, that One of the most frequently mentioned limitations of is, longer sustained attention to positive than to negative the dot-probe and other reaction time (RT) based ABM pictures. More specifically, it trained participants to more paradigms is the low reliability of these tasks (e.g., quickly disengage their attention from negative stimuli and Schmukle 2005; Waechter and Stolz 2015). More impor- direct their attention to positive pictures. No such changes tantly, however, the suitability of the dot-probe task has were found in the negative training group. Although the especially been doubted in the context of depression, as training affected mood directly, with the negative group it is not clear which component of attention is measured showing a stronger increase in negative mood in response and trained (Leyman et al. 2007). It has been argued that to the training than the positive group, the training did not with the longer stimulus durations commonly used in differentially affect emotional reactions to a subsequent labo- depressed populations (Beevers et al. 2015; Wells and ratory stressor. Beevers 2010), participants may shift their attention back Although this first study provides promising evidence and forth between stimuli before the target appears. This that depression-relevant attentional processes can be altered in turn leaves undetected whether the task is tapping into by means of this novel ABM paradigm, several questions heightened vigilance for negative stimuli or the depres- remain to be answered. First, in the previous study, the sion-characteristic slowed disengagement from negative positive training was compared to a negative training in 1 3 410 Cognitive Therapy and Research (2018) 42:408–420 order to maximize group differences, hence it remains to Nijmegen, the Netherlands, with elevated depression scores be established whether the positive training is also superior on Beck’s Depression Inventory (BDI-II; Beck et al. 1996; to a placebo control condition in changing the attentional M = 15.57, SD = 7.13) participated in return for course credit processes. Second, before applying this training in clinically or 20 €. In order to test participants with at least mild depres- depressed populations, it should first be tested whether the sive symptoms, we only invited students with BDI-II scores results can be replicated in a sample with subclinical levels higher than 8, which is in line with previous studies (Mas- of depression, which is assumed to show a stronger pre- tikhina and Dobson 2017; Wells and Beevers 2010). This existing negative attentional bias. Finally, training effects in low cutoff score was chosen to maximize sensitivity of the the first study were only tested with the same set of stimuli BDI-II (Sprinkle et al. 2002). Of 841 students who signed as used during the training, leaving it open whether training up, 96 were invited to the experiment and participated. They effects are restricted to the specific stimuli being used, or were randomly allocated to either the PT (n = 50) or the ST whether the attentional processing of positive and negative (n = 46). stimuli in general is affected. Hence, the primary aim of this study was to replicate the Instruments and Materials findings of the previous study in an emotionally vulnerable sample, with a sham training as control condition and dif- Baseline Questionnaires ferent stimulus sets in training and assessment. To this end, we randomly assigned participants with elevated depression All of the following questionnaires were administered in the scores to one of two training conditions. Half of the partici- participants’ dominant language (i.e., German or Dutch). pants received the positive training (PT) in which they were Depression levels were assessed with the revised version trained to direct their gaze away from negative pictures and of Beck’s Depression Inventory (BDI-II, Beck et al. 1996). towards positive pictures. The other half received a sham The internal consistency of the BDI in the current sample training (ST), where no valence-specific gaze patterns were was excellent (α = .97). To be able to control for possible reinforced. Importantly, at the beginning of the experiment, baseline differences in trait anxiety, the trait subscale of the a negative mood was induced by means of a sad movie. This State-Trait Anxiety Inventory (STAI-T; Spielberger 1989) allows for re-activation of otherwise latent depressogenic was administered. The internal consistency of the STAI was structures in emotionally vulnerable individuals (e.g., Beck excellent (α = .97). Moreover, individual differences in rumi- 1967) and hence may serve to elicit a negative attentional native thinking were assessed with the Ruminative Response bias in our sample (for a detailed description of the proce- Scale (RRS; Treynor et al. 2003). As the German and the dure, see Scher et al. 2005). We expected that, compared to Dutch version slightly differ in the number of items, we cal- participants in the ST, (1) participants in the PT would show culated a mean score instead of a sum score. Both, the Dutch an increase in positive attentional bias (i.e., relatively longer and the German version of the RRS showed a good internal fixations on positive than on negative pictures), and that (2) consistency (Dutch: α = .89; German: α = .86). participants in the PT would specifically learn to faster dis- engage their attention from negative pictures. Mood State To get a first indication of the potential therapeutic effects of the training, we additionally explored participants’ mood Throughout the experiment, participants were asked to rate reactivity and recovery in response to a laboratory stressor their current general mood state (i.e., “How is your mood at at the end of the experiment. In their eye-tracking experi- this moment?”) on a 100 mm Visual Analogue Scale (VAS), ment, Sanchez et al. (2013) showed that specifically the ranging from 0 (very bad) to 100 (very good). In order to slowed disengagement from negative information is related assess changes in mood state in response to the stress task, to impaired mood recovery after stress in depressed indi- we additionally presented Likert scales as used by Sanchez viduals. In accordance with these findings, we expected that et al. (2013). Each Likert scale measured mood with three participants in the PT would show better stress recovery than items: happy mood (happy, optimistic, joyful), anxious participants in the ST. mood (nervous, tense, anxious), and sad mood (depressed, upset, sad). All items were rated on a scale ranging from 0 (not at all) to 10 (very much). One extra item was added to Methods assess the tiredness of participants (0 = not at all; 10 = very much). Participants Eighty-four female and 12 male undergraduate students (mean age = 21.67, SD = 4.66), of Radboud University 1 3 Cognitive Therapy and Research (2018) 42:408–420 411 State Rumination Fig. 1 Schematic overview of the task design. On each trial of the positive training (PT), a fixation cross is presented. Upon fixation (500 ms), two negative and two positive pictures appear. a The free To assess state rumination in response to the stress task, the viewing task (assessment) is similar to the training, however, all trials Momentary Ruminative Self-focus Inventory (MRSI; Mor last 3000 ms and no probe is presented. b, c Display a sample trial of et al. 2013) was administered. The MRSI contains six items, the PT. b On negative (PT: disengagement) trials, participants have to disengage their attention from the fixated negative picture and fix- measuring momentary self-focused rumination (e.g., “Right ate one of the two positive pictures. c On positive (PT: maintained now, I am thinking about the possible meaning of the way attention) trials, attention has to be maintained at the fixated positive I feel”). Items are rated on a 7-point scale, ranging from picture or at the other positive picture. b, c Upon fixation of a posi- “totally not agree” to “totally agree”. tive picture for 1000 ms, all pictures disappear and an arrow replaces the fixated picture. Participants respond to arrow direction by press- ing a key. The arrow then disappears and a new trial starts. d An Negative Mood Induction example of four consecutive trials in the sham training (ST), with the first trial requiring a fixation of a picture in the upper right corner. On each consecutive trial participants have to fixate a picture in a dif- Negative mood was induced by means of two sequences (i.e., ferent corner, while the relevant corner changes in clockwise order 20 min in total) of the movie “Sophie’s choice”. The two (2nd trial: lower right corner; 3rd trial: lower left corner; 4th trial: sequences have been shown to effectively induce negative upper left corner). Upon fixation of the picture in the correct corner for 1000 ms, all pictures disappear and an arrow replaces the fixated mood in previous studies (e.g., Randall and Cox 2001), and picture. Participants respond to arrow direction by pressing a key. The they have been used before to elicit a negative mood state arrow then disappears and a new trial starts. Note This figure con- in healthy and dysphoric individuals prior to CBM training tains sample images, which have not been used in the current study. (Becker et al. 2016). All images were obtained from Flickr and were published under a Creative Commons license. The formats of the images were slightly adapted for this figure. Credits: top left, Joe deSousa, CC0 1.0; top ETABM T ‑ ask right, West Point—The U.S. Military Academy, CC BY 2.0; bottom left, Steven Depolo, CC BY 2.0; bottom right, bettyx1138, CC BY 2.0. For license terms see, CC0 1.0 (https://creativecommons.org/ Stimuli publicdomain/zero/1.0/); CC BY 2.0 (https://creativecommons.org/ licenses/by/2.0/) In line with the study of Ferrari et al. (2016), a broad range of disorder-nonspecic fi picture stimuli representing die ff rent disappeared and a set of four pictures was presented. Placing categories (e.g., people, animals, objects; Nencki Affective the fixation cross into one of the quadrants (instead of the Picture System; Marchewka et al. 2014) was selected. Ninety screen center) and making sure that it was actually fixated, positive and ninety negative pictures (14.3 cm × 10.7 cm) allowed for reliably manipulating which picture was fixated were used during the training phase. A further set of 90 first. positive and 90 negative pictures was used exclusively dur- The training contained two different types of trials: trials ing the assessment phases, to allow for testing generalization on which a negative picture replaced the cross (i.e., negative of training effects to untrained pictures. trials) and trials on which a positive picture replaced the For each phase, training and assessment, 45 picture cross (i.e., positive trials). Participants in the positive train- sets were created, always containing two positive and two ing (PT) had to disengage attention from negative pictures negative pictures matched on content. The pictures were and shift it to positive pictures, and to maintain attention arranged in a 2 × 2 grid, which separated the screen into to positive pictures. On negative trials, participants had to four quadrants. The location at which each picture appeared look away from the fixated negative picture and fixate one (upper/lower and left/right part of the grid) was counterbal- of the two positive pictures for 1000 ms. Upon a sufficiently anced across trials. The stimuli were displayed on a black long fixation of a positive picture, all pictures disappeared 53.2 cm × 29.9 cm computer screen (BenQ XL2420Z), with and this previously fixated picture was replaced by a probe 1 cm distance between the pictures. Participants were seated (i.e., an arrow pointing left or right, with the direction of the about 60 cm away from the screens’ center. arrow being counterbalanced across picture valence). Par- ticipants had to identify the direction the arrow was point- Task Design ing to by pressing a computer key, upon which the probe disappeared and a new trial started. On positive trials, the The eye-tracking task was adapted from Ferrari et al. (2016) trial continued only if participants kept looking at the fix- and consisted of pre-assessment, training and post-assess- ated positive picture for 1000 ms, or if they fixated the other ment. For a graphic illustration of the task design, see Fig. 1. positive picture in the picture set for 1000 ms. On each trial of the task, a white fixation-cross appeared In the sham training (ST), a different attentional pattern in the middle of one of the four quadrants of the grid. As was trained, which was independent of the valence of the soon as the participant had fixated the cross for 500 ms, it 1 3 412 Cognitive Therapy and Research (2018) 42:408–420 1 3 Cognitive Therapy and Research (2018) 42:408–420 413 Fig. 1 (continued) stimuli, such that neither disengagement from negative stim- of the probe. The training contained 270 training trials dis- uli nor maintained attention to positive stimuli was selec- tributed across 3 blocks, during which each of the 45 picture tively reinforced. Instead, participants were trained to show sets was presented 6 times, in a new random order for each a clockwise viewing pattern of the presented picture sets. participant. This means that, if on the first trial, the picture in the upper The pre- and post-assessment consisted of a free viewing left part of the grid had to be x fi ated, the picture in the upper task similar to the training and was introduced to partici- right part had to be fixated on the next trial, and the picture pants as a calibration procedure. Different from the training, in the lower right part had to be fixated afterwards, and so all picture sets were presented for 3000 ms, independently on. The location of fixation cross, positive and negative pic- of participants’ viewing patterns, and no probe followed. tures was counterbalanced, such that on negative (positive) During assessment, the 45 picture assessment sets were pre- trials, attention had to be disengaged from negative (posi- sented twice (90 trials), once as positive and once as nega- tive) pictures as often as it had to be maintained to negative tive trials. During each assessment-phase, the location of (positive) pictures. As in the PT, after a sufficiently long the fixation cross was counterbalanced across valences and fixation of the correct picture, all pictures disappeared and a grid positions. The entire task took approximately 45 min, probe replaced the previously fixated picture. In both groups, depending on how quickly participants learned the required the participants’ gaze pattern thus controlled the appearance viewing patterns. 1 3 414 Cognitive Therapy and Research (2018) 42:408–420 Eye-Tracking Device has previously been used to assess the association between negative mood recovery after stress and attentional bias in Monocular gaze data of the dominant eye were obtained at depressed individuals (Sanchez et al. 2013). Participants a frequency of 500 Hz, by means of the iView X Hi Speed were informed that their speech would be video-recorded so system by SMI, a video based eye-tracking system. that it later can be evaluated on its quality by two independ- ent researchers. To increase stress levels, participants were Calculation of Attentional Indices not allowed to take any notes during preparation and a clock on the computer screen signaled how much time was left. In line with previous studies (Ferrari et al. 2016; Sanchez After 1 min, a “beep” sound occurred. The experimenter et al. 2013), only fixation durations of at least 100 ms were then entered the room, started the video-recording, asked considered. About 1.5 percent of all trials at pre- and post- participants to deliver their speech into a webcam, and left. assessment were deleted, due to poor tracking quality or After 3 min, the experimenter entered the room, stopped fixation durations shorter than 100 ms. The remaining trials the video recording and left again for a 5-min resting phase. were used to calculate three attentional indices separately Participants were instructed to sit down quietly and relax. for pre- and post-assessment: a “sustained attention bias” A clock on the screen again signaled the time left of the score, reflecting the proportion of the total fixation time on resting phase. positive compared to negative pictures, as well as the two relevant attentional components, that is “disengagement Procedure from negative pictures” (short: negative disengagement) and “maintained attention to positive pictures” (short: positive Prior to the experimental procedure, potential participants maintained attention). were pre-screened by means of the BDI via an online screen- The sustained attention bias score was calculated in line ing system of Radboud University Nijmegen. Only individu- with Ferrari et al. (2016): In the first step, we calculated two als with a score higher than 8 were invited for participation sum scores per trial, reflecting the total time participants fix- in the experiment, which took place within one week after ated positive and negative pictures. Based on these scores, pre-screening. These participants were tested individually medians were calculated for each participant, representing in a cubicle of the Behavioural Science Institute of the Uni- the median fixation time on positive and negative pictures. In versity. After providing informed consent, they were ran- the last step, we calculated the bias score: (Median fixation domly assigned in a double-blind fashion to one of the two time on positive pictures)/(median fixation time on positive training conditions (i.e., PT or ST). The experimenter then pictures + median fixation time on negative pictures). Scores determined the participants’ dominant eye and subsequently larger than 0.5 are indicative of a more positive sustained calibrated the eye-tracker. Next, participants were seated in attention bias (relatively longer fixations on positive pic- front of a computer, where they filled in the baseline ques- tures), while scores smaller than 0.5 are indicative of a more tionnaires and mood state measures (T0: baseline). These negative sustained attention bias. measures were followed by a negative mood induction to re- Negative disengagement scores and positive maintained activate the latent depressogenic schemas of the dysphoric attention scores were derived from negative and positive sample (e.g., Beck 1967), before we assessed mood state a trials respectively. We calculated the median fixation dura- second time (T1: pre-training). Participants were then seated tion on the first (positive or negative) picture until the first in front of the eye-tracker, where a brief calibration proce- attentional shift to (and fixation of) one of the pictures of dure was started, which was followed by pre-assessment, the opposite valence. Hence, on negative trials, longer fixa- training and post-assessment of the ET-ABM task. If nec- tion durations on negative pictures reflect a slower negative essary, the calibration procedure was repeated before each disengagement, while on positive trials, longer fixation dura- training block and post-assessment. Afterwards, participants tions on positive pictures reflect prolonged positive main- were again seated in front of the other computer, where the tained attention. stress task was presented. Throughout this task, participants were asked to rate their mood state on the VAS scales: at Stress Task the beginning of the task (T2: pre-stress), after the speech instructions were provided (T3: anticipatory stress), retro- In line with the study by Ferrari et al. (2016), we used an spectively during the speech (T4: during stress), after speech adapted speech task by Amir et al. (2008), in order to inves- delivery (T5: post-stress), and after the 5-min resting period tigate training effects on emotional reactivity in response to (T6: recovery). In addition to the VAS scales, we presented a laboratory stressor. Via the computer, participants were the Likert mood-scales at T2 (pre-stress), at T3 (anticipa- informed that they would get 1 min to prepare a 3-min- tory stress), and at T5 (post-stress). To assess possible train- speech on the topic “Why am I a good friend”. This topic ing effects on state rumination, the MRSI was administered 1 3 Cognitive Therapy and Research (2018) 42:408–420 415 Table 1 Group differences on demographic variables and baseline lower than 9. Another 3 participants were excluded due questionnaires to a lack of task adherence, and another 5 due to extreme values on the eye-tracking indices (i.e., data points more PT (n = 40) ST (n = 35) than 1.5 interquartile ranges below the first or above the Age 21.18 (2.36) 21 (2.76) t(73) = 0.3, p = .771 third quartile). Due to skewness of the data, the attentional Gender χ (1) = 5.2, p = .023 indices “negative disengagement” and “positive maintained Male 8 1 attention” were log-transformed. Due to missing data of Female 32 34 23 participants on a single item of the Likert mood scales, Nationality χ (1) = 0.14, p = .711 we calculated means scores per participant instead of sum Dutch 20 19 scores. The resulting groups (PT: n = 40, ST: n = 35), did German 20 16 not differ on any of the trait variables (p > .6) or mood state BDI-II 17.53 (6.44) 16.64 (6.25) t(73) = 0.73, p = .654 (p = .95) at baseline. Moreover, the groups did not differ STAI-T 49.4 (9.33) 51.29 (9.82) t(73) = 0.85, p = .952 on any of the demographic variables besides gender, with RRS 2.3 (0.56) 2.14 (0.53) t(71) = 1.29, p = .762 significantly more men in the PT (8 males) than in the ST (1 male), χ (75) = 5.2, p = .023 (see Table 1). A manipula- RRS scores of two participants were missing tion check of the mood induction showed that VAS scores PT positive training, ST sham training, BDI-II Becks Depression Inventory Second Edition, STAI-T Spielberger Trait Anxiety Inven- dropped from before to after the negative mood induction tory, RRS Ruminative Response Scale 2 procedure, F(1, 72) = 140.04, p < .001, η = .66, similarly for both groups, F(1, 72) = .04, p = .839. directly after the resting period. After having filled in the questionnaire, a sequence of a happy movie (Jungle book) Attentional Processes at Baseline was shown to elevate participants’ mood, followed by a final mood rating (VAS scale; T7). At the end of the experiment, Sustained Attention Bias participants filled in an awareness check, where we asked them to indicate on a 100 mm VAS scale, in how far they felt A univariate ANOVA of the sustained attention bias at the able to exert control on the eye-tracking task. Finally, par- pre-assessment revealed no significant group difference, F( 1, ticipants were compensated for participation and debriefed. 73) = .54, p = .466. A subsequent one-sample t-test indicated The entire procedure took about 120 min. that the bias score did not deviate significantly from zero, indicating that before the training, participants had neither a tendency to attend towards positive nor towards negative Results pictures, t(74) = 1.11, p = .271. Preliminary Analyses and Group Characteristics Positive Maintained Attention and Negative Disengagement In the analyses, we included only participants for whom both pre-assessment and post-assessment data of the eye-tracking Two separate ANOVAs comparing the log-transformed task were available (n = 96). Of those, 13 participants had attentional indices between the two groups, showed that to be excluded from the analyses because of a BDI score PT and ST group did not differ from each other regarding Table 2 Mean fixation times PT ST (with standard deviations) in milliseconds during the free Pre-training Post-training Pre-training Post-training viewing task, and the resulting Fixation time on positive pictures 1434 (270) 1797 (400) 1499 (273) 1658 (402) attentional bias scores Fixation time on negative pictures 1413 (301) 1000 (373) 1378 (281) 1189 (372) Sustained attention bias score 0.51 (0.1) 0.64 (0.13) 0.52 (0.1) 0.58 (0.13) Disengagement from negative pictures 711 (217) 555 (169) 643 (178) 745 (342) Maintained attention for positive pictures 624 (202) 802 (531) 595 (160) 801 (408) Sustained attention bias score: Proportion of fixation time on positive pictures compared to negative pic- tures; Disengagement from negative pictures: Latency of the first shift from a negative picture until fixation of a positive picture; Maintained attention for positive pictures: Latency of the first shift from a positive picture until fixation of a negative picture PT positive training, ST sham training 1 3 416 Cognitive Therapy and Research (2018) 42:408–420 the attentional components at baseline (positive maintained the crucial 3-way interaction was significant as well, attention: F(1, 74) = 0.26, p = .612; negative disengagement: F(1, 73) = 8.94, p = .004, η = .11 indicating that the two F(1, 74) = 1.71, p = .195). A subsequent paired-samples groups showed differential changes in positive maintained t-test revealed that at baseline, participants showed slower attention and negative disengagement. Subsequent paired- disengagement from negative pictures than from positive samples t-tests revealed that participants in the PT learned to pictures, t(74) = 3.09, p = .003, d = .33. For descriptives, see disengage more quickly from negative pictures, t(39) = 4.78, Table 2. p < .001, d = .73, and to longer fixate positive pictures, t(39) = 2.11, p = .041, d = .36. Participants in the ST became Training Eec ff ts on Attentional Processes generally slower with disengaging attention from both types of pictures (positive maintained attention: t(34) = 3.56, Changes in Sustained Attention Bias p = .001, d = .69; negative disengagement: t(34) = 2.13, p = .041, d = .36). While at pre-assessment, participants A 2 (group: PT, ST) × 2 (time: pre, post) repeated-measures had the tendency to disengage from negative pictures more (RM) ANOVA of the sustained attention bias scores revealed slowly than from positive pictures, this bias disappeared at a main effect of time, F (1, 73) = 92.18, p < .001, η = .56, post-assessment in the ST, t(34) = 0.47, p = .639, and was which was moderated by the training, F(1, 73) = 14.64, even reversed in the PT, t(39) = 3.48, p = .001, d = .57. See p < .001, η = .17. As indicated by a subsequent paired- Table 2 for untransformed scores of the eye-tracking indices. samples t-test, both groups showed an increase in sustained attention bias for positive pictures (PT: t(39) = 9.46, p < .001, Training Eec ff ts on Mood d = 1.54; ST: t (34) = 4.21, p < .001, d = .76). The PT, how- ever, showed a stronger bias after the training than the ST, Direct Eec ff ts on Mood t(73) = 2.83, p = .041, d = .48. For means, see Table 2. An independent-samples t-test on the VAS revealed that PT Changes in Maintained Attention and Negative and ST did not differ in mood state right before the training, Disengagement t(73) = 0.18, p = .859. A subsequent 2 (group: PT, ST) × 2 (time: pre-assessment, post-assessment) RM ANOVA of the A 2 (group: PT, ST) × 2 (time: pre, post) × 2 (valence: VAS revealed a recovery from the negative mood induction positive, negative) RM ANOVA on the log-transformed procedure during the training, F (1, 72) = 106.96, p < .001, attentional indices revealed a marginally significant effect of η = .6, which was not significantly different for the two time, F(1, 73) = 3.12, p = .081, η = .04, which was moder- groups, F (1, 72) = 1.3, p = .257. For means and SDs, see ated by valence, F(1, 73) = 27.17, p < .001, η = .27, as well Table 3. as by group, F(1, 73) = 8.14, p = .005, η = .1. Importantly, Table 3 Mean mood scores (with standard deviations) for all assessment points T0: baseline T1: pre-training T2: pre-stress T3: anticipatory T4: during T5: post-stress T6: recovery T7: mood stress stress induction PT VAS 56.85 (19.6) 27.95 (17.14) 56.63 (14.92) 48.24 (23.41) 46.76 (24.72) 56.42 (22.37) 55.71 (17.69) 68.18 (17.27) Likert Happy – – 6.26 (1.48) 6 (1.84) – 5.89 (1.62) – – Anxious – – 3.64 (1.47) 4.66 (2.36) – 3.88 (2.2) – – Sad – – 3.58 (1.54) 3.86 (1.84) – 3.4 (1.88) – ST VAS 57.15 (21.33) 27.24 (17.13) 48.53 (16.87) 44.56 (22.71) 39.81 (25.95) 49.06 (20.06) 50.84 (16.34) 65.28 (18.86) Likert Happy – – 5.42 (1.89) 5.39 (2.14) – 5.41 (2.05) – – Anxious – – 4.07 (2.14) 5.25 (2.36) – 4.24 (2.47) – – Sad – – 3.88 (1.84) 4.18 (2.28) – 3.53 (2.11) – – PT positive training, ST sham training, VAS Visual Analogue Scale of general mood state, Likert Likert mood scales = happy mood (items: happy, optimistic, joyful), anxious mood (items: nervous, tense, anxious), sad mood (items: depressed, upset, sad), T1 pre-training directly after the negative mood induction, Pre-stress directly after the training 1 3 Cognitive Therapy and Research (2018) 42:408–420 417 Eec ff ts on Mood Reactivity and Recovery Contingency Awareness in Response to Stress The awareness check revealed that relatively more partici- VAS pants in the PT became aware of the training contingency than in the ST, 55% versus 12%; χ (74) = 15.07, p < .001. A 2 (group: PT, ST) × 4 (time: pre-speech, announcement, Moreover, participants in the PT reported more perceived during-speech, post-speech) RM ANOVA on the VAS, control over the eye-tracking task than participants in the revealed a significant time effect, F (3, 66) = 9.7, p < .001, ST, t(73) = 3.98, p < .001. Therefore, we repeated the main η = .31. Within-subject contrasts revealed that the speech analysis with awareness as an additional between-subjects task had its intended effects: The announcement of the factor. The 2 (group: PT, ST) × 2 (time: pre, post) × 2 (con- task resulted in a drop in mood, F(1, 68) = 6.16, p = .016, tingency awareness: yes, no) RM ANOVA of the sustained η = .08, which remained low during the speech task, attention bias revealed no 3-way interaction effect involv - F(1,68) = 2.61, p = .111, while mood increased again after ing awareness, F(1, 70) = 2.66, p = .107. The 2 (group: PT, the speech task, F(1, 68) = 28.78, p < .001, η = .3. For ST) × 2 (time: pre, post) × 2 (valence: positive, negative) × 2 means, see Table 3. However, the crucial time-by-group (contingency awareness: yes, no) RM ANOVA on the log- interaction was not significant, F (3, 66) = 0.46, p = .710, transformed attention bias indices did not indicate a signifi- indicating that PT and ST did not differentially affect cant 4-way interaction with awareness either, F(1,70) = 3.36, mood reactivity or recovery in response to the stress task. p = .071. Likert Scales Discussion The 2 (group: PT, ST) × 4 (time: pre-speech, announce- To address the limitations of previously used RT-based ment, during-speech, post-speech) RM ANOVA was ABM paradigms, a novel ABM paradigm based on eye- repeated for each of the three Likert scales (i.e., hap- tracking was recently developed (i.e., the ET-ABM; Fer- piness, anxiety and sadness). While happiness ratings rari et al. 2016). This paradigm was specifically designed to remained unaffected by the speech task, F (2, 67) = 0.7 assess and target the attentional components that are biased p = .449, anxiety and sadness ratings changed through- in depression: the disengagement from negative stimuli and out the speech task (Anxiety: F(2,67) = 14.25, p < .001, the maintained attention to positive stimuli. A first proof-of 2 2 η = .3; Sadness: F (2,67) = 10.03, p < .001, η = .23). principle study with healthy students showed that, compared Inspection of the means suggests that anxiety and sadness to a negative training version, the ET-ABM can induce a increased in response to the announcement (Anxiety: T2 positive sustained attention bias as well as faster disengage- M = 3.84 (SD = 1.8), T3 M = 4.93 (SD = 2.36); Sadness: ment from negative stimuli. The aim of the present study T2 M = 3.72 (SD = 1.68), T3 M = 4.01 (SD = 2.05)), and was to replicate these promising findings in an emotionally dropped again after the speech task (Anxiety: T5 M = 4.05 vulnerable sample of dysphoric students, with a placebo (SD = 2.32); Sadness T5 M = 3.46 (SD = 1.97)). However, sham-training as control condition. in line with the analysis of the VAS, the crucial interac- In line with the findings of Ferrari et al. (2016), the PT tion effect was not significant, suggesting that changes in induced a positive sustained attention bias (i.e., longer fixa- mood in response to the speech task did not differ between tions on positive than on negative stimuli). Notably, both PT groups (Anxiety: F(2, 67) = 0.19, p = .827; Sadness: F(2, and ST showed an increase in positive sustained attention 67) = 0.32, p = .73). bias. However, this increase was stronger in the PT group, supporting the effectiveness of the training in modifying Training Eec ff ts on State‑Rumination attentional processes in dysphoric individuals. As in the previous study, these general training effects were again We compared MRSI scores after the stress task between driven by reduced disengagement latencies from negative the two groups by means of an independent-samples t-test. stimuli in the PT group. Beyond that, the PT group also This analysis revealed no significant training effect on state showed an increase in maintained attention to the first fix- rumination (PT: M = 20.32 (SD = 6.85); ST: M = 21.91 ated positive pictures, suggesting that the training may also (SD = 6.69); t(86) = 0.98, p = .332). affect the initial processing of positive stimuli. In contrast, the ST did not induce any valence-specific attentional view - ing patterns. Instead, participants in this group became generally slower with directing their gaze away from the initially fixated pictures, resulting in slower disengagement 1 3 418 Cognitive Therapy and Research (2018) 42:408–420 from both negative and positive pictures. Summarizing, during the stress task either. In line with the previous study while the PT was effective in reversing an initially negative (Ferrari et al. 2016), PT and ST groups did not differ in their attentional bias into a positive attentional bias, character- mood reactivity or recovery from the speech challenge. It ized by relatively longer fixations on positive pictures and was therefore speculated that stress-attenuating effects of by relatively quicker disengagement from negative pictures, the training may be restricted to emotionally vulnerable sam- the ST resulted in a decline of the negative bias. A potential ples, as suggested by previous CBM research (Becker et al. explanation for the observed effect in the ST group might be 2016). Our sample did consist of individuals with elevated that the ST was possibly more difficult than the PT. Remark - depression scores. Hence, one possible interpretation of our ably, more participants in the PT became aware of the rein- results may be that in depression, the modification of atten- forced training pattern, and the PT group experienced a tional processes does not affect mood reactivity or recovery. stronger feeling of control over the task. It is likely that In the context of anxiety, the link between attentional bias the ST therefore was more tiring than the PT, resulting in and emotional vulnerability has been investigated in a range slower latencies in general. of studies (for a review, see Clarke et al. 2014), whereas In a previous study (Ferrari et al. 2016), the PT did not only a few studies have addressed this topic in depression. modify initial maintained attention to positive pictures. It Although Sanchez et al. (2013) did not experimentally had been suggested that the temporal criteria defining a manipulate attentional processes, they found that slower fixation as sufficiently long to continue a training trial (i.e., disengagement from negative stimuli predicted lower mood 1000 ms) might not be appropriate to induce “longer” main- recovery after stress. Together with the few studies show- tained attention to positive stimuli. As our main goal was ing that reducing a negative attentional bias may attenuate to replicate the earlier training effects on general sustained depressive symptoms (Browning et al. 2012; Wells and attention and attentional disengagement from negative stim- Beevers 2010; Yang et al. 2014), this provides evidence uli, we did not increase the required fixation duration on supporting a causal link between attentional bias and main- positive pictures. Nevertheless, in the current study, training tenance of depressed mood. It is important to note here, that effects were partially driven by the increased initial main - the latter studies all made use of multiple training sessions, tained attention to positive stimuli. This might be explained distributed over a longer period of time. However, given by the different stimulus sets used in the two studies. In the the limited number of studies and the contradicting conclu- previous study, negative and positive pictures were matched sions from meta-analyses regarding number of sessions as a on their valence ratings, whereas this was not done in the moderator of training effects (Beard et al. 2012; Cristea et al. current study. As a result, more extreme positive and nega- 2015; Hallion and Ruscio 2011; Mogoaşe et al. 2014), it tive pictures might have been presented during the training, remains to be investigated whether training effects on stress which possibly resulted in a better contrast between these reactivity and recovery, or even depressive symptoms, can valences. This change in contrast might have helped par- be achieved by increasing the number of training sessions. ticipants to more easily identify the two different responses Moreover, we would like to emphasize that the stress task required to react to the two different picture types, resulting employed in this study may not be optimal for measuring in the modic fi ation of both indices, disengagement from neg - transfer effects of the training to mood responses, which ative pictures and maintained attention to positive pictures. might be the reason why we found no relation between Although this explanation remains speculative, the current attentional processing and emotional reactivity. Although findings provide promising evidence that the ET-ABM may Sanchez et al. (2013) found a significant association of actually directly tap into both components of attention that slow disengagement from negative stimuli with lower mood are relevant in depression. Importantly, as different picture recovery after a speech challenge as used in our study, one sets were used during training and assessments, we may may question whether a performance-related speech-chal- further conclude that the observed training effects are not lenge can be considered a relevant stressor in the context merely the result of stimulus-specific response patterns, of depression. In fact, the link with attentional processes but reflect a modified attentional processing of emotionally was exclusively found for “sad” mood recovery. Hence, valenced information in general. before drawing firm conclusions about the causal role of To get a first indication of the potential therapeutic effects attentional bias in depressed mood, future research should of the ET-ABM, we additionally explored changes in mood. consider to increase the number of training sessions and In contrast to the first study, the PT and ST did not differ - investigate its effect on mood recovery after a depression- entially affect participants’ mood state. In general, positive relevant stressor. Such a stressor could for instance involve mood increased indistinguishably in both groups from before a video-clip that induces sad mood, as used at the beginning to after the training, possibly pointing to recovery from the of our experiment. negative mood induction at the beginning of the experiment. Subsequent studies using this paradigm might also want More importantly, the training did not affect mood changes to include other measurement instruments that can detect 1 3 Cognitive Therapy and Research (2018) 42:408–420 419 far transfer effects. Even though this study administered a Compliance with Ethical Standards free-viewing assessment task with different pictures than Conflict of Interest Martin Möbius, Gina Ferrari, Robin van den used during training, the free-viewing task shares several Bergh, Eni S. Becker, and Mike Rinck declare that they have no con- characteristics with the training paradigm. To rule out that flict of interest. only a single task-relevant component has been trained, we recommend to make additional use of alternative bias meas- Informed Consent All procedures performed in the study involving human participants were in accordance with the ethical standards of the ures, such as a spatial-cueing task (Baert et al. 2010) or the institutional research committee of the Behavioural Science Institute engagement-disengagement eye-tracking task by Sanchez of Radboud University, Nijmegen, the Netherlands, and with the 1964 et al. (2013). Moreover, as the current study does not allow Helsinki declaration and its later amendments or comparable ethical to attribute training effects to either of the two attentional standards. Informed consent was obtained from all individual partici- pants included in the study. components (i.e., negative disengagement or positive main- tained attention), follow-up studies are required to disen- Animal Rights No animal studies were carried out by the authors for tangle the specific working mechanism of the ET-ABM. this article. Finally, a measurement-only control condition might be a useful addition for future research, as sham-training pro- Open Access This article is distributed under the terms of the Creative cedures as implemented in this study may have training Commons Attribution 4.0 International License (http://creativecom- unspecific effects as well (e.g., Gladwin 2017; Wells and mons.org/licenses/by/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate Beevers 2010). credit to the original author(s) and the source, provide a link to the For these follow-up studies, we would strongly recom- Creative Commons license, and indicate if changes were made. mend to take all three attentional components into account (i.e., sustained attention, negative disengagement, positive maintained attention). The current study found no sustained attention bias at baseline, which might have been expected References based on the extensive literature on the existence of nega- tive attentional biases in depression (e.g., Peckham et al. Amir, N., Weber, G., Beard, C., Bomyea, J., & Taylor, C. T. (2008). The effect of a single-session attention modification program 2010). However, the fact that we found a slower disengage- on response to a public-speaking challenge in socially anxious ment from negative than from positive pictures points to the individuals. Journal of Abnormal Psychology, 117(4), 860–868. importance of looking into the different attentional com- https://doi.org/10.1037/a0013445. ponents separately, rather than exclusively investigating a Armstrong, T., & Olatunji, B. O. (2012). Eye tracking of attention in the affective disorders: A meta-analytic review and synthe- general bias. Differentiating between negative disengage- sis. Clinical Psychology Review, 32(8), 704–723. https://doi. ment and positive maintained attention may allow for a more org/10.1016/j.cpr.2012.09.004. sensitive measurement of attentional bias in depression. Baert, S., De Raedt, R., Schacht, R., & Koster, E. H. W. (2010). Atten- To summarize, this study provides further support for tional bias training in depression: Therapeutic effects depend on depression severity. Journal of Behavior Therapy and Experi- the effectiveness of the ET-ABM in modifying the specific mental Psychiatry, 41(3), 265–274. https://doi.org/10.1016/j. attentional components assumed to be causally involved in jbtep.2010.02.004. the development and maintenance of depression. In order Beard, C., Sawyer, A. T., & Hofmann, S. G. (2012). Efficacy of atten- to investigate whether ABM can indeed alleviate emotional tion bias modification using threat and appetitive stimuli: A meta- analytic review. Behavior Therapy, 43(4), 724–740. https://doi. vulnerability, it has been suggested that we need refined org/10.1016/j.beth.2012.01.002. or new paradigms which reliably assess and modify these Beck, A. T. (1967). Depression: Clinical, experimental, and theoretical processes (Clarke et al. 2014). This study suggests that the aspects. New York: Hoeber. ET-ABM task may indeed be such a paradigm that facili- Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York: International University Press. tates further progress in this field of research. Fortunately, Beck, A. T., Steer, R., & Brown, G. (1996). Beck depression inventory- the technical developments of the past decade made eye- II. San Antonio, TX: Psychological Corporation. tracking devices accessible at reasonable prices and suitable Becker, E. S., Ferentzi, H., Ferrari, G., Möbius, M., Brugman, S., for the use in hospitals without requiring the expertise of Custers, J., … Rinck, M. (2016). Always approach the bright side of life: A general positivity training reduces stress reactions in technicians. Therefore, the next step should be to investigate vulnerable individuals. Cognitive Therapy and Research, 40(1), the beneficial effects of the ET-ABM on clinically relevant 57–71. https://doi.org/10.1007/s10608-015-9716-2. measures in a patient sample. Beevers, C. G., Clasen, P. C., Enock, P. M., & Schnyer, D. M. (2015). Attention bias modification for major depressive disorder: Acknowledgements We would like to thank Sebastian Herrmann, Lea Effects on attention bias, resting state connectivity, and symp- Koch, Marleen Kragting, Lea Naczenski, Linda Neuburg, Lena Schäfer, tom change. Journal of Abnormal Psychology, 124, 463–475. and Kaspar Pel for helping with the data collection. The Behavioural https://doi.org/10.1037/abn0000049. Science Institute, Radboud University Nijmegen, supported this project Browning, M., Holmes, E. A., Charles, M., Cowen, P. J., & Harmer, financially. C. J. (2012). Using attentional bias modification as a cognitive 1 3 420 Cognitive Therapy and Research (2018) 42:408–420 vaccine against depression. Biological Psychiatry, 72(7), 572– and Anxiety, 27(12), 1135–1142. https://doi.org/10.1002/ 579. https://doi.org/10.1016/j.biopsych.2012.04.014. da.20755. Cisler, J. M., & Koster, E. H. W. (2010). Mechanisms of attentional Randall, D. M., & Cox, W. M. (2001). Experimental mood inductions biases towards threat in anxiety disorders: An integrative in persons at high and low risk for alcohol problems. American review. Clinical Psychology Review, 30(2), 203–216. https:// Journal of Drug and Alcohol Abuse, 27(1), 183–187. doi.org/10.1016/j.cpr.2009.11.003. Sanchez, A., Vazquez, C., Marker, C., LeMoult, J., & Joormann, J. Clarke, P. J. F., Notebaert, L., & MacLeod, C. (2014). Absence of (2013). Attentional disengagement predicts stress recovery in evidence or evidence of absence: Reflecting on therapeutic depression: An eye-tracking study. Journal of Abnormal Psychol- implementations of attentional bias modification. BMC Psy - ogy, 122(2), 303–313. https://doi.org/10.1037/a0031529. chiatry, 14, 8. https://doi.org/10.1186/1471-244X-14-8. Scher, C. D., Ingram, R. E., & Segal, Z. V. (2005). Cognitive reac- Cristea, I. A., Kok, R. N., & Cuijpers, P. (2015). Efficacy of cogni- tivity and vulnerability: Empirical evaluation of construct acti- tive bias modification interventions in anxiety and depression: vation and cognitive diatheses in unipolar depression. Clinical Meta-analysis. The British Journal of Psychiatry, 206(1), 7–16. Psychology Review, 25(4), 487–510. https://doi.org/10.1016/j. https://doi.org/10.1192/bjp.bp.114.146761. cpr.2005.01.005. Ellis, A. J., Beevers, C. G., & Wells, T. T. (2010). Attention allo- Schmukle, S. C. (2005). Unreliability of the dot probe task. European cation and incidental recognition of emotional information in Journal of Personality, 19(7), 595–605. https://doi.org/10.1002/ dysphoria. Cognitive Therapy and Research, 35(5), 425–433. per.554. https://doi.org/10.1007/s10608-010-9305-3. Sears, C. R., Thomas, C. L., LeHuquet, J. M., & Johnson, J. C. Ferrari, G. R. A., Möbius, M., van Opdorp, A., Becker, E. S., & S. (2010). Attentional biases in dysphoria: An eye-track- Rinck, M. (2016). Can’t look away: An eye-tracking based ing study of the allocation and disengagement of atten- attentional disengagement training for depression. Cognitive tion. Cognition & Emotion, 24(8), 1349–1368. https://doi. Therapy and Research, 40(5), 672–686. https://doi.org/10.1007/ org/10.1080/02699930903399319. s10608-016-9766-0. Seemüller, F., Meier, S., Obermeier, M., Musil, R., Bauer, M., Adli, Gladwin, T. E. (2017). Negative effects of an alternating-bias train- M., … Möller, H. J. (2014). Three-Year long-term outcome of ing aimed at attentional flexibility: A single session study. 458 naturalistically treated inpatients with major depressive epi- Health Psychology and Behavioral Medicine, 5(1), 41–56. sode: Severe relapse rates and risk factors. European Archives of https://doi.org/10.1080/21642850.2016.1266634. Psychiatry and Clinical Neuroscience, 264, 567–575. https://doi. Hallion, L. S., & Ruscio, A. M. (2011). A meta-analysis of the effect org/10.1007/s00406-014-0495-7. of cognitive bias modification on anxiety and depression. Psy - Spielberger, C. D. (1989). State-trait anxiety inventory: Bibliography chological Bulletin, 137(6), 940–958. https://doi.org/10.1037/ (2nd ed.). Palo Alto, CA: Consulting Psychologists Press. a0024355. Sprinkle, S. D., Lurie, D., Insko, S. L., Atkinson, G., Jones, G. L., Kellough, J. L., Beevers, C. G., Ellis, A. J., & Wells, T. T. (2008). Time Logan, A. R., & Bissada, N. N. (2002). Criterion validity, sever- course of selective attention in clinically depressed young adults: ity cut scores, and test-retest reliability of the Beck Depression An eye tracking study. Behaviour Research and Therapy, 46(11), Inventory-II in a university counseling center sample. Jour- 1238–1243. https://doi.org/10.1016/j.brat.2008.07.004. nal of Counseling Psychology, 49(3), 381–385. https://doi. Leyman, L., De Raedt, R., Schacht, R., & Koster, E. H. W. (2007). org/10.1037//0022-0126.96.36.1991. Attentional biases for angry faces in unipolar depression. Psy- Steinert, C., Hofmann, M., Kruse, J., & Leichsenring, F. (2014). The chological Medicine, 37(3), 393–402. https://doi.org/10.1017/ prospective long-term course of adult depression in general prac- S003329170600910X. tice and the community. A systematic literature review. Journal MacLeod, C., Rutherford, E., Campbell, L., Ebsworthy, G., & Holker, of Affective Disorders, 152–154, 65–75. https://doi.org/10.1016/j. L. (2002). Selective attention and emotional vulnerability: Assess- jad.2013.10.017. ing the causal basis of their association through the experimental Teasdale, J. D. (1988). Cognitive vulnerability to persistent depres- manipulation of attentional bias. Journal of Abnormal Psychology, sion. Cognition & Emotion, 2(3), 247–274. https://doi. 111(1), 107–123. https://doi.org/10.1037//0021-843x.111.1.107. org/10.1080/02699938808410927. Marchewka, A., Zurawski, Ł, Jednoróg, K., & Grabowska, A. (2014). Treynor, W., Gonzalez, R., & Nolen-Hoeksema, S. (2003). Rumi- The Nencki Affective Picture System (NAPS): Introduction to nation reconsidered: A psychometric analysis. Cognitive a novel, standardized, wide-range, high-quality, realistic picture Therapy and Research, 27(3), 247–259. https://doi.org/10.102 database. Behavior Research Methods, 46(2), 596–610. https:// 3/A:1023910315561. doi.org/10.3758/s13428-013-0379-1. Waechter, S., & Stolz, J. A. (2015). Trait anxiety, state anxiety, and Mastikhina, L., & Dobson, K. (2017). Biased attention retraining in attentional bias to threat: Assessing the psychometric properties of dysphoria: A failure to replicate. Cognition and Emotion, 31(3), response time measures. Cognitive Therapy and Research, 39(4), 625–631. https://doi.org/10.1080/02699931.2015.1136270. 441–458. https://doi.org/10.1007/s10608-015-9670-z. Mathews, A., & MacLeod, C. (2005). Cognitive vulnerability to emo- Wells, T. T., & Beevers, C. G. (2010). Biased attention and dysphoria: tional disorders. Annual Review of Clinical Psychology, 1, 167– Manipulating selective attention reduces subsequent depressive 195. https://doi.org/10.1146/annurev.clinpsy.1.102803.143916. symptoms. Cognition & Emotion, 24(4), 719–728. https://doi. Mogoaşe, C., David, D., & Koster, E. H. W. (2014). Clinical efficacy org/10.1080/02699930802652388. of attentional bias modification procedures: An updated meta- World Health Organization (WHO). (2017). Depression [Fact sheet]. analysis. Journal of Clinical Psychology, 70(12), 1133–1157. Retrieved May 5, 2017, from http://www.who.int/mediacentre/ https://doi.org/10.1002/jclp.22081. factsheets/fs369/en/. Mor, N., Marchetti, I., & Koster, E. H. W. (2013). The Momentary Yang, W., Ding, Z., Dai, T., Peng, F., & Zhang, J. (2014). Attention bias Ruminative Self-focus Inventory (MRSI): Validation and psycho- modification training in individuals with depressive symptoms: metric evaluation. Manuscript submitted for publication. A randomized controlled trial. Journal of Behavior Therapy and Peckham, A. D., McHugh, R. K., & Otto, M. W. (2010). A meta-analy- Experimental Psychiatry. Retrieved from http://www.sciencedi- sis of the magnitude of biased attention in depression. Depression rect.com/science/article/pii/S0005791614000755. 1 3
Cognitive Therapy and Research – Springer Journals
Published: Jan 16, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera