Content uploaded by Alvaro Sanchez-Lopez
Author content
All content in this area was uploaded by Alvaro Sanchez-Lopez on Jun 06, 2016
Content may be subject to copyright.
RUNNING HEAD: attention training and reappraisal
Attention training through gaze-contingent feedback:
Effects on reappraisal and negative emotions
Alvaro Sanchez, Jonas Everaert, and Ernst H. W. Koster
Ghent University, Belgium
Key words: Selective attention; attentional control; reappraisal; negative emotions; attention
bias modification
Abstract: 224 words
Article body (Introductory and discussion materials, and acknowledgments): 1,966 words
References: 40
Number of tables: 4
Number of figures: 7
Supplemental material: 1
* Corresponding author:
Alvaro Sanchez
Ghent University
Department of Experimental Clinical and Health Psychology
Henri Dunantlaan 2
B-9000 Ghent
Belgium
Tel: +0032 09 264 91 05
Fax: +0032 09 264 64 89
E-mail: alvaro.sanchezlopez@ugent.be
ATTENTION TRAINING AND REAPPRAISAL 2
Abstract
Reappraisal is central to emotion regulation but its mechanisms are unclear. This study tested the
theoretical prediction that emotional attention bias is linked to reappraisal of negative emotion-
eliciting stimuli and subsequent emotional responding using a novel attentional control training.
Thirty-six undergraduates were randomly assigned to either the control or the attention training
condition and were provided with different task instructions while they performed an
interpretation task. Whereas control participants freely created interpretations, participants in the
training condition were instructed to allocate attention toward positive words to efficiently create
positive interpretations (i.e., recruiting attentional control) while they were provided with gaze-
contingent feedback on their viewing behavior. Transfer to attention bias and reappraisal success
was evaluated using a dot-probe task and an emotion regulation task which were administered
before and after the training. The training condition was effective at increasing attentional
control and resulted in beneficial effects on the transfer tasks. Analyses supported a serial
indirect effect with larger attentional control acquisition in the training condition leading to
negative attention bias reduction, in turn predicting greater reappraisal success which reduced
negative emotions. Our results indicate that attentional mechanisms influence the use of
reappraisal strategies and its impact on negative emotions. The novel attention training highlights
the importance of tailored feedback to train attentional control. The findings provide an
important step toward personalized delivery of attention training.
ATTENTION TRAINING AND REAPPRAISAL 3
Introduction
Reappraising the meaning of an emotion-eliciting event to decrease its negative impact is
a powerful regulatory process integral to healthy as well as distorted emotional functioning
(Gross, 2014). In nonclinical samples, reappraisal is effective at increasing positive and
decreasing negative emotions (Augustine & Hemenover, 2009; Webb, Miles, & Sheeran, 2012)
and has been associated with better interpersonal functioning (Gross & John, 2003) as well as
enhanced stress recovery (Jamieson, Nock, & Mendes, 2012). Failures to reappraise have been
reported in depressed individuals (Aldao, Nolen-Hoeksema, & Schweizer, 2010; Johnstone, van
Reekum, Urry, Kalin, & Davidson, 2007). Models of reappraisal suggest that attentional
mechanisms determine reappraisal success (Joormann & D’Avanzato, 2010; Sheppes, Suri, &
Gross, 2015) and that unsuccessful reappraisal in depression can be related to attentional
impairments and biases in processing emotional material (De Raedt & Koster, 2010).
An upsurge of research examining the role of attention allocation in reappraisal of
emotional material has yielded equivocal findings. Some investigations have found that
participants spent less time looking at emotional regions of negative material when down-
regulating negative emotions (Manera, Samson, Pehrs, Lee, & Gross, 2014; van Reekum et al.,
2007), and that such shorter viewing times mediate the effect of reappraisal on the negative
emotions experienced (Manera et al., 2014). Other studies, however, have reported the reverse,
namely longer viewing times for emotional content during reappraisal (Bebko, Franconeri,
Ochsner, & Chiao, 2011) and that constraining participants’ gaze toward neutral vs. negative
regions of emotional pictures does not alter reappraisal success (Bebko, Franconeri, Ochsner, &
Chiao, 2014; Urry, 2010). One explanation for the mixed findings could be that effective
ATTENTION TRAINING AND REAPPRAISAL 4
reappraisal can be achieved in multiple ways, differentially recruiting attentional resources
(Morris, Leclerc, & Kensinger, 2014).
Thus far, research has been limitedly successful at characterizing the role of emotional
attention (bias) in reappraisal, especially because most studies are cross-sectional and preclude
claims regarding causality (but see Bebko et al., 2014). How attentional mechanisms directly
contribute to reappraisal has not been investigated thoroughly. One common technique to assess
and manipulate attentional allocation draws from the emotional dot-probe paradigm (MacLeod,
Rutherford, Campbell, Ebsworthy, & Holker, 2002). Here, participants learn to allocate attention
toward or away from threats via experimental contingencies between the presentation of stimuli
(either threatening or neutral) and a to-be-detected target. Although initial reports on the
effectiveness of dot-probe training were promising (Hakamata et al., 2010), recent meta-analyses
indicate that the training has only modest effects in modifying biases in attentional allocation
(Beard, Sawyer, & Hofmann, 2012; Mogoașe, David, & Koster, 2014). Factors limiting the
effectiveness of training are the limited benefit of picking up the contingency between a
particular emotional picture and the subsequently presented target in a dot-probe training task as
well as the lack of individualized feedback on trainees’ performance during attention training.
Initial empirical evidence suggests that providing online feedback during training increases
awareness of emotional biases in attention allocation, which can in turn increase regulatory
control (Bernstein & Zvielli, 2014; Schnyer et al., 2015). Novel attentional control training
techniques could improve training effects via tailored feedback to take full advantage of a
trainee’s ability to implement self-regulatory control in redirecting attention (i.e., attentional
control).
ATTENTION TRAINING AND REAPPRAISAL 5
The present study investigated the direct influence of emotional attention (bias) in
reappraisal using a novel attentional control training methodology implementing gaze-contingent
feedback. For this purpose, we modified a recently designed method to investigate attentional
bias during interpretation (Everaert, Duyck, & Koster, 2014; Sanchez, Everaert, De Putter,
Mueller, & Koster, 2015) into an attentional control training task. In the basic task design,
participants unscrambled emotional sentences (e.g., “future dismal very my bright looks”) using
five of the six words into either a positive or a negative sentence, while their gaze behavior was
recorded. In the training variant, participants were instructed to guide attention allocation toward
positive words to create positive sentences and were provided with gaze-contingent feedback on
their viewing times to positive vs. negative words of the emotional scrambled sentences.
Therefore, participants in the attentional control condition were explicitly instructed to
implement top-down regulation of their attention patterns (i.e., increase visual fixation to
positive over negative words), and received gaze-contingent feedback in order to maximize such
regulation of attention according to the explicitly instructed pattern. To assess the effectiveness
of the attentional control training in modifying attentional bias and reappraisal, we administered
a dot-probe task and an emotion regulation task before and after the modification procedure. We
hypothesized that gaze-contingent feedback in the training condition would increase attentional
control (i.e., significant changes in attention bias in the training group from a baseline to a
modification phase during the training), transferring to attentional bias on the dot-probe task, and
affecting reappraisal success in the emotion regulation task.
ATTENTION TRAINING AND REAPPRAISAL 6
Method
Participants
To obtain large variability in negative attentional bias, participants with minimal and
severe depressive symptoms were sampled from the Ghent University research participant pool
based on a prescreening measure (Mood and Anxiety Symptom Questionnaire; Watson, Clark,
Weber, Assenheimer, Strauss, & McCormick, 1995). At testing, 40 participants (33 women; 18-
29 years) reported a broad range of depressive symptom severity levels (range: 0–42, M=13.28,
SD=9.15) on the Beck Depression Inventory-II (Beck, Steer, & Brown, 1996; Van der Does,
2002). Sample size in the study was determined based on previous research reporting effects of
attentional control training procedures in attention bias modification (Bernstein & Zvielli, 2014).
Participants were paid 15 euro. The institutional review board approved the study protocol.
Design Overview
Figure 1 depicts the sequence of tasks. All participants completed a baseline phase
followed by either a training or control modification phase that was determined by random
assignment. Before the baseline and after the modification phase, participants completed the
emotion regulation and dot-probe tasks. An experimental session lasted approximately 85 min.
Training Procedure
Baseline phase. Eye movements were monitored via eye tracking while participants
completed an interpretation task, the Scrambled Sentences Test (Everaert, Duyck, et al., 2014).
On each trial (Figure 2), a neutral (e.g., “the I theatre visit cinema often”) or emotional (e.g., “am
winner born loser a I”) scrambled sentence was displayed following fixation (left-aligned to
elicit left-to-right reading). While the item was on-screen, participants were instructed to
ATTENTION TRAINING AND REAPPRAISAL 7
unscramble the sentence to form a grammatically correct and meaningful statement using five of
the six words as quickly as possible and within a time limit of 8000 ms (e.g., “I am a born
winner”). Upon completion, they pressed a button to continue and report their solution using the
numbers linked to the words of a scrambled sentence.
After a 3-trial practice phase with neutral scrambled sentences, 12 emotional scrambled
sentences were presented in random order. Participants then completed 6 filler neutral scrambled
sentences before starting the modification phase.
Modification phase. In both the training and control condition of the modification phase,
participants completed 8 blocks of 6 randomly presented emotional scrambled sentences. Again,
eye movements were registered while participants unscrambled the sentences. While the task in
the control condition was identical to the baseline task phase, several manipulations were made
in the training condition (Figure 3). First, participants were instructed to unscramble all
sentences into positive self-statements (Sanchez et al., 2015) and to focus attention on positive
words, as this would help to identify and form positive meanings more efficiently. Second,
participants received online feedback about their attentional deployment while unscrambling the
sentences. A red or green square respectively framed the negative or positive target each time the
eye-tracker detected a fixation. This online feedback aimed to help participants to quickly
disengage from negative information and maintain attention to positive information. Finally,
after each training block, participants received feedback comparing their gaze behavior during
the last block (e.g., “You looked 54% of the time at the positive word”) with gaze behavior
during the baseline phase (e.g., “You looked 42% of the time at the positive word”). This
procedure intends to increase awareness of the progress made in the training condition compared
with baseline.
ATTENTION TRAINING AND REAPPRAISAL 8
Stimuli. Sixty-nine scrambled sentences (60 emotional, 9 neutral sentences) were drawn
from a prior study (Everaert, Duyck, et al., 2014). All scrambled sentences were self-referent and
6 words long. Negative and positive target words in each emotional sentence (e.g., “winner” and
“loser” in “am winner born loser a I”) were matched between valence categories on word length,
word class, and word frequency (Duyck, Desmet, Verbeke, & Brysbaert, 2004), all F-s<1. Word
position within each scrambled sentence was randomized with the constraint that target words
occurred neither next to each other nor as the first or last word within a scrambled sentence.
Positive and negative target word order in emotional sentences was counterbalanced. The same
criteria were applied to target words in the neutral sentences.
Dependent variables. The eye-tracker recorded online the total fixation times (sum of
durations across fixations) on positive and negative target words in emotional scrambled
sentences during both the baseline and modification phase. An index of attention bias in
processing positive vs. negative material was computed by dividing the total fixation time on
positive words by the total fixation time on emotional (positive and negative) words (Everaert,
Duyck, et al., 2014; Sanchez et al., 2015) separately for each training phase (i.e., baseline phase
vs. modification phase). These attention bias indices served to test the hypothesis that
participants would implement attentional regulation (i.e., attentional control acquisition) in the
training condition, by showing significant increases in attention bias to positive over negative
material from the baseline to the modification phase.
Similarly, an index of interpretation bias was computed by dividing the number of
positively unscrambled sentences by the total number correctly completed emotional (positive
and negative) sentences (Everaert, Duyck, et al., 2014; Sanchez et al., 2015) separately for the
ATTENTION TRAINING AND REAPPRAISAL 9
baseline and modification phases, which served to test the change in interpretation bias from the
baseline to the modification phase.
Transfer of training
Attention bias
An emotional dot-probe task (MacLeod, Mathews, & Tata, 1986) indexed transfer of the
attentional control training to attention bias (Figure 4). After central fixation, each trial
simultaneously presented two words (positive–negative or neutral–neutral pairs) for 1000 ms at
either side of fixation (above vs. below fixation). After offset, a probe (“X”) appeared with equal
probability at the location of one of the stimuli. Participants were instructed to locate the probe
as quickly and accurately as possible by pressing the corresponding buttons.
Forty-eight positive–negative and 24 neutral–neutral word pairs were selected. Positive,
negative, and neutral words were matched on word length, word class, and word frequency
(Duyck et al., 2004), all F-s<1. The total set of 288 trials (72 word pairs × 2 word locations × 2
probe locations) was divided to create two dot-probe versions. Each version contained 144 trials
(96 positive-negative trials, 48 neutral-neutral trials) with word and probe location
counterbalanced. The two versions served as pre- and post-training procedure measures of
attention bias. Administration of the versions was counterbalanced across participants.
Data from the dot-probe tasks were trimmed to minimize the influence of outliers. Errors
and RTs < 150 ms and > 1500 ms were removed and then RTs falling more than 3 SDs from
each participant's mean RT were excluded (Everaert, Mogoase, David, & Koster, 2014).
Analyses were conducted on 98% of the data. An index of negative attention bias was calculated
for pre- and post-training task versions. RTs on trials with probes replacing negative words (i.e.,
congruent trials) were subtracted from RTs on trials with probes replacing positive words (i.e.
ATTENTION TRAINING AND REAPPRAISAL 10
incongruent trials). Higher scores indicate a stronger attentional bias for negative words
(MacLeod et al., 1986).
Emotion Regulation
An emotion regulation task (Figure 5) assessed transfer to reappraisal (Vanderhasselt,
Kühn, & De Raedt, 2013). On each trial, a negative picture was presented and, after 2000 ms,
participants rated their negative emotional experience on a 10-point scale (0 – ‘not at all’ to 9 –
‘very much’). A cue subsequently prompted them to appraise or reappraise the picture’s
meaning. When instructed to appraise, participants were asked to look at the picture and freely
experience the elicited feelings. When instructed to reappraise, participants were asked
reinterpret the picture’s meaning in a less negative way by changing the emotions, actions, and
outcomes of individuals depicted in the picture (Ochsner, Bunge, Gross, & Gabrieli, 2002). After
10 s, participants’ negative emotional experiences were reassessed using the same 9-point rating
scale. When instructed to reappraise, participants also provided a description of how they
reappraised the picture.
Stimuli and task versions. Thirty-two negative IAPS pictures (Lang, Bradley, &
Cuthbert, 2008) depicting depression-relevant themes (e.g., crying people, loneliness) were
selected based on arousal (M<4, range 4.30-7.93) and valence ratings (M<4, range 1.37-3.72).
Two sets of 16 pictures were created that differed neither on valence nor on arousal, all p-s > .05.
One set was presented before and the other after the training procedure, counterbalanced across
participants. In each task version, half of the pictures were appraised and the other half
reappraised. Pictures and regulatory instructions were randomly presented with the constraint
that maximum 2 pictures with the same regulatory instruction occurred consecutively.
ATTENTION TRAINING AND REAPPRAISAL 11
Dependent variables. First, reappraisal scores were computed using narrative
descriptions provided by participants. Two blind raters evaluated whether participants were
successful at generating reappraisals of negative scenes using a 5-point scale (0–No Description,
1–Not at all, 2–A little, 3–Good, 4–Very good). An intra-class correlation of .90 (p=.001)
indicated high inter-rater agreement. Reappraisal scores were computed by averaging the blind
raters' scores separately for the pre- and post-training emotion regulation tasks. Higher scores
indicate better reappraisal. Second, negative emotion scores were computed by averaging the
emotion ratings indicating the degree of negative emotions after viewing pictures, for the
appraisal and reappraisal trials.
Eye-tracker
A Tobii TX300 eye-tracker recorded gaze behavior during the dot-probe tasks and the
training procedure, with eye-gaze coordinates sampling at 300 Hz. Participants were seated
approximately 60 cm from the eye tracker. Visual fixations were considered when longer than
100 ms. Stimulus presentation and eye movement recording were controlled by E-prime
Professional software. E-prime extensions (TET and Clearview PackageCalls) converted eye
movement signals to visual fixation data, and computed and presented fixation time scores in the
training condition.
Results
Sample Characteristics
Four participants were excluded from the analyses due to problems in the detection of
gaze position, low quality of eye-tracking recordings (valid samples < 75%), or lack of fluency
in Dutch. The final sample size was 36 individuals (30 women; 18-29 years). The control and
ATTENTION TRAINING AND REAPPRAISAL 12
training condition did not significantly differ in age, t(33)=0.22, p=.83, gender ratio, χ²(1)=0.17,
p=.68, nor depressive symptoms, t(33)=-1.42, p=.16. Table 1 presents descriptives on all study
variables.
Training Effectiveness
A series of 2 (Condition: Training, Control) x 2 (Phase: Baseline, Modification) mixed-
design ANOVAs were conducted to examine effects of the attentional control training on
changes in interpretation bias and attention bias across the training procedure (i.e., from the
baseline to the modification phase). The first ANOVA employed the measures of interpretation
bias at each phase as dependent variable, whereas the second ANOVA employed the measures of
attention bias at each phase as dependent variables (see Table 1). Regarding interpretation bias
change, analyses revealed a significant main effect of Phase, F(1,34)=18.76, p=.001, 2=.36,
qualified by a Condition × Phase interaction, F(1,34)=18.20, p=.001, 2=.35. Follow-up
Bonferroni-corrected comparisons showed no differences between conditions at the Baseline
phase, F(1,34)=0.16, p=.69, 2=.01. At the Modification phase the training condition reported a
higher positive interpretation bias compared to the control condition, F(1,34)=37.38, p=.001,
2=.52. As expected, a significant increase in positive interpretations was found from Baseline to
Modification in the training, F(1,34)=36.96, p=.001, 2=.52, but not in the control condition,
F(1,34)=0.01, p=.96, 2=.01.
Regarding attention bias change, analyses revealed a main effect of Phase,
F(1,34)=15.74, p=.001, 2=.32, qualified by a Condition × Phase interaction, F(1,34)=31.14,
p=.001, 2=.48. Follow-up tests showed no differences between conditions in attention bias at
the Baseline phase, F(1,34)=1.62, p=.21, 2=.04, but significant differences at the Modification
phase, F(1,34)=38.76, p=.001, 2=.53, with participants in the training condition showing a
ATTENTION TRAINING AND REAPPRAISAL 13
larger positive attention bias (i.e. more time attending to positive over negative words in the
modification phase in comparison to participants in the control condition). As expected, there
was a significant increase in attention bias from Baseline to Modification phase in the training,
F(1,34)=45.58, p=.001, 2=.57, but not in the control condition, F(1,34)=1.30, p=.26, 2=.04
(i.e., significant increase in the time attending to positive over negative material from the
baseline to the modification phase in the training condition). Further comparisons of each
attention bias score at each phase of the procedure against a value of 0.5 (i.e., indicative of
absence of bias) showed no biases during the baseline phase in either the training or the control
condition, p=.51 and p=.99, respectively. In contrast, whereas the control group continued
showing absence of bias during the modification phase, p=.61, the increase in the trained pattern
in the training group was qualified by a bias to fixate more time in positive than in negative
material during the modification phase, p=.001. Overall, these results suggest that the attentional
control training was effective in increasing attention toward positive relative to negative
information by implementing attentional control on the trained pattern.
1
Transfer of Training
Given that prior research has revealed marked individual differences in the malleability of
attention bias through training (Clarke, Chen, & Guastella, 2012; Clarke, Macleod, & Shirazee,
2008; Everaert et al., 2014) and that there also was substantial variability in attention bias scores
both in the attentional control training and in the dot-probe task in the current study (see Table
1
Further analyses were conducted to establish whether the manipulations introduced in the training condition during
the modification phase would also lead to a faster performance on the task (i.e., faster times to unscramble sentences
at the modification phase in comparison to the baseline phase in the training condition). A 2 (Condition: Training,
Control) x 2 (Phase: Baseline, Modification) mixed-design ANOVA with the mean time to unscramble the sentences
as dependent variable showed a marginally significant Condition by Phase interaction, F(1,32)=3.35, p=.07, 2=.09.
Bonferroni-corrected comparisons showed a significant reduction in the time to perform the task from the Baseline
to Modification Phase in the training condition, Baseline: 4728 ms (SD= 1107), Modification: 4111 ms (SD=1089),
p=.001, whereas a similar trend did not reach significance in the control condition, Baseline: 4571 ms (SD= 1190),
Modification: 4335 ms (SD=913), p=.11. These results suggest that the training group increased their efficiency to
perform the task as the result of the manipulation procedures introduced during the modification phase.
ATTENTION TRAINING AND REAPPRAISAL 14
1), we focused the statistical analyses on the individual differences when evaluating transfer of
the attentional control training to the dot-probe and emotion regulation tasks.
Individual differences were analyzed via residualized change scores, constructed using
simple linear regression models (Segal et al., 2006). First, attentional control acquisition in the
training procedure was indexed by computing change scores in attention bias from the baseline
to the modification phase. Attention bias scores during the baseline phase were entered in a
simple regression model as predictor of attention bias scores during the modification phase. The
resulting standardized residuals served as a measure of attentional control acquisition. Second, in
a similar way, changes in attention bias in the dot-probe task, and in reappraisal and negative
emotions after reappraisal in the emotion regulation task were indexed by computing change
scores from the pre- to the post-training for each of these measures separately. Each simple
regression model regressed the post-training score on the pre-training score (i.e., time 1 score
predict time 2 score, repeated for: attention bias in the dot-probe task, reappraisal and negative
emotions after reappraisal in the emotion regulation task). The resulting standardized residuals of
each regression model served as change scores. Using standardized residuals is a reliable method
to control for variability among differences in the baseline scores (Segal et al., 2006). Table 2
presents correlations among the change scores.
Serial mediation models were used to examine effects of the modification condition (i.e.
training vs. control) on changes in dot-probe attention bias, reappraisal, and negative emotions,
via attentional control acquisition. After testing the significance of the total and direct effects, the
significance of the indirect effect for each model was tested using a 5000 samples bias-corrected
bootstrapping procedure (Preacher & Hayes, 2008). Bootstrapping is a nonparametric approach
to estimate the magnitude and significance of indirect effects and is recommended for use with
ATTENTION TRAINING AND REAPPRAISAL 15
small samples (Preacher & Hayes, 2008). The estimated 95% bootstrap confidence intervals
should not contain 0 to be significant (Preacher & Hayes, 2008). The effect size of each indirect
effect model was derived by computing partially standardized indirect effects. This approach is
indicated when the predictor variable is a dichotomous variable in which the two groups differ
by one unit ( i.e., 0 – Control condition, 1 – Modification condition; see Preacher & Kelley,
2011). Partially standardized effect sizes can be interpreted as the number of standard deviations
in the outcome that the groups differed on average as result of the indirect mechanisms tested.
Further results from single-step multiple regression analyses testing each mediational model are
provided as supplemental material.
Transfer to the dot-probe task
A first mediation model examined the effect of modification condition (training vs.
control) on dot-probe attention bias change via attentional control acquisition. Neither the total
effect, c=-.29 (SE=.34), t=-0.85, p=.40, 95%-CI: [-.9834, .4033], nor the direct effect, c’=.66
(SE=.42), t=-1.57, p=.13, 95%-CI: [-.1952, 1.5203], were significant. The indirect effect was
negative (coefficient=-.95, SE=.63) and statistically different from zero, 95%-CI: [-2.4692, -
.0092], supporting the model. Partially standardized indirect effect of the model was -.93
(SE=.52; 95%-CI: [-2.0829, -.0050]), showing that the training modification condition was
associated with decreases of .93 standard deviations in negative attention bias on the dot-probe
via its effect on attentional control acquisition (Figure 6).
Transfer to emotion regulation
Effects on reappraisal. We tested a serial mediation model in which the modification
condition predicts attentional control acquisition which in turn predicts attention bias change in
the dot-probe and this predicts reappraisal change (outcome variable). The total effect, c=.06
ATTENTION TRAINING AND REAPPRAISAL 16
(SE=.33), t=0.18, p=.86, 95%-CI: [-.6092, .7283], and direct effect, c’=.40 (SE=.44), t=0.90,
p=.37, 95%-CI: [-.4965, 1.2907], were not significant. The indirect effect was positive
(coefficient=.47, SE=.41) and statistically different from zero, 95%-CI: [.0061, 1.6238].
Importantly, neither of the alternative models, where mediators were removed one-by-one (i.e.,
becoming covariates and, therefore, controlling for their influence on the other predictors), were
significant (Table 3). Therefore, the only statistically supported indirect effect path was the one
hypothesized. Partially standardized indirect effect of the supported model was .49 (SE=.41;
95%-CI: [.0038, 1.5871]). Thus, the training modification condition was indirectly associated
with 0.49 standard deviations of reappraisal improvement via its effect on attention control
acquisition, which in turn predicted attention bias change, which was associated with reappraisal
change (Figure 7).
Effects on negative emotion. A final serial mediation model was tested adding negative
emotional state after reappraisal to the previously validated indirect effect model: modification
condition → attentional control acquisition → attention bias change → reappraisal change →
negative emotional state after reappraisal (Figure 7). The total effect, c=.18 (SE=.29), t=0.62,
p=.54, 95%-CI: [-.4198, .7911], and direct effect, c’=.39 (SE=.40), t=0.97, p=.34, 95%-CI: [-
.4307, 1.2111], were not significant. The indirect effect was negative (coefficient =-.22, SE=.22)
and statistically different from zero, 95%-CI: [.-9512, -.0045], whereas neither of the alternative
models were significant (Table 4). Therefore, the only statistically supported indirect effect path
was the one hypothesized. The partially standardized indirect effect of the supported model was -
.25 (SE=.26; 95%-CI: [-1.0763, -.0034]). Therefore, the training modification condition
indirectly led to decreases by 0.25 standard deviations in negative emotion after reappraisal (i.e.,
better emotion regulation), via its influence in attention control acquisition, the influence of
ATTENTION TRAINING AND REAPPRAISAL 17
attentional control acquisition on attention bias change in the dot-probe, the influence of
attention bias change on reappraisal change, and the influence of reappraisal change on negative
emotion change.
Discussion
This study tested whether alteration of attention bias via a novel attentional control
training task with gaze-contingent feedback would influence attention bias and reappraisal
success assessed by transfer tasks. The results indicate that the training vs. control condition had
an indirect effect on negative emotion repair: attentional control training led to attentional
control acquisition during the training procedure, which predicted attention bias change in the
dot-probe task, which in turn was associated with reappraisal change, which regulated negative
emotion change. These transfer effects of attentional control training were specifically observed
for individuals in the training and not in the control condition. The large individual differences in
attentional control (acquisition) during the training indicated that the training was particularly
effective for a subset of the trained individuals. The results support the proposal that attentional
control training can modify attention bias, which in turn influences the use of cognitive
reappraisal to decrease negative emotions.
Theoretical models of emotion regulation hypothesize that attentional mechanisms are
causally related to emotion regulation strategies, including reappraisal and its impact on negative
emotions (De Raedt & Koster, 2010; Joormann & D’Avanzato, 2010; Sheppes et al., 2015).
Consistent with the theoretical predictions and prior research suggesting that down-regulating
negative emotions is guided by less attention toward negative information (Manera et al., 2014;
van Reekum et al., 2007), the present study showed that modifying attention bias via attentional
ATTENTION TRAINING AND REAPPRAISAL 18
control training influences reappraisal and negative emotionality. However, the present findings
are in contrast with studies that have reported longer viewing times for negative content during
reappraisal (Bebko et al., 2011) and no effects of attention manipulations on reappraisal success
(Bebko et al., 2014; Urry, 2010). While there could be multiple routes to effective reappraisal
(Morris et al., 2014), another explanation for this inconsistency is that these studies have only
targeted overt attentional processes (i.e., fixating gaze position to a certain region of a negative
picture) and do not control covert attentional shifts which may also explain reappraisal success.
In the present study, both overt (i.e., eye movement indices) and covert (i.e., RTs on the dot-
probe task) attentional shifts were indexed and related to reappraisal success. The established
relation between attention and emotion reappraisal points to the importance of considering
attentional mechanisms in understanding (and treating disordered) emotional wellbeing.
The novel training paradigm applied in this study may provide a promising tool to
improve emotion regulation difficulties in remitted and/or clinically depressed individuals
(Aldao et al., 2010; Johnstone et al., 2007). Interestingly, previous research (Sanchez et al.,
2015) has shown that ability-related processes recruiting attentional control act as an intervening
variable in the relation between depressive symptom levels and interpretation bias. Therefore,
procedures increasing attentional control may help to reduce emotional dysregulation related to
depression by affecting attentional mechanisms involved in successful reappraisal. Here, two
aspects of the developed training procedure seem to be critical towards explaining its beneficial
effects. First, different from the presentation of words or pictures in a standard dot-probe
training, the present training procedure provided trainees with specific contexts (i.e., the content
of the scrambled sentences), instructions, and feedback to help them in considering positive
meanings in a self-referent manner. Second, the current training procedure provided
ATTENTION TRAINING AND REAPPRAISAL 19
individualized feedback on trainees’ attention allocation performance during the training in order
to maximize the regulation of attention according to an explicitly instructed pattern (i.e.,
intentionally (re)direct attention to positive information to form positive self-referent meanings).
Both individualized feedback procedures involved the use of voluntary top-down strategies
according to the instruction (online feedback: intentionally inhibit attention from negative words
when they are fixated and maximize intentional visual search of positive words; feedback on
performance between blocks: increase awareness on emotional biases to increase regulatory
control in redirecting attention in subsequent trials; Bernstein & Zvielli, 2014; Schnyer et al.,
2015). Additionally, online gaze-contingent feedback may also tap into stimulus-driven bottom-
up factors that have been found to be relevant for the modification of emotional biases in
attention (see Price, Greven, Siegle, Koster & De Raedt, 2016). Further studies will require
disentangling the specific effects of each of the feedback procedures comprising this new
approach. It is noteworthy that the training and control condition differed on a number of
elements, with the training including not only (1) different types of feedback, but also (2) the
instruction to form positive sentences. One might wonder which components are most important
in training attention. Noteworthy, in a previous study we have used a version of the SST where
we instructed to form positive sentences without providing gaze-contingent feedback (in relation
to a different research question; Sanchez et al., 2015). We found that attention did not change
substantially in the SST where individuals were only instructed to form positive sentences,
suggesting that feedback is a crucial element. Further studies should test the contribution of each
of the feedback procedures in comparison to a condition only instructing to form positive scenes
(without receiving feedback), helping to disentangle the specific adding of each feedback
procedure to the explicit instruction.
ATTENTION TRAINING AND REAPPRAISAL 20
Broadly, this training is an important step towards more personalized and advanced
cognitive training where biological indices directly provide feedback about performance. Future
research will require extending the present findings by testing the transfer effects of this
intervention to other sources of emotional information (e.g., effects in attention bias for
emotional faces and scenes) using different methodologies to target covert and overt processes
sub-serving attention bias as well as different indices of emotional functioning (i.e., self-reported
subjective mood and objective physiological indicators of emotional functioning).
Some limitations should be acknowledged. First, the study was conducted in a nonclinical
sample, which may limit generalizability of the findings to clinical samples. Given that attention
biases observed in nonclinical samples of individuals with varying depression levels often differ
from clinical samples in terms of degree rather than type (Baert, De Raedt, & Koster, 2010; Beck
& Haigh, 2014), it can be expected that clinically depressed individuals experiencing profound
attentional control and emotion regulation impairments (De Raedt & Koster, 2010; Joormann &
D’Avanzato, 2010) may benefit more from the attentional control training tested in this study.
Future research, however, needs to address this open question. Second, these results show that
attentional mechanisms contribute to reappraisal and emotional experiences, but we did not
assess clinical outcomes. An extended multiple session variant of the attentional control training
may need to examine the long-term endurance of the effects observed in the current study as well
as to test its effectiveness in improving depressive symptom severity and quality of life. Finally,
the large individual differences in attentional control acquisition and the transfer of training
showed that a subset of individuals took advantage of the training. Future research efforts need to
identify (cognitive) markers to preselect individuals who may benefit from attentional control
training procedures.
ATTENTION TRAINING AND REAPPRAISAL 21
In this study, attentional control training modified emotional attention bias, which in turn
was associated with reappraisal and a reduction in experienced negative emotions. Our study
provides clear evidence on the link between attentional mechanisms and reappraisal. Moreover,
our new training provides an important step to personalized cognitive training of attention.
ATTENTION TRAINING AND REAPPRAISAL 22
Acknowledgements
This research was supported by a grant of the Research Foundation Flanders (FWO,
reference 117438) awarded to Alvaro Sanchez and a grant BOF10/GOA/014 for a Concerted
Research Action of Ghent University awarded to Ernst Koster. We are very grateful to Igor
Marchetti and Megan T. deBettencourt for their assistance and comments on earlier versions of
the manuscript. We also thank Jeroen Barbé, Naomi De Borger, and Ena Coenen for their help in
the data collection.
ATTENTION TRAINING AND REAPPRAISAL 23
References
Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emotion-regulation strategies across
psychopathology: A meta-analytic review. Clinical Psychology Review, 30(2), 217–237.
http://doi.org/10.1016/j.cpr.2009.11.004
Augustine, A. a., & Hemenover, S. H. (2009). On the relative effectiveness of affect regulation
strategies: A meta-analysis. Cognition & Emotion, 23(6), 1181–1220.
http://doi.org/10.1080/02699930802396556
Baert, S., De Raedt, R., & Koster, E. (2010). Depression-related attentional bias: The influence
of symptom severity and symptom specificity. Cognition & Emotion, 24(6), 1044–1052.
http://doi.org/10.1080/02699930903043461
Beard, C., Sawyer, A. T., & Hofmann, S. G. (2012). Efficacy of attention bias modification using
threat and appetitive stimuli: A meta-analytic review. Behavior Therapy, 43, 724–740.
http://doi.org/10.1016/j.beth.2012.01.002
Bebko, G. M., Franconeri, S. L., Ochsner, K. N., & Chiao, J. Y. (2011). Look before you
regulate: Differential perceptual strategies underlying expressive suppression and cognitive
reappraisal. Emotion (Washington, D.C.), 11(4), 732–742. http://doi.org/10.1037/a0024009
Bebko, G. M., Franconeri, S. L., Ochsner, K. N., & Chiao, J. Y. (2014). Attentional Deployment
Is Not Necessary for Successful Emotion Regulation via Cognitive Reappraisal or
Expressive Suppression. Emotion (Washington, D.C.), 14(3), 504–512.
http://doi.org/10.1037/a0035459
Beck, A. T., & Haigh, E. A. P. (2014). Advances in cognitive theory and therapy: The generic
cognitive model. Annual Review of Clinical Psychology, 10, 1–24.
http://doi.org/10.1146/annurev-clinpsy-032813-153734
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory -
II. San Antonio, TX: Psychological Corporation.
Bernstein, A., & Zvielli, A. (2014). Attention Feedback Awareness and Control Training (A-
FACT): Experimental test of a novel intervention paradigm targeting attentional bias.
Behaviour Research and Therapy, 55(1), 18–26. http://doi.org/10.1016/j.brat.2014.01.003
Clarke, P. J. F., Chen, N. T. M., & Guastella, A. J. (2012). Prepared for the best: Readiness to
modify attentional processing and reduction in anxiety vulnerability in response to therapy.
Emotion, 12(3), 487–494. http://doi.org/10.1037/a0025592
Clarke, P., Macleod, C., & Shirazee, N. (2008). Prepared for the worst: readiness to acquire
threat bias and susceptibility to elevate trait anxiety. Emotion (Washington, D.C.), 8(1), 47–
57. http://doi.org/10.1037/1528-3542.8.1.47
ATTENTION TRAINING AND REAPPRAISAL 24
De Raedt, R., & Koster, E. H. W. (2010). Understanding vulnerability for depression from a
cognitive neuroscience perspective: A reappraisal of attentional factors and a new
conceptual framework. Cognitive, Affective & Behavioral Neuroscience, 10(1), 50–70.
http://doi.org/10.3758/CABN.10.1.50
Duyck, W., Desmet, T., Verbeke, L. P. C., & Brysbaert, M. (2004). WordGen: A tool for word
selection and nonword generation in Dutch, English, German, and French. Behavior
Research Methods, Instruments, & Computers, 36(3), 488–499.
http://doi.org/10.3758/BF03195595
Everaert, J., Duyck, W., & Koster, E. H. W. (2014). Attention, interpretation, and memory biases
in subclinical depression: A proof-of-principle test of the combined cognitive biases
hypothesis. Emotion, 14(2), 331–340. http://doi.org/10.1037/a0035250
Everaert, J., Mogoase, C., David, D., & Koster, E. H. W. (2014). Attention bias modification via
single-session dot-probe training: Failures to replicate. Journal of Behavior Therapy and
Experimental Psychiatry, 1–8. http://doi.org/10.1016/j.jbtep.2014.10.011
Gross, J. J. (2014). Emotion Regulation: Conceptual and Empirical Foundations. In J. J. Gross
(Ed.), Handbook of emotion regulation (pp. 3-24). New York, NY: The Guilford Press.
Gross, J. J., & John, O. P. (2003). Individual differences in two emotion regulation processes:
implications for affect, relationships, and well-being. Journal of Personality and Social
Psychology, 85(2), 348–362.
Hakamata, Y., Lissek, S., Bar-Haim, Y., Britton, J. C., Fox, N. a., Leibenluft, E., … Pine, D. S.
(2010). Attention bias modification treatment: A meta-analysis toward the establishment of
novel treatment for anxiety. Biological Psychiatry, 68(11), 982–990.
http://doi.org/10.1016/j.biopsych.2010.07.021
Jamieson, J. P., Nock, M. K., & Mendes, W. B. (2012). Mind over matter: Reappraising arousal
improves cardiovascular and cognitive responses to stress. Journal of Experimental
Psychology: General General, 141(3), 417–22. http://doi.org/10.1037/a0025719
Johnstone, T., van Reekum, C. M., Urry, H. L., Kalin, N. H., & Davidson, R. J. (2007). Failure
to regulate: counterproductive recruitment of top-down prefrontal-subcortical circuitry in
major depression. The Journal of Neuroscience : The Official Journal of the Society for
Neuroscience, 27(33), 8877–84. http://doi.org/10.1523/JNEUROSCI.2063-07.2007
Joormann, J., & D’Avanzato, C. (2010). Emotion regulation in depression: Examining the role of
cognitive processes. Cognition & Emotion, 24(6), 913–939.
http://doi.org/10.1080/02699931003784939
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2008). International affective picture system
(IAPS): Affective ratings of pictures and instruction manual. Gainesville, FL: University of
Florida.
ATTENTION TRAINING AND REAPPRAISAL 25
MacLeod, C., Mathews, A., & Tata, P. (1986). Attentional bias in emotional disorders. Journal
of Abnormal Psychology, 95(1), 15–20. http://doi.org/10.1037/0021-843X.95.1.15
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(1),
107–123. http://doi.org/10.1037/0021-843X.111.1.107
Manera, V., Samson, A. C., Pehrs, C., Lee, I. A., & Gross, J. J. (2014). The eyes have it: The role
of attention in cognitive reappraisal of social stimuli. Emotion, 14(5), 833–839.
http://doi.org/10.1037/a0037350
Mogoașe, C., David, D., & Koster, E. H. W. (2014). Clinical efficacy of attentional bias
modification procedures: An updated meta-analysis. Journal of Clinical Psychology,
70(12), 1133–1157. http://doi.org/10.1002/jclp.22081
Morris, J. a, Leclerc, C. M., & Kensinger, E. a. (2014). Effects of valence and divided attention
on cognitive reappraisal processes. Social Cognitive and Affective Neuroscience, 1–10.
http://doi.org/10.1093/scan/nsu004
Ochsner, K. N., Bunge, S. a, Gross, J. J., & Gabrieli, J. D. E. (2002). Rethinking feelings: an
FMRI study of the cognitive regulation of emotion. Journal of Cognitive Neuroscience,
14(8), 1215–1229. http://doi.org/10.1162/089892902760807212
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and
comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3),
879–91. http://doi.org/10.3758/BRM.40.3.879
Preacher, K. J., & Kelley, K. (2011). Effect size measures for mediation models: quantitative
strategies for communicating indirect effects. Psychological Methods, 16(2), 93–115.
http://doi.org/10.1037/a0022658
Price, R. B., Greven, I. M., Siegle, G. J., Koster, E. H. W., & De Raedt, R. (2016). A novel
attention training paradigm based on operant conditioning of eye gaze: Preliminary
findings. Emotion, 16(1), 110–116. doi:10.1037/emo0000093
Sanchez, A., Everaert, J., De Putter, L., Mueller, S., & Koster, E. H. W. (2015). Life is … great !
Emotional attention during instructed and uninstructed ambiguity resolution in relation to
depressive symptoms. Biological Psychology, 109, 67-72.
http://doi.org/10.1016/j.biopsycho.2015.04.007
Schnyer, D. M., Beevers, C. G., deBettencourt, M. T., Sherman, S. M., Cohen, J. D., Norman, K.
a, & Turk-Browne, N. B. (2015). Neurocognitive therapeutics: from concept to application
in the treatment of negative attention bias. Biology of Mood & Anxiety Disorders, 5(1), 16–
19. http://doi.org/10.1186/s13587-015-0016-y
ATTENTION TRAINING AND REAPPRAISAL 26
Segal, Z. V, Kennedy, S., Gemar, M., Hood, K., Pedersen, R., & Buis, T. (2006). Cognitive
reactivity to sad mood provocation and the prediction of depressive relapse. Archives of
General Psychiatry, 63(7), 749–755. http://doi.org/10.1001/archpsyc.63.7.749
Sheppes, G., Suri, G., & Gross, J. J. (2015). Emotion Regulation and Psychopathology. Annual
Review of Clinical Psychology, 11, 379–405. http://doi.org/10.1146/annurev-clinpsy-
032814-112739
Urry, H. L. (2010). Seeing, thinking, and feeling: emotion-regulating effects of gaze-directed
cognitive reappraisal. Emotion (Washington, D.C.), 10(1), 125–135.
http://doi.org/10.1037/a0017434
Van der Does, A. J. W. (2002). Handleiding bij de Nederlandse versie van de Beck Depression
Inventory - second edition (BDI-II-NL). [The Dutch version of the Beck Depresion
Inventory-II]. Lisse, The Netherlands: Swets & Zeitlinger.
Van Reekum, C. M., Johnstone, T., Urry, H. L., Thurow, M. E., Schaefer, H. S., Alexander, A.
L., & Davidson, R. J. (2007). Gaze fixations predict brain activation during the voluntary
regulation of picture-induced negative affect. NeuroImage, 36(3), 1041–1055.
http://doi.org/10.1016/j.neuroimage.2007.03.052
Vanderhasselt, M. A., Kühn, S., & De Raedt, R. (2013). “Put on your poker face”: Neural
systems supporting the anticipation for expressive suppression and cognitive reappraisal.
Social Cognitive and Affective Neuroscience, 8(8), 903–910.
http://doi.org/10.1093/scan/nss090
Webb, T. L., Miles, E., & Sheeran, P. (2012). Dealing with feeling: A meta-analysis of the
effectiveness of strategies derived from the process model of emotion regulation.
Psychological Bulletin, 138(4), 775–808. http://doi.org/10.1037/a0027600
Watson, D., Clark, L. A., Weber, K., Assenheimer, J. S., Strauss, M. E., & McCormick R. A.
(1995) Testing a tripartite model: I. Evaluating the convergent and discriminant validity of
anxiety and depression symptom scales. Journal of Abnormal Psychology, 104, 3–14.
http://doi.org/10.1037/0021-843X.104.1.3
ATTENTION TRAINING AND REAPPRAISAL 27
Table 1. Sample characteristics and descriptive statistics.
Variables
Training (N=18)
M SD
Control (N=18)
M SD
Gender (male/female)
3/15
3/15
Age
21.06
3.26
21.35
4.77
BDI-II
14.78
10.61
10.35
7.38
Attentional control training indices
Interpretation bias baseline phase (prop)
0.72
0.24
0.74
0.20
Interpretation bias modification phase (prop)
0.99
0.01
0.75
0.17
Attention bias baseline phase (prop)
0.50
0.04
0.52
0.04
Attention bias modification phase (prop)
0.63
0.08
0.50
0.03
Pre-training indices
Dot-probe RT congruent trials (ms)
525.34
101.07
479.43
62.11
Dot-probe RT incongruent trials (ms)
522.61
112.97
479.75
67.03
Dot-probe attention bias (d score in ms)
-2.72
34.07
0.32
16.28
Reappraisal (range 0-4)
2.10
0.71
1.90
0.77
Negative emotion after appraisal (range 0-9)
5.92
1.56
5.90
1.75
Negative emotion after reappraisal (range 0-9)
4.62
1.24
4.85
1.76
Post-training indices
Dot-probe RT congruent trials (ms)
506.52
116.53
437.71
53.14
Dot-probe RT incongruent trials (ms)
494.12
126.40
436.63
52.61
Dot-probe attention bias (d score in ms)
-12.39
43.94
-1.08
17.88
Reappraisal (range 0-4)
2.11
0.75
1.97
0.77
Negative emotion after appraisal (range 0-9)
5.51
1.84
5.46
2.08
Negative emotion after reappraisal (range 0-9)
4.74
1.66
4.64
1.74
Notes. M = Mean; SD = Standard deviation; RT = reaction time; ms = millisecond; prop = proportion
ATTENTION TRAINING AND REAPPRAISAL 28
Table 2. Correlations between the training condition and change scores.
Variables
1
2
3
4
5
1. Training Condition
(0 – Control; 1 – Training)
–
.70**
-.14
.03
.11
2. Attentional control acquisition (training)
–
-.45**
.02
.04
3. Attention bias change (dot-probe)
–
-.39*
.08
4. Reappraisal change
–
-.45**
5. Negative Emotion after Reappraisal change
–
Note. +p<.10; *p<.05; **p<.01
ATTENTION TRAINING AND REAPPRAISAL 29
Table 3. Indirect effect models tested with condition as predictor, reappraisal change as outcome, and attentional control acquisition
and attention bias change as potential mediators
Indirect Effect
Model
Effect (a x b)
SE
CI (lower)
CI (upper)
Total
-.3376
.3653
-1.1590
.2418
1 (Alternative)
-.4821
.3502
-1.2867
.0621
2 (Hypothesized)
.4747
.4060
.0061
1.6238
3 (Alternative)
-.3302
.3349
-1.3122
.0509
Indirect Effect Models:
1: Condition → Attentional control acquisition → Reappraisal change; n.s.
2 (Hypothesized): Condition → Attentional control acquisition → Attention bias change → Reappraisal change; p < .05
3: Condition → Attention bias change → Reappraisal change; n.s.
Notes: SE = Standard error; CI (lower) = lower bound of a 95% confidence interval; CI (upper) = upper bound; → = affects; n.s. = non-significant
ATTENTION TRAINING AND REAPPRAISAL 30
Table 4. Indirect effect models tested with Condition as predictor, Negative Emotions after Reappraisal change as outcome, and
Attentional Control Acquisition, Attention bias change and Reappraisal change as potential mediators
Indirect Effect
Model
Effect (a x b)
Boot SE
Boot LLCI
Boot ULCI
Total
-.2045
.4740
-1.3209
.4598
1 (Alternative)
-.2189
.4671
-1.6448
.3342
2 (Alternative)
.1394
.3022
-.1305
1.3130
3 (Alternative)
.2276
.1768
-.0066
.7078
4 (Hypothesized)
-.2241
.2247
-.9512
-.0045
5 (Alternative)
-.0970
.2491
-1.1755
.0879
6 (Alternative)
.1559
.1860
-.0188
7989
7 (Alternative)
-.1875
.2420
-.7696
.1492
Indirect Effect Models:
1: Condition → Attentional control acquisition → Negative emotions change; n.s.
2: Condition → Attentional control acquisition → Attention bias change → Negative emotions change; n.s.
3: Condition → Attentional control acquisition → Reappraisal change → Negative emotions change; n.s.
4 (Hypothesized): Condition → Attentional control acquisition → Attention bias change → Reappraisal change → Negative emotions change; p < .05
5: Condition → Attention bias change → Negative emotions change; n.s.
6: Condition → Attention bias change → Reappraisal change → Negative emotions change; n.s.
7: Condition → Reappraisal change → Negative emotions change; n.s.
Notes: SE = Standard error; CI (lower) = lower bound of a 95% confidence interval; CI (upper) = upper bound; →= affects; n.s. = non significant
ATTENTION TRAINING AND REAPPRAISAL 31
Figure 1. Schematic on the task sequence during the experimental session, and overview of indices computed in each task
Notes. IB = Interpretation bias; AB = Attention bias; T1 = Time 1 (pre-training); T2 = Time 2 (post-training)
ATTENTION TRAINING AND REAPPRAISAL 32
Figure 2. Schematic on the basic trial sequence in the SST combined with ET.
Notes. SST = Scramble Sentence Task; ET = Eye-tracker; sec = seconds
ATTENTION TRAINING AND REAPPRAISAL 33
Figure 3. Schematic overview of the attentional control training procedure
ATTENTION TRAINING AND REAPPRAISAL 34
Figure 4. Flow of trial events in the dot-probe task
Notes. AB = Attention bias; RT = Reaction time
ATTENTION TRAINING AND REAPPRAISAL 35
Figure 5. Flow of trial events in the emotion regulation task.
Note: ms = milisesonds
ATTENTION TRAINING AND REAPPRAISAL 36
Figure 6. Relationship among attention control acquisition in the training and attention bias
change levels in the dot-probe at each modification condition.
Notes. AC = attention control (training); AB = attention bias (dot-probe)
AC acquisition and AB change are measured by standardized residuals (positive scores = increases from Time 1 to
Time 2; negative scores = decreases from T1 to T2)
In the control condition, neither AC acquisition in the training nor AB changes in the dot probe task were observed.
In the training condition, individual differences in AC acquisition in the training predicted the level of AB change in
the dot probe task. Specific dispersion area on the association between AC acquisition increases and AB decreases
in the Training condition is highlighted. The highlighted dispersion area indicates that only the participants in the
Training condition who showed the largest AC acquisition were the ones who also showed a reduction in the
negative AB.
ATTENTION TRAINING AND REAPPRAISAL 37
Figure 7. Overview of the indirect effect model supported: Attention training leads to increases
in AC acquisition (in the training); larger AC acquisition leads to larger AB reduction (in the dot-
probe); larger AB reduction leads to larger reappraisal increases; larger reappraisal increases
leads to larger reductions in negative emotion after reappraisal
Notes. AC = attention control (training); AB = attention bias (dot-probe)
B = Beta; SE = Standardized error; CI = Confidence Interval
