1 Accepted for publication in “Behaviour Research and Therapy”
Note: This is an uncorrected version of an author’s manuscript accepted for publication.
Copyediting, typesetting, and review of the resulting proofs will be undertaken on this manuscript before final publication. During production and pre-press, errors may be discovered that could affect the content.
Looking for carrots, watching out for sticks: A gaze-contingent approach towards training contextual goal-dependent affective attention flexibility
Malvika Godara1*, Alvaro Sanchez-Lopez2, Chris Baeken3, 4, 5, 6 & Rudi De Raedt1
1 Department of Experimental Clinical and Health Psychology, Ghent University, Belgium 2 Department of Personality, Evaluation and Psychological Treatment, Complutense
University of Madrid, Spain
3 Department of Head and Skin, Faculty of Medicine and Health Sciences, Ghent University, Belgium
4 Department of Psychiatry, University Hospital Brussels (UZBrussel), Brussels, Belgium 5 Ghent Experimental Psychiatry (GHEP) lab, Ghent University, Ghent, Belgium 6 Eindhoven University of Technology, Department of Electrical Engineering, Eindhoven, the
Netherlands
*Corresponding Author:
Malvika Godara
Department of Experimental, Clinical & Health Psychology Ghent University
Henri Dunantlaan 2 Ghent 9000
Belgium
Word Count: 6494 words
Keywords: depression, flexibility, training, goal-switching, eye-tracking Declaration of interests: None
3 Highlights
• Lower affective flexibility is linked to greater rumination and lower resilience
• We trained affective attention flexibility using context-dependent goals
• Gaze behavior towards emotional material was reinforced with music and white noise
• Active compared to control training condition displayed a greater positivity bias
• Active condition displayed goal-dependent flexible attention for emotional material
4 Abstract
The ability to flexibly shift between changing goals is crucial to develop an adaptive response to life stressors. Accordingly, lower affective attention flexibility, i.e. the ability to shift attention flexibly between goals, is associated with low reappraisal ability, high levels of rumination, and lower levels of resilience. However, attempts to manipulate affective attention flexibility with current attention training procedures have seen limited success. In the current study, we tested a novel attention flexibility training paradigm using eye-tracking, wherein dysphoric participants had to switch between different context-dependent goals.
Attention towards goal-relevant emotional stimuli was reinforced using music and harsh sound. We found that participants in the training condition became significantly faster in switching attention from negative to positive goal-related stimuli pre- to post-training on an attention flexibility task, compared to the control condition. On a transfer task where participants had to give a presentation to an audience, we found that participants in the training condition displayed a more flexible pattern of attending to emotional faces during their presentation, depending upon which goal was activated. Comparatively, participants in the control condition remained rigidly attentive towards negative faces despite the goal activated. Our findings provide preliminary evidence for the role of context-based attention flexibility in dysphoria, and support scope for future clinical applications.
5 Introduction
Dynamically changing environments or situations, in daily life, require flexible cognitive control over stimuli in the environment. This requires the ability to shift between changing mental sets and goals, which is termed as cognitive flexibility (Lezak, 1995).
Affective attention flexibility, a sub-component of cognitive flexibility, can be defined as the ability to flexibly switch between emotional and non-emotional goals (Genet, Malooly, &
Siemer, 2013; Malooly, Genet, & Siemer, 2013). It involves the flexible shifting of attention to goal-relevant affective information and inhibition of attention towards goal-irrelevant non- affective information or vice versa, i.e. shifting towards goal-relevant non-affective
information and inhibiting attention towards goal-irrelevant affective information. For example, Malooly et al. (2013) instructed participants to shift between sorting affective pictures according to an emotional (positive vs. negative) or non-emotional (one person vs.
greater than one person) rule. Subsequently, the ability to respond appropriately to encountered affective stimuli, and the capacity to flexibly adapt responses over time depending upon the environment, are considered essential for successful social and interpersonal functioning (Demaree, Schmeichel, Robinson, & Everhart, 2004; Keltner &
Gross, 1999).
However, mood disorders such as major depression are characterized by a relative inability to modulate responses to affective stimuli in the environment and, hence, with a lack of affective attention flexibility (Bylsma, Morris, & Rottenberg, 2008; Chida & Hamer, 2008;
Friedman, 2007; Goeleven, De Raedt, Baert, & Koster, 2006; Nolen‐Hoeksema, 1991; Siegle, Granholm, Ingram, & Matt, 2001; Thayer & Friedman, 2002). Accordingly, a hallmark feature of depression is attentional rigidity towards negative affective stimuli (De Raedt &
Koster, 2010), and thereby the lack of attention flexibility towards positive stimuli. Further, recent empirical work has demonstrated that deficits in flexible allocation of attention to
6 affective material are associated with emotion regulation difficulties (De Lissnyder, Koster, Derakashan & De Raedt, 2010; Malooly, Genet & Siemer, 2013), and are influenced by mood fluctuations (Grol & De Raedt, 2018). A recent study by Wen and Yoon (2019), specifically linked the faster switching towards and the inability to switch away from negative aspects of emotional material, i.e. lowered affective flexibility, to be uniquely predictive of prospective development of depressive symptomatology after a stress episode. Although, evidently, it is key to emotion regulation and depressive symptomatology, hardly any studies have sought to experimentally modulate affective attention flexibility so far (Malooly, 2016). Therefore, in the present study we aimed to train attention flexibility towards affective material in a sample of individuals displaying elevated negative mood symptoms.
In order to accurately modulate affective attention flexibility, it is necessary that the precise mechanisms underlying flexible attention behavior are manipulated. Affect-cognition models have posited that the motivation to approach or avoid certain consequences or objects drives attention towards affective stimuli in a top-down goal-directed manner, inducing an attentional narrowing towards goal-relevant affective stimuli (Gable & Harmon-Jones, 2008;
2010). For example, the goal to have fun might induce attention narrowing towards positive stimuli while the goal to protect self from harm might perhaps induce a prioritization of negative stimuli in the environment. This view of attention prioritization by goal-relevant stimuli has seen considerable empirical support (Jiang, 2018; van der Laan, Papies, Hooge, &
Smeets, 2017; Vogt, De Houwer & Crombez, 2011; Vogt, De Houwer & Moors, 2011; Vogt, De Houwer, Moors, Van Damme, & Crombez, 2010). Relatedly, recent research has shown that attention bias towards threat, i.e. sustained attention towards threat, can be modulated using reward- and punishment-related top down goals, in healthy and anxious samples (Padmala, Sirbu, & Pessoa, 2017; Vromen, Lipp, & Remington, 2015; Vromen, Lipp, Remington, & Becker, 2016; Vogt, De Houwer, Crombez, & Van Damme, 2013; Vogt,
7 Koster, & De Houwer, 2016). Consequently, attention training paradigms incorporating reward-associated top down goals have been used successfully to reduce attention bias for threat (Lazarov, Pine, & Bar-Haim, 2017; Mogg, Waters, & Bradley, 2017; Sigurjónsdóttir, Björnsson, Ludvigsdóttir, & Kristjánsson, 2015). Correspondingly, recent studies employing dysphoric individuals found that attention bias for negative stimuli, i.e. difficulty to disengage from negative material, could be modulated using music-rewarded and white noise-punished top down goals (Godara, Sanchez-Lopez, De Raedt, 2019; 2021). Therefore, in the current study we used music reward and white noise punishment reinforced top-down goals to train attention towards affective stimuli in dysphoric individuals.
Importantly, although attention is often directed to stimuli in a top-down, goal-driven manner, goal activation at any given moment will depend upon the ongoing context of the individual. According to the current context, goals can be deployed which direct attention to either positive or negative affective stimuli. Subsequently, as the context changes, the goals might shift, directing attention towards differently-valenced affective stimuli. For example, being at a party and then leaving and walking home on a dark street, the goal would change from having fun at the party to getting home safely. Accordingly, in this example, attention would be directed towards positive stimuli initially, then switching to negative stimuli as the context-dependent goal changes. As such, attention processes can be seen as more of a flexibility mechanism, which depends upon the goals within the current context of the individual. From this perspective, affective attention rigidity towards negative stimuli in depressed individuals can be seen as a lack of affective attention flexibility as they remain rigidly unable to disengage from negative stimuli despite changes in the environment or context and corresponding goals (Roethermund, Voss, & Wentura, 2008). This view is in line with literature that individuals in depressive states tend to have difficulties in attending to relevant emotional information since they have broader reductions in sensitivity to emotional
8 contexts and changing emotional environments (Rottenberg, 2005; Rottenberg, 2007;
Rottenberg, Gross & Gotlib, 2005; Rottenberg & Hindash, 2015). Therefore, it is likely that contextual factors will play a key role in the affective attention flexibility mechanism and might be crucial to its modulation.
This context-dependent view of affective attention flexibility that provides the basis for our training approach is aligned with recent models of emotion regulation that view context-appropriate emotion regulation as adaptive in nature and highlight its role in promotion of long-term resilience (Aldao, 2013; Bonanno & Burton, 2013; Bonanno et al., 2004; Kashdan & Rottenberg, 2010). Accordingly, Stange et al. (2017) proposed that deficits in affective cognitive flexibility can be implicated in increasing vulnerability to depression as a result of context change. This view was supported by the findings of Everaert et al. (2018), who show that depression severity is associated with more inflexible interpretation patterns, tending to more negative and less positive interpretations. Accordingly, a recent study tested whether contextual changes impact affective attention flexibility in a sample of dysphoric and non-dysphoric individuals (Godara, Sanchez-Lopez, Liefooghe, & De Raedt, 2020). In this study, participants were introduced to different social contexts which required them to deploy various goals which directed attention towards positive or negative affective stimuli.
Participants were then presented this information in a reaction-time task, wherein they had to either repeat or switch between different contexts and accordingly affective stimuli. Godara et al. (2020) found that contextual changes facilitated switching, exacerbating attention towards negative goal-relevant stimuli in dysphoric individuals and positive goal-relevant stimuli in non-dysphoric individuals. These results suggest that contextual changes play a crucial role in guiding affective attention flexibility, and therefore incorporating contextual changes in training it might be important.
9 In the present study, we tested a novel eye-tracking based training paradigm of
context-dependent affective attention flexibility. In this paradigm, we used reinforcers to train attention towards positive and negative affective stimuli in a contextual goal-directed manner.
For the purpose of this study, we developed several different social contexts, each of which consisted of two distal goals (Godara et al., 2020). For example, the social context “Friend’s party” has two goals, “To enjoy the party” and “To solve argument with friends” These distal goals direct attention towards goal-relevant stimuli through one of two proximal goals of attending to positive or negative face expressions. Thus, the distal goal “To enjoy the party”
functions through the proximal goal of “attend to positive”, and the goal “To solve argument with friends” functions through the proximal goal of “attend to negative”. The proximal goals are each associated with one of two shapes (blue square or blue circle), which change
meaning based on the context. For example, in the “Friend’s party” context, blue square is associated with the “attend to positive” goal, whereas in the “Work Presentation” context, blue square denotes “attend to negative” goal. Depending upon which context and which goal is activated, participants must direct their attention to positive or negative faces in an 8-face matrix. When the goal activated requires participants to attend to positive faces, participants are exposed to music if they keep their eye-gaze on the positive faces in the matrix. If they disengage from positive faces and engage with negative faces, the music stops. Meanwhile, when the goal activated requires participants to attend to negative faces, participants can avoid listening to harsh white noise if they keep their eye-gaze on the negative faces in the matrix. If the participants disengage from negative faces, the white noise is delivered.
Hypotheses
We expected that participants who undergo the flexibility training would become faster in switching attention towards goal-relevant positive and negative information. Further, we hypothesized that participants in the active, compared to control, condition would display
10 more flexibility in attention (i.e., faster attentional switching) to affective material depending upon the contextual goal that is activated. The impact of the training was assessed using a close transfer task and a far transfer task.
Close transfer. The close transfer task was a contextual attention flexibility task (Godara et al., 2020) wherein participants had to repeat and switch attention towards positive and negative stimuli according to the current vs. immediately precedent context-dependent goal that was activated. Accordingly, we expected that participants in the active condition would show a faster repetition and switching of attention towards both positive and negative goal-relevant stimuli at post-training compared to baseline, whereas we did not expect any such differences in the control condition (first hypothesis).
Far transfer. As a measure of attention flexibility in an ecologically valid context, we administered a speech task (Sanchez, Romero, & De Raedt, 2017). In this task, participants were asked to give a speech while they received false feedback in forms of positive and negative avatars on a computer screen. Participants were asked to switch between different valenced goals, where it would be instrumental to either look at the positive or negative facial feedback depending on the goal, thereby directing attention positive and negative stimuli depending upon the goal activated. We expected that participants in the active, compared to control, training condition would show greater attention to relevant emotional avatars depending upon the goal activated (second hypothesis).
Method Participants
A total of 101 individuals (83 Females, Mage = 22.19 years, SD = 3.13, 18 – 37 years) participated in the current study. Participants were invited to take part in the study based on their pre-screening scores on the Anhedonic Depression (AD) subscale of the Mood and
11 Anxiety Symptoms Questionnaire – D30 (MASQ–D30; Wardenaar et al., 2010). We recruited individuals who displayed elevated mood symptoms or dysphoric patterns as indicated by a score of 25 or greater on AD subscale of MASQ-D30 (in our sample MMASQ-AD = 31.43, SD = 6.22). Participants were excluded if they had any current psychiatric problems which was identified using the MINI-International neuropsychiatric interview (Sheehan et al., 1998). The MINI was used to confirm the absence of any psychiatric conditions which the participants had not revealed, and which could otherwise interfere with our study. Further, the present study served as proof-of-concept, and therefore we did not want to include participants who had current psychiatric disorders. Participants were also requested to refrain from
consumption of drugs and alcohol for 6 hours prior to participation. All participants had normal or corrected-to-normal vision. The study was approved by the medical ethical committee of Ghent University Hospital and was conducted in accordance with the declaration of Helsinki (2004). All participants provided informed consent and were compensated for their time (€15).
Materials
Questionnaires. Participants completed the Mood and Anxiety Symptoms Questionnaire – D30 (Wardenaar et al., 2010). The short adaptation of the self-report questionnaire consists of 30 items, out of which 10 items are depression-specific on the AD subscale. Participants respond to statements using a 5-point scale, wherein ‘1’ is “Not at all”
and ‘5’ is “Extremely”. The items on the AD subscale are reverse scored. MASQ – D30 has a Cronbach’s alpha of .92 in young adults, and the internal consistency in our sample was .91.
Apparatus. The stimulus presentation of all procedures in the experiment was programmed using E-prime Professional 2.0 (2008). The stimulus presentation was
implemented on a 23-inch high screen, with a resolution of 1920 x 1080 pixels and luminance of 300 cd/m2. Participants were seated 59 – 65 cm from the screen, and we used Tobii TX 300
12 eye tracker, at a sampling rate of 300 Hz (binocular). Since all tasks in the current study employed eye-tracking, participants performed a 9-point calibration procedure prior to each evaluation task or training.
Pictures. Emotional stimuli used in the attention flexibility training procedure were obtained from the Radboud Faces Database (RaFD; Langner et al., 2010). Thirty-nine positive category (“happy” emotion) and 39 negative category (“disgust” and “angry” emotions) pictures were obtained from RaFD (Caucasian, Adult, Frontal subset), based on high scores for intensity (MNegative = 4.10, SDNegative = 0.33; MPositve = 4.10, SDPositive = .04) and valence (MNegative = 2.03, SDNegative = 0.20; MPositve = 4.20, SDPositive = 0.30). Two context images represented an associated context text and did not consist of human figures. Context pictures were obtained from Google Images by filtering the search results according to usage rights, and pictures marked “Labeled for reuse” and “Labeled for noncommercial reuse” were used.
Audible material. During the training, gaze behavior was reinforced using rewarding music of participants’ choice and harsh white noise. The music used during the training was selected by participants from a pre-selected list of 100 songs. The songs comprised 8 different genres of music and were selected based on the top charts of Belgium at the time of testing. A standard white noise signal was used in the present study. All audible material was
administered through headphones at 65 dB(A).
Attention flexibility training. Two social contexts adapted from Godara et al. (2020) were used in the training procedure: Friend’s party, and Work presentation (see Appendix A for contexts presented for the training and the close transfer task). The contexts consisted of social situations and described what was taking place in the scene. These contexts were chosen as they are ecologically valid and can be seen as inherently valence-neutral, describing only neutral, semantic situations, with the valence being subjective based upon the individual.
For example, a birthday celebration with family members may be viewed as a positive context
13 by individuals who enjoy the company of their families, while this context might hold a negative value for individuals who do not get along with their families. Each context was coupled with a context image which depicted the text in pictorial form, without the presence of any human figures.
Each context consisted of two goals, one which prompted the participants to direct attention towards positive stimuli in the environment, e.g. the goal to enjoy the party, and one which required them to look at negative stimuli, e.g. the goal to solve conflict with friends.
Both the goals had an associated goal object, a blue square or blue circle, indicating which goal had to be activated in the particular context. For example, a blue circle in the friend’s party context indicated that participants had to activate the goal of looking towards negative stimuli in the context. Even though only two goal-cueing shapes were used in both contexts (square and circle), the shape associated with the valenced-goals differed for both contexts.
Thus, in the party context, square indicated that participants had to look towards positive stimuli, but in the presentation context, the square denoted the goal to look towards negative stimuli. This was done in order to avoid a single shape getting associated with a particular valence, allowing us to ascertain that participants processed the context to make their decision to look towards positive or negative stimuli.
Each trial in the training procedure started with a black fixation cross (8mm) in the center of a white screen, for 500ms. Next, a single context image (1024 x 682) appeared on the screen with a goal-cueing shape (square or circle; 100 x 100) placed on top of it in the center of the screen. At the bottom of the screen, a sentence asked participants to “Press spacebar” in order to view the affective faces. At this point, participants could take as much time as they needed to recall the correct response (i.e., look towards positive or towards negative faces) according to the context image and the goal-cueing shape presented on the screen. Upon pressing the button, a 2×4 matrix of 8 emotional faces (341 x 513) appeared on
14 the screen. The following parameters were followed: each actor appeared only once in a matrix; each matrix contained four male and four female faces; half of the faces showed a negative expression, and half showed a positive expression; with both male and female faces displaying two positive and two negative expression each; and both the rows in each matrix had two positive and two negative face expressions each (see Figure 1 for an example of a trial).
Participants had to direct their eye gaze toward the goal-relevant emotional stimuli (i.e., positive or negative faces), based on the goal activated. If the goal activated required participants to attend to positive faces, participants were rewarded with music when they maintained their eye-gaze to positive face expression. When the participants disengaged from positive faces and/or fixated on negative faces, the music (i.e., reward) stopped. If the goal activated required participants to attend to negative face expressions, participants avoided listening to punishing white noise when they maintained their eye-gaze to negative faces.
When the participants disengaged from negative faces and/or fixated on positive faces, white noise (i.e., punishment) was administered. In a control condition of the training, participants were randomly administered music or white noise, when positive and negative goals were activated respectively. The control condition was yoked to the degree that the music or white noise patterns (duration of administration across a trial and number of pauses across a trial) were programmed to play identically to the music and white noise patterns recorded for the participants in the active training condition (Bosmans, Sanchez-Lopez, Finet, & De Raedt, 2019). The duration of time for which the reinforcer was administered in the control condition was on average same as the amount of time the reinforcer was presented in the active
condition. However, the gaze behavior did not determine the delivery of the reinforcer in the control condition. For each trial in the training, the face matrix was displayed for 10000ms, upon which the trial elapsed.
15 Participants performed 64 trials of the training procedure, and performed equal
number of “repeat” (i.e., the goal and/or context was the same as in the previous trial) and
“switch” trials (i.e. the goal and/or context was different to the one in the previous trial) for all the different combinations of goals (positive and negative) and contexts. We had 8 trials each for the following combinations: repeat context-repeat negative goal, repeat context-repeat positive goal, repeat context-switch to negative goal, repeat context-switch to positive goal, switch context-repeat negative goal, switch context-repeat positive goal, switch context- switch to negative goal, and switch context-switch to positive goal. These trials were presented in a pseudo-randomized manner, such that there were 8 blocks (consisting of 8 predefined “repeat” or “switch” trials), which were presented in a random order.
Transfer of training effects
Close transfer – Attention flexibility task. Participants performed the attention flexibility task (Godara et al., 2019b) to evaluate attention switching based on the activation of context-dependent goals. The attention flexibility task was similar to the attention
flexibility training and was performed prior to and after the training. In this task, akin to the training, participants were introduced to two social contexts (two unique contexts both at baseline and post-training, different to the ones used in the training task), and two goals were associated with each context which required gaze behavior towards positive or negative faces, respectively, depending on two goal-cueing shapes (blue square or circle). This information was then presented to participants in a reaction-time task format. Each trial in the task began with a white fixation cross (8mm) in the center of a black screen, for 500ms (see Figure 2).
This was followed by a single context image (1024 x 682) with a goal-cueing shape (square or circle; 100 x 100) in the center of the screen, presented for 3000ms. Upon fixation on the goal-cueing shape for 100ms, a sentence appeared at the bottom of the screen which required participants to “Press spacebar for faces”. Participants could take as long as they needed to
16 recall the correct response (i.e., look towards positive or towards negative faces) according to the context image and the goal-cueing shape presented on the screen. Upon button-press, 8 emotional faces (4 positive and 4 negative) were presented around a fixation cross for
3000ms. Participants had to direct their eye gaze to the correct goal-relevant emotional stimuli (i.e., positive or negative) as quickly as possible. Participants performed 64 trials of the
attention flexibility task, both pre- and post-training (see Figure 2 for trial procedure). The trials were presented in a pseudo-randomized order, such that there were equal number of
“repeat” and “switch” trials for all of the 8 switching conditions mentioned in the “Attention flexibility training” section above. We obtained the “First fixation to goal-relevant emotion”
attention index for each trial, i.e. how quickly participants looked towards the given goal- relevant emotion after the emotional faces had been presented on the screen. These first fixations were used to calculate the 8 attention switching conditions: repeat context-repeat negative goal, repeat context-repeat positive goal, repeat context-switch to negative goal, repeat context-switch to positive goal, switch context-repeat negative goal, switch context- repeat positive goal, switch context-switch to negative goal, and switch context-switch to positive goal.
Far transfer - Speech task. After performing the post-training measure of attention flexibility task, participants performed the speech task (Sanchez et al., 2017). In addition to the close transfer evaluation described above, we wanted to evaluate how the training effects translate to an actual social context wherein participants had to switch between positively and negatively valenced goals. We adopted an identical procedure to Sanchez et al. (2017).
Participants were asked to prepare and deliver a 5-minute speech on the topic “Why are you a good friend?” to a 4-member jury. Participants were told that they would be delivering the speech through a web-camera, and 4 expert psychologist jury members would evaluate their speech for persuasiveness, coherence and clarity. They were also informed that they would
17 receive real-time feedback on their performance during the speech delivery, consisting of avatar faces representing evaluators’ expressions presented on the screen. Although participants were told that they were being evaluated by jury members, the feedback presented on the screen were pre-programmed emotional avatars, and there were no real people evaluating the speech. False feedback comprised a total of 4 faces, positive (smiling) and negative (disgusted), with 2 male and 2 female avatars. The false feedback was presented in a pre-determined manner with 3 unique false feedback slides: Balanced (50% positive and 50% negative), Positive (25% negative and 75% positive), and Negative (25% positive and 75% negative). The false feedback slides were presented in the following order: Balanced – Positive – Balanced – Negative – Balanced (see Figure 3). Each slide was presented for 15 seconds, preceded by a fixation oval for 15 seconds. The entire order of the slides was presented twice. Participants were told that they had to deploy two goals while delivering the speech: to be confident about their speech (best to attend to positive faces), and to solve concerns of the jury (best to attend to negative faces). Participants began the speech with one of the goals, and 2 minutes and 30 seconds (halfway) into the speech they were asked to switch and deploy the other goal. They were also told that their eye movements would be recorded throughout the speech. We measured the fixation duration towards both positive and negative faces, i.e. how long participants looked at positive and negative faces, for each feedback slide during both goal conditions. In total, we obtained 20 fixation duration indices (10 positive and 10 negative; 10 for each goal condition).
Procedure
The experiment began with participants signing the informed consent, followed by the administration of the MINI neuropsychiatric interview. The included participants then
performed a baseline measure of the attention flexibility task consisting of 64 trials. Upon completion of the baseline measure, participants were assigned to either the active or the
18 control condition of the training. Participants then performed 64 trials of the training task, followed by completion of a post-measure of the close transfer attention flexibility task (64 trials). Next, participants were introduced to, and performed the far transfer speech task.
Lastly, participants filled out the questionnaires, and were finally debriefed.
Results
Close transfer – Attention flexibility task. We obtained 8 attention switching variables each for both baseline and post-training measures of attention flexibility task. First, we conducted a 2 (Context Switch) x 2 (Goal Switch) x 2 (Goal valence) x 2 (Condition) mixed measures ANOVA using baseline attention switching scores. Context Switch (Repeat vs.
Switch Context), Goal Switch (Repeat vs. Switch Goal), and Goal Valence (Positive vs.
Negative) were the within-subjects factors, and Condition (Active vs. Control) was the between-subjects factor. At baseline, we found a significant Context Switch x Goal Switch x Goal Valence interaction, F(1, 96) = 9.97, p = .002, hp2 = .09. There was no main effect of Condition, F(1, 96) = .44, p = .51, hp2 = .01. The lack of main effect for, and interaction with, the between-group factor indicates that there were no baseline group differences between active and control conditions. Bonferroni-corrected pairwise comparisons clarify that dysphoric participants, regardless of the training condition, were faster at baseline in both repeating and switching attention towards negative over positive goals (all ps < .003). Further, they found it easier to repeat, compared to switch, attention towards negative goals when the context was repeating (p = .04). No difference was observed between repeating and switching attention to negative goals when the context was switching (p = .24).
Next, we conducted a 2 (Time) x 2 (Context Switch) x 2 (Goal Switch) x 2 (Goal valence) x 2 (Condition) mixed measures ANOVA, using attention switching indices obtained for baseline and post-training as dependent variable. Time (Pre- vs. Post-Training), Context
19 Switch (Repeat vs. Switch Context), Goal Switch (Repeat vs. Switch Goal), and Goal Valence (Positive vs. Negative) were the within-subjects factors, and Condition (Active vs. Control) was the between-subjects factor.
First, we found a significant Time x Goal Valence interaction, F(1, 96) = 52.76, p <
.001, hp2 = .36. Pairwise comparisons revealed that all participants were faster in repeating and switching attention towards negative goals, compared to positive, at pre-training (B = .24, p < .001, 95% CI[ .17, .31]). However, at post-training all participants were faster in repeating and switching attention towards positive, compared to negative, goals (B = .13, p < .001, 95%
CI[ .06, .20]; see Figure 4). Second, there was a significant Goal Valence x Condition interaction, F(1, 96) = 11.57, p = .001, hp2 = .11, that was partially accounted by a near significant Time x Goal Valence x Condition interaction, F(1, 96) = 3.05, p = .06, hp2 = .03.
A full decomposition of the 3-way interaction is provided in Appendix B. Specific pairwise comparisons concerning our first hypothesis showed that specifically participants in the active training condition, compared to control, were significantly faster in repeating and switching attention to positive faces at post-training, (B = .14, p = .006, 95% CI[ .04, .25]; see Figure 5).
However, there were no significant differences between active and control conditions in repeating or switching attention towards negative goals at post-training (p > .05). As for the other 3-way, 4-way, and 5-way interactions included in the design of the paradigm, no
interaction relevant for the specific hypothesis in the study reached significance, all Fs < 1, all ps > .05, all hp2s < .01.
Overall, we found that all participants became significantly faster in repeating and switching attention towards positive but not negative goals from pre- to post-training.
However, participants in the active, compared to control, condition were significantly faster in
20 repeating and switching attention to positive goal-relevant stimuli at post-training, but no such significant differences were observed in fixations towards negative goals.”
Far transfer – Speech task. We obtained 20 fixation duration indices for analyses, 5 positive and 5 negative indices each for both the goals activated. We conducted a 2
(Condition) x 2 (Goal Valence) x 2 (Emotion) x 5 (Feedback) mixed measures ANOVA. Goal Valence (Positive vs. Negative), Emotion (Positive vs. Negative avatars), and Feedback (Balanced 1 vs. Positive vs. Balanced 2 vs. Negative vs. Balanced 3) were the within-subjects factors. Condition (Active vs. Control) was the between-subjects factor. We found a
significant main effect of Emotion, F(1, 95) = 6.91, p = .01, hp2 = .07, and a significant Condition x Emotion interaction, F(1, 95) = 114.35, p < .001, hp2 = .55. Follow-up comparisons revealed that participants in the active, compared to the control, condition fixated longer on the positive avatars (B = 2.47, p < .001, 95% CI[ 1.72, 3.23]). Meanwhile, participants in the control condition, compared to active, fixated significantly longer on negative avatars (B = 1.55, p < .001, 95% CI[ .86, 2.23]).
However, these effects were eclipsed by a significant Condition x Goal Valence x Emotion x Feedback interaction, F(4, 380) = 3.14, p = .03, hp2 = .03. Please see Appendix C for a decomposition of the interaction. Pairwise comparisons showed that individuals in the active training condition fixated significantly longer on positive avatars when positive
compared to negative valence goal was activated during Balanced 2 (B = 1.54, p < .001, 95%
CI[ .76, 2.32]), and Negative (B = 1.51, p = .001, 95% CI[.83, 2.19]) feedbacks (see Figure 6). Similarly, we found that individuals in the active training condition had significantly longer dwell times on negative avatars when negative compared to positive valence goal was activated during Balanced 2 (B = .94, p = .02, 95% CI[ .16, 1.72]), Negative (B = 1.25, p = .002, 95% CI[ .48, 2.03]), and Balanced 3 (B = .79, p = .01, 9% CI[ .18, 1.40]) feedbacks (see Figure 7). In the control condition, participants fixated significantly longer on positive avatars
21 when the positive versus the negative valence goals was activated during Positive feedback (B
= 1.40, p = .02, 95% CI[.20, 2.61]). Vice versa, they fixated significantly longer on negative avatars when the negative versus positive goal was activated during Positive feedback (B = .93, p = .04, 95% CI[.03, 1.84]). However, no other goal-based differential attention to positive nor negative avatars was observed (all p’s > .08). Overall, as accounted in the two- way interaction, participants in the control condition fixated significantly longer on negative over positive avatars regardless of the feedback presented and the goal activated. These results are in line with our second hypothesis, such that participants in the active, compared to control, training condition overall show greater attention to goal-relevant emotional avatars depending upon the goal activated.
Discussion
In the present study, we investigated the efficacy of a novel affective attention
flexibility training in a sample of dysphoric individuals. In the training paradigm, attention is trained towards positive and negative stimuli in a contextual goal-directed manner. This is in line with the recent models of emotion regulation which articulate a greater role of context- appropriate reactivity (Aldao, 2013; Bonanno & Burton, 2013). Further, our training draws from theoretical models which support the existence of emotion context insensitivity in depressogenic states (Rottenberg, 2005; Rottenberg, 2007; Rottenberg, Gross & Gotlib, 2005). As such, we aimed to train switching and repetition of attention towards positive and negative stimuli depending upon the contextual goal activated, and reinforcing goal-directed behavior using rewarding music and punishing white noise.
We found that dysphoric individuals, characterized by faster repetition and switching of attention toward negative goals at baseline, displayed faster switching and repetition of attention towards positive goals on a close transfer task after receiving the training. Further, on a far transfer task we were able to show that individuals in the active training condition had
22 significantly longer dwell times on positive and negative stimuli depending upon which goal was activated. Meanwhile, individuals in a control training condition showed similar
improvements as the active condition in attention switching capacities towards positive goals on the close transfer task, but significantly less pronounced. Therefore, although individuals in both conditions improved significantly in repeating and switching attention to positive goals, this effect was markedly conspicuous in the active condition. Moreover, in the far transfer task individuals in the control training condition largely fixated more towards negative faces regardless of the goal activated. Thus, in the ecologically valid context of presenting a speech, a decidedly unique pattern of goal-dependent flexibility emerged for individuals who received the active training condition, compared to the negative rigidity exhibited by participants in the control condition.
Although our hypothesis for the far transfer task was confirmed, we were only partially able to confirm our hypotheses for the near transfer task. We had expected that participants in the active condition would become faster in repeating and switching attention to both positive and negative goals from pre- to post-training. Further, we expected this effect to be more pronounced in the active over the control condition. However, we found this to be the case only for positive goals. For negative goals, there was no significant difference in attention switching capacities from pre- to post-training. One explanation for this finding could be the fact that individuals in depressive states already have a greater attention towards negative stimuli than healthy controls (De Raedt & Koster, 2010; Koster, De Raedt, Leyman,
& De Lissnyder 2010; Sears, Thomas, LeHuquet, & Johnson, 2010). Indeed, an efficient pattern of attentional processing of (repeat) and flexibility for (switching) negative goals was evidenced by the dysphoric participants in the study, in the close transfer task completed at baseline. As such, it can be assumed that training is not required to improve their capabilities in guiding attention towards negative material above and beyond what they already possess.
23 This suggests that the attention flexibility training was able to induce an equilibrium between underlying approach and avoidance goals to yield balanced attention towards different goals (Trew, 2011).
Furthermore, we had expected that participants in the control condition would not show any improvement in switching capacities towards positive or negative goals from baseline to post-training. However, we found an improvement in switching and repeating attention towards positive goals, but not towards negative goals, in control condition. A possible explanation for this effect could be task-specific learning (Schubert, Strobach, &
Karbach, 2014). Both the training and the close transfer task were very similar in nature, and the flexibility task was performed twice. As such, it is possible that participants in both conditions improved as a result of performing similar types of trials over a period of time.
However, since there is a valence-specific improvement in both groups, it could be attributed to the impact of music as a motivator for goal-driven behavior. Music has been shown to be a therapeutic for a variety of psychopathologies (Wheeler, 2016), and has been successfully used in a sample of socially anxious individuals to modulate gaze behavior towards
threatening stimuli (Lazarov, Pine, & Bar-Haim, 2017). Participants in our study could choose their preferred track to play during the training, likely increasing the subjective value of the reward for all. Another possible explanation for the effect could be attributed to nonspecific placebo effects, owing to similarities in the nature of both active and control training, and testing conditions (Baskin, Tierney, Minami, & Wampold, 2003; Hróbjartsson, & Gøtzsche, 2002).
Regardless, our findings provide proof-of-concept for the successful modulation of affective biases in depressed states as a function of the current goals of the individuals. Using our attention flexibility training paradigm, we were able to modulate attention for affective material in a context-dependent manner in an ecologically valid setting. Our results add to the
24 growing evidence for a dysfunctional flexibility mechanism in depression (Everaert et al., 2018; Stange et al., 2017). As such, from our findings we understand that affective attention flexibility in depression could be affected by dysfunction at three distinct levels: the level of processing contextual changes, the level of deploying larger distal goals motivated by
approach and avoidance motivation, and the level of applying the proximal goals which serve as a function of the distal goals. Accordingly, the present study suggests that attention
problems in depression might be the outcome of difficulties in the ability to process
contextual changes which exacerbate the ability to deploy relevant approach and avoidance goals, leading to greater number of avoidance goals activated regardless of the context (Rottenberg, 2007; Trew, 2011). This could further enhance the difficulties in switching between the proximal goals of directing attention towards negative or positive stimuli, creating difficulties in flexible shifting between differently valenced emotional stimuli and precipitating a bias towards negative stimuli (De Raedt & Koster, 2010; Hollenstein, 2015).
The difficulties in shifting between proximal goals of attending to emotional information might in turn lead to depressed individuals getting stuck with distal goals motivated by avoidance, leading to context-inappropriate goal deployment and difficulties in processing shift in contexts. We surmise that these different components of affective attention flexibility likely reinforce each other and precipitate greater depressive symptomatology in the long-run.
However, our results go even further to provide evidence for an induction of greater affective attention flexibility in our training group, indicating a scope for clinical applications. This is a step forward considering that previous studies have been unable to find evidence for
modulation of affective flexibility (Malooly, 2016). A novel feature of the current training paradigm is the combination of cognitive control style (top-down regulation) training with attention bias modification approaches (attention training using eye-gaze contingent
25 procedures). This allows for simultaneous training of more overt attention processes while also tackling more covert attention switching capacities.
Moreover, our findings also have relevance in the clinical context. Current attention bias modification (ABM) procedures aimed at reducing attention bias for negative
information in depression have shown inadequate success (Beard, Sawyer, & Hofmann, 2012;
Cristea, Kok, & Cuijpers, 2015; Mogoaşe, David, & Koster, 2014). In addition to being tedious, these procedures suffer from a lack of motivational component. Our novel attention flexibility paradigm considers the underlying approach and avoidance goals to train attention.
Further, current ABM procedures train attention towards neutral or positive stimuli, and away from negative stimuli. However, this would likely not promote resilience in the long run, as ABM procedures train individuals to attend to positive information even in situations where it might not be the most adaptive response. For example, not engaging with negative feedback at a job performance review might inhibit performance improvement over time, which could be crucial to job retention. Consequently, maladaptive positive responsivity in real-life contexts would also likely lead to negative consequences, followed by initiation of negative mood state and activation of negative self-schemas, which would ultimately redirect attention to negative stimuli (Disner, Beevers, Haigh, & Beck, 2011; Farb, Irving, Anderson, & Segal, 2015; Watkins & Nolen-Hoeksema, 2014). Therefore, a key novel feature of our flexibility training paradigm is the inclusion of contextual factors, and eventually context-dependent training of attention using ecologically valid social contexts. Although we do not examine longitudinal changes in the current study, we expect that training affective attention biases and attention flexibility in this manner will likely lead to greater control over emotional material in the long run.
Despite the strengths of the attention flexibility training paradigm, some limitations of the study must be acknowledged. First, the present study investigates modulation of affective
26 attention flexibility in depressive states, but we do not employ a clinical sample. In order to provide a proof-of-concept, the present study was conducted in a sample of subclinical individuals who at the time of testing displayed elevated negative mood symptoms. As such, the findings of the current study must be replicated in a clinical sample before the modulatory efficacy of the current training paradigm for depression can be absolutely claimed. Second, majority of our sample were college students, which may be problematic due to the
heterogeneity of the sample and the lack of generalizability to the population (Hanel & Vione, 2016). Further, our sample suffers from overreliance on female participants. Previous research has shown that women tend to have stronger response to goal-tracking cues, while men seem to hold stronger salience for sign-tracking (Burns & Domjan, 1996; DiFeliceantonio &
Berridge, 2012). Therefore, the present study must be replicated in a community sample to have more diversity in terms of age and sex, which could be crucial moderators for affective attention flexibility. Lastly, the present study did not include measures of emotion regulation.
It may be important to see how well the results transfer to emotion and stress regulation capabilities as well as other forms of attentional and cognitive control. As such, future studies should employ heart rate variability or performance on emotion regulation tasks as a marker of stress regulation, and attention bias tasks which can clarify the impact of the training on specific depression-related impaired attentional disengagement mechanisms (Sanchez et al., 2017).
In conclusion, in the present study we tested the effectiveness of a novel gaze-
contingent affective attention flexibility training which incorporates context-dependent goals.
We found that the training improved attention switching capacity towards positive stimuli on a close transfer flexibility task. Further, we found that in an ecologically valid social context, individuals who underwent the training displayed a more flexible pattern of attention to
27 affective stimuli, depending upon the current goal activated. Our results highlight the promise of this type of approach as a potential future clinical application.
28 Acknowledgements
This research was supported by Grant BOF16/GOA/017 for a Concerted Research Action of Ghent University awarded to Rudi De Raedt. Alvaro Sanchez-Lopez is supported by the Program for the Attraction of Scientific Talent of the Community of Madrid. The funding authorities were not involved in the conceptualization of this study, nor in the collection, analysis or writing up of the data. Further, the funding authorities were not involved in the decision to submit this manuscript for publication.
29 References
Aldao, A. (2013) The future of emotion regulation research: Capturing context.
Perspectives on Psychological Science, 8, 155–72.
Baskin, T. W., Tierney, S. C., Minami, T., & Wampold, B. E. (2003). Establishing specificity in psychotherapy: a meta-analysis of structural equivalence of placebo controls.
Journal of Consulting and Clinical Psychology, 71, 973. Doi: 10.1037/0022-006X.71.6.973 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.
doi: 10.1016/j.beth.2012.01.002
Bonanno, G. A., & Burton, C. L. (2013). Regulatory flexibility: An individual differences perspective on coping and emotion regulation. Perspectives on Psychological Science, 8(6), 591-612. doi: 10.1177/1745691613504116.
Bonanno, G. A., Papa, A., Lalande, K., Westphal, M. & Coifman, K. (2004) The
importance of being flexible: The ability to both enhance and suppress emotional expression predicts long-term adjustment. Psychological Science, 15, 482–87. doi:
10.1111/j.09567976.2004.00705.x.
Bosmans, G., Sanchez-Lopez, A., Finet, C., & De Raedt, R. (2019). Attachment-related attention bias plays a causal role in trust in maternal support. Journal of Experimental Child Psychology, 185, 176-190. doi: 10.1016/j.jecp.2019.04.017
Burns, M., & Domjan, M. (1996). Sign tracking versus goal tracking in the sexual conditioning of male Japanese quail (Coturnix japonica). Journal of Experimental Psychology: Animal Behavior Processes, 22, 297. Doi : 10.1037/0097-7403.22.3.297
30 Bylsma, L. M., Morris, B. H., & Rottenberg, J. (2008). A meta-analysis of emotional reactivity in major depressive disorder. Clinical Psychology Review, 28, 676-691. Doi : 10.1016/j.cpr.2007.10.001
Chida, Y., & Hamer, M. (2008). Chronic psychosocial factors and acute physiological responses to laboratory-induced stress in healthy populations: a quantitative review of 30 years of investigations. Psychological Bulletin, 134, 829. Doi : 10.1037/a0013342
Cristea, I. A., Kok, R. N., & Cuijpers, P. (2015). Efficacy of cognitive bias modification interventions in anxiety and depression: meta-analysis. The British Journal of Psychiatry, 206(1), 7-16. doi: 10.1192/bjp.bp.114.146761
De Lissnyder, E., Koster, E. H., Derakshan, N., & De Raedt, R. (2010). The association between depressive symptoms and executive control impairments in response to emotional and non-emotional information. Cognition and Emotion, 24(2), 264-280.
De Raedt, R., & Koster, E. H. (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, 50-70. doi:
10.3758/CABN.10.1.50
DiFeliceantonio, A. G., & Berridge, K. C. (2012). Which cue to ‘want’? Opioid
stimulation of central amygdala makes goal-trackers show stronger goal-tracking, just as sign- trackers show stronger sign-tracking. Behavioural Brain Research, 230, 399-408. Doi:
10.1016/j.bbr.2012.02.032
Disner, S. G., Beevers, C. G., Haigh, E. A., & Beck, A. T. (2011). Neural mechanisms of the cognitive model of depression. Nature Reviews Neuroscience, 12, 467. doi:
10.1038/nrn3027.
31 Demaree, H., Schmeichel, B., Robinson, J., & Everhart, D. E. (2004). Behavioural,
affective, and physiological effects of negative and positive emotional exaggeration.
Cognition & Emotion, 18, 1079-1097. Doi: 10.1080/02699930441000085
Everaert, J., Bronstein, M. V., Cannon, T. D., & Joormann, J. (2018). Looking through tinted glasses: Depression and social anxiety are related to both interpretation biases and inflexible negative interpretations. Clinical Psychological Science, 1-12. doi:
10.1177/2167702617747968
Farb, N. A., Irving, J. A., Anderson, A. K., & Segal, Z. V. (2015). A two-factor model of relapse/recurrence vulnerability in unipolar depression. Journal of Abnormal Psychology, 124, 38. doi: 10.1037/abn0000031.
Gable, P. A., & Harmon-Jones, E. (2008). Approach-motivated positive affect reduces breadth of attention. Psychological Science, 19, 476-482. Doi: 10.1111/j.1467-
9280.2008.02112.x
Gable, P., & Harmon-Jones, E. (2010). The blues broaden, but the nasty narrows:
Attentional consequences of negative affects low and high in motivational intensity.
Psychological Science, 21, 211-215. Doi: 10.1177/0956797609359622
Genet, J. J., Malooly, A. M., & Siemer, M. (2013). Flexibility is not always adaptive:
Affective flexibility and inflexibility predict rumination use in everyday life. Cognition &
emotion, 27(4), 685-695. doi: 10.1080/02699931.2012.733351
Godara, M., Sanchez-Lopez, A., & De Raedt, R. (2019). Music to my ears, goal for my eyes? Music reward modulates gaze disengagement from negative stimuli in dysphoria.
Behaviour Research and Therapy, 120, 103434. doi: 10.1016/j.brat.2019.103434
32 Godara, M., Sanchez-Lopez, A., & De Raedt, R. (2021). Manipulating avoidance
motivation to modulate attentional bias for negative information in dysphoria. Journal of Behavior Therapy and Experimental Psychiatry, 70, 101613. doi:
10.1016/j.jbtep.2020.101613
Godara, M., Sanchez-Lopez, A., Liefooghe, B., & De Raedt, R. (2020). Contextual changes influence attention flexibility towards new goals. Cognitive Therapy and Research, 44, 327-344. doi: 10.1007/s10608-019-10072-5
Goeleven, E., De Raedt, R., Baert, S., & Koster, E. H. (2006). Deficient inhibition of emotional information in depression. Journal of Affective Disorders, 93, 149-157. Doi:
10.1016/j.jad.2006.03.007
Grol, M., & De Raedt, R. (2018). The effect of positive mood on flexible processing of affective information. Emotion, 18, 819. Doi: 10.1037/emo0000355
Hanel, P. H., & Vione, K. C. (2016). Do student samples provide an accurate estimate of the general public?. PloS One, 11, e0168354. Doi: 10.1371/journal.pone.0168354
Hollenstein, T. (2015). This time, it’s real: Affective flexibility, time scales, feedback loops, and the regulation of emotion. Emotion Review, 7, 308-315. doi:
10.1177/1754073915590621
Hróbjartsson, A., & Gøtzsche, P. C. (2002). Placebo treatment versus no treatment.
Cochrane Database of Systematic Reviews, 4. Doi: 10.1002/14651858.CD003974 Jiang, Y. V. (2018). Habitual versus goal-driven attention. Cortex, 102, 107-120. Doi:
10.1016/j.cortex.2017.06.018
33 Joormann, J., & Tanovic, E. (2015). Cognitive vulnerability to depression: examining cognitive control and emotion regulation. Current Opinion in Psychology, 4, 86-92. doi:
10.1016/j.copsyc.2014.12.006
Kashdan, T. B., & Rottenberg, J. (2010). Psychological flexibility as a fundamental aspect of health. Clinical Psychology Review, 30, 865-878. doi: 10.1016/j.cpr.2010.03.001
Keltner, D., & Gross, J. J. (1999). Functional accounts of emotions. Cognition & Emotion, 13, 467-480. Doi: 10.1080/026999399379140
Koster, E. H., De Raedt, R., Leyman, L., & De Lissnyder, E. (2010). Mood-congruent attention and memory bias in dysphoria: Exploring the coherence among information- processing biases. Behaviour Research and Therapy, 48, 219-225. Doi:
10.1016/j.brat.2009.11.004
Langner, O., Dotsch, R., Bijlstra, G., Wigboldus, D. H., Hawk, S. T., & Van Knippenberg, A. D. (2010). Presentation and validation of the Radboud Faces Database. Cognition and Emotion, 24(8), 1377-1388. doi: 10.1080/02699930903485076
Lazarov, A., Pine, D. S., & Bar-Haim, Y. (2017). Gaze-contingent music reward therapy for social anxiety disorder: A randomized controlled trial. American Journal of Psychiatry, 174, 649-656. doi: 10.1176/appi.ajp.2016.16080894
Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.). New York: Oxford University Press.
Malooly, A. M. (2016). Training Affective Flexibility: Effects of an Affective-Control Training Task on Emotion Regulation. (Unpublished Doctoral Dissertation). University of Miami, Coral Gables, FL. Retrieved from: Open Access Dissertations. 1758.
34 Malooly, A. M., Genet, J. J., & Siemer, M. (2013). Individual differences in reappraisal effectiveness: The role of affective flexibility. Emotion, 13(2), 302. doi: 10.1037/a0029980
Mogg, K., Waters, A. M., & Bradley, B. P. (2017). Attention bias modification (ABM):
Review of effects of multisession ABM training on anxiety and threat-related attention in high-anxious individuals. Clinical Psychological Science, 5, 698-717. Doi:
10.1177/2167702617696359
Mogoaşe, C., David, D., & Koster, E. H. (2014). Clinical efficacy of attentional bias modification procedures: An updated meta-analysis. Journal of Clinical Psychology, 70, 1133-1157. doi: 10.1002/jclp.22081.
Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the duration of depressive episodes. Journal of Abnormal Psychology, 100, 569. Doi: 10.1037/0021-
843X.100.4.569
Padmala, S., Sirbu, M., & Pessoa, L. (2017). Potential reward reduces the adverse impact of negative distractor stimuli. Social, Cognitive and Affective Neuroscience, 12, 1402-1413.
Doi: 10.1093/scan/nsx067
Psychology Software Tools, Inc. [E-Prime 2.0.8.22]. (2008). Retrieved from http://www.pstnet.com
Rothermund, K., Voss, A., & Wentura, D. (2008). Counter-regulation in affective attentional biases: a basic mechanism that warrants flexibility in emotion and motivation.
Emotion, 8, 34. Doi: 10.1037/1528-3542.8.1.34
Rottenberg, J. (2005). Mood and emotion in major depression. Current Directions in Psychological Science, 14, 167-170. Doi: 10.1111/j.0963-7214.2005.00354.x
35 Rottenberg, J. (2007). Major Depressive Disorder: Emerging Evidence for Emotion
Context Insensitivity. In J. Rottenberg & S. L. Johnson (Eds.), Emotion and psychopathology:
Bridging affective and clinical science (pp. 151-165). Washington, DC, US: American Psychological Association. doi: 10.1037/11562-007
Rottenberg, J., Gross, J. J., & Gotlib, I. H. (2005). Emotion context insensitivity in major depressive disorder. Journal of Abnormal Psychology, 114, 627. Doi: 10.1037/0021-
843X.114.4.627
Rottenberg, J., & Hindash, A. C. (2015). Emerging evidence for emotion context insensitivity in depression. Current Opinion in Psychology, 4, 1-5. doi:
10.1016/j.copsyc.2014.12.025
Sanchez-Lopez, A., Romero, N., & De Raedt, R. (2017). Depression-related difficulties disengaging from negative faces are associated with sustained attention to negative feedback during social evaluation and predict stress recovery. PloS one, 12, e0175040. doi:
10.1371/journal.pone.0175040
Schubert, T., Strobach, T., & Karbach, J. (2014). New directions in cognitive training: on methods, transfer, and application. Psychological Research, 78, 749-755. Doi:
10.1007/s00426-014-0619-8
Sears, C. R., Thomas, C. L., LeHuquet, J. M., & Johnson, J. C. (2010). Attentional biases in dysphoria: An eye-tracking study of the allocation and disengagement of attention.
Cognition and Emotion, 24, 1349-1368. doi: 10.1080/02699930903399319
Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Amorim, P., Janvas, J., Weiller, E., …, &
Dunbar, G. C. (1998). The Mini-International Neuropsychiatric Interview (MINI): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59, 22-33
36 Siegle, G. J., Granholm, E., Ingram, R. E., & Matt, G. E. (2001). Pupillary and reaction time measures of sustained processing of negative information in depression. Biological Psychiatry, 49, 624-636. Doi: 10.1016/S0006-3223(00)01024-6
Sigurjónsdóttir, O., Björnsson, A. S., Ludvigsdóttir, S. J., & Kristjánsson, A. (2015).
Money talks in attention bias modification: Reward in a dot-probe task affects attentional biases. Visual Cognition, 23, 118-132. doi:10.1080/13506285.2014.984797
Stange, J. P., Alloy, L. B., & Fresco, D. M. (2017). Inflexibility as a vulnerability to depression: A systematic qualitative review. Clinical Psychology: Science and Practice, 24, 245-276. doi: 10.1111/cpsp.12201
Thayer, J. F., & Friedman, B. H. (2002). Stop that! Inhibition, sensitization, and their neurovisceral concomitants. Scandinavian Journal of Psychology, 43, 123-130. Doi:
10.1111/1467-9450.00277
Trew, J. L. (2011). Exploring the roles of approach and avoidance in depression: An integrative model. Clinical Psychology Review, 31, 1156-1168. doi:
10.1016/j.cpr.2011.07.007
Van der Laan, L. N., Papies, E. K., Hooge, I. T., & Smeets, P. A. (2017). Goal-directed visual attention drives health goal priming: An eye-tracking experiment. Health Psychology, 36, 82. Doi: 10.1037/hea0000410
Vogt, J., De Houwer, J., & Crombez, G. (2011). Multiple goal management starts with attention: Goal prioritizing affects the allocation of spatial attention to goal-relevant events.
Experimental Psychology, 58, 55-61. doi: 10.1027/1618-3169/a000066.
37 Vogt, J., De Houwer, J., Crombez, G., & Van Damme, S. (2013). Competing for
attentional priority: Temporary goals versus threats. Emotion, 13, 587-598.
doi:10.1037/a0027204
Vogt, J., De Houwer, J., & Moors, A. (2011). Unintended allocation of spatial attention to goal- relevant but not to goal-related events. Social Psychology, 42, 48–55. doi:10.1027/1864- 9335/a000042
Vogt, J., De Houwer, J., Moors, A., Van Damme, S., & Crombez, G. (2010). The
automatic orienting of attention to goal-relevant stimuli. Acta Psychologica, 134, 61-69. Doi:
10.1016/j.actpsy.2009.12.006
Vogt, J., Koster, E. H.W. and De Houwer, J. (2016) Safety first: instrumentality for reaching safety determines attention allocation under threat. Emotion, 17, 528-537. doi:
10.1037/emo0000251
Vromen, J. M. G., Lipp, O. V., & Remington, R. W. (2015). The spider does not always win the fight for attention: Disengagement from threat is modulated by goal set. Cognition and Emotion, 29, 1185-1196. doi: 10.1080/02699931.2014.969198
Vromen, J. M., Lipp, O. V., Remington, R. W., & Becker, S. I. (2016). Threat captures attention, but not automatically: Top-down goals modulate attentional orienting to threat distractors. Attention, Perception, & Psychophysics, 78, 2266-2279. Doi: 10.3758/s13414- 016-1142-3
Wardenaar, K. J., van Veen, T., Giltay, E. J., de Beurs, E., Penninx, B. W., & Zitman, F.
G. (2010). Development and validation of a 30-item short adaptation of the Mood and Anxiety Symptoms Questionnaire (MASQ). Psychiatry research, 179(1), 101-106. doi:
10.1016/j.psychres.2009.03.005
38 Watkins, E. R., & Nolen-Hoeksema, S. (2014). A habit-goal framework of depressive rumination. Journal of Abnormal Psychology, 123, 24. Doi: 10.1037/a0035540
Wen, A., & Yoon, K. L. (2019). Depression and affective flexibility: A valence-specific bias. Behaviour Research and Therapy, 123, 103502. doi: 10.1016/j.brat.2019.103502
Wheeler, B. L. (2016). Research in music therapy. In S. Hallam, I. Cross, & M. Thaut (Eds.), Oxford library of psychology. The Oxford handbook of music psychology (p. 837–
855). Oxford University Press.
39 Figure Captions:
Figure 1: Trial sequence in the attention flexibility training paradigm. The trial sequence for both active and control conditions remained the same.
Figure 2: Trial sequence in the attention flexibility task. All times are in milliseconds.
Figure 3: The order of presentation of the false feedback slides for each of the two goals activated during the task. This order was presented twice, once when each goal was activated.
Figure 4: Post-training repeat and switch first fixations towards positive and negative goals when context is switching for both active and control conditions. All times are presented in seconds.
Figure 5: Repeat and switch first fixations in active condition when context is switching for both pre- and post-training. All times are presented in seconds.
Figure 6: Fixation duration times towards positive faces in active condition for both positive and negative valence goals. All times presented in seconds.
Figure 7: Fixation duration times towards negative faces in active condition for both positive and negative valence goals. All times presented in seconds.
40 Appendix A
Context Goal Context Picture
Friend’s party:
To enjoy the party, it is best to pay attention to people who look happy.
To solve an argument with your friends, it is best to pay attention to people who look angry or sad.
To enjoy the party
To solve argument with friends
Work presentation:
To remain confident about your presentation, it is best to pay attention to people who look happy.
To solve doubts of the audience about your presentation, it is best to pay attention to people who look upset.
To appear confident
To solve doubts
Cousin’s birthday party:
To enjoy the birthday party, it is best to pay attention to people who look happy.
To solve a conflict you have with your family members, it is best to pay attention to people who look angry or sad.
To enjoy the Birthday party
To solve a conflict you had with your family
