Preventing Recurrence of Depression: Long-Term Effects of a Randomized Controlled Trial on Cognitive Control Training for Remitted Depressed Patients

Abstract

Previous studies suggest that cognitive control training (CCT) shows potential as a preventive intervention for depression. In this study, the first to examine long-term preventive effects of CCT, we examined effects on (a) task-specific cognitive transfer at 1-year follow-up, (b) recurrence of depression, and (c) functioning over the course of a year. Each of 92 remitted depressed patients were randomly assigned to a CCT condition or an active control condition (ACT). Effects of training were monitored using weekly assessments of emotion regulation, cognitive complaints, depressive symptoms, and resilience (brief weekly questionnaire). At 1-year follow-up, participants completed a structured clinical interview, cognitive transfer task, and questionnaires. We observed task-specific cognitive transfer (p < .001, d = 1.23) and lower recurrence rates in the CCT condition (p = .04; odds ratio = 0.38). However, no long-term beneficial effects of training were observed on the weekly ratings of functioning, and groups did not differ in performance on the self-report questionnaires at 1-year follow-up.

Keywords

cognitive control cognitive training depression remission prevention open data preregistered

Depression is one of the world’s most debilitating disorders, affecting millions of people worldwide. On top of the major impact on the patients’ well-being, it is also associated with high societal costs (Ekman et al., 2013; Kessler, 2012). Depression is often chronic in nature: The risk of recurrence following treatment is estimated to be between 40% and 60% (Bockting et al., 2015). Crucially, relapse rates remain high despite treatment and are gradually climbing with increasing number of previous episodes (Bockting et al., 2006, 2010, 2015). Investigating innovative techniques that can reduce risk for recurrence is therefore of high priority. Given the well-documented presence of cognitive impairments after remission (e.g., Levens & Gotlib, 2015) that influence psychosocial functioning (Weightman et al., 2019), one such innovative technique is cognitive control training (CCT).

CCT refers to the use of computerized tasks to remediate executive functions, allowing flexible adaptation of one’s thoughts and behavior as a function of one’s goals (Braver et al., 2002; J. Cohen, 2017; Miller & Cohen, 2001). This may be of particular interest to patients following remission of depression given the importance of cognitive control for daily functioning. In particular, cognitive control deficits have repeatedly been linked to impaired emotion-regulation processes (for reviews, see Joormann & D’Avanzato, 2010; Joormann & Vanderlind, 2014; Mor & Daches, 2015), reflecting an increased cognitive and neurobiological vulnerability for recurrence of depression (De Raedt & Koster, 2010).

Among the various tasks that can be used to train cognitive control in the context of depression, a significant amount of the evidence stems from the adaptive paced auditory serial-addition task (aPASAT; Siegle et al., 2007; for a review, see Koster et al., 2017). On a neurobiological level, it has been suggested that to successfully perform the PASAT, one needs to recruit the dorsolateral prefrontal cortex (DLPFC) while downplaying limbic interference (e.g., because of the challenging nature of the task; Siegle et al., 2007). Given the importance of these regions for successful regulation of emotions (e.g., Buhle et al., 2014; De Raedt et al., 2010), experimental manipulation of cognitive control may beneficially affect emotion regulation (e.g., fostering disengagement from irrelevant repetitive negative thoughts, allowing one to override habitual ruminative response tendencies; Watkins & Roberts, 2020) and, in turn, decrease the likelihood of ending up in a state of sustained negative affect (De Raedt & Koster, 2010; Koster et al., 2011, 2017; Watkins & Roberts, 2020).

Initial neuroimaging studies suggest that CCT may increase connectivity between prefrontal and limbic regions (N. Cohen et al., 2016), decreasing disruptions at the level of the DLPFC and amygdala (N. Cohen et al., 2016; Siegle et al., 2007). On a behavioral level, aPASAT training has been shown to beneficially affect rumination (Siegle et al., 2007, 2014; Vanderhasselt et al., 2015) and depressive symptoms (Brunoni et al., 2014; Siegle et al., 2007) in patients suffering from major depressive disorder (MDD). Moreover, findings from a recent double-blind randomized controlled trial (RCT) study (Hoorelbeke & Koster, 2017) suggest that CCT shows potential as a preventive intervention following remission from depression. That is, in a remitted depressed (RMD) sample, it was found that 10 sessions of the aPASAT resulted in a decrease in depressive symptoms, less use of maladaptive emotion-regulation strategies (e.g., depressive rumination), and improved resilience (Hoorelbeke & Koster, 2017). Compared with an active control condition, these effects were most pronounced at 3-month follow-up. Moreover, initial gains in cognitive control predicted lower depressive symptoms at 3-month follow-up. This relation was partially mediated by immediate improvements in emotion regulation following CCT (Hoorelbeke & Koster, 2017). Furthermore, earlier work with undergraduate students scoring high on trait rumination also suggests that aPASAT training may affect stress reactivity and depressive rumination (Hoorelbeke et al., 2015). In addition, the immediate effects of aPASAT training was investigated in daily life using experience sampling methodology, suggesting effects on emotion regulation in response to daily life stressors in a convenience sample (Hoorelbeke et al., 2016) and RMD patients (Hoorelbeke et al., 2021). These findings suggest that the aPASAT may buffer against use of rumination, possibly because of a more efficient use of working memory and attentional processes, which might, in turn, allow for more successful downregulation of negative thoughts (Van den Bergh et al., 2018).

Findings from a recent systematic review (Koster et al., 2017) and meta-analysis (Motter et al., 2016) suggest that cognitive training has added value in the context of depression. However, several key questions remain to be addressed. For instance, prior clinical studies have typically evaluated short-term cognitive and emotional transfer effects of CCT. Several studies suggest that cognitive-training procedures may have a long-lasting impact on cognitive functioning (e.g., Schmiedek et al., 2014; Wilkinson & Yang, 2016). However, in the context of depression, repeated assessments of cognitive functioning following aPASAT training have typically been limited in time and exploring cognitive transfer immediately following training (Hoorelbeke et al., 2015, 2016; Siegle et al., 2007), at 3-month follow-up (Hoorelbeke & Koster, 2017), or 6-month follow-up (Vervaeke, Hoorelbeke, Baeken, & Koster, 2021).

Related to this, assessments of effects of aPASAT training on emotional functioning have typically also been limited in time (e.g., Hoorelbeke & Koster, 2017; Vervaeke et al., 2020), providing only indirect evidence for the potential of CCT as a preventive intervention for depression (for a review, see Koster et al., 2017). The value of CCT in terms of preventing recurrence of depression remains to be tested. Hence, we conducted an RCT to address these questions, exploring long-term effects of aPASAT training in RMD patients using a multimethod approach.

For this purpose, following an experimental manipulation of cognitive control (10 sessions of CCT vs. ACT) and evaluation of immediate effects of CCT on emotion-regulation dynamics using an experience sampling procedure of which the results are reported elsewhere (Hoorelbeke et al., 2021), participants were instructed to fill out a weekly questionnaire over a period of 1 year. This brief questionnaire contained indicators of severity of depressive symptoms, cognitive complaints, occurrence of stressful experiences, and indicators of emotion regulation over the past week. In addition, at 1-year follow-up, participants were reinvited to the lab where cognitive and emotional functioning was reassessed. At that time, using a structured clinical interview, we assessed recurrence of depressive episodes during the 1-year follow-up period. We had three aims:

Establish whether training-related gains on a closely related cognitive control task would be observed at 1-year follow-up. For this purpose, we reassessed a nonadaptive version of the PASAT task (i.e., the standardized PASAT task in which task difficulty is not tailored to the performance of the individual; Gronwall, 1977) as an indicator of task-specific transfer.

Evaluate effects of CCT in terms of prevention of recurrence of depression at 1-year follow-up.

Evaluate long-term effects of CCT on self-reported rumination, positive appraisal, cognitive complaints, depressive symptoms, and resilience. For this purpose, these variables were assessed over the course of a year using a brief weekly questionnaire in addition to standardized questionnaires at baseline and 1-year follow-up.

Method

Design

In this article, we report the findings of a single-blind RCT that was preregistered. The study was approved by the local ethics committee, and written informed consent was obtained for all participants. This trial consisted of (a) a screening phase; (b) a baseline clinical assessment; (c) a cognitive-training procedure (experimental manipulation) during which (d) immediate effects were assessed in daily life over a period of 4 weeks using experience sampling methodology; (e) a weekly monitoring of functioning over a period of 1 year using a brief questionnaire that was administered using participants’ smartphones; and (f) a follow-up assessment including the nonadaptive PASAT, questionnaires, and a clinical interview to reassess the state of stable remission at 1-year follow-up. Here, we report on the long-term effects of CCT, including the weekly assessments over a period of 1 year (weekly questionnaire) and the follow-up assessment 1 year after training. Results pertaining to the immediate effects of training in daily life (the experience sampling phase) are reported elsewhere (Hoorelbeke et al., 2021).¹ Each participant was randomly assigned to the experimental condition following the prescreening phase using RandList software (https://randomisation.eu). Participants were blind to the experimental condition. In addition, for evaluation of effects of CCT on recurrence of depression, we used independent raters to assess the interrater reliability for the clinical interviews (see below). The latter were also blind to the training condition of participants.

Participants

RMD patients were recruited via advertisements in local newspapers and on social media. The study was conducted at the Faculty of Psychology and Educational Sciences of Ghent University. Participants were considered eligible for participation in the study if they (a) were 23 to 65 years old, (b) reported a history of depression, (c) possessed a smartphone with a data plan, and (d) did not meet criteria for an ongoing depressive episode or report current substance abuse or presence of a psychotic disorder. Eligibility was assessed using a two-step screening protocol. Potential eligible participants first completed a brief telephone screening in which the relevant modules of the Mini-International Neuropsychiatric Interview (MINI; Sheehan et al., 1998; Van Vliet & De Beurs, 2007) were used. This is a structured clinical interview based on the Diagnostic and Statistical Manual of Mental Disorders (DSM). Eligible participants were invited to Ghent University for a second screening that included the mood disorders module of the MINI, which was administered by a licensed clinical psychologist, to establish or reestablish a lifetime history of depression and current remission. Given the power analyses reported in prior preventive CCT studies (Hoorelbeke & Koster, 2017), we aimed to include 68 RMD patients in the analysis of short- and long-term effects of CCT. For this purpose, we oversampled because we expected significant dropout given the initial phase of intensive daily monitoring, the experimental manipulation, and the 1-year follow-up period. Participants enrolled in the study between March and July 2017.

Materials

Experimental manipulation

The experimental manipulation was identical to the procedure set out by Hoorelbeke and Koster (2017). Participants were randomized over two conditions, the CCT condition (experimental group) or a closely matched active control condition (ACT). In the CCT condition, participants performed the aPASAT task. During the aPASAT, participants were presented with a continuous stream of auditory digits ranging from one to nine. Participants were instructed to continuously respond to the sum of the last two heard digits. Responses were made by clicking on the corresponding number in a fixed grid; responses ranged from two to 18. To train cognitive functioning, we adapted the pace of the task to the performance of the participant. That is, each session started with an intertrial interval (ITI) of 3,000 ms. After every four consecutive correct responses, this interval decreased by 100 ms, and after four consecutive incorrect responses, this interval increased by 100 ms, providing a stressful training context. During the aPASAT, participants need to recruit the DLPFC while downplaying limbic interference (Siegle et al., 2007). That is, this task loads heavily on working memory functioning such that it is assumed to train the ability to select goal-relevant information and responses in the presence of task-irrelevant information (i.e., distractors). Thus, this task is assumed to train central cognitive control features such as updating and interference control/inhibition (Van den Bergh et al., 2018).

Participants in the ACT performed a low cognitive load version of the aPASAT. That is, participants in the ACT group were presented with digits ranging from one to 18 and instructed to continuously click on the last heard digit (see Hoorelbeke et al., 2016). In line with the aPASAT task, the ITI decreased/increased by 100 ms following every four correct/incorrect responses. Thus, participants in the ACT condition performed a speed-of-response training. This task was designed to draw less heavily on working memory processes than the aPASAT while keeping all other characteristics similar to the experimental condition. This allowed us to control for motivational effects of cognitive training as well as effects of broader attentional processes that are likely to be trained in both conditions (e.g., sustained attention). Comparison between both conditions thus allowed us to evaluate effects of the working memory component of cognitive control.

In both training conditions, participants were instructed to complete 10 online sessions of the corresponding cognitive training task using the Millisecond software web application (https://www.millisecond.com/). Each session contained 400 trials. To foster task engagement (Siegle et al., 2014), we gave both groups standardized psycho-education regarding the cognitive-training procedure (Hoorelbeke et al., 2015).

Screening and outcome measures

Structured clinical interview

History of depression and recurrence of depression during the follow-up period were evaluated according to the criteria of the fifth edition of the DSM (DSM-5; American Psychiatric Association, 2013). In particular, we used the Dutch translation of the MINI (Sheehan et al., 1998; Van Vliet & De Beurs, 2007) during the screening phase to assess participants’ eligibility for the study. The Dutch version of this measure was developed on the basis of the DSM-IV. Thus, before this study, each relevant item of the Dutch version of the MINI was carefully checked for compliance with the DSM-5 criteria for mood disorders (American Psychiatric Association, 2013) and adjusted as necessary to keep wording between the items of the MINI and the novel criteria of the DSM-5 consistent.

Moreover, state of remission was reassessed at 1-year follow-up using the mood disorders module of the MINI. In particular, we evaluated recurrence of depression during the period of 1-year follow-up according to the DSM-5 criteria for depression (American Psychiatric Association, 2013). The interviews were conducted by K. Hoorelbeke and N. Van den Bergh, who also provided technical assistance to participants during the training phase and thus were not blind to the condition of participants. To prevent potential bias and to be able to evaluate the quality of the assessments, we recorded all interviews, and they were reassessed by two independent raters who were blind to the experimental condition. Each independent rater rated approximately 50% of the recorded interviews. This was conducted in a separate room, where the audio recordings were provided to the independent raters without further communication regarding the interviews between the researchers involved in this project and the independent raters during this stage of the project. In total, all interviews were rated twice. In case of disagreement between the independent raters and the researchers, we followed the categorization of recurrence made by the independent rater.

Interrater agreement between the researchers and independent raters was perfect (κ = 1.00; Orwin & Vevea, 2009). For each of the participants completing the follow-up assessment, the researchers (i.e., K. Hoorelbeke and N. Van den Bergh) and independent raters agreed in terms of diagnosis of absence or recurrence of depression. One participant was lost during the phase of follow-up because of suicide and thus did not complete the interview. Both the independent raters and researchers each independently categorized this as indicative for recurrence of depression.

Questionnaires

The following questionnaires were administered at baseline and at follow-up:

Residual depressive symptoms were assessed using the 21-item Beck Depression Inventory (BDI-II-NL; range = 0–63; Beck et al., 1996; Van der Does, 2002) and the Remission of Depression Questionnaire (RDQ; 41 items, range = 0–82; Zimmerman et al., 2013a, 2013b).

Positive and negative affect were assessed using the 20-item trait version of the Positive and Negative Affect Scales (PANAS; Engelen et al., 2006; Watson et al., 1988), with a range of 10 to 50 for both positive and negative affect.

Adaptive and maladaptive emotion regulation were assessed using the Ruminative Response Scale (RRS; Nolen-Hoeksema & Morrow, 1991; Treynor et al., 2003) and the Cognitive Emotion Regulation Questionnaire (CERQ; Garnefski et al., 2001). In particular, we used the total score of the RRS (22 items, range = 22–88; Treynor et al., 2003) and computed a compound score for use of adaptive emotion-regulation strategies (20 items, range = 20–100) and maladaptive emotion-regulation strategies (16 items, range = 16–80) using the CERQ.

We used the Global Executive Scale of the Behavior Rating Inventory of Executive Function Adult Version (BRIEF-A; 75 items, range = 70–210; Scholte & Noens, 2011) as an indicator of cognitive complaints.

Resilience was assessed using the Resilience Scale (RS; 25 items, range = 25–100; Portzky, 2008; G. Wagnild & Young, 1993).

To control for motivational effects, we also included the Credibility/Expectancy Questionnaire (CEQ; six items; Devilly & Borkovec, 2000) at baseline and at 1-year follow-up. In addition, at follow-up, participants filled out an item assessing whether they had the feeling that they were randomized to a placebo condition (placebo: “yes” or “no”; see Hoorelbeke & Koster, 2017).

Furthermore, at 1-year follow-up, participants completed several items pertaining to the need for treatment during the past year. Two items assessed potential changes in experienced need for pharmacological or psychotherapeutic treatment for depression. In particular, participants indicated whether their use of psychotherapy/medication remained stable, increased, or decreased over the past year. In addition, participants reported number of psychotherapeutic consults and days administered to a psychiatric hospital because of depression over the past year.

Finally, participants also completed the List of Threatening Experiences (LTE; range = 0–13; Brugha & Cragg, 1990; Rosmalen et al., 2012)), allowing us to take into account the occurrence of life events during the follow-up phase. For the BDI-II-NL, RDQ, PANAS Negative Affect Scale, RRS, CERQ compound score for maladaptive emotion regulation, BRIEF-A, and LTE, a high score is indicative of more psychopathological/maladaptive processes. In contrast, for the PANAS Positive Affect Scale, the CERQ compound score for adaptive emotion regulation, the RS, and the CEQ, a high score is indicative for more adaptive processes.

Weekly questionnaire

We developed a brief questionnaire to monitor effects of the experimental manipulation on functioning over a period of 1-year follow-up. For this purpose, seven items were reassessed on a weekly basis using participants’ own smartphones. On a fixed day of the week, participants would receive a text message (sent using SurveySignal software; Hofmann & Patel, 2015) containing a unique link directing them to an online Limesurvey questionnaire (https://www.limesurvey.org/).

The online questionnaire contained seven items measuring six constructs:

Stressors. Participants were instructed to rate the extent to which, in the previous week, the most influential event or events that occurred were unpleasant rather than pleasant events (“Think about the most influential event you experienced over the past week. How (un)pleasant was this?”; 1 = very unpleasant, 100 = very pleasant).

Rumination (two items). Participants were instructed to rate the extent to which, in the previous week, they had engaged in repetitive negative thinking regarding their (a) feelings and (b) problems, as an indicator of depressive rumination (“Focused on feelings,” “Focused on problems”; Moberly & Watkins, 2008). These two rumination items were developed by Moberly and Watkins (2008) to measure ruminative self-focus in a way that is consistent with the response styles theory (Moberly & Watkins, 2008; Nolen-Hoeksema & Morrow, 1991).

Positive appraisal. Participants were instructed to rate the extent to which, in the previous week, they had deployed positive appraisal, cognitively appraising or reappraising events in a positive manner (“Focused on a positive meaning”). The positive-appraisal item was proposed by Hoorelbeke et al. (2016) as an adaptive counterpart, aiming to assess positive appraisal as an important adaptive emotion-regulation strategy in the context of depression vulnerability.

Resilience. Participants rated the extent to which they “felt resilient.” In line with the operationalization of resilience in the RS (the perceived ability to cope with adversity, including aspects such as inner strength and optimism; G. Wagnild & Young, 1993; R. N. Wagnild, 2009), resilience was defined as the extent to which participants considered themselves able to effectively cope with adversity in daily life.

Cognitive complaints. Participants also rated the extent to which they experienced cognitive complaints (“Experienced cognitive complaints”), which referred to experiencing working memory complaints or complaints related to broader executive functions such as attention deficits. These deficits typically remain present following remission of depression (Semkovska et al., 2019).

Depression. Participants also rated the extent to which they experienced depressive symptoms (“Experienced depressive symptoms”), which referred to main depressive features described in the DSM (e.g., being in a state of sustained negative affect, experiencing anhedonic features such as loss of interest). This item—in line with the depression item proposed by Richmond et al. (2015)—aimed to capture a wide spectrum of experiences relating to depression.

The three emotion-regulation items (1, 2, and 3, above) have been used in previous studies to model effects of CCT on emotion-regulation dynamics in daily life in a convenience sample (Hoorelbeke et al., 2016).

With the exception of occurrence of pleasant or unpleasant events, all items were rated on a scale from 1 (not at all) to 100 (very much). Each assessment started with the item measuring the occurrence of stressful/pleasant events, after which all other items were administered in a random order (to prevent order effects and sustain motivation to complete the repeated assessments).

To familiarize participants with the items of the weekly assessment, we asked the participants to respond to these items for the first time during the baseline lab visit after they received written and oral instructions regarding the content of the items. The weekly online assessments started 4 weeks after the baseline assessment, allowing participants to complete the experimental manipulation. The assessments continued on a weekly basis for a period of 1 year following the experimental manipulation and ended during the follow-up visit. At that point, the items were reassessed for the last time. The day of the week at which the text message was sent varied across participants and was determined by the initial date of the baseline assessment. Signals were always sent at 6 p.m. Participants were requested to complete the questionnaire within 24 hr after receiving the first signal. If participants did not complete the questionnaire within 1 hr, a reminder signal was sent. After 24 hr, the link timed out, and participants could no longer respond to the questionnaire.

Analysis of the associations between responses on the measures used in the brief weekly questionnaire to assess emotion regulation, cognitive complaints, depressive symptoms, and resilience and the corresponding standardized self-report measures provides support for the validity of the brief weekly questionnaire. In addition, the test–retest reliability and internal consistency of the items were adequate. For a more detailed discussion of the psychometric properties of the brief weekly questionnaire, please see the Supplemental Material available online.

Task-specific cognitive transfer

In line with Hoorelbeke and Koster (2017), we relied on the nonadaptive PASAT task (Gronwall, 1977) to assess task-specific cognitive transfer. The nonadaptive or “standardized” PASAT task has been used to assess executive functioning in a multitude of clinical conditions, including traumatic brain injury, psychosis, and mood disorders (for a review, see Tombaugh, 2006). This task differs from the training task in that it forms an assessment task. That is, in contrast to the training task in which task difficulty is continuously modified according to one’s response (i.e., the aPASAT), the nonadaptive PASAT relies on ITIs that are fixed within blocks to assess executive functioning. Specifically, following a practice phase consisting of 10 trials, participants completed three blocks with increasing difficulty using an ITI of 3,000 ms, 2,000 ms, and 1,500 ms. Each block contained 60 trials, resulting in a total of 180 trials. The nonadaptive PASAT task was administered at baseline and at 1-year follow-up using the INQUISIT Millisecond software package.

Procedure

Following the two-phased screening procedure (which included a structured clinical interview), eligible RMD participants provided informed consent and completed a baseline assessment. This contained multiple questionnaires, a behavioral measure for cognitive control (nonadaptive PASAT), and psycho-education regarding the weekly questionnaire assessments, the cognitive-training procedure, and its rationale. At the end of the baseline assessment, participants’ smartphones were registered in SurveySignal. In the period of 4 weeks following the baseline assessment, participants completed the online training procedure and were intensively monitored using experience-sampling methodology. Following training, participants entered a follow-up phase during which functioning was repeatedly assessed on a weekly basis over a period of 1 year using a brief online questionnaire. At 1-year follow-up, participants were reinvited to the lab for a follow-up assessment of functioning. This included a reassessment of the baseline questionnaire and the cognitive transfer task. Most importantly, at 1-year follow-up, recurrence of depression was assessed using a structured clinical interview. At the end of this session, participants were debriefed and received financial reimbursement for participation in the study.

Results

Group characteristics

After the screening phase, each of 92 participants was randomly assigned to the CCT or ACT condition (for the CONSORT patient flow diagram, see Fig. 1). The two groups did not significantly differ before training on age, t(87.35) = 0.04, p = .97; gender, χ²(1) = 0.87, p = .39; indicators of socioeconomic status (education: p = .20; employment status: p = .34; Fisher’s exact tests), use of medication (all ps > .27, Fisher’s exact tests), type of mood disorder (MDD vs. bipolar disorder: p > .99, Fisher’s exact test), or history of depression, measured by number of previous episodes: χ²(3) = 0.62, p = .92; mean episode length: t(90) = 0.49, p = .63; and time in remission: t(90) = 0.28, p = .78; for descriptive statistics, see Table 1). In addition, no significant group differences emerged at the level of the baseline questionnaires or performance on the cognitive transfer task (Table 2; all ts < 1.81).

Fig. 1.

CONSORT flow diagram. Participants who did not complete at least 50% of the weekly assessments or were lost to follow-up did not significantly differ from participants who were included for the analyses in terms of level of baseline functioning or clinical characteristics. However, we did observe a tendency for participants who did not complete the weekly assessments to report lower levels of maladaptive emotion regulation at baseline (as measured using the Cognitive Emotion Regulation Questionnaire (Garnefski et al., 2001; completers: M = 40.20, SD = 8.41; noncompleters: M = 35.45, SD = 7.46), t(90) = 1.78, p = .08; all other ps ≥ .12.

Table 1.

Group Characteristics

Characteristic	Training condition
Characteristic	CCT (N = 47)	ACT (N = 45)
Age	M = 45.74 (SD = 14.21)	M = 45.84 (SD = 11.40)
Gender
Male	19	14
Female	28	31
Highest degree obtained
No higher education	13	7
Bachelor’s degree	13	20
Master’s degree	20	18
PhD	1	0
Employment
Full-time	17	16
Part-time	11	17
Unemployed	3	2
Retired	6	4
Incapacitated	6	6
Student	4	0
Number of previous episodes
1	9	8
2	10	7
3	10	11
≥ 4	18	19
Mean self-reported episode length (months)	M = 7.27(SD = 9.94)	M = 6.36(SD = 7.74)
Mean self-reported time since previous episode (months)	M = 31.80(SD = 44.08)	M = 29.29(SD = 40.96)
RMDs reporting . . .
Depressive episodes within a bipolar disorder (%)	6	7
Current use of antidepressant medication (%)	51	49
Current use of benzodiazepines (%)	13	22
Current use of lithium (%)	6	4
Current use of antipsychotic or neuroleptic medication (%)	2	2

Note: Values are ns unless otherwise specified. CCT = cognitive control training; ACT = active control condition; RMD = remitted depressed participant.

Table 2.

Descriptive Information for Cognitive-Task Performance and Self-Report Questionnaires per Condition Over Time

Variables	CCT				ACT
	Time 1 (N = 47)		Time 2 (N = 43)		Time 1 (N = 45)		Time 2 (N = 39)
	M	SD	M	SD	M	SD	M	SD
Cognitive control (nonadaptive PASAT)	0.55	0.17	0.81	0.13	0.56	0.17	0.62	0.18
Depressive symptoms (BDI-II-NL)	11.28	8.58	9.51	8.75	13.07	9.66	11.38	10.16
Residual symptoms (RDQ)	22.40	16.49	20.37	15.45	24.87	14.38	23.38	17.60
Positive affect (PANAS)	34.77	7.12	35.12	8.46	33.33	7.20	33.28	7.25
Negative affect (PANAS)	23.15	6.70	23.30	7.43	25.42	6.07	23.67	6.96
Rumination (RRS)	49.96	12.57	45.95	14.44	53.98	12.83	48.90	13.30
Adaptive emotion regulation (CERQ)	65.23	13.27	67.51	14.46	61.33	16.73	64.41	16.33
Maladaptive emotion regulation (CERQ)	38.85	7.61	36.12	8.05	40.44	9.19	39.38	9.44
Resilience (RS)	73.06	11.64	75.53	12.69	74.42	11.13	76.49	11.58
Cognitive complaints (BRIEF-A)	123.70	20.22	118.91	22.85	131.33	20.25	125.13	25.21
Credibility (CEQ)	0.02	2.47	0.16	2.40	−0.02	2.27	−0.17	2.90
Expectancy (CEQ)	0.15	2.67	−0.04	2.80	−0.16	2.59	0.05	2.83

Note: CCT = cognitive control training; ACT = active control condition; PASAT = paced auditory serial-addition task (Siegle et al., 2007); BDI-II-NL = Beck Depression Inventory (Beck et al., 1996; Van der Does, 2002); RDQ = Remission of Depression Questionnaire (Zimmerman et al., 2013a, 2013b); PANAS = Positive and Negative Affect Scales (Engelen et al., 2006; Watson et al., 1988); RRS = Ruminative Response Scale (Nolen-Hoeksema & Morrow, 1991; Treynor et al., 2003); CERQ = Cognitive Emotion Regulation Questionnaire (Garnefski et al., 2001); RS = Resilience Scale (Portzky, 2008; G. Wagnild & Young, 1993); BRIEF-A = Behavior Rating Inventory of Executive Function Adult Version (Scholte & Noens, 2011); CEQ = Credibility/Expectancy Questionnaire (Devilly & Borkovec, 2000).

Participant blinding

We conducted two 2 (time: baseline vs. follow-up) × 2 (group: CCT vs. ACT) mixed analyses of variance (ANOVAs) to test for motivational effects of undergoing the experimental manipulation. In line with our expectations, no significant interaction effects were found for credibility and expectancy of the intervention (CEQ; see Table 3; for descriptives, see Table 2). In addition, only a limited number of participants reported perceiving the intervention as being a placebo (placebo:no-placebo ratio = 6:37 in the CCT group and 9:30 in the ACT group). Note that this proportion did not significantly differ between the two groups (χ² = 1.14, p = .29). Together, these findings suggest that blinding of participants was successful.

Table 3.

Evidence for Training Effects From Mixed Analyses of Variance

Variable	Main effect of time			Main effect of group			Time × Group interaction
Variable	F	p	η_p²	F	p	η_p²	F	p	η_p²
Cognitive control (nonadaptive PASAT)	168.53	< .001	.68	5.51	.021	.06	78.41	< .001	.50
Credibility (CEQ)	0.06	.80	< .01	0.06	.82	< .01	0.39	.54	< .01
Expectancy (CEQ)	< 0.01	.99	< .01	0.03	.88	< .01	0.23	.63	< .01
Depressive symptoms (BDI-II-NL)	1.49	.23	.02	0.77	.38	.01	0.10	.75	< .01
Residual symptoms (RDQ)	0.88	.35	.01	0.41	.52	< .01	0.31	.58	< .01
Positive Affect (PANAS)	0.18	.68	< .01	0.96	.33	.01	0.21	.65	< .01
Negative Affect (PANAS)	1.37	.25	.02	0.81	.37	.01	1.44	.23	.02
Rumination (RRS)	12.75	.001	.14	1.87	.18	.02	0.24	.63	< .01
Adaptive emotion regulation (CERQ)	3.12	.081	.04	0.71	.40	.01	0.14	.71	< .01
Maladaptive emotion regulation (CERQ)	5.13	.026	.06	2.87	.094	.04	0.34	.56	< .01
Resilience (RS)	4.94	.029	.06	0.30	.58	< .01	0.18	.67	< .01
Cognitive complaints (BRIEF-A)	7.49	.008	.09	2.12	.15	.03	0.03	.86	< .01

Note: In all cases, the degrees of freedom for the F test were (1, 80). PASAT = paced auditory serial-addition task (Siegle et al., 2007); CEQ = Credibility/Expectancy Questionnaire (Devilly & Borkovec, 2000); BDI-II-NL = Beck Depression Inventory (Beck et al., 1996; Van der Does, 2002); RDQ = Remission of Depression Questionnaire (Zimmerman et al., 2013a, 2013b); PANAS = Positive and Negative Affect Scales (Engelen et al., 2006; Watson et al., 1988); RRS = Ruminative Response Scale (Nolen-Hoeksema & Morrow, 1991; Treynor et al., 2003); CERQ = Cognitive Emotion Regulation Questionnaire (Garnefski et al., 2001); RS = Resilience Scale (Portzky, 2008; G. Wagnild & Young, 1993); BRIEF-A = Behavior Rating Inventory of Executive Function Adult Version (Scholte & Noens, 2011).

Question 1: task-specific cognitive transfer

Task-specific transfer effects were assessed using a nonadaptive version of the PASAT task that was administered at baseline and at 1-year follow-up. Using a 2 (time: baseline vs. follow-up) × 2 (group: CCT vs. ACT) mixed ANOVA, we observed a significant interaction for nonadaptive PASAT task performance (Table 3). Results from follow-up paired samples t tests suggest that both groups showed a significant increase in cognitive-task performance from baseline to 1-year follow-up: CCT, t(42) = 13.88, p < .001, 95% confidence interval (CI) = [0.23, 0.31], d_av = 1.80; ACT, t(38) = 3.49, p = .001, 95% CI = [0.02, 0.08], d_av = 0.28.² In absence of baseline group differences, t(90) = 0.35, p = .73, 95% CI = [−0.08, 0.06], d = 0.06, participants in the CCT condition significantly outperformed participants in the ACT condition at 1-year follow-up, t(68.30) = 5.29, p < .001, 95% CI = [0.12, 0.26], d = 1.23 (for descriptives, see Table 2).

Question 2: recurrence of depression

Recurrence of depressive episodes was retrospectively assessed at 1-year follow-up using a structured clinical interview. One participant in the ACT condition was lost to follow-up because of suicide. This tragic event was categorized as recurrence of depression. Thus, this analysis is based on 40 participants in the ACT group and 43 participants in the CCT group (Fig. 1). Overall, we observed high recurrence rates, and 36% (N = 30) of the participants met criteria for a depressive episode throughout the course of the follow-up period. Results from a Pearson χ² test suggest a significant effect of training condition on recurrence of depression (χ² = 4.31, p = .04, Cramér’s V = .23), demonstrating reduced recurrence of depressive episodes in the CCT condition (25.58%) compared with the ACT group (47.50%; nonrecurrence:nonrecurrence ratio = 11:32 in the CCT group, 19:21 in the ACT group; odds ratio [OR] = 0.38). Thus, the odds of recurrence of depression during 1-year follow-up were 2.63 times higher for participants in the ACT condition than for participants in the CCT condition.³

In addition, at 1-year follow-up, several indirect indicators of recurrence were assessed, including self-reported need for pharmacological or psychotherapeutic treatment for depression, number of psychotherapeutic consults, and days administered to a psychiatric hospital because of depression over the past year (CCT: N = 43; ACT: N = 39). Training condition did not affect experienced need for pharmacological treatment for depression (χ² = 0.79, p = .68, Cramér’s V = .10). That is, in both groups, most participants reported that use of pharmacological treatment for depression remained stable over the past year (CCT = 79%; ACT = 74%). A limited portion of participants reported a reduced (CCT = 5%, ACT = 3%) or increased (CCT = 16%, ACT = 23%) need for pharmacological treatment over the past year. In contrast, a significant effect of training condition emerged at the level of reported need for psychotherapeutic treatment for depression (χ² = 7.35, p = .03, Cramér’s V = .30). Specifically, in the CCT group, 35% of the participants reported a reduced need for treatment. Fifty-one percent reported that need of psychotherapeutic treatment for depression remained stable over the past year. An additional 14% reported an increased need for treatment in the CCT condition. In contrast, only 13% of the ACT condition reported a reduced need for psychotherapy over the past year, whereas 79% and 8% of the ACT condition reported stable or increased need, respectively, for psychotherapeutic treatment over the past year.

However, no significant group differences emerged at the level of consumption of health care services. For instance, according to an independent-samples t tests, number of therapeutic consults used during the past year did not significantly differ between both conditions (CCT: M = 6.47, SD = 9.52; ACT: M = 7.13, SD = 9.75), t(80) = 0.31, p = .76, 95% CI = [−4.90, 3.58], d = 0.07. Moreover, only one participant was hospitalized during the course of the follow-up period. Thus, the groups did not significantly differ in the number of participants that were hospitalized because of recurrence of depression over the past year (hospitalized:nonhospitalized ratio = 0:43 in the CCT group and 1:38 in the ACT group; p = .48, Fisher’s exact test). Thus, although relatively more participants in the CCT group reported experiencing a decreased need for psychotherapy compared with the ACT condition, this was not reflected by lower actual use of health care services over the past year.

Question 3a: weekly ratings of functioning

Eighty-one participants completed at least 50% of the weekly assessments over the period of 1-year follow-up and were included for these analyses (Fig. 1). The exact length of the weekly follow-up period was determined by the 1-year follow-up visit in the lab, after which participation in the study terminated. This was situated between Week 52 and Week 60 following the baseline assessment (Mdn = 55). Given that the number of participants in each training condition rapidly declined after assessment Week 52, we included only the first 52 weeks for analysis of effects of CCT on weekly ratings of functioning at group level. Given the nature of the design, no weekly assessments took place between baseline and Week 4. This resulted in a maximum of 4,050 unique assessments. Including the baseline assessment, participants completed 3,670 assessments; average response was 90.62%.

We used hierarchical linear modeling to examine effects of the experimental manipulation on rumination (intraclass correlation coefficient [ICC] = .60), use of positive appraisal (ICC = .58), cognitive complaints (ICC = .56), depressive symptoms (ICC = .42), and self-reported resilience (ICC = .51) over the course of 1 year. At Level 1, we modeled how variation in the weekly assessments of the variable of interest was an effect of time (number of weeks since baseline assessment) by allowing modeling of linear and quadratic changes for each of the variables of interest over time. At the second level, we modeled the effect of training condition (CCT or ACT) on the slope of time (linear) and time² (quadratic), allowing random effects for the intercept and slope of time and time². We relied on an autoregressive AR(1) structure to account for the temporal correlation in an individual. Separate hierarchical linear models were fitted for each of the variables of interest. Results of the analyses are reported in Table 4.

Table 4.

Evidence for Training Effects in Daily Life From Hierarchical Linear Modeling

Variable and term	γ	SE	t
Pleasant or unpleasant events
Time	−0.203	0.213	t(3585) = −0.95
Time²	0.003	0.004	t(3585) = 0.80
Group	2.073	4.410	t(79) = 0.47
Time × Group	−0.071	0.297	t(3585) = −0.24
Time² × Group	< 0.001	0.005	t(3585) = 0.03
Rumination
Time	−0.123	0.188	t(3585) = −0.66
Time²	0.003	0.003	t(3585) = 0.82
Group	−6.214	4.307	t(79) = −1.44
Time × Group	0.138	0.261	t(3585) = 0.53
Time² × Group	−0.003	0.005	t(3585) = −0.53
Positive appraisal
Time	−0.270	0.158	t(3585) = −1.70^†
Time²	0.006	0.003	t(3585) = 1.95^†
Group	−0.868	4.610	t(79) = −0.19
Time × Group	0.006	0.220	t(3585) = 0.03
Time² × Group	−0.001	0.004	t(3585) = −0.24
Cognitive complaints
Time	−0.399	0.199	t(3584) = −2.00*
Time²	0.007	0.004	t(3584) = 2.05*
Group	−1.099	3.907	t(79) = −0.28
Time × Group	0.108	0.277	t(3584) = 0.39
Time² × Group	−0.003	0.005	t(3584) = −0.61
Depressive symptoms
Time	0.248	0.201	t(3585) = 1.24
Time²	−0.004	0.004	t(3585) = −1.16
Group	0.348	3.684	t(79) = 0.09
Time × Group	−0.232	0.280	t(3585) = −0.83
Time² × Group	0.004	0.005	t(3585) = 0.79
Resilience
Time	−0.026	0.142	t(3585) = −0.18
Time²	0.001	0.003	t(3585) = 0.48
Group	1.471	3.844	t(79) = 0.38
Time × Group	−0.283	0.198	t(3585) = −1.43
Time² × Group	0.004	0.004	t(3585) = 1.12

†

p < .10. *p < .05.

In contrast to our hypotheses, no significant Time × Group interaction effects emerged for the variables of interest. Moreover, none of the main effects reached significance, with the exception of a significant linear and quadratic function for change in self-reported cognitive complaints over time. That is, independent of the cognitive-training procedure used (CCT or ACT), participants reported an initial decrease in cognitive complaints that remained fairly stable as time advanced (see Fig. S1 in the Supplemental Material). In addition, we observed a tendency for an initial decrease in use of positive appraisal (see Fig. S2 in the Supplemental Material). We observed no significant main or interaction effects for the weekly ratings of pleasant or unpleasant events (Table 4).

Question 3b: self-report questionnaires at 1-year follow-up

Several questionnaires were administered at baseline and 1-year follow-up. Effects of training condition over time were assessed using 2 (time: baseline vs. follow-up) × 2 (group: CCT vs. ACT) mixed ANOVAs. The results of these analyses are reported in Table 3 (for descriptives, see Table 2). Overall, in contrast to our hypotheses, no significant Time × Group interaction effects were observed. For several indicators of emotion regulation, a main effect of time was observed. Participants reported a decrease in rumination (RRS) and maladaptive emotion regulation in general (CERQ). Furthermore, participants reported an increase in resilience (RS) over time and a reduction in cognitive complaints (BRIEF). However, no differential effects of training condition emerged at 1-year follow-up.

In addition, participants were asked to report life events that had occurred over the past year using the LTE. Effects of an independent-samples t test suggest that the groups did not differ in terms of occurrence of life events (CCT: M = 1.84, SD = 1.36; ACT: M = 1.82, SD = 1.35), t(80) = 0.06, p = .96, 95% CI = [−3.11, 3.15], d < 0.01.

Discussion

The aim of the current study was to examine the long-term effects of CCT for remitted depressed patients. In particular, we aimed to examine cognitive and emotional transfer effects over a period of 1-year follow-up. Given prior work suggesting that task-specific transfer effects remain present until 3 months after training (Hoorelbeke & Koster, 2017), our first aim was to examine task-specific transfer at 1-year follow-up. In line with our hypothesis, participants in the CCT condition showed a large increase in cognitive-task performance from baseline to 1-year follow-up, which significantly differed from the ACT condition.

At 1-year follow-up, recurrence rates were found to be significantly lower in the CCT condition compared with the ACT group, suggesting that CCT may indeed be effective as a preventive intervention for RMD patients. This finding is in line with previous studies suggesting that CCT may reduce depression vulnerability, as shown by previously observed short-term beneficial effects of aPASAT training in healthy (e.g., Hoorelbeke et al., 2016), at-risk (e.g., Calkins et al., 2015; Hoorelbeke et al., 2015; Hoorelbeke & Koster, 2017), and clinically depressed patient samples (e.g., Brunoni et al., 2014; Segrave et al., 2014; Siegle et al., 2007, 2014; Vanderhasselt et al., 2015). Note that the obtained OR of 0.38 corresponded to prior reported effects of widely used preventive interventions in the context of depression. For instance, in a meta-analysis comparing effects of antidepressant medication such as venlafaxine in terms of prevention of recurrence of depression with a placebo intervention, Bauer et al. (2009) reported an OR of 0.37 in favor of venlafaxine. Likewise, Cox and colleagues (2012) reported an OR of 0.34 in a meta-analysis of effects of antidepressants on recurrence of depression in children and adolescents.

Furthermore, the obtained effect of CCT on risk for recurrence of depression was of similar magnitude as previously reported preventive effects of mindfulness-based cognitive therapy (MBCT). For instance, Teasdale and colleagues (2000) compared effects of MBCT + treatment as usual (TAU) with a TAU control group and reported beneficial effects of MBCT in terms of prevention of recurrence of depression that were moderated by number of prior depressive episodes. However, when not taking into account the number of prior depressive episodes in RMD patients, the overall observed OR for MBCT + TAU compared with TAU was 0.57 (Teasdale et al., 2000), suggesting that risk for recurrence of depression was 1.75 times higher in the TAU group compared with the MBCT + TAU group, respectively. In a follow-up study, Ma and Teasdale (2004) replicated the beneficial additive effects of MBCT to TAU in terms of prevention of recurrence of depression, resulting in an OR of similar magnitude (OR = 0.39). These findings highlight the clinical potential of CCT as a preventive intervention for remitted depressed patients. Compared with existing preventive interventions for depression, one of the main advantages of CCT is that it can be easily disseminated online, offering the possibility to reach a large population in a cost-effective manner. In addition, CCT lends itself to be integrated within existing treatment protocols and is less time consuming.

In addition to the observed beneficial effects in terms of reduced risk for recurrence of depression, participants in the CCT condition also reported having experienced a reduced need for psychotherapeutic treatment over the past year compared with the ACT condition. This may be indicative for increased feelings of agency in the CCT condition. However, neither condition significantly differed in terms of absolute number of psychotherapeutic consults or in occurrence of hospitalization because of recurrence of depression. Note that in this context, number of psychotherapeutic consults includes maintenance sessions, including sessions aimed at monitoring the use of antidepressant medication. However, these findings may also suggest that CCT as a stand-alone intervention did not have a clinically significant impact on daily life functioning over the course of 1 year. Related to this, neither group differed in terms of estimated need for psychopharmacological treatment for depression in that most participants reported having experienced a stable need for use of preventive psychopharmacological treatment for depression. This is in line with previous findings (Siegle et al., 2014) suggesting no effects of CCT on number of outpatient psychotherapeutic consults or on medication management visits. In the latter study, only an effect of CCT was found on number of intensive outpatient day-treatment visits (Siegle et al., 2014). It would be interesting for future studies to follow up on this finding and monitor quality of life, use of health care services, and related costs in a more rigid manner, not strictly relying on retrospective self-report measures. This would allow evaluation of the potential of CCT from a health-economic perspective.

A third aim of the study was to model long-term effects of CCT on indicators of functioning. Although previous findings suggest that beneficial effects of CCT on indicators of vulnerability for depression may continue until 3 months following training (e.g., Hoorelbeke & Koster, 2017), no differential effects of training condition were found on self-report questionnaires for mood, emotion regulation, residual symptoms, and resilience during the assessment of functioning at 1-year follow-up. Instead, both groups demonstrated a significant decrease in rumination, use of general maladaptive emotion regulation strategies, and cognitive complaints. Furthermore, we observed a significant increase in resilience over time in both groups. This may be due to overlap between the domains of influence of both training procedures (i.e., potentially affecting broader attentional processes). This result is in line with a recent replication study providing mixed evidence for the emotional transfer effects of aPASAT training compared with a similar active control training in RMD patients (Vervaeke et al., 2020). That is, in contrast to Hoorelbeke and Koster (2017), Vervaeke and colleagues (2020) did not observe beneficial effects of aPASAT training on self-reported emotion regulation or resilience over a 6-month follow-up period. Instead, Vervaeke et al. observed a general reduction in rumination over time. Vervaeke et al. did, however, observe beneficial effects of aPASAT training on depressive symptoms, which became apparent at 3-month follow-up and remained present until 6-month follow-up.

To shed further light on how emotional transfer effects of CCT develop over time, we monitored participants’ functioning over a period of 1 year using weekly assessments of functioning. In particular, use of adaptive or maladaptive emotion-regulation strategies, self-reported cognitive complaints, depressive symptoms, and resilience were reassessed weekly over the period of 1 year. Here, we were interested in emotional transfer effects over time, given the high clinical value of such data (e.g., allowing one to detect when booster sessions may be needed). In line with previous studies with RMD patients (e.g., Weissman et al., 1978), the subjects included in the current study showed considerable fluctuations in residual symptoms. However, using the weekly assessments of functioning, our analyses of impact of training on severity of residual symptoms, emotion regulation, and resilience over the period of 1 year suggested no differential long-term effects of CCT. Instead, both groups showed an initial decrease in cognitive complaints, which was maintained throughout the follow-up period. The lack of differential effects of CCT on the weekly ratings and standardized questionnaires that were reassessed at 1-year follow-up seems to be at odds with previously reported findings in the literature (for a recent review, see Koster et al., 2017) and the observed effect of CCT on risk for recurrence of depression.

Identifying which factors may have contributed to the observed discrepancy between the analyses pertaining to effects of training condition on cognitive transfer, functioning throughout the course of a year, and recurrence risk is challenging given the vast differences between the methods that support these results. Thus, several factors may have contributed to this. For instance, for the weekly ratings, participants were required to indicate to what extent they had experienced “depressive symptoms” in the past week on a scale ranging from 1 to 100 using a one-item question. This offers a cruder assessment of severity of depressive symptoms. During the structured clinical interview, in contrast, participants were asked in depth whether specific depressive symptoms were present most of the time, for at least two consecutive weeks, sometime in the past year. Related to this, a potential source of discrepancy is located at the level of data analysis. That is, our analyses of functioning over the course of a year aimed to model increases or decreases at group level in symptoms over time. Consequently, this does not take into account the temporary chunking of elevated depressive symptoms over time at an individual level, which is key to the diagnosis of recurrence of depression (the 2-week criteria). Thus, to clarify the observed discrepancies between findings stemming from analysis of the weekly ratings of depressive symptoms and the diagnosis of recurrence of depression at 1-year follow-up, an alternative data-analytical approach than the approach that was preregistered is necessary.

For this purpose, we conducted post hoc survival analyses on the weekly data using the area under the curve indicator to statistically inform us of the ideal cutoff value for recurrence of depression according to the weekly ratings (taking into account temporal chunking of the data to meet the 2-week criteria). Results of these analyses are presented in the Supplemental Material. In line with the statement that CCT may have beneficial effects on recurrence and risk for recurrence of depression, we observed a tendency for both survival distributions to differ (see the Supplemental Material). In particular, a discrepancy between the CCT and ACT condition seemed to develop around the period of 3 months after the intervention. This is in line with recent findings suggesting a delayed effect of CCT on depressive symptoms in RMD patients (e.g., Vervaeke et al., 2020). On the basis of the obtained cutoff for the weekly ratings, estimated recurrence rates were 31.82% in the CCT condition compared with 51.22% in the ACT condition. This resembles the recurrence rates from the structured clinical interview at 1-year follow-up (CCT = 25.58%; ACT = 47.50%). Although these findings suggest that CCT may affect risk for recurrence of depression, note that the findings stemming from the standardized questionnaires suggest that this did not result in reduced severity of residual depressive complaints in the CCT condition at 1-year follow-up.

Our results also suggest that RMD patients show substantial fluctuations in self-reported emotion regulation and depressive symptoms. These findings may account for inconsistencies within the CCT literature. That is, although multiple studies suggest the potential of CCT in reducing depression vulnerability (e.g., Brunoni et al., 2014; Calkins et al., 2015; Martin et al., 2018; Segrave et al., 2014; Siegle et al., 2007, 2014; Vanderhasselt et al., 2015), note that several studies have failed to identify short-term emotional transfer effects after aPASAT training or other CCT procedures. For instance, comparing effects of CCT with an active control training in participants with an euthymic or depressed mood, Moshier et al. (2015) observed no beneficial effects of aPASAT and Wells’s attention training on depressive symptoms. Likewise, Wanmaker et al. (2015) observed no emotional transfer effects following an intensive combined adaptive dual n-back and symmetry span training in patients with MDD and/or anxiety. Moreover, in a recent study in which Moshier and Otto (2017) evaluated the efficacy of aPASAT and Wells’s attention training as an add-on intervention for behavioral activation in MDD patients, no beneficial effects of adjunctive CCT were observed. In this context, note that previous studies have typically relied on one or two posttraining assessments to evaluate effects of CCT (e.g., Hoorelbeke & Koster, 2017). However, our findings indicate the importance of using multiple assessments when evaluating effects of interventions in clinical samples.

Furthermore, to date, we have limited knowledge regarding the mechanisms underlying previously reported beneficial effects of CCT, which suggests a strong need for mechanistic research exploring how cognitive control is causally related to risk for depression. In this context, exploring how variability in training-task progress, cognitive gains, and emotional transfer relate may provide a first step forward. For instance, exploratory analyses suggest that in the CCT condition, training-task progress was related to cognitive transfer. However, training-task progress was unrelated to change in the weekly ratings of functioning over the 1-year follow-up period or change in the self-report measures from baseline to 1-year follow-up. In line with our main analyses suggesting beneficial effects of CCT on cognitive-task performance and risk for recurrence of depression, cognitive transfer emerged as a marginal significant predictor for recurrence of depression (for a more detailed discussion of these findings, see the Supplemental Material).

To the best of our knowledge, this preregistered study is the first to directly examine beneficial effects of CCT on recurrence of depression. A major strength of this study is the intensive multimethod procedure used to assess effects of CCT and the relatively long follow-up period, allowing us to gain a better understanding of how effects of CCT develop over time. This offers an important advantage given that previous studies have typically explored short-term effects of CCT using a limited number of assessments. In line with our hypotheses, our findings suggest that task-specific transfer effects remain present until at least 1 year after the experimental manipulation. We observed a significant effect of training condition, suggesting a reduced risk for recurrence of depression in the CCT condition compared with the ACT condition. This effect was of comparable strength of widely accepted preventive interventions for depression. However, the beneficial effect of CCT on recurrence of depression was at least partially inconsistent with our indicators of functioning (questionnaire data). Overall, this suggests that the long-term benefits of CCT may be limited and that future research is necessary to identify under which conditions CCT may yield emotional transfer effects and whether booster sessions are necessary to maintain treatment gains.

Some limitations need to be mentioned regarding the discrepancies between the different outcome measures. First, in line with previous clinical studies in which the aPASAT was used to train cognitive control (e.g., Hoorelbeke & Koster, 2017; Siegle et al., 2007), cognitive transfer was assessed using a cognitive task that showed high resemblance to the training procedure. Although this allows one to evaluate task-specific cognitive transfer, no conclusions can be drawn regarding more distal cognitive transfer effects. That is, effects may be task specific or affected by confounds (e.g., because of strategy learning; Shipstead et al., 2012). It would be interesting for researchers of future studies to provide an in-depth exploration of cognitive transfer effects following aPASAT training by relying on a multitude of behavioral measures.

Related to this, a second limitation of this study is that task-specific cognitive transfer effects were assessed only at 1-year follow-up. Likewise, the standardized self-report questionnaires and the structured clinical interview were reassessed only at 1-year follow-up, although our brief questionnaire/mobile assessments allowed to continuously model effects of CCT from baseline until 1-year follow-up. Absence of a cognitive assessment immediately following training does not allow one to evaluate the course of the task-specific cognitive transfer effects over the 1-year follow-up period. In this context, previous studies suggest a slight decrease in task-specific cognitive transfer over time following CCT in RMD patients (e.g., comparing posttraining performance with performance on the nonadaptive PASAT task at 3-month follow-up; Hoorelbeke & Koster, 2017). Taking this into account, the magnitude of the observed task-specific cognitive transfer effect at 1-year follow-up in the current study is noteworthy.

Third, we kept the number of items in the weekly questionnaire at a minimum to foster compliance to the assessment procedure. Choice of items was based on previous studies in which positive effects of CCT were obtained (Hoorelbeke et al., 2016). Despite the fact that one-item solutions for depression have previously been validated against standardized questionnaires (Richmond et al., 2015), it is likely that our one-item measure for depression did not allow us to represent the full complexity of depressive symptoms. Note that the items used in our brief weekly questionnaire show adequate reliability and validity (see the Supplemental Material). However, using multiple items to assess the constructs of interest may yield a more detailed understanding of training effects, for instance, using the recently validated measures for repetitive negative thinking (e.g., Rosenkranz et al., 2019).

Fourth, the current study compared effects of CCT with an ACT condition in which an adaptive speed-of-response task was used. Although this allows controlling for motivational effects of undergoing training, at the same time, this forms a very strict control condition. That is, it is likely that broader executive functions (e.g., attentional processes) are also trained to some extent in the ACT condition, reducing the likelihood to detect differential training effects of CCT. In addition, the training procedure used in the current study relies on neutral stimuli, whereas it has been argued that CCT procedures using emotional stimuli may be more effective (e.g., Iacoviello et al., 2014; Schweizer et al., 2011; for a review, see Koster et al., 2017).

Fifth, recurrence of depression was assessed retrospectively at 1-year follow-up using a structured clinical interview. Related to this, we did not assess time to recurrence during this interview. We would suggest that researchers of future studies monitor recurrence continuously in daily life, which would allow taking into account time until recurrence.

Sixth, in line with the preregistration, the main analyses presented in the current article mostly rely on a nomothetic approach. However, given the strong heterogeneity observed in RMD patients, it would be informative for future studies to yield a ideographic approach (e.g., Bosley et al., 2018; Fisher et al., 2018). Finally, although we used mixed methods to better understand the mechanisms of training effects, the current study failed to clarify why training yielded beneficial effects on some of the variables of interest but not on others. This may be considered a weakness, yet we feel that such data are crucial to better understand causal mechanisms of change in which CCT did not influence some of the expected processes (e.g., rumination). Similar to the cognitive bias modification field, the CCT field may require additional tolerance for mixed findings to advance (Koster & Bernstein, 2015).

Conclusion

Given the promising short-term effects of CCT in the literature, we set out to examine long-term effects of CCT on indicators of cognitive and emotional transfer. For this purpose, participants were monitored using a brief weekly questionnaire over a period of 1 year following the training, which concluded with a follow-up assessment in the lab. At 1-year follow-up, task-specific cognitive transfer effects were observed; the CCT condition showed a considerable increase in cognitive-task performance compared with the ACT condition. Moreover, CCT was related to reduced recurrence of depressive episodes during the period of 1-year follow-up. This effect was of similar magnitude of previous findings reported in the literature stemming from widely accepted preventive interventions for depression. These findings suggest that CCT may indeed show clinical potential as a preventive intervention following remission from depression. However, no differential effects of training condition emerged for the self-report questionnaires at 1-year follow-up. Instead, we observed a general decrease in cognitive complaints, improvements in emotion regulation, and an increase in self-reported resilience for both the CCT condition and the ACT condition. Moreover, CCT did not affect level of depressive symptoms at 1-year follow-up. In addition, the weekly ratings of functioning in daily life did not suggest long-term beneficial effects of CCT. Together, these findings indicate that the emotional transfer effects of CCT should be interpreted with caution.

Supplemental Material

sj-pdf-1-cpx-10.1177_2167702620979775 – Supplemental material for Preventing Recurrence of Depression: Long-Term Effects of a Randomized Controlled Trial on Cognitive Control Training for Remitted Depressed Patients

Supplemental material, sj-pdf-1-cpx-10.1177_2167702620979775 for Preventing Recurrence of Depression: Long-Term Effects of a Randomized Controlled Trial on Cognitive Control Training for Remitted Depressed Patients by Kristof Hoorelbeke, Nathan Van den Bergh, Rudi De Raedt, Marieke Wichers and Ernst H. W. Koster in Clinical Psychological Science

Footnotes

Acknowledgements

We thank Bo Swinnen, Lazlo Van Praet, Maarten Annot, and Yana Hertegonne for practical assistance. We thank Evelien Snippe and Laura Bringmann for their helpful advice regarding analysis of the time-series data.

Transparency

Action Editor: Michael F. Pogue-Geile

Editor: Kenneth J. Sher

Author Contributions

K. Hoorelbeke and N. Van den Bergh contributed equally to the article and share first authorship. K. Hoorelbeke developed the study concept. All of the authors contributed to the study design. Data collection was performed by K. Hoorelbeke and N. Van den Bergh. K. Hoorelbeke analyzed the data. K. Hoorelbeke and N. Van den Bergh drafted the manuscript. All of the authors provided critical revisions. All of the authors approved the final manuscript for submission.

ORCID iDs

Kristof Hoorelbeke

Marieke Wichers

Supplemental Material

Additional supporting information can be found at

Notes

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.).

Bauer

Tharmanathan

Volz

H.-P.

Moeller

H-J.

Freemantle

(2009). The effect of venlafaxine compared with other antidepressants and placebo in the treatment of major depression. European Archives of Psychiatry and Clinical Neuroscience, 259(3), 172–185.

Beck

A. T.

Steer

R. A.

Brown

G. K.

(1996). Manual for the Beck Depression Inventory-II. Psychological Corporation.

Bockting

C. L.

Hollon

S. D.

Jarrett

R. B.

Kuyken

Dobson

(2015). A lifetime approach to major depressive disorder: The contributions of psychological interventions in preventing relapse and recurrence. Clinical Psychology Review, 41, 16–26.

Bockting

C. L.

Koeter

Schene

(2010). Prediction of recurrence in recurrent depression: A 5.5-year prospective study. The Journal of Clinical Psychiatry, 71(8), 984–991.

Bockting

C. L.

Spinhoven

Koeter

M. W.

Wouters

L. F.

Schene

A. H.

(2006). Prediction of recurrence in recurrent depression and the influence of consecutive episodes on vulnerability for depression: A 2-year prospective study. The Journal of Clinical Psychiatry, 67(5), 747–755.

Bosley

H. G.

Fisher

A. J.

Taylor

C. B.

(2018). Differential responses of positive affect, negative affect, and worry in CBT for generalized anxiety disorder: A person-specific analysis of symptom course during therapy. Psychotherapy Research, 28(4), 630–642.

Braver

T. S.

Cohen

J. D.

Barch

D. M.

(2002). The role of the prefrontal cortex in normal and disordered cognitive control: A cognitive neuroscience perspective. Oxford University Press.

Brugha

T. S.

Cragg

(1990). The list of threatening experiences: The reliability and validity of a brief life events questionnaire. Acta Psychiatrica Scandinavica, 82(1), 77–81.

10.

Brunoni

Boggio

De Raedt

Benseñor

Lotufo

Namur

Vanderhasselt

M.-A.

(2014). Cognitive control therapy and transcranial direct current stimulation for depression: A randomized, double-blinded, controlled trial. Journal of Affective Disorders, 162, 43–49.

11.

Buhle

J. T.

Silvers

J. A.

Wager

T. D.

Lopez

Onyemekwu

Kober

Ochsner

K. N.

(2014). Cognitive reappraisal of emotion: A meta-analysis of human neuroimaging studies. Cerebral Cortex, 24, 2981–2990. https://doi.org/10.1093/cercor/bht154

12.

Calkins

A. W.

McMorran

K. E.

Siegle

G. J.

Otto

M. W.

(2015). The effects of computerized cognitive control training on community adults with depressed mood. Behavioural and Cognitive Psychotherapy, 43(5), 578–589.

13.

Cohen

(2017). Cognitive control: Core constructs and current considerations (The Wiley Handbook of Cognitive Control). John Wiley.

14.

Cohen

Margulies

D. S.

Ashkenazi

Schaefer

Taubert

Henik

Okon-Singer

(2016). Using executive control training to suppress amygdala reactivity to aversive information. NeuroImage, 125, 1022–1031. https://doi.org/10.1016/j.neuroimage.2015.10.069

15.

Cox

G. R.

Fisher

C. A.

De Silva

Phelan

Akinwale

O. P.

Simmons

M. B.

Hetrick

S. E.

(2012). Interventions for preventing relapse and recurrence of a depressive disorder in children and adolescents. Cochrane Database of Systematic Reviews, 11, Article CD007504. https://doi.org/10.1002/14651858.CD007504.pub2

16.

Cumming

(2012). Understanding the new statistics: Effect sizes, confidence intervals, and meta-analysis. Routledge

17.

De Raedt

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, 50–70. https://doi.org/10.3758/cabn.10.1.50

18.

De Raedt

Leyman

Baeken

Van Schuerbeek

Luypaert

Vanderhasselt

M.-A.

Dannlowksi

(2010). Neurocognitive effects of HF-rTMS over the dorsolateral prefrontal cortex on the attentional processing of emotional information in healthy women: An event-related fMRI study. Biological Psychology, 85, 487–495. https://doi.org/10.1016/j.biopsycho.2010.09.015

19.

Devilly

G. J.

Borkovec

T. D.

(2000). Psychometric properties of the credibility/expectancy questionnaire. Journal of Behavior Therapy and Experimental Psychiatry, 31(2), 73–86.

20.

Ekman

Granström

Omérov

Jacob

Landén

(2013). The societal cost of depression: Evidence from 10,000 Swedish patients in psychiatric care. Journal of Affective Disorders, 150(3), 790–797.

21.

Engelen

De Peuter

Victoir

Van Diest

Van den Bergh

(2006). Verdere validering van de Positive and Negative Affect Schedule (PANAS) en vergelijking van twee Nederlandstalige versies [Further validation of the Positive and Negative Affect Schedule (PANAS) and comparison of two Dutch versions]. Gedrag en gezondheid, 34(2), 61–70.

22.

Fisher

A. J.

Medaglia

J. D.

Jeronimus

B. F.

(2018). Lack of group-to-individual generalizability is a threat to human subjects research. Proceedings of the National Academy of Sciences, USA, 115(27), E6106–E6115.

23.

Garnefski

Kraaij

Spinhoven

(2001). Negative life events, cognitive emotion regulation and emotional problems. Personality and Individual Differences, 30(8), 1311–1327.

24.

Gronwall

(1977). Paced auditory serial-addition task: A measure of recovery from concussion. Perceptual and Motor Skills, 44(2), 367–373.

25.

Hofmann

Patel

P. V.

(2015). SurveySignal: A convenient solution for experience sampling research using participants’ own smartphones. Social Science Computer Review, 33(2), 235–253.

26.

Hoorelbeke

Faelens

Behiels

Koster

E. H.

(2015). Internet-delivered cognitive control training as a preventive intervention for remitted depressed patients: Protocol for a randomized controlled trial. BMC Psychiatry, 15(1), Article 125. https://doi.org/10.1037/ccp0000128

27.

Hoorelbeke

Koster

E. H.

(2017). Internet-delivered cognitive control training as a preventive intervention for remitted depressed patients: Evidence from a double-blind randomized controlled trial study. Journal of Consulting and Clinical Psychology, 85(2), 135–146.

28.

Hoorelbeke

Koster

E. H.

Demeyer

Loeys

Vanderhasselt

M.-A.

(2016). Effects of cognitive control training on the dynamics of (mal)adaptive emotion regulation in daily life. Emotion, 16(7), 945–956.

29.

Hoorelbeke

Koster

E. H.

Vanderhasselt

M.-A.

Callewaert

Demeyer

(2015). The influence of cognitive control training on stress reactivity and rumination in response to a lab stressor and naturalistic stress. Behaviour Research and Therapy, 69, 1–10.

30.

Hoorelbeke

Van den Bergh

De Raedt

Wichers

Albers

C.J.

Koster

E. H. W.

(2021). Regaining control of your emotions? Investigating the mechanisms underlying effects of cognitive control training for remitted depressed patients [Manuscript submitted for publication]. Department of Experimental Clinical and Health Psychology, Ghent University.

31.

Hoorelbeke

Van den Bergh

Wichers

Koster

E. H. W.

(2019). Between vulnerability and resilience: A network analysis of fluctuations in cognitive risk and protective factors following remission from depression. Behaviour Research and Therapy, 116, 1–9. https://doi.org/10.1016/j.brat.2019.01.007

32.

Iacoviello

B. M.

Alvarez

Huryk

Collins

K. A.

Murrough

J.W.

Iosifescu

D.V.

Charney

D.S.

(2014), Cognitive-emotional training as an intervention for major depressive disorder. Depression & Anxiety, 31(8), 699–706. https://doi.org/10.1002/da.22266

33.

Joormann

D’Avanzato

(2010). Emotion regulation in depression: Examining the role of cognitive processes. Cognition & Emotion, 24, 913–939. https://doi.org/10.1080/02699931003784939

34.

Joormann

Vanderlind

M. W.

(2014). Emotion regulation in depression: The role of biased cognition and reduced cognitive control. Clinical Psychological Science, 2, 402–421. https://doi.org/10.1177/2167702614536163

35.

Kessler

R. C.

(2012). The costs of depression. Psychiatric Clinics, 35(1), 1–14.

36.

Koster

E. H.

Bernstein

(2015). Introduction to the special issue on cognitive bias modification: Taking a step back to move forward? Journal of Behavior Therapy and Experimental Psychiatry, 49, 1–4.

37.

Koster

E. H.

De Lissnyder

Derakshan

De Raedt

(2011). Understanding depressive rumination from a cognitive science perspective: The impaired disengagement hypothesis. Clinical Psychology Review, 31, 138–145. https://doi.org/10.1016/j.cpr.2010.08.005

38.

Koster

E. H.

Hoorelbeke

Onraedt

Owens

Derakshan

(2017). Cognitive control interventions for depression: A systematic review of findings from training studies. Clinical Psychology Review, 53, 79–92.

39.

Lakens

(2013). Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs. Frontiers in Psychology, 4, Article 863. https://doi.org/10.3389/fpsyg.2013.00863

40.

Levens

S. M.

Gotlib

I. H.

(2015). Updating emotional content in recovered depressed individuals: Evaluating deficits in emotion processing following a depressive episode. Journal of Behavior Therapy and Experimental Psychiatry, 48, 156–163.

41.

S. H.

Teasdale

J. D.

(2004). Mindfulness-based cognitive therapy for depression: Replication and exploration of differential relapse prevention effects. Journal of Consulting and Clinical Psychology, 72(1), 31–40.

42.

Martin

D. M.

Teng

J. Z.

T. Y.

Alonzo

Goh

Iacoviello

B. M.

Loo

C. K.

(2018). Clinical pilot study of transcranial direct current stimulation combined with cognitive emotional training for medication resistant depression. Journal of Affective Disorders, 232, 89–95. https://doi.org/10.1016/j.jad.2018.02.021

43.

Miller

E. K.

Cohen

J. D.

(2001). An integrative theory of prefrontal cortex function. Annual Review of Neuroscience, 24, 167–202.

44.

Moberly

N. J.

Watkins

E. R.

(2008). Ruminative self-focus and negative affect: An experience sampling study. Journal of Abnormal Psychology, 117(2), 314–323.

45.

Mor

Daches

(2015). Ruminative thinking: Lessons learned from cognitive training. Clinical Psychological Science, 3, 574–592. https://doi.org/10.1177/2167702615578130

46.

Moshier

S. J.

Molokotos

Stein

A. T.

Otto

M. W.

(2015). Assessing the effects of depressed mood and cognitive control training on memory confidence and accuracy following repeated checking. International Journal of Cognitive Therapy, 8, 1–16. https://doi.org/10.1521/ijct_2015_8_02

47.

Moshier

S. J.

Otto

M. W.

(2017). Behavioral activation treatment for major depression: A randomized trial of the efficacy of augmentation with cognitive control training. Journal of Affective Disorders, 210, 265–268. https://doi.org/10.1016/j.jad.2017.01.003

48.

Motter

J. N.

Pimontel

M. A.

Rindskopf

Devanand

D. P.

Doraiswamy

P. M.

Sneed

J. R.

(2016). Computerized cognitive training and functional recovery in major depressive disorder: A meta-analysis. Journal of Affective Disorders, 189, 184–191.

49.

Nolen-Hoeksema

Morrow

(1991). A prospective study of depression and posttraumatic stress symptoms after a natural disaster: The 1989 Loma Prieta Earthquake. Journal of Personality and Social Psychology, 61(1), 115–121.

50.

Orwin

R. G.

Vevea

J. L.

(2009). Evaluating coding decisions. In Cooper

Larry

H. V.

Jeffrey

V. C.

(Eds.), The handbook of research synthesis and meta-analysis (2nd ed., pp. 177–203). Russell Sage Foundation.

51.

Portzky

(2008). RS-nl: Resilience Scale-Nederlandse versie. Harcourt Test Publishers.

52.

Richmond

S. J.

Keding

Hover

Gabe

Cross

Torgerson

MacPherson

(2015). Feasibility, acceptability and validity of SMS text messaging for measuring change in depression during a randomised controlled trial. BMC Psychiatry, 15(1), Article 68. https://doi.org/10.1186/s12888-015-0456-3

53.

Rosenkranz

Takano

Watkins

E. R.

Ehring

(2019). Assessing repetitive negative thinking in daily life: Development of an ecological momentary assessment paradigm. PsyArXiv. https://doi.org/10.31234/osf.io/nj73c

54.

Rosmalen

Bos

De Jonge

(2012). Validation of the Long-term Difficulties Inventory (LDI) and the List of Threatening Experiences (LTE) as measures of stress in epidemiological population-based cohort studies. Psychological Medicine, 42(12), 2599–2608.

55.

Schmiedek

Lovden

Lindenberger

(2014). Younger adults show long-term effects of cognitive training on broad cognitive abilities over 2 years. Developmental Psychology, 50, 2304–2310. https://doi.org/10.1037/a0037388

56.

Scholte

Noens

(2011). BRIEF-A Executieve functies gedragsvragenlijst voor volwassenen [BRIEF-A Executive functions behavioral questionnaire for adults]. In Roth

R. M.

Isquith

P. K.

Gioia

G. A.

(Eds.), Vragenlijst executieve functies voor volwassenen. Handleiding [Executive functions questionnaire for adults. Instruction manual] (pp. 1–54). Hogrefe Uitgevers.

57.

Schweizer

Hampshire

Dalgleish

(2011). Extending brain-training to the affective domain: Increasing cognitive and affective executive control through emotional working memory training. PLOS ONE, 6, Article e24372. https://doi.org/10.1371/journal.pone.0024372

58.

Segrave

R. A.

Arnold

Hoy

Fitzgerald

P. B.

(2014). Concurrent cognitive control training augments the antidepressant efficacy of tDCS: A pilot study. Brain Stimulation, 7(2), 325–331. https://doi.org/10.1016/j.brs.2013.12.008

59.

Semkovska

Quinlivan

O’Grady

Johnson

Collins

O’Connor

Gload

(2019). Cognitive function following a major depressive episode: A systematic review and meta-analysis. The Lancet Psychiatry, 6, 851–861. https://doi.org/10.1016/S2215-0366(19)30291-3

60.

Sheehan

D. V.

Lecrubier

Sheehan

K. H.

Amorim

Janavs

Weiller

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. The Journal of Clinical Psychiatry, 59, 22–33.

61.

Shipstead

Redick

T. S.

Engle

R. W.

(2012). Is working memory training effective? Psychological Bulletin, 138(4), 628–654. https://doi.org/10.1037/a0027473

62.

Siegle

G. J.

Ghinassi

Thase

M. E.

(2007). Neurobehavioral therapies in the 21st century: Summary of an emerging field and an extended example of cognitive control training for depression. Cognitive Therapy and Research, 31(2), 235–262.

63.

Siegle

G. J.

Price

R. B.

Jones

N. P.

Ghinassi

Painter

Thase

M. E.

(2014). You gotta work at it: Pupillary indices of task focus are prognostic for response to a neurocognitive intervention for rumination in depression. Clinical Psychological Science, 2(4), 455–471.

64.

Teasdale

J. D.

Segal

Z. V.

Williams

J. M. G.

Ridgeway

V. A.

Soulsby

J. M.

Lau

M. A.

(2000). Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. Journal of Consulting and Clinical Psychology, 68(4), 615–623.

65.

Tombaugh

T. N.

(2006). A comprehensive review of the Paced Auditory Serial Addition Test (PASAT). Archives of Clinical Neuropsychology, 21, 53–76. https://doi.org/10.1016/j.acn.2005.07.006

66.

Treynor

Gonzalez

Nolen-Hoeksema

(2003). Rumination reconsidered: A psychometric analysis. Cognitive Therapy and Research, 27(3), 247–259.

67.

Van den Bergh

Hoorelbeke

De Raedt

Koster

E. H.

(2018). Remediation of depression-related cognitive impairment: Cognitive control training as treatment augmentation. Expert Review of Neurotherapeutics, 18(12), 907–913.

68.

Van der Does

(2002). BDI-II-NL. Handleiding. De Nederlandse versie van de Beck depression inventory [BDI-II-NL. Instruction manual. The Dutch version of the Beck depression inventory]. Harcourt Test Publishers.

69.

Vanderhasselt

M.-A.

De Raedt

Namur

Lotufo

P. A.

Bensenor

I. M.

Boggio

P. S.

Brunoni

A. R.

(2015). Transcranial electric stimulation and neurocognitive training in clinically depressed patients: A pilot study of the effects on rumination. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 57, 93–99.

70.

Van Vliet

De Beurs

(2007). The MINI-International Neuropsychiatric Interview. A brief structured diagnostic psychiatric interview for DSM-IV en ICD-10 psychiatric disorders. Tijdschrift voor psychiatrie, 49(6), 393–397.

71.

Vervaeke

Hoorelbeke

Baeken

Koster

E. H. W.

(2021). Online cognitive control training for remitted depressed individuals: A replication and extension study. [Manuscript submitted for publication.] Department of Experimental Clinical and Health Psychology, Ghent University.

72.

Vervaeke

Hoorelbeke

Baeken

Van Looy

Koster

E. H. W.

(2020). Transfer and motivation after cognitive control training for remitted depression in healthy sample. Journal of Cognitive Enhancement, 4, 49–61.

73.

Wagnild

Young

(1993). Development and psychometric evaluation of the Resilience Scale. Journal of Nursing Measurement, 1(2), 165–178.

74.

Wagnild

R. N.

(2009). A review of the Resilience Scale. Journal of Nursing Measurement, 17, 105–113.

75.

Wanmaker

Geraerts

Franken

I. H.

(2015). A working memory training to decrease rumination in depressed and anxious individuals: A double-blind randomized controlled trial. Journal of Affective Disorders, 175, 310–319. https://doi.org/10.1016/j.jad.2014.12.027

76.

Watkins

E. R.

Roberts

(2020). Reflecting on rumination: Consequences, causes, mechanisms and treatment of rumination. Behaviour Research and Therapy, 127, Article 103573. https://doi.org/10.1016/j.brat.2020.103573

77.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070.

78.

Weightman

M. J.

Knight

M. J.

Baune

B. T.

(2019). A systematic review of the impact of social cognitive deficits on psychosocial functioning in major depressive disorder and opportunities for therapeutic intervention. Psychiatry Research, 274, 195–212.

79.

Weissman

M. M.

Prusoff

B. A.

Klerman

G. L.

(1978). Personality and the prediction of long-term outcome of depression. The American Journal of Psychiatry, 135(7), 797–800.

80.

Wilkinson

A. J.

Yang

(2016). Long-term maintenance of inhibition training effects in older adults: 1- and 3-year follow-up. The Journals of Gerontology: Series B, 71, 622–629. https://doi.org/10.1093/geronb/gbu179

81.

Zimmerman

Martinez

J. H.

Attiullah

Friedman

Toba

Boerescu

D. A.

Ragheb

(2013a). A new type of scale for determining remission from depression: The remission from depression questionnaire. Journal of Psychiatric Research, 47(1), 78–82.

82.

Zimmerman

Martinez

J. H.

Attiullah

Friedman

Toba

Boerescu

D. A.

Ragheb

M. V.

(2013b). [Dutch translation of Remission from Depression Questionnaire] ( Peeters

Nicolson

Wichers

Hacker

, Trans.). [Unpublished instrument]. Department of Psychiatry and Psychology, Maastricht University. (Original work published 2013.)

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.31 MB