psychology

lable at ScienceDirect

Behaviour Research and Therapy 89 (2017) 1e13

Contents lists avai

Behaviour Research and Therapy

journal homepage: www.elsevier.com/locate/brat

The effects of adaptive working memory training and mindfulness meditation training on processing efficiency and worry in high worriers*

Jenna Course-Choi, Harry Saville, Nazanin Derakshan*

Department of Psychological Sciences, Birkbeck, University of London, Malet Street, London, WC1E 7HX, UK

a r t i c l e i n f o

Article history: Received 14 January 2016 Received in revised form 1 November 2016 Accepted 6 November 2016 Available online 10 November 2016

Keywords: Attentional control Worry Working memory training

* Self-reported data collected pre- and post-trainin is included in Table 1 of Supplementary Information differ at baseline on self-report measures, except for * Corresponding author. Centre for Risk and Re

Department of Psychological Sciences, Birkbeck Unive London, UK.

E-mail address: n.derakhshan@bbk.ac.uk (N. Derak

http://dx.doi.org/10.1016/j.brat.2016.11.002 0005-7967/Crown Copyright © 2016 Published by Els

a b s t r a c t

Worry is the principle characteristic of generalised anxiety disorder, and has been linked to deficient attentional control, a main function of working memory (WM). Adaptive WM training and mindfulness meditation practice (MMP) have both shown potential to increase attentional control. The present study hence investigates the individual and combined effects of MMP and a dual adaptive n-back task on a non- clinical, randomised sample of high worriers. 60 participants were tested before and after seven days of training. Assessment included self-report questionnaires, as well as performance tasks measuring attentional control and working memory capacity. Combined training resulted in continued reduction in worry in the week after training, highlighting the potential of utilising n-back training as an adjunct to established clinical treatment. Engagement with WM training correlated with immediate improvements in attentional control and resilience, with worry decreasing over time. Implications of these findings and suggestions for future research are discussed.

Crown Copyright © 2016 Published by Elsevier Ltd. All rights reserved.

1. Introduction

Worry has been defined as a stream of negative, uncontrollable thoughts and images that represent attempts to manage or avoid future threats and negative outcomes (Borkovec, Robinson, Pruzinsky, & DePree, 1983). Moderate levels of worry can be constructive, encouraging action against threatening or unpleasant stimuli (McCaul, Mullens, Romanek, Erickson, & Gatheridge, 2007) and facilitating problem solving (Szabo & Lovibond, 2002). How- ever, excessive worry is an inefficient coping strategy (Borkovec, Hazlett, & Diaz, 1999) associated with depression and anxiety (Andrews & Borkovec, 1988; Starcevic, 1995), increased negative affect (Delgado et al., 2009) and impaired cognitive function (Hayes, Hirsch, & Mathews, 2008).

Worry has most often been studied in the context of generalised

g and at one-week follow up . Groups did not significantly scores of rumination. silience in Psychopathology, rsity of London, Malet Street,

shan).

evier Ltd. All rights reserved.

anxiety disorder (GAD), of which it is considered to be a primary attribute (APA, 1994). Cognitive theories of both anxiety (Berggren & Derakshan, 2013; Derakshan & Eysenck, 2009; Eysenck, Derakshan, Santos, & Calvo, 2007) and depression (Joormann & D'Avanzato, 2010; De Raedt & Koster, 2010) posit deficits in atten- tional control are a central feature of anxiety and depression maintenance and recurrence. Attentional control has been defined as the efficiency with which we regulate attention towards relevant and away from irrelevant material, and is a key function of working memory (Duncan & Humprheys, 1989; Unsworth, Redick, Spillers, & Brewer, 2012). Attentional control is closely linked to the concept of working memory capacity (WMC) which according to recent research is the efficacy by which we attend to and maintain goal relevant information and resist distraction from task irrelevant material (Shipstead, Tyler & Engle, 2015). Recent conceptualisa- tions go as far as to propose a causal role for attentional control in predicting anxiety and depressive-linked vulnerability (Sari, Koster, Pourtois & Derakshan, in press; Koster, Hoorelbeke, Onraedt, Owens & Derakshan, under review), with poor attentional control resulting in increased worry and rumination. It is thought the development of greater attentional control may therefore reduce anxiety and depression. Accordingly, and in line with studies sug- gesting plasticity of WMC and executive function (e.g. Klingberg,

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e132

2010), there has been a burgeoning interest in the potential of cognitive training as a means to improve WMC and potentially alleviate clinical symptoms (e.g. Bomyea & Amir, 2011; Cohen, Mor, & Henik, 2015; Wanmaker, Geraerts, & Franken, 2015). We first summarise attentional control theory (Eysenck et al., 2007), upon which the study is based, and then review extant research of WM training and mindfulness meditation practice.

1.1. Attentional control theory

The central tenet of attentional control theory (ACT) is that anxiety impacts performance via its negative effects on attentional control. The exercise of attentional control involves the activation of two subsystems of attention: one top-down, goal-driven and controlled, the other bottom-up, stimulus-driven, and reflexive (Corbetta & Shulman, 2002). When these systems function effec- tively, goal-relevant information is selectively maintained and held readily available in WM, while irrelevant information is filtered so it does not distract. ACT holds that anxiety upsets the balance be- tween these subsystems, reducing top-down processes through biasing increased bottom processes of attention (Miyake, Friedman, Emerson, Witzki, & Howerter, 2000). There is now substantial ev- idence showing an association between anxiety and an attentional bias for threat-related stimuli (see Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg, & van IJzendoorn, 2007, for a review) as well as evidence linking anxiety to inefficient recruitment of pre- frontal mechanisms heavily implicated in attentional control (Ansari & Derakshan, 2011a, 2011b; Basten, Stelzel, & Fiebach, 2011, 2012). Both behavioural and neural evidence hence provide impetus for the assertion that anxiety heightens attention to task- irrelevant stimuli, leaving fewer resources available for concurrent task demands (see Berggren & Derakshan, 2013, for a review).

ACT suggests a possible mechanism by which anxiety reduces attentional control is through the impact of internal as well as external distractions e namely, negative self-dialogue or worry. Recent research has shown worry is associated with reduced cognitive control and fewer attentional control resources (Stefanopoulou, Hirsch, Hayes, Adlam, & Coker, 2014), and ineffi- cient filtering of irrelevant information from WM (Stout, Shackman, Johnson, & Larson, 2014). Worry-linked vulnerability has been found to modulate the effects of cognitive control on cognitive load, necessitating greater use of cognitive resources to accomplish tasks involving heavy WM use (Owens, Derakshan, & Richards, 2015), with a recent study finding direct evidence for active worrying to reduce WMC (Sari et al., in press). Thus, reduced processing effi- ciency in worry is associated with a compensatory mechanism that necessitates the greater recruitment of prefrontal resources in achieving task outcomes, reducing attentional control. Elsewhere it has been documented that reduced attentional control may also maintain worry, directing resources towards worry thoughts in an attempt to manage a perceived threat (Hirsch & Mathews, 2012). Daches and Mor (2014) recently confirmed the effect of attentional control on excessive negative thought, demonstrating that a cognitive training protocol which promoted inhibition of irrelevant material resulted in a reduction of rumination. It seems high- worriers may become trapped in a cycle of cognitive impairment and negative bias not dissimilar to that identified in depressive rumination (Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008). It follows that increasing attentional control should improve cogni- tive efficiency and reduce worry.

1.2. Working memory training

One potential method for increasing attentional control is WM training, a relatively new mode of low intensity cognitive

treatment. The underlying mechanisms of WM training and transfer are still unclear (Buschkuehl, Jaeggi, & Jonides, 2012), but Engle and colleagues posit attentional control processes, including inhibition, modulate individual differences in WMC (Engle, 2002; Kane, Bleckley, Conway, & Engle, 2001). Inhibitory-related func- tion has been shown to correlate highly with WMC in both healthy and dysphoric populations (Owens, Koster, & Derakshan, 2012; Vogel, McCollough, & Machizawa, 2005). Owens, Koster, and Derakshan (2013) therefore suggest WMC improvements following WM training are indicative of an underlying improve- ment to inhibitory processes, making such training a promising method for improving cognitive deficits associated with depression and anxiety.

One of the most commonly used WM training paradigms is the adaptive dual n-back training paradigm first employed by Jaeggi, Buschkuehl, Jonides, and Perrig (2008). It requires participants to process simultaneously-presented auditory and visual information and to determine whether either the current auditory or visual stimuli match those presented a specific number of trials (n) back in the sequence. After each sequence, the level of n increases, de- creases or stays the same, depending on participant performance, so that as performance improves, the task becomes increasingly difficult. There is evidence linking n-back training to the improve- ment of a variety of executive processes, including focus of atten- tion (Lilienthal, Tamez, Shelton, Myerson, & Hale, 2013), and filtering of irrelevant information in dysphoric individuals, with transfer to both behavioural and neural measures of WMC (Owens et al., 2013), but see Onraedt and Koster (2014) for failures of transfer-related gains of training on unrelated tasks, which contests to more research needed to establish the reliable transference of training-related gains to unrelated tasks. An affective version of the dual n-back task using emotionally valenced stimuli has been found to enhance WM and affective cognitive control (Schweizer, Grahn, Hampshire, Mobbs, & Dalgleish, 2013). Other adaptive WM training has also been found to reduce depressive symptomatology in depressed samples (e.g. Brunoni et al., 2014), with long-term ef- fects: Siegle et al. (2014) found a combination of treatment as usual and cognitive control training in a clinical sample resulted in reduced need for outpatient services one year later. These findings indicate targeting improvements in cognitive processes can lead to a reduction in depressive symptoms. Early research investigating the effects of such training in the context of anxiety is also prom- ising. Sari et al. (in press) tested high trait anxious individuals before and after a three-week adaptive n-back training interven- tion, and found attentional control improved, with transfer to neural and behavioural measures. As yet, no current research has looked into sustained effects of inhibitory control post-treatment, a factor the current study investigates.

The clinical implications of such adaptive, systematic training are substantial - if WM training results in sustainable improvement in attentional control, it could complement existing treatments for anxiety and depression, including mindfulness-based and cognitive behavioural therapy. Online training programs such as the n-back task are low cost, easily accessible, and easily monitored. Surpris- ingly, however, no study of which the authors are aware has yet compared the effects of WM training against the effects of other interventions, or examined the potential of utilising WM training as an adjunct to established clinical treatment. Could mindfulness practice, another form of training thought to utilise and increase attentional control, stand to benefit from the effects of WM training?

1.3. Mindfulness training

Over the past 20 years, clinicians have increasingly incorporated

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e13 3

mindfulness-based interventions into treatment, and there is sub- stantial evidence of its beneficial impact on a wide range of psy- chological disorders, including chronic pain (Kabat-Zinn, 1984), substance abuse (Marlatt, 1994), and anxiety and depressive symptoms (Coehlo, Canter, & Ernst, 2007; Kim et al., 2009; for meta-analyses see Baer, 2003; Grossman, Niemann, Schmidt, & Walach, 2004). Described as “paying attention in a particular way, on purpose, in the present moment, and nonjudgmentally” (Kabat- Zinn, 1994, p.4), its concept is rooted in Buddhist philosophy, which suggests the state of mindfulness is developed through meditation practice. Mindfulness meditation practice (MMP) focuses attention on the sensations of the breath and the body, fostering passive observation of internal and external phenomena (Wallace & Shapiro, 2006). Operational definitions of mindfulness have pro- duced conflicting conceptualisations (e.g. Baer, Smith, & Allen, 2004; Bishop et al., 2004; Lau et al., 2006), but the two facets emphasised by most are nonjudgment of, and complete attention to, the present moment. In this latter aspect, a fundamental component of MMP is attentional training. Bishop et al. posit clinical benefits associated with MMP, such as reductions in rumi- nation and avoidance (Kumar, Feldman, & Hayes, 2008), may be linked to improvements in attentional control and the inhibition of unnecessary elaborative processing. Several processes have been proposed to account for this. Accepting, rather than judging, thought processes could disengage typical cognitive biases and defences. This may increase cognitive flexibility (Roemer & Orsillo, 2003) and reduce reactivity to negative emotions (Baer, 2003). MMP may also promote emotional stability through its emphasis of non-judgemental observation of present-moment phenomena, without avoidance or over-involvement (Carmody, 2009). Delgado et al. assert these processes “are clearly opposite to those of chronic worry” (2010, p. 874), and could therefore act as a mechanism to counter it.

Studies measuring MMP's effect on self-reported anxiety and depressive symptoms support these theories (Evans et al., 2008; Segal, Williams, & Teasdale, 2002). MMP training has been shown to reduce self-reported worry in older adults (Lenze et al., 2014) and physiological measures of worry in non-clinical high worriers (Delgado et al., 2010). Self-reported depressive rumination has been reduced in healthy participants after a 10-day MMP retreat (Chambers, Chuen Yee Lo, & Allen, 2008), while just four or five 20- min MMP sessions have produced a significant drop in anxiety (Tang et al., 2007; Zeidan, Johnson, Diamond, David, & Goolkasian, 2010). Reductions in affective disorder symptomatology also appear to sustain after treatment has finished (Baer, 2003), with improvements remaining intact three months post-treatment (Kabat-Zinn et al., 1992; Miller, Fletcher, & Kabat-Zinn, 1995).

However, research investigating cognitive effects of MMP thought to mediate the above results suggests length of meditation practice is key. An association between long-term MMP and various aspects of attention has been documented, including sustained attention (Brefczynski-Lewis, Lutz, Schaefer, Levinson, & Davidson, 2007; Jha, Krompinger, & Baime, 2007), enhanced attention switching (Hodgins & Adair, 2010), selective attention and execu- tive attention (Chan & Woollacott, 2007). Yet studies of briefer in- terventions have been mixed, with many failing to observe group differences on these measures post-intervention (Anderson, Lau, Segal, & Bishop, 2007; Polak, 2010; Tang et al., 2007). Short-term MMP interventions measuring effects on WMC have also pro- duced conflicting findings, some reporting significant WMC im- provements (Mrazek, Franklin, Phillips, Baird, & Schooler, 2013), others yielding mixed results (Zeidan et al., 2010). The possibility that WM training might boost the efficacy of short-term MMP, or act as a catalyst by promoting attentional control processes, is an exciting prospect for proponents of MMP.

1.4. The current study

Although a combination of cognitive training and pharmaceu- tical treatments has proven fruitful (Siegle et al., 2014), the present study is the first to investigate the effects of WM training in conjunction with another low intensity intervention believed to affect the same cognitive mechanisms. The study aims to examine the individual and combined effects of a well-established form of WM training and MMP on attentional control/WMC and negative symptomatology in a population of high worriers. Previous MMP studies have often featured non-randomised designs (e.g. Jha et al., 2007; for a review, see; Chiesa, Calati, & Serretti. 2011), while studies of WM training have produced conflicting results (e.g. Borella, Carretti, Riboldi, & De Beni, 2010; Chein & Morrison, 2010), leading some to question its efficacy and transferability (see Shipstead, Redick, & Engle, 2012). The present study is randomised with an active control, aiming to produce robust findings. We also wanted to consider the importance of magnitude of progress dur- ing WM training as a potential mechanism underlying WM training transfer. Training improvement is not always taken into account (e.g. Brehmer, Westerberg, & Backman, 2012), but was highlighted in von Bastian and Oberauer’s (2014) recent review of training studies as an important consideration that future research should take into account. Evidence for this has begun to surface in the literature e Sari et al. (in press) found training-related gains were associated with reductions in self-reported levels of trait anxiety e and will be further examined here.

Here, we investigated the effects of seven days' training comprising one of the following regimes: 1) a non-adaptive dual 1- back task, which served as an active control, 2) WM training, in the form of an adaptive dual n-back task, 3) MMP, in the form of a guided sitting meditation, and 4) combined adaptive dual n-back task and guided sitting meditation. The neutral dual n-back task (as in Owens et al., 2013; Sari et al., in press) was used. The choice for seven days of continuous training was motivated by previous research showing reliable transfer effects with similar (e.g. Owens et al., 2013), or much shorter lengths (e.g. Siegle et al., 2014). Self- report measures pre- and post-training were used to investigate effects on emotional vulnerability and resilience. As very little research includes resilience-related measures to examine transfer of cognitive-related benefits it was decided to include a measure of resilience in the current study. These measures were also admin- istered one week after training, to examine longer-term effects of training after consolidation with the environment. Transfer of training-related gains on cognitive performance was examined using the antisaccade task, a process-pure measure of attentional control measuring inhibition (see Friedman & Miyake, 2004) and a visual change detection task (CDT) measuring WMC. The anti- saccade task (Hallet, 1978) is a widely used measure of attentional control in healthy individuals (Hutton & Ettinger, 2006) as well as clinical and subclinical anxiety and depression (see Ainsworth & Garner, 2013; Berggren & Derakshan, 2013, for reviews). Success- ful performance on this task requires top-down attentional control for the inhibition of a reflexive saccade towards a sudden peripheral stimulus, while at the same time executing a voluntary saccade in the opposite direction as quickly as possible. Correct antisaccade latencies index processing efficiency, as slower reaction times are believed to reflect the utilisation of greater processing resources to inhibit reflexive saccades towards the stimulus (Olk & Kingstone, 2003). The effect of anxiety on antisaccade latencies has been confirmed repeatedly (see Berggren & Derakshan, 2013).

The CDT is a standard paradigm for measuring visual WMC, which in this task is positively correlated with the ability to filter task-irrelevant information from WM (Vogel et al., 2005; Owens et al., 2012, 2013). Previous research has found that depressive

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e134

vulnerability (Owens et al., 2012, 2013) is linked with poor WMC and active worrying reduces WMC (Sari et al., in press) as assessed by the CDT task. The CDT was the same used in Owens et al. (2012, 2013), and Sari et al. (in press).

In line with previous research (Owens et al., 2013; Sari et al., in press), we predicted participants who undertook WM training would show improvement in WM performance over the training period with transferable gains on self-reported worry, as well as other measures of emotional vulnerability and resilience, and cognitive performance, relative to the control group. We also pre- dicted that these changes should be greater with combined WM training and MMP, relative to the group who undertook mindful- ness training alone.

2. Method

2.1. Participants

A sample of high worriers was recruited via advertisements on the campus of Birkbeck, University of London, as well as online. Participants were offered £20, or course credits, for their partici- pation for approximately 5 experimental hours. Participants had to be over the age of 17, and were preselected based on their scores on the Pennsylvania State Worry Questionnaire (PSWQ; Meyer, Miller, Metzger, & Borkovec, 1990). Participants with PSWQ scores of 45 or more were eligible for the study (National IAPT Programme Team, 2011). Of a total of 86 individuals invited to participate, 18 were excluded due to no response or unavailability, 4 because they did not complete enough training sessions, 2 because they requested to be withdrawn after the first testing session, 1 due to technical failures, and 1 due to a lower PSWQ score (<45) at time of pre- intervention.

A final sample of 60 participants (15 male) with normal or corrected-to-normal vision was analysed. Participants were randomly assigned to one of the four training conditions at baseline (after exclusion criteria were considered) resulting in 15 partici- pants in each group. A sample size of 15 in each group is compatible with previous research using similar group sizes yielding moderate to high effect size (e.g. Owens et al., 2013; Sari et al., in press). Groups did not significantly differ from each other on age (Control, M ¼ 27.33, SD ¼ 3.96; N-back, M ¼ 27.93, SD ¼ 7.29; MMP, M ¼ 30.67, SD ¼ 8.91; Combined, M ¼ 28.73, SD ¼ 9.48; Welch's F(3,29) < 1), and had a similar gender distribution, (Control, 3 males-12 females; N-back, 5 males-10 females; MMP, 4 males-11 females, Combined, 3 males-12 females; p ¼ 0.91, two-tailed Fisher's exact test). Groups did not significantly differ from each other on the pre-selection measure of PSWQ scores at pre-test, (Control, M ¼ 55.80, SD ¼ 11.30; N-back, M ¼ 55.87, SD ¼ 8.53; MMP, M ¼ 61.73, SD ¼ 7.15; Combined, M ¼ 56.93, SD ¼ 5.89; F(3,56) ¼ 1.66, ns). The study was approved by Birkbeck's ethical committee.

2.2. Materials and tasks

2.2.1. Self-report measures At pre-intervention, participants completed self-report mea-

sures of state anxiety (STAI-SA; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983), resilience (Connor-Davidson Resilience Scale [CD-RISC]; Connor & Davidson, 2003), rumination (Rumina- tive Response Scale [RRS]; Nolen-Hoeksema & Morrow, 1991) and current mood (Positive And Negative Affect Schedule [PANAS]; Watson, Clark, & Tellegen, 1988). Baseline PSWQ scores were taken from screening, unless the screening survey was completed over one calendar month prior to the first session. In such cases, the PSWQ was re-administered alongside the other questionnaires (10

participants).

2.2.2. Anti-saccade and pro-saccade tasks This was modelled on Derakshan, Ansari, Hansard, Shoker, &

Eysenck, 2009 (Exp1). Eye-movements were recorded using an SR Research Eyelink 1000 eye-tracker (SR Research, ON, Canada). Only one eye was tracked during the experiment. Calibration involved tracking nine points across the computer screen to ensure tracking accuracy was within 1� of visual angle. Images were presented on a 2100 Mitsubishi Diamond Pro 2070 CRT monitor (85 Hz) and a chinrest was used to ensure a constant viewing distance of 60 cm. The experiment was designed and presented using the SR Research Experiment Builder software. The stimulus used for the antisaccade and prosaccade tasks consisted of a white oval-shaped object (created in simple graphics design application) subtending 2.58� � 4.77� and measuring 35 � 63 mm in dimension, presented on a black background. This oval shape served as a ‘‘Target’’.

Each trial started with a fixation cross subtending 0.95� � 0.95� and measuring 12 � 12 mm presented in the center of the screen for 1000 ms. Participants were instructed to fixate on the cross until it disappeared. After a short gap (200 ms), the stimulus (the white oval-shaped object) was presented 11� from the center of the screen, for 600 ms, either to the left or right along the horizontal axis, with equal probability. Participants were required to direct their gaze as quickly as possible either “AT” the cue (prosaccade task) or “AWAY FROM” the cue (antisaccade task), before returning to the fixation cross, which reappeared at the start of the next trial. There were four experimental blocks (two antisaccade and two prosaccade) of 36 trials, resulting in a total of 144 trials. Block order was counterbalanced between participants.

2.2.3. Change detection task This task was adapted from Owens et al. (2012, 2013) and pro-

grammed in E-prime. Stimuli were shown on a 1700 monitor. Par- ticipants were positioned 60 cm from the monitor and instructed to focus on a central fixation cross throughout the task. Trials began with a white arrow shown above the fixation, pointing left or right and presented for 700 ms. Participants were instructed to attend to the side the arrow pointed to. Two arrays were then shown in quick succession, featuring a combination of red (target) and blue (dis- tractor) rectangles on both sides of the fixation. Participants were required to remember the orientation of red rectangles on the side the arrow had pointed towards in the first array, and to compare it to the second array, which was presented after a 900 ms retention gap. During the presentation of the second array, participants recorded whether the orientation of any of the target red rectangles was altered or whether they remained identical by pressing one of two keys.

There were three array conditions: on both sides of the fixation, each array featured either two red rectangles, four red rectangles, or two red rectangles and two blue rectangles. The first array was presented for 100 ms. Stimuli disappeared for 900 ms before the second array was presented for 2000 ms. In half the trials, no change in the orientation of any rectangles occurred. In the other half, the orientation of one red rectangle differed between the two arrays. Rectangles were oriented randomly and were either vertical, horizontal, 45� left or 45� right. Array condition, arrow direction and change or no-change trials were randomised. Participants' re- sponses were recorded.

2.2.4. Adaptive dual n-back training task The dual n-back task was similar to that of Owens et al. (2013).

Participants were presented with a 3 � 3 grid of nine squares. A fixation cross occupied the central cell. In each trial, one of the remaining eight cells turned green. Simultaneously, one of eight

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e13 5

letters (c, h, k, l, q, r, s, t) was spoken by a female automated voice. Participants were instructed to remember both the letter spoken and the location of the green square and to compare the current letter/location to that presented n number of trials back in the sequence. If there was a letter match, they were asked to respond by pressing the “L” key. If there was a location match, they were asked to respond by pressing the “A” key. Where both letter and location matched, participants were able to press either or both keys, and for no match, no response was necessary. Trials lasted 3000 ms each (500 ms stimuli presentation, 2500 ms inter-stimuli interval). Participants in the n-back training group completed 20 blocks, each comprising 20 þ n number of trials, while participants in the combined training group completed 15 blocks. The training adapted to the level of the participant, with the level of n increasing when a participant achieved over 95% accuracy averaged across audio and visual conditions, maintaining when response accuracy was between 95% and 75%, and decreasing when response accuracy was below 75%. Level of ‘n’ could increase to a maximum of 4-back. Each block contained matches for four letters, four locations, and two both, randomly distributed within the block. Blocks were separated by short breaks of 20 s.

2.2.5. Control 1-back task As in Owens et al. (2013), the control group completed a 1-back

version of the task, identical in length and design to that of the n- back group, but without the adaptive function: participants only ever had to compare the stimuli to what was presented in the previous trial, regardless of performance level.

2.2.6. Mindfulness audio The audio clip used was from The Free Mindfulness Project, an

online collaboration providing access to mindfulness meditation exercises. The clip is free to download and distribute non- commercially. The audio clip was a guided seated meditation ses- sion lasting 21:03 min from the UCSD Center For Mindfulness, a program of the UC San Diego Center for Integrative Medicine and Department of Psychiatry, which promotes mindfulness-based techniques and initiatives. As is typical with seated meditations, the central focus of the session is the breath. Awareness of thoughts, feelings, sounds and sensations also feature.

2.3. Procedure

Participants completed pre- and post-intervention tasks and self-report measures in the MERLiN labs in the Department of Psychological Sciences at Birkbeck. After signing consent forms, they completed the questionnaires. They were then seated in front of the eye tracker at a viewing distance of 60 cm, with their heads in a fixed position using a chin and forehead rest to complete the anti and prosaccade tasks. The lights were dimmed and the eye tracker calibration procedure was run. This required participants to fixate on a series of nine points on the screen. Following successful cali- bration, participants were instructed not to move their head. Spoken instructions were delivered by the researcher, with a basic summary appearing on screen. Speed and accuracy of response were emphasised, as was the need to keep as still as possible. There were 16 practice trials without on screen feedback, though the researcher used live eye-tracking data to check the participant was performing the task sufficiently, and reiterated the instructions if necessary. The main experimental task was then run.

After this, participants moved to another dimly lit experiment room to complete the CDT task, where instructions were emphas- ised. A series of 24 practice trials was run with the researcher in the room to ensure that performance was above 50% - if not, the practice was run again. A total of 160 experimental trials was

divided into 4 blocks of 40 trials, with a short break after each block. The task lasted approximately 8 min.

Following the CDT task, the online WM training and guided MMP was explained to the participants (as appropriate). For the WM training task participants practiced a few trials of the n-back task with the experimenter in the lab to get familiar with the task and clarify any questions. They were instructed to complete the task for a continuous number of 7 day at approximately the same time every day. Participants could see a summary of their daily performance and progress at the end of the training session. They were told that the experimenter would be tracking their perfor- mance and completion rates each day. All data concerning partic- ipant performance on the training task (e.g., login, missed trials, time spent on task) was recorded, with the experimenter moni- toring performance remotely. Participants who missed training on one day were permitted to continue and complete the training regime one day later. MMP was introduced to the participants in the lab following the first CDT task, as appropriate. Participants were told to access the online mindfulness audio tape via a specific link issued for the participant. Participants logged into a secure online platform which took them through the mindfulness exer- cise, after which they indicated using the link when they had completed the session and login and logout details as well as time spent on mindfulness were recorded. The combined group (MMP and dual n-back) did both tasks at the same time with the sequence counterbalanced across participants, to avoid order effects. Partic- ipants began the online training the day after the pre-intervention session. After training participants were then re-tested on the CDT and antisaccade tasks the day after their last training session. Seven days after the second test session, self-report questionnaires were administered a third time, using the online survey platform Survey Monkey.

2.4. Data preparation

2.4.1. Training improvement The prime measure of improvement was training slope. For this,

regression was used to produce a coefficient which represented the rate of each participant's improvement and amount of engagement across training days.

2.4.2. Antisaccade measures There were two main dependent variables: latency of correct

antisaccades, and error rate (percentage of incorrect antisaccades). Percentage of saccadic errors and mean latency of correct saccades were calculated on an individual basis. Latency was defined as the elapsed time between onset of the target and the first saccade in the correct direction. In keeping with Derakshan et al. (2009), saccades occurring less than 83 ms after target onset were considered anticipatory and were excluded from analysis (0.69% of antisaccade trials). Incorrect saccades were defined as the first saccade after target onset towards the target in the antisaccade condition or away from the target in the prosaccade condition. Trials where no eye movements were made or where the eye- tracker failed to record data (“no saccade” trials) were discarded (1.37% of antisaccade trials).

2.4.3. Working memory capacity Participants' WMC was estimated from their results on the CDT

using a well-established formula for this paradigm (Cowan, 2001; Owens et al., 2013; Vogel et al., 2005). The formula used is K ¼ S(H�F), where K is WMC, S is the array size (4 or 2), H is the hit rate, or proportion of accurate responses when a change has occurred, and F the false alarm rate, or proportion of erroneous responses when a change has occurred.

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e136

3. Results

3.1. Training improvement

Performance on the adaptive dual n-back task for the two groups who underwent n-back training over the seven-day training period can be seen in Fig. 1, which shows the average level of dif- ficulty achieved.

Higher levels of difficulty (level of n) achieved by the end of the week indicate mean performance on the training task improved from beginning (day 1) (M ¼ 1.54, SD ¼ 0.47) to end of training (day 7) (M ¼ 2.15, SD ¼ 0.96), t(14) ¼ 3.19, p < 0.008, in the n-back group. The slope of improvement was significantly different from zero, t(14) ¼ 3.14, p < 0.008. The combined n-back and MMP group also improved: mean performance level of n rose from 1.66 (SD ¼ 0.43) at day 1e2.45 (SD ¼ 0.67) at day 7, t(14) ¼ 4.98, p < 0.001. Slope of improvement was also significantly different from zero in this group, t(14) ¼ 5.29, p < 0.001.

The control group attained 94.62% accuracy in 1-back training over the testing period, and percentage accuracy scores remained level from the first (M ¼ 94.27, SD ¼ 3.87) to last (M ¼ 94.65, SD ¼ 4.94) day of training.

3.2. Change detection task (CDT)

Working memory capacity (K) increased from pre-intervention (M ¼ 1.27, SD ¼ 0.83) to post-intervention (M ¼ 1.53, SD ¼ 0.79). A mixed ANOVA with Group (control, n-back, MMP, and combined) and Time (pre intervention e post intervention) confirmed this through a main effect of Time, F(1,56) ¼ 8.95, p ¼ 0.004, h2p ¼ 0.14. The main effect of Group was not significant, F < 1, and neither was the interaction of Time X Group, F(3,56) ¼ 1.27, p ¼ 0.29, h2p ¼ 0.06.

3.3. Antisaccade task

The groups did not differ on anticipatory saccades at pre- or post-intervention (both Fs < 1, ns), and the same was found for no saccade trials (both Fs < 1).

3.3.1. Saccadic latencies Saccadic latencies were subjected to a 2 (Time: pre-intervention,

Fig. 1. Mean n-back level achieved by n-back and combined groups across the training period.

post-intervention) X 2 (Task: antisaccade, prosaccade) X 4 (Group: control, MMP, n-back and combined) mixed ANOVA. Latencies got faster at post- (M ¼ 230.1, SD ¼ 25.47) vs. pre-intervention (M ¼ 235.91, SD ¼ 30.5), as demonstrated by a main effect of Time, F(1, 56) ¼ 4.98, p ¼ 0.03, h2p ¼ 0.08. A Time X Task interaction was observed, F(1, 56) ¼ 4.62, p ¼ 0.03, h2p ¼ 0.08, which showed that while antisaccade latencies got faster from pre- (M ¼ 268.54, SD ¼ 39.82) to post-intervention (M ¼ 258.86, SD ¼ 32.57), pro- saccade latencies did not, (pre-intervention, M ¼ 203.30, SD ¼ 30.65; post-intervention M ¼ 201.30, SD ¼ 29.57). The main effect of Group, the interaction of Time X Group and the three-way interaction of Time X Group X task were not significant: all Fs < 1, ns.

However, consistent with our predictions, we found that slope of improvement (degree of engagement with training) correlated with improvements on antisaccade latencies (pre- to post- intervention), in those who undertook n-back training (combined and n-back groups): r ¼ �0.36, N ¼ 30, p ¼ 0.05 (see Fig. 2), such that greater levels of improvement were met with better perfor- mance in antisaccade latencies.

3.3.2. Saccadic error rates Saccadic error rates were subjected to a 2 (Time: pre-

intervention, post-intervention) X 2 (Task: antisaccade, pro- saccade) X 4 (Group: control, MMP, n-back and combined) mixed ANOVA. Error rates reduced post-intervention (M ¼ 6.14, SD ¼ 5.85) relative to pre-intervention (M ¼ 8.78, SD ¼ 8.43), as revealed by a main effect of Time, F(1,56) ¼ 8.56, p ¼ 0.005, h2p ¼ 0.13. A main effect of Task was also observed, F(1, 56) ¼ 40.39, p < 0.001, h2p ¼ 0.41, that demonstrated antisaccade errors (M ¼ 11.02, SD ¼ 9.56) were greater than prosaccade errors (M ¼ 3.91, SD ¼ 5.17). No other main effects or interactions reached signifi- cance, all Fs < 1, ns.

3.4. Self-reported symptomatology

3.4.1. Trait worry Worry scores significantly reduced from pre-intervention

(M ¼ 57.58, SD ¼ 8.60) to post-intervention (M ¼ 51.70, SD ¼ 10.84), with sustained reductions at one week follow-up

Fig. 2. Relationship between n-back training (slope of improvement) and change in antisaccade latency (negative values indicate improvement at post- vs pre- intervention).

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e13 7

(M ¼ 51.17, SD ¼ 11.46). A mixed ANOVA with Group and Time (3 levels) confirmed a main effect of Time, F(2,112) ¼ 25.95, p < 0.001, and a significant Group X Time interaction, F(5,112) ¼ 3.47, p < 0.007. Fig. 3 shows that, while reductions in worry were observed from pre- to post-intervention in the n-back and MMP groups (corrected ps < 0.01), the combined group showed the biggest decrease from pre-intervention to follow-up, t(14) ¼ 4.17, p ¼ 0.001. This group demonstrated significant reductions from post-intervention to follow-up as well, t(14) ¼ 5.35, p < 0.001, suggesting the combination of n-back training and MMP yielded longer-term positive effects.

3.5. Additional analysis

3.5.1. Training improvement and change in self-reported symptomatology

In line with previous findings on the effects of training improvement on anxiety-linked symptomatology (Sari et al., 2016), and recent recommendations to examine training-related gains as a function of training engagement/improvement on training tasks (von Bastian & Oberauer, 2014), we examined how worry scores changed as a function of training engagement in the n-back group. Change in worry from post-intervention to one week follow-up was marginally correlated with training engagement on the n-back task, as measured by the slope of improvement: r ¼ �0.48, p ¼ 0.07, with 22.9% shared variability, showing greater improvements met with lower scores at follow-up (see Fig. 4a). Given the small sample size, this result is suggestive of a meaningful relationship with a sub- stantial effect.

Correlational analysis of difference scores for all other self- report measures (state anxiety, positive affect, negative affect, rumination and resilience, see Supplementary Material for descriptive statistics) and training improvement in the dual n-back group demonstrated only one other significant relationship. This was between changes in resilience from pre- to post-intervention and training slope: r ¼ 0.53, p ¼ 0.04, indicating that the greater the level of engagement/improvement, the greater the resilience scores (see Fig. 4b). Additionally, improvements in resilience scores at post-intervention were significantly correlated with reductions in worry at follow-up, relative to post-intervention, r ¼ 0.64, N ¼ 15, p ¼ 0.01, suggesting a possible link between resilience and worry (see Fig. 4c). This pattern was also confirmed in the combined training group, who practiced the dual n-back and MMP: r ¼ 0.58,

Fig. 3. Mean self-reported worry scores at pre-intervention, post-intervention, and one week follow-up.

N ¼ 30, p < 0.002 (see Fig. 4c). No other correlations reached significance.

4. Discussion

The present study aimed to investigate the combined and in- dividual effects of WM training and MMP on attentional control/ WMC in a sample of high worriers, and to examine whether these effects transferred to measures of worry and other self-reported symptomatology. The role of attentional control and WMC in excessive anxiety and worry has been highlighted in recent theo- retical models (see Berggren & Derakshan, 2013). However, despite burgeoning interest in WM training and widespread use of mindfulness-based therapies in clinical practice, individual and combined effects have yet to be investigated in a single study. Considering the debate surrounding the efficacy of WM training (e.g. Shipstead et al., 2012), and the lack of clarity regarding the cognitive underpinnings of MMP (e.g. Chiesa, Calati, & Serretti, 2011), investigation of transfer of training-related gains onto other cognitive measures is highly apposite. Transfer to self- reported symptomatology measures would also have important clinical implications.

Our initial and principal analytic approach concerning transfer- related gains to antisaccade and CDT performance did not find that improved working memory performance was modulated by group. All participants demonstrated enhanced cognitive control, as measured by the antisaccade and CDT task. This was somewhat unexpected, given the outcomes of similar studies (Owens et al., 2013, Sari et al., 2016), and hence contrary to our hypotheses. However, as previously noted, results of cognitive training studies have varied, with some failing to observe transfer-related gains to unrelated tasks (Onraedt & Koster, 2014). Our study, in this respect, continues the ongoing debate regarding working memory training (see Melby-Lervag & Hulme, 2013, for a meta-analytic review), showing an absence of effect. Furthermore, as this lack of between- group difference in WMC is observed alongside a significant reduction in self-reported worry between groups, it also appears to be at odds with attentional control theory.

Our secondary, within-group, findings, however, add important nuance to these points and merit discussion. These findings suggest the relationship between WM training and WMC may be more complex than first posited in the literature. By considering outcome measures as a function of training improvement in our two adap- tive n-back groups, we observed three correlations of note. Firstly, it appears that there is a meaningful relationship between level of WM training-related improvement and improved attentional con- trol, as assessed by the antisaccade task in individuals who un- dertook adaptive dual n-back training on its own and in combination with MMP. Secondly, individuals who completed the combined MMP and n-back training showed the greatest re- ductions in self-reported worry pre-intervention to one week follow-up post-intervention. Thirdly, there was a trend towards an association between WM training-related improvement and longer-term reduction in levels of self-reported worry, a relation- ship that could be influenced by increased resilience as a function of training.

It is prudent to note that without the within-group analysis, our results could have been construed as disappointing, as they found no between-group effect on either of our cognitive measures. Because our original design primarily focused on between-group analysis, our correlational results are limited by lack of random- isation, and we acknowledge the potential for known and unknown confounds. However, this study is not the first to consider within- group analysis in this context (e.g. Sari et al., in press; von Bastian & Oberauer, 2014). What has emerged from our findings is that

Fig. 4. a) correlation between change in worry and slope of training improvement in the n-back group, b) correlation between change in resilience and slope of training improvement in the n-back group, c) correlation between change in resilience and change in worry in the n-back and combined groups.

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e138

between-group analysis may be unable to take into account considerable heterogeneity in individual engagement with our working memory training task. We find a tentative link between WM training improvement, cognitive improvement, and reduction in worry that is worthy of further investigation, and suggest these results proffer provisional support for the notion there are indi- vidual differences in attentional control plasticity, and that these differences extend to training transfer effects. Individual differ- ences in rates of improvement have been posited as a contributing factor in discrepancies in WM training experiments by Moreau (2014), who encourages caution when interpreting seemingly incompatible findings.

With the above caveats in mind, we consider the implications of our three findings below.

4.1. Training-related gains on attentional control

WM training-related gains on attentional control were investi- gated via an antisaccade task. Post-intervention, all groups improved on both antisaccade latency and error rate measures across time, initially indicating no specific training effects. How- ever, examination of the n-back and combined groups' training performance showed a relationship between training improvement and changes in antisaccade latencies. Higher engagement with training was associated with improved antisaccade performance.

This is indicative of a meaningful relationship, and although pre- liminary, it is an important finding. Mrazek et al. (2013) assert the best evidence for improved cognitive ability stems from studies featuring a training task dissimilar to the outcome measure. The n- back task is designed to improve fundamental processes related to WM, targeting a general pool of resources which is not modality specific. We have demonstrated transfer to a specific task which is considered a process pure measure of inhibition (Friedman & Miyake, 2004), and which cannot be accounted for by similarity in training and testing contexts. This finding contrasts with a recent study of high trait anxiety individuals which did not observe training-related gains to performance on the antisaccade task (Sari et al., in press). The authors posit the use of emotional targets in their version of the antisaccade task, as opposed to neutral stimuli, may have activated specific processes related to selective attention and the inhibition of threat-related stimuli. Here, our use of neutral stimuli avoided the complicating factor of threat-response to valenced material and enabled us to demonstrate a clear effect towards transfer-related gains after just seven days of training.

4.2. Effects of mindfulness meditation practice

The MMP group significantly reduced worry scores pre- intervention to follow-up, replicating previous findings (Delgado et al., 2010; Lenze et al., 2014) and suggesting MMP can be an

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e13 9

effective method of reducing worry. There was no transfer to either of the cognitive measures, which is consistent with previous studies of similar length, which found no significant improvements in top-down attentional processes post-intervention (Anderson et al., 2007; Tang et al., 2007). Our findings, together with these studies, suggest short-term MMP may improve aspects of well- being, but is not sufficient to impact cognitive functioning in the way long-term MMP seems to (e.g. Hodgins & Adair, 2010).

The results of the combined training group, however, which undertook both n-back and MMP, present an exciting prospect for clinical practitioners working with anxiety disorders. This group demonstrated significant improvement in working memory over the course of the week, and the impressive mean levels of n-back achieved indicated high levels of engagement with the WM training. Combined training was associated with a significant reduction in worry from post-intervention to follow up one week later, suggesting individuals participating in MMP may benefit from longer-lasting reduction in negative symptomatology if they engage in concurrent cognitive training with the n-back task. One explanation for this is that MMP utilises, and its effects depend upon, pre-frontal mechanisms. This is consistent with Bishop et al.'s (2004) conceptualisation of mindfulness, which links it to attentional control and the inhibition of unnecessary elaborative processing. It appears successful MMP requires attentional control, which n-back training is known to boost, and that increasing attentional control increases the positive effects of MMP. According to Teasdale, Segal, and Williams (1995), these effects include teaching individuals to identify destructive thought patterns sooner, and to process this information in a neutral way that fa- cilitates cognitive flexibility, enabling the individual to reduce worry and self-criticism. In this way, n-back training potentially facilitates MMP, increasing its effectiveness and positive impact on emotional vulnerability.

4.3. Training-related gains on self-report measures

Participants in all conditions experienced decreases in worry pre- to post-intervention. This finding is consistent with other studies that have noted anxiety reduces over time (Ramsawh, Raffa, Edelen, Rende, & Keller, 2009; Wanmaker et al., 2015). However, individuals in the n-back group showed a trend toward correlation between training slope improvements on the n-back task and better follow up worry scores relative to post-intervention, as well as a significant correlation between training slope and resilience difference scores. Participants in the n-back group who engaged more with training reported higher levels of resilience post- intervention than pre-intervention, suggesting n-back training has broader beneficial effects than on worry alone.

Together, these findings are a compelling argument for the importance of considering engagement levels when reporting ef- fects of cognitive training. They also corroborate the results of Sari et al. (in press), who found a significant effect of task engagement (indexed by level of training improvement) on change in trait anxiety. In their study, participants who engaged more with 15 days of training showed a significantly greater decrease on trait anxiety scores than those who engaged less. Siegle et al. (2014) emphasises the importance of engagement in training for clinical benefits, pointing out that targeting a specific process via a cognitive task is likely to be dependent on ability to engage with the task. Our re- sults support this notion: it appears engagement with n-back training is crucial if people are to see both cognitive transfer and improvement on measures of wellbeing. This is an important consideration from a clinical perspective: people who don't improve on the task may be discouraged, which may maintain or even exacerbate negative symptomatology. Confirmation that

comprehension of the task is sufficient, provision of ongoing sup- port during training, and assessment of motivation prior to training may help to ensure individuals experience the full benefits of n- back training.

In the n-back training group, improvement in worry scores over the second week of the study also significantly correlated with improvement in resilience scores over the first week. The timeline of observed correlations prompts us to speculate that resilience, a measure of stress-coping ability (Connor & Davidson, 2003), may be an important mechanism by which the relationship between training gains and reductions in worry can be explained. Therefore, training related gains are associated with greater changes in resil- ience and a greater reduction in worry over time. There is evidence to support this assertion - hope, a factor of resilience (Lloyd & Hastings, 2009), has been shown to function as a protective factor against psychological distress, and higher hope has been associated with lower worry (Ogston, Mackintosh, & Myers, 2011). Our results corroborate this, and present the possibility that training which increases one's ability to cope with stress may decrease levels of worry.

4.4. Directions for future research and limitations

In line with Bishop et al.'s (2004) theory of mindfulness, our results suggest MMP utilises pre-frontal mechanisms, including attentional control. We observed reductions in worry were greater and longer-lasting when participants engaged in WM training alongside MMP, indicating mindfulness may involve processing efficiency. WM training's potential to improve processing efficiency therefore makes it an excellent candidate for further investigation as an experimental tool that may enhance the clinical outcomes of not only MMP, but other clinical treatments which stand to benefit from greater attentional control. These include cognitive behav- ioural therapy (CBT), the success of which has been predicted by pre-frontal regions implicated in attentional control. Klumpp, Fitzgerald, Angstadt, Post, and Phan (2014) reported patients with an anxiety disorder who exhibited greater attentional control pre- treatment were more likely to benefit from CBT. Siegle, Ghinassi, and Thase (2007) have highlighted the advantages of adjunctive interventions which target prefrontal control, rather than specific symptoms, noting they could “improve the efficacy of conventional therapies … by allowing patients to overcome specific roadblocks to their success in these therapies” (p. 238). As decreased prefrontal control has been observed in a host of conditions, ranging from depression (Mayberg et al., 1999) to obsessive compulsive disorder (van den Heuvel et al., 2005) and addiction (Goldstein et al., 2004), the current study presents exciting avenues for future research to investigate WM training as an adjunct to an abundance of clinical treatments.

Our results indicate n-back training has the potential to impact attentional control and aspects of wellbeing when participants engage with the training, inviting further research as to what level of task improvement is sufficient to facilitate transfer to cognitive and emotional vulnerability measures. Researchers could also look to increase motivation to engage by introducing game elements and frequent, immediate feedback about performance (e.g. reward points), which have been shown to positively impact motivation (Prins, Dovis, Ponsioen, ten Brink, & van der Oord, 2011).

There are several ways in which the current study could be improved. It will be fruitful for future studies to conduct training interventions with follow-up periods beyond one week, to inves- tigate the stability of transfer effects. Additionally, studies would benefit from a larger sample size. With 15 participants per group in the current study, individual differences may have impacted results at group level. This may account for the lack of group differences

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e1310

between the control and experimental groups post-intervention e a weakness of our findings when considered from a traditional analytic approach. However, as research begins to show the importance of analysing results as a function of training improve- ment/engagement (e.g. Sari et al., in press), we predict training studies will shift increasingly towards within-group analysis. With this in mind, future research could attempt to index mindfulness, so that MMP training-related gains may be examined. Self-report measures of MMP are a topic of contention in the literature (see Grossman, 2011), and a recent systematic review found important limitations in the field remain (Park, Reilly-Spong, & Gross, 2013). However, efforts to refine measures are ongoing, and are worth considering in future studies (Brown, Ryan, Loverich, Biegel, & West, 2011). Finally, given the small sample size and the fact we did not exclude any data, we consider the observed trends towards correlations between training improvement and attentional con- trol/vulnerability outcomes to be worthy of discussion. Replication is necessary to further elucidate the relationship between training improvement, cognitive and wellbeing measures.

It is clear that longer training times are associated with larger training gains for both n-back training (Jaeggi et al., 2008) and MMP (e.g. Hodgins & Adair, 2010). Owens et al. (2013) used a similar amount of n-back training days to the current study, but utilised breaks over weekends to spread training over 2 weeks. This may account for our failure to replicate findings that n-back training improved performance on a measure of WMC (Owens et al., 2013). Longer training times coupled with breaks in training may produce the best results for clinical intervention. Training-related gains did not transfer to our measure of WMC in the same way as they transferred to the antisaccade task. This was unexpected, as per- formance on the CDT has been shown to correlate with the ability to filter irrelevant information (Vogel et al., 2005), and hence is thought to be indicative of executive control abilities. It is possible the CDT and antisaccade task tap different facets of attentional control (Banich, 2009; Friedman et al., 2008). Another possibility is the CDT did not feature large enough set sizes for differences to be discerned. Shin, Lee, Yoo, and Chong (2015) examined the effects of inhibition training on WMC measured via a change detection task and found the training group's WMC increase was significantly higher for the larger change detection set sizes of 4, 5, 6 and 7. They conclude “filtering training was particularly effective for larger memory loads (i.e., more difficult conditions).” (p. 11). Pailian and Halberda's (2015) study of the CDT as an estimate for visual WMC also found much larger variance in K scores where arrays had a set size of eight rectangles, as compared to a set size of four. They posit, “The variability in performance at set size 8 … may reflect indi- vidual differences in the ability to effectively organise large amounts of information at encoding” (p. 397). Their findings sug- gest a version of this task with a condition featuring a set size of 8, with greater numbers of distracting stimuli, may provide a more sensitive measure of executive control abilities. This is consistent with Kane and Engle’s (2003) observation that without distraction or interference, there tend not to be individual differences in task performance as a function of WMC.

4.5. Conclusions

In sum, the conclusions of this study are important in two ways. Firstly, our findings indicate practitioners of mindfulness medita- tion stand to benefit from simultaneous WM training. They support the theoretical assumption that MMP utilises the same pre-frontal mechanisms that the n-back task has been shown to tap, including attentional control, and suggest that the n-back task's positive impact on these mechanisms facilitates the effects of MMP, resulting in greater positive effects on wellbeing. In the same way,

WM training may serve as a catalyst for other clinical treatment, including cognitive behavioural therapy. Secondly, we have demonstrated engagement in the n-back task is crucial if in- dividuals are to see transfer to both cognitive performance and measures of symptomatology. Our evidence, together with recent research (Siegle et al., 2014; Sari et al., in press), provides compel- ling support for the notion that training studies must identify improvement/engagement levels and take these into consideration when drawing conclusions about the efficacy of a training inter- vention. From a clinical perspective, and with due caution given the limited sample size, we suggest participants who are able to engage with WM training stand to benefit from boosted effects of MMP and improved clinical outcomes, with exciting implications for other forms of clinical treatment.

Competing interests

The authors declare no conflict of interest.

Sources of funding

This research was supported by post-graduate funding from the Department of Psychological Sciences at Birkbeck University of London.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.brat.2016.11.002.

References

Ainsworth, B., & Garner, M. (2013). Attention control in mood and anxiety disor- ders: Evidence from the antisaccade task. Human Psychopharmacology: Clinical & Experimental, 28(3), 274e280. http://dx.doi.org/10.1002/hup.2320.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington DC: Author.

Anderson, N. D., Lau, M. A., Segal, Z. V., & Bishop, S. R. (2007). Mindfulness-based stress reduction and attentional control. Clinical Psychology & Psychotherapy, 14(6), 449e463. http://dx.doi.org/10.1002/cpp.544.

Andrews, V. H., & Borkovec, T. D. (1988). The differential effects of inductions of worry, somatic anxiety, and depression on emotional experience. Journal of Behavior Therapy and Experimental Psychiatry, 19(1), 21e26. http://dx.doi.org/ 10.1016/0005-7916(88)90006-7.

Ansari, T. L., & Derakshan, N. (2011a). The neural correlates of cognitive effort in anxiety: Effects on processing efficiency. Biological Psychology, 86(3), 337e348. http://dx.doi.org/10.1016/j.biopsycho.2010.12.013.

Ansari, T. L., & Derakshan, N. (2011b). The neural correlates of impaired inhibitory control in anxiety. Neuropsychologia, 49(5), 1146e1153. http://dx.doi.org/ 10.1016/j.neuropsychologia.2011.01.019.

Baer, R. A. (2003). Mindfulness training as a clinical intervention: A conceptual and empirical review. Clinical Psychology: Science and Practice, 10(2), 125e143. http://dx.doi.org/10.1093/clipsy/bpg015.

Baer, R. A., Smith, G. T., & Allen, K. B. (2004). Assessment of mindfulness by self- report: The Kentucky inventory of mindfulness skills. Assessment, 11(3), 191e206. http://dx.doi.org/10.1177/1073191104268029.

Banich, M. T. (2009). Executive function: The search for an integrated account. Current Directions in Psychological Science (Wiley-Blackwell), 18(2), 89e94. http://dx.doi.org/10.1111/j.1467-8721.2009.01615.x.

Bar-Haim, Y., Lamy, D., Pergamin, L., Bakermans-Kranenburg, M. J., & van IJzendoorn, M. H. (2007). Threat-related attentional bias in anxious and non- anxious individuals: A meta-analytic study. Psychological Bulletin, 133(1), 1e24. http://dx.doi.org/10.1037/0033-2909.133.1.1.

Basten, U., Stelzel, C., & Fiebach, C. J. (2011). Trait anxiety modulates the neural efficiency of inhibitory control. Journal of Cognitive Neuroscience, 23(10), 3132e3145.

Basten, U., Stelzel, C., & Fiebach, C. J. (2012). Trait anxiety and the neural efficiency of manipulation in working memory. Cognitive, Affective & Behavioral Neurosci- ence, 12(3), 571e588. http://dx.doi.org/10.3758/s13415-012-0100-3.

von Bastian, C. C., & Oberauer, K. (2014). Effects and mechanisms of working memory training: A review. Psychological Research, 78(6), 803e820. http:// dx.doi.org/10.1007/s00426-013-0524-6.

Berggren, N., & Derakshan, N. (2013). Attentional control deficits in trait anxiety: Why you see them and why you don't. Biological Psychology, 92(3), 440e446. http://dx.doi.org/10.1016/j.biopsycho.2012.03.007.

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e13 11

Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. D., Carmody, J., … Devins, G. (2004). Mindfulness: A proposed operational defini- tion. Clinical Psychology: Science and Practice, 11(3), 230e241. http://dx.doi.org/ 10.1093/clipsy.bph077.

Bomyea, J., & Amir, N. (2011). The effect of an executive functioning training pro- gram on working memory capacity and intrusive thoughts. Cognitive Therapy & Research, 35(6), 529e535. http://dx.doi.org/10.1007/s10608-011-9369-8.

Borella, E., Carretti, B., Riboldi, F., & De Beni, R. (2010). Working memory training in older adults: Evidence of transfer and maintenance effects. Psychology and Aging, 25(4), 767e778. http://dx.doi.org/10.1037/a0020683.

Borkovec, T. D., Hazlett-Stevens, H., & Diaz, M. L. (1999). The role of positive beliefs about worry in generalized anxiety disorder and its treatment. Clinical Psy- chology & Psychotherapy, 6(2), 126e138.

Borkovec, T. D., Robinson, E., Pruzinsky, T., & DePree, J. A. (1983). Preliminary exploration of worry: Some characteristics and processes. Behaviour Research and Therapy, 21(1), 9e16. http://dx.doi.org/10.1016/0005-7967(83)90121-3.

Brefczynski-Lewis, J. A., Lutz, A., Schaefer, H. S., Levinson, D. B., & Davidson, R. J. (2007). Neural correlates of attentional expertise in long-term meditation practitioners. PNAS Proceedings of the National Academy of Sciences of the United States of America, 104(27), 11483e11488. http://dx.doi.org/10.1073/ pnas.0606552104.

Brehmer, Y., Westerberg, H., & B€ackman, L. (2012). Working-memory training in younger and older adults: Training gains, transfer, and maintenance. Frontiers in Human Neuroscience, 6. http://dx.doi.org/10.3389/fnhum.2012.00063.

Brown, K. W., Ryan, R. M., Loverich, T. M., Biegel, G. M., & West, A. M. (2011). Out of the armchair and into the streets: Measuring mindfulness advances knowledge and improves interventions: Reply to Grossman (2011). Psychological Assess- ment, 23(4), 1041e1046. http://dx.doi.org/10.1037/a0025781.

Brunoni, A. R., Boggio, P. S., De Raedt, R., Bense~nor, I. M., Lotufo, P. A., Namur, V., …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, 43e49. http:// dx.doi.org/10.1016/j.jad.2014.03.026.

Buschkuehl, M., Jaeggi, S. M., & Jonides, J. (2012). Neuronal effects following working memory training. Developmental Cognitive Neuroscience, 2(Suppl 1), S167eS179. http://dx.doi.org/10.1016/j.dcn.2011.10.001.

Carmody, J. (2009). Evolving conceptions of mindfulness in clinical settings. Journal of Cognitive Psychotherapy, 23(3), 270e280. http://dx.doi.org/10.1891/0889- 8391.23.3.270.

Chambers, R., Chuen Yee Lo, B., & Allen, N. B. (2008). The impact of intensive mindfulness training on attentional control, cognitive style, and affect. Cognitive Therapy & Research, 32, 303e322. http://dx.doi.org/10.1007/s10608-007-9119- 0.

Chan, D., & Woollacott, M. (2007). Effects of level of meditation experience on attentional focus: Is the efficiency of executive or orientation networks improved? The Journal of Alternative and Complementary Medicine, 13(6), 651e657. http://dx.doi.org/10.1089/acm.2007.7022.

Chein, J. M., & Morrison, A. B. (2010). Expanding the mind's workspace: Training and transfer effects with a complex working memory span task. Psychonomic Bulletin & Review, 17(2), 193e199. http://dx.doi.org/10.3758/PBR.17.2.193.

Chiesa, A., Calati, R., & Serretti, A. (2011). Does mindfulness training improve cognitive abilities? A systematic review of neuropsychological findings. Clinical Psychology Review, 31(3), 449e464. http://dx.doi.org/10.1016/j.cpr.2010.11.003.

Coelho, H. F., Canter, P. H., & Ernst, E. (2007). Mindfulness-based cognitive therapy: Evaluating current evidence and informing future research. Journal of Clinical and Consulting Psychology, 75(6), 1000e1005.

Cohen, N., Mor, N., & Henik, A. (2015). Linking executive control and emotional response: A training procedure to reduce rumination. Clinical Psychological Science, 3(1), 15e25. http://dx.doi.org/10.1177/2167702614530114.

Connor, K. M., & Davidson, J. R. T. (2003). Development of a new resilience scale: The Connor-Davidson Resilience Scale (CD-RISC). Depression & Anxiety (1091e4269), 18(2), 76.

Corbetta, M., & Shulman, G. L. (2002). Control of goal-directed and stimulus-driven attention in the brain. Nature Reviews Neuroscience, 3(3), 215e229. http:// dx.doi.org/10.1038/nrn755.

Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral & Brain Sciences, 24(1), 87e185. http:// dx.doi.org/10.1017/S0140525X01003922.

Daches, S., & Mor, N. (2014). Training ruminators to inhibit negative information: A preliminary report. Cognitive Therapy and Research, 38, 160e171. http:// dx.doi.org/10.1007/s10608-013-9585-5.

De Raedt, R., & Koster, E. W. (2010). Understanding vulnerability for depression from a cognitive neuroscience perspective: A reappraisal of attentional factors and a new conceptual framework. Cognitive, Affective & Behavioral Neuroscience, 10(1), 50e70. http://dx.doi.org/10.3758/CABN.10.1.50.

Delgado, L. C., Guerra, P., Perakakis, P., Mata, J. L., P�erez, M. N., & Vila, J. (2009). Psychophysiological correlates of chronic worry: Cued versus non-cued fear reaction. International Journal of Psychophysiology, 74(3), 280e287. http:// dx.doi.org/10.1016/j.ijpsycho.2009.10.007.

Delgado, L. C., Guerra, P., Perakakis, P., Vera, M. N., Del Paso, G. R., & Vila, J. (2010). Treating chronic worry: Psychological and physiological effects of a training programme based on mindfulness. Behaviour Research & Therapy, 48(9), 873e882. http://dx.doi.org/10.1016/j.brat.2010.05.012.

Derakshan, N., Ansari, T. L., Hansard, M., Shoker, L., & Eysenck, M. W. (2009). Anxiety, inhibition, efficiency, and effectiveness: An investigation using the

antisaccade task. Experimental Psychology, 56(1), 48e55. http://dx.doi.org/ 10.1027/1618-3169.56.1.48.

Derakshan, N., & Eysenck, M. W. (2009). Anxiety, processing efficiency, and cogni- tive performance: New developments from attentional control theory. European Psychologist, 14(2), 168e176. http://dx.doi.org/10.1027/1016-9040.14.2.168.

Duncan, J., & Humprheys, G. W. (1989). Visual search and stimulus similarity. Psy- chological Review, 96(3), 433e458.

Engle, R. W. (2002). Working memory capacity as executive attention. Current Di- rections in Psychological Science (Wiley-Blackwell), 11(1), 19.

Evans, S., Ferrando, S., Findler, M., Stowell, C., Smart, C., & Haglin, D. (2008). Mindfulness-based cognitive therapy for generalized anxiety disorder. Journal of Anxiety Disorders, 22(4), 716e721. http://dx.doi.org/10.1016/ j.janxdis.2007.07.005.

Eysenck, M. W., Derakshan, N., Santos, R., & Calvo, M. G. (2007). Anxiety and cognitive performance: Attentional control theory. Emotion, 7(2), 336e353. http://dx.doi.org/10.1037/1528-3542.7.2.336.

Friedman, N. P., & Miyake, A. (2004). The relations among inhibition and interfer- ence cognitive functions: A latent variable analysis. Journal of Experimental Psychology: General, 133(1), 101e135. http://dx.doi.org/10.1037/0096- 3445.133.1.101.

Friedman, N. P., Miyake, A., Young, S. E., DeFries, J. C., Corley, R. P., & Hewitt, J. K. (2008). Individual differences in executive functions are almost entirely genetic in origin. Journal of Experimental Psychology: General, 137(2), 201e225. http:// dx.doi.org/10.1037/0096-3445.137.2.201.

Goldstein, R. Z., Leskovjan, A. C., Hoff, A. L., Hitzemann, R., Bashan, F., Khalsa, S. S., …Volkow, N. D. (2004). Severity of neuropsychological impairment in cocaine and alcohol addiction: Association with metabolism in the prefrontal cortex. Neuropsychologia, 42(11), 1447e1458. http://dx.doi.org/10.1016/ j.neuropsychologia.2004.04.002.

Grossman, P. (2011). Defining mindfulness by how poorly I think I pay attention during everyday awareness and other intractable problems for psychology's (re)invention of mindfulness: Comment on Brown et al. (2011). Psychological Assessment, 23, 1034e1040. http://dx.doi.org/10.1037/a0022713.

Grossman, P., Niemann, L., Schmidt, S., & Walach, H. (2004). Mindfulness-based stress reduction and health benefits: A meta-analysis. Journal of psychosomatic research, 57(1), 35e43.

Hallet, P. E. (1978). Primary and secondary saccades to goals defined by instructions. Vision Research, 18(10), 1279e1296. http://dx.doi.org/10.1016/0042-6989(78) 90218-3.

Hayes, S., Hirsch, C., & Mathews, A. (2008). Restriction of working memory capacity during worry. Journal of Abnormal Psychology, 117(3), 712e717. http://dx.doi.org/ 10.1037/a0012908.

van den Heuvel, O. A., Veltman, D. J., Groenewegen, H. J., Cath, D. C., van Balkom, A. M., van Hartskamp, J., … van Dyck, R. (2005). Frontal-striatal dysfunction during planning in obsessive-compulsive disorder. Archives of General Psychiatry, 62(3), 301e310. http://dx.doi.org/10.1001/archpsyc.62.3.301.

Hirsch, C. R., & Mathews, A. (2012). A cognitive model of pathological worry. Behaviour Research and Therapy, 50(10), 636e646. http://dx.doi.org/10.1016/ j.brat.2012.06.007.

Hodgins, H. S., & Adair, K. C. (2010). Attentional processes and meditation. Con- sciousness and Cognition: An International Journal, 19(4), 872e878. http:// dx.doi.org/10.1016/j.concog.2010.04.002.

Hutton, S. B., & Ettinger, U. (2006). The antisaccade task as a research tool in psy- chopathology: A critical review. Psychophysiology, 43(3), 302e313. http:// dx.doi.org/10.1111/j.1469-8986.2006.00403.x.

Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid in- telligence with training on working memory. Proceedings of the National Academy of Sciences of the United States of America, 105(19), 6829e6833. http:// dx.doi.org/10.1073/pnas.0801268105.

Jha, A. P., Krompinger, J., & Baime, M. J. (2007). Mindfulness training modifies subsystems of attention. Cognitive, Affective & Behavioral Neuroscience, 7(2), 109e119. http://dx.doi.org/10.3758/CABN.7.2.109.

Joormann, J., & D'Avanzato, C. (2010). Emotion regulation in depression: Examining the role of cognitive processes. Cognition & Emotion, 24(6), 913e939. http:// dx.doi.org/10.1080/02699931003784939.

Kabat-Zinn, J. (1984). An outpatient program in behavioural medicine for chronic pain patients based on the practice of mindfulness meditation: Theoretical considerations and preliminary results. Revision, 7(1), 71e72.

Kabat-Zinn, J. (1994). Wherever you go, there you are: Mindfulness meditation for everyday life. New York: Hyperion.

Kabat-Zinn, J., Massion, A. O., Kristeller, J., Peterson, L. G., Fletcher, K. E., Pbert, L., … Santorelli, S. F. (1992). Effectiveness of a meditation-based stress reduction program in the treatment of anxiety disorders. American Journal of Psychiatry, 149(7), 936e943.

Kane, M. J., Bleckley, M. K., Conway, A. A., & Engle, R. W. (2001). A controlled- attention view of working-memory capacity. Journal of Experimental Psychology: General, 130(2), 169e183. http://dx.doi.org/10.1037/0096-3445.130.2.169.

Kane, M. J., & Engle, R. W. (2003). Working-memory capacity and the control of attention: The contributions of goal neglect, response competition, and task set to Stroop interference. Journal of Experimental Psychology, 132, 47e70 Accessed at: http://englelab.gatech.edu/2003/working-memory-capicity-and-the- control-of-attention.pdf.

Kim, Y. W., Lee, S. H., Choi, T. K., Suh, S. Y., Kim, B., Kim, C. M., …Yook, K. H. (2009). Effectiveness of mindfulness-based cognitive therapy as an adjuvant to phar- macotherapy in patients with panic disorder or generalized anxiety disorder.

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e1312

Depression and Anxiety, 26(7), 601e606. Klingberg, T. (2010). Training and plasticity of working memory. Trends in Cognitive

Sciences, 14(7), 317e324. http://dx.doi.org/10.1016/j.tics.2010.05.002. Klumpp, H., Fitzgerald, D. A., Angstadt, M., Post, D., & Phan, K. L. (2014). Neural

response during attentional control and emotion processing predicts improvement after cognitive behavioral therapy in generalized social anxiety disorder. Psychological Medicine, 44(14), 3109e3121. http://dx.doi.org/10.1017/ S0033291714000567.

Koster, E.H.W., Hoorelebeke, K., Onraedt, T., Owens, M., & Derakshan, N. (under review). Cognitive control interventions for depression: A systematic review of findings from training studies

Kumar, S., Feldman, G., & Hayes, A. (2008). Changes in mindfulness and emotion regulation in an exposure-based cognitive therapy for depression. Cognitive Therapy & Research, 32(6), 734e744. http://dx.doi.org/10.1007/s10608-008- 9190-1.

Lau, M. A., Bishop, S. R., Segal, Z. V., Buis, T., Anderson, N. D., Carlson, L., … Devins, G. (2006). The Toronto mindfulness scale: Development and validation. Journal of Clinical Psychology, 62(12), 1445e1467. http://dx.doi.org/10.1002/jclp.20326.

Lenze, E. J., Hickman, S., Hershey, T., Wendleton, L., Ly, K., Dixon, D., …Wetherell, J. L. (2014). Mindfulness-based stress reduction for older adults with worry symp- toms and co-occurring cognitive dysfunction. International Journal of Geriatric Psychiatry, 29(10), 991e1000. http://dx.doi.org/10.1002/gps.4086.

Lilienthal, L., Tamez, E., Shelton, J. T., Myerson, J., & Hale, S. (2013). Dual n-back training increases the capacity of the focus of attention. Psychonomic Bulletin & Review, 20(1), 135e141. http://dx.doi.org/10.3758/s13423-012-0335-6.

Lloyd, T. J., & Hastings, R. (2009). Hope as a psychological resilience factor in mothers and fathers of children with intellectual disabilities. Journal of Intel- lectual Disability Research, 53(12), 957e968. http://dx.doi.org/10.1111/j.1365- 2788.2009.01206.x.

Marlatt, G. A. (1994). Addiction, mindfulness, and acceptance. Content, context, and the types of psychological acceptance. In S. C. Hayes, N. S. Jacobson, V. M. Follette, & M. J. Dougher (Eds.), Acceptance and change: Content and context in psychotherapy (pp. 175e197). Reno, NV: Context Press.

Mayberg, H. S., Liotti, M., Brannan, S. K., McGinnis, S., Mahurin, R. K., Jerabek, P. A., … Fox, P. T. (1999). Reciprocal limbic-cortical function and nega- tive mood: Converging PET findings in depression and normal sadness. The American Journal of Psychiatry, 156(5), 675e682.

McCaul, K. D., Mullens, A. B., Romanek, K. M., Erickson, S. C., & Gatheridge, B. J. (2007). The motivational effects of thinking and worrying about the effects of smoking cigarettes. Cognition & Emotion, 21(8), 1780e1798. http://dx.doi.org/ 10.1080/02699930701442840.

Melby-Lervåg, M., & Hulme, C. (2013). Is working memory training effective? A meta-analytic review. Developmental Psychology, 49, 270e291. http://dx.doi.org/ 10.1037/a0028228.

Meyer, T. J., Miller, M. L., Metzger, R. L., & Borkovec, T. D. (1990). Development and validation of the Penn state worry questionnaire. Behaviour Research and Therapy, 28, 487e495. http://dx.doi.org/10.1016/0005-7967(90)90135-6.

Miller, J. J., Fletcher, K., & Kabat-Zinn, J. (1995). Three-year follow-up and clinical implications of a mindfulness meditation-based stress reduction intervention in the treatment of anxiety disorders. General Hospital Psychiatry, 17(3), 192e200.

Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., & Howerter, A. (2000). The unity and diversity of executive functions and their contributions to complex 'frontal lobe' tasks: A latent variable analysis. Cognitive Psychology, 41(1), 49e100. http://dx.doi.org/10.1006/cogp.1999.0734.

Moreau, D. (2014). Making sense of discrepancies in working memory training experiments: A Monte Carlo simulation. Frontiers in Systems Neuroscience, 8, 161. http://dx.doi.org/10.3389/fnsys.2014.00161.

Mrazek, M. D., Franklin, M. S., Phillips, D. T., Baird, B., & Schooler, J. W. (2013). Mindfulness training improves working memory capacity and GRE perfor- mance while reducing mind wandering. Psychological Science, 24(5), 776e781. http://dx.doi.org/10.1177/0956797612459659.

National IAPT Programme Team. (2011). The IAPT data handbook: Guidance on recording and monitoring outcomes to support local evidence-based practice. Retrieved from http://www.iapt.nhs.uk/silo/files/the-iapt-data-handbook.pdf.

Nolen-Hoeksema, S., & Morrow, J. (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, 115e121.

Nolen-Hoeksema, S., Wisco, B. E., & Lyubomirsky, S. (2008). Rethinking rumination. Perspectives on Psychological Science (Wiley-Blackwell), 3(5), 400e424. http:// dx.doi.org/10.1111/j.1745-6924.2008.00088.x.

Ogston, P. L., Mackintosh, V. H., & Myers, B. J. (2011). Hope and worry in mothers of children with an autism spectrum disorder or Down syndrome. Research in Autism Spectrum Disorders, 5(4), 1378e1384. http://dx.doi.org/10.1016/ j.rasd.2011.01.020.

Olk, B., & Kingstone, A. (2003). Why are antisaccades slower than prosaccades? A novel finding using a new paradigm. Neuroreport: For Rapid Communication of Neuroscience Research, 14(1), 151e155. http://dx.doi.org/10.1097/00001756- 200301200-00028.

Onraedt, T., & Koster, E. H. W. (2014). Training working memory to reduce rumi- nation http://dx.doi.org/10.1371/journal.pone.0090632.

Owens, M., Derakshan, N., & Richards, A. (2015). Trait susceptibility to worry modulates the effects of cognitive load on cognitive control: An ERP study. Emotion. http://dx.doi.org/10.1037/emo0000052.

Owens, M., Koster, E. W., & Derakshan, N. (2012). Impaired filtering of irrelevant

information in dysphoria: An ERP study. Social Cognitive & Affective Neurosci- ence, 7(7), 752e763.

Owens, M., Koster, E. W., & Derakshan, N. (2013). Improving attention control in dysphoria through cognitive training: Transfer effects on working memory capacity and filtering efficiency. Psychophysiology, 50(3), 297e307. http:// dx.doi.org/10.1111/psyp.12010.

Pailian, H., & Halberda, J. (2015). The reliability and internal consistency of one-shot and flicker change detection for measuring individual differences in visual working memory capacity. Memory & Cognition, 43(3), 397e420. http:// dx.doi.org/10.3758/s13421-014-0492-0.

Park, T., Reilly-Spong, M., & Gross, C. R. (2013). Mindfulness: A systematic review of instruments to measure an emergent patient-reported outcome (PRO). Quality of Life Research, 22(10), 2639e2659. http://dx.doi.org/10.1007/s11136-013- 0395-8.

Polak, E. L. (2010). Impact of two sessions of mindfulness training on attention. Dissertation Abstracts International, 70, 5200.

Prins, P. J., Dovis, S., Ponsioen, A., ten Brink, E., & van der Oord, S. (2011). Does computerized working memory training with game elements enhance moti- vation and training efficacy in children with ADHD? Cyberpsychology, Behavior & Social Networking, 14(3), 115e122. http://dx.doi.org/10.1089/cyber.2009.0206.

Ramsawh, H. J., Raffa, S. D., Edelen, M. O., Rende, R., & Keller, M. B. (2009). Anxiety in middle adulthood: Effects of age and time on the 14-year course of panic dis- order, social phobia and generalized anxiety disorder. Psychological Medicine, 39(4), 615e624. http://dx.doi.org/10.1017/S0033291708003954.

Roemer, L., & Orsillo, S. M. (2003). Mindfulness: A promising intervention strategy in need of further study. Clinical Psychology: Science and Practice, 10(2), 172e178. http://dx.doi.org/10.1093/clipsy/bpg020.

Sari, B., Koster, E.H., Pourtois, G., & Derakshan, N., Training working memory to improve attentional control in anxiety: A proof-of-principle study using behavioural and electrophysiological measures. Biological Psychology. in press, http://www.sciencedirect.com/science/article/pii/S0301051115300570.

Schweizer, S., Grahn, J., Hampshire, A., Mobbs, D., & Dalgleish, T. (2013). Training the emotional brain: Improving affective control through emotional working memory training. The Journal of Neuroscience, 33(12), 5301e5311.

Segal, Z. V., Williams, J. M. G., & Teasdale, J. D. (2002). Mindfulness-based cognitive therapy for depression: A new approach to preventing relapse. New York: Guilford Publications.

Shin, E., Lee, H., Yoo, S., & Chong, S. C. (2015). Training improves the capacity of visual working memory when it is adaptive, individualised, and targeted. Plos One, 10(4), 1e14. http://dx.doi.org/10.1371/journal.pone.0121702.

Shipstead, Z., Harrison, T. L., & Engle, R. (2015). Working memory capacity and the scope of attention control. Attention, Perception & Psychophysics, 77(2). http:// dx.doi.org/10.3758/s13414-015-0899-0.

Shipstead, Z., Redick, T. S., & Engle, R. W. (2012). Is working memory training effective? Psychological Bulletin, 138(4), 628e654. http://dx.doi.org/10.1037/ a0027473.

Siegle, G. J., Ghinassi, F., & 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 & Research, 31(2), 235e262. http://dx.doi.org/10.1007/s10608-006-9118-6.

Siegle, G. J., Price, R. B., Jones, N. P., Ghinassi, F., Painter, T., & 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), 455e471. http://dx.doi.org/10.1177/2167702614536160.

Spielberger, C. C., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychol- ogists Press.

Starcevic, V. (1995). Pathological worry in major depression: A preliminary report. Behaviour Research & Therapy, 33(1), 55e56. http://dx.doi.org/10.1016/0005- 7967(93)E0028-4.

Stefanopoulou, E., Hirsch, C. R., Hayes, S., Adlam, A., & Coker, S. (2014). Are atten- tional control resources reduced by worry in generalized anxiety disorder? Journal of Abnormal Psychology, 123(2), 330e335. http://dx.doi.org/10.1037/ a0036343.

Stout, D. M., Shackman, A. J., Johnson, J. S., & Larson, C. L. (2014). Worry is associated with impaired gating of threat from working memory. Emotion, 15(1), 6e11. http://dx.doi.org/10.1037/emo00000015.

Szabo, M., & Lovibond, P. F. (2002). The cognitive content of naturally occuring worry episodes. Cognitive Therapy and Research, 26(2), 167e177.

Tang, Y., Ma, Y., Wang, J., Fan, Y., Feng, S., Lu, Q., … Posner, M. I. (2007). Short-term meditation training improves attention and self-regulation. PNAS Proceedings of the National Academy of Sciences of the United States of America, 104(43), 17152e17156. http://dx.doi.org/10.1073/pnas.0707678104.

Teasdale, J. D., Segal, Z., & Williams, J. G. (1995). How does cognitive therapy prevent depressive relapse and why should attentional control (mindfulness) training help? Behaviour Research & Therapy, 33(1), 25e39. http://dx.doi.org/10.1016/ 0005-7967(94)E0011-7.

Unsworth, N., Redick, T. S., Spillers, G. J., & Brewer, G. A. (2012). Variation in working memory capacity and cognitive control: Goal maintenance and microadjust- ments of control. Quarterly Journal of Experimental Psychology, 65, 326e355.

Vogel, E., McCollough, A., & Machizawa, M. (2005). Neural measures reveal indi- vidual differences in controlling access to working memory. Nature, 438, 500e503.

Wallace, B. A., & Shapiro, S. L. (2006). Mental balance and well-being: Building bridges between Buddhism and Western psychology. American Psychologist,

J. Course-Choi et al. / Behaviour Research and Therapy 89 (2017) 1e13 13

61(7), 690e701. http://dx.doi.org/10.1037/0003-066X.61.7.690. Wanmaker, S., Geraerts, E., & Franken, I. A. (2015). A working memory training to

decrease rumination in depressed and anxious individuals: A double-blind randomized controlled trial. Journal of Affective Disorders, 175, 310e319. http://dx.doi.org/10.1016/j.jad.2014.12.027.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of

Personality and Social Psychology, 54(6), 1063e1070. http://dx.doi.org/10.1037/ 0022-3514.54.6.1063.

Zeidan, F., Johnson, S. K., Diamond, B. J., David, Z., & Goolkasian, P. (2010). Mind- fulness meditation improves cognition: Evidence of brief mental training. Consciousness and Cognition: An International Journal, 19(2), 597e605. http:// dx.doi.org/10.1016/j.concog.2010.03.014.