Elsevier

Clinical Neurophysiology

Volume 127, Issue 2, February 2016, Pages 1307-1320
Clinical Neurophysiology

EEG alpha power during maintenance of information in working memory in adults with ADHD and its plasticity due to working memory training: A randomized controlled trial

https://doi.org/10.1016/j.clinph.2015.10.032Get rights and content

Highlights

  • Randomized controlled clinical trial investigating neural changes in alpha power after working memory training.

  • Differences in alpha power between ADHD and non-ADHD students, and changes after treatment, were marginally significant and had low effect sizes.

  • Findings suggest that alpha power could not reliably distinguish between ADHD and non-ADHD, nor trace treatment effects, in the present sample.

Abstract

Objective

The present study examined whether neural indices of working memory maintenance differ between young adults with ADHD and their healthy peers (Study 1), and whether this neural index would change after working memory training (Study 2).

Methods

Study 1 involved 136 college students with ADHD and 41 healthy peers (aged 18–35 years) and measured their posterior alpha activity during a visual delayed-match-to-sample task using electroencephalography (EEG). Study 2 involved 99 of the participants with ADHD who were randomized into a standard-length or shortened-length Cogmed working memory training program or a waitlist control group.

Results

The ADHD group tended to be less accurate than the peers. Similarly, the ADHD group exhibited lower posterior alpha power at a trend level compared to their healthy peers. There were no training effects on participants’ performance and only marginal increases in posterior alpha power in training groups compared to the waitlist group.

Conclusions

Considering that the training effects were small and there was no load and dose effect, we conclude that the current study provides no convincing evidence for specific effects of Cogmed.

Significance

These findings provide unique insights into neuroplasticity, or lack thereof, with near-transfer tasks in individuals with ADHD.

Introduction

Attention-Deficit/Hyperactivity Disorder (ADHD) is considered a neurobiological disorder often characterized by difficulties with working memory functioning (Barkley, 1997, Martinussen et al., 2005). The symptoms of a poor working memory resemble many of the everyday life problems individuals with ADHD experience with organization, distraction, and sustained concentration (Gray et al., 2015, Hervey et al., 2004). To improve working memory functioning in ADHD, software programs were developed to specifically target working memory using adaptive and intensive training (Klingberg et al., 2005). Although the current literature has shown that working memory improvements from such programs do not appear to transfer or generalize to other domains of functioning in ADHD populations, they have been found to improve performance in working memory tasks similar to those practiced – in other words, near transfer effects (Rapport et al., 2013, Mawjee et al., 2014, Melby-Lervåg and Hulme, 2013).

However, research into the effects of working memory training on the brain’s response to near transfer effects has been scarce. It is possible that neural measures are better able to capture subtle processing differences between individuals, or that result from training, which may not be detected by performance measures alone. Examining neural training effects can contribute to our basic understanding of neural plasticity related to working memory.

We conducted two studies. The objective of study 1 was to compare neural activity during the maintenance phase of working memory in young adults with and without ADHD; whereas the objective of study 2 was to determine whether intensive computerized working memory training (Cogmed working memory training, CWMT) would alter the neural activity during the maintenance phase in those with ADHD. To do so we used EEG with a visual delayed match-to-sample task which constitutes a type of span task that was comparable to, but not identical with those used as CWMT training tasks and likely relies on similar or overlapping neural networks. Accordingly, it may be conceptualized as a ‘near transfer’ working memory task.

A delayed match-to-sample task (see Haenschel et al., 2009, Kim et al., 2014) was used to investigate changes in neural responses during the maintenance phase of working memory. The task was visual in nature and manipulated the amount of representations that had to be held in mind. In this sense, this task was similar to the working memory processes consistently trained in the CWMT program. The vast majority of the CWMT tasks focus on increasing working memory capacity for visuo-spatial tasks by training the amount of information that can be maintained in working memory for several seconds (see Mawjee et al., 2014, for more details). Although this task has not been used previously in CWMT studies, we believed it was likely that it would be sensitive to near transfer effects. This task was used before in a previous study of college students with ADHD that focused on the encoding stage (Kim et al., 2014).

Recent theoretical insights on the role of attention in working memory, spurred by neuroscientific findings, have suggested that working memory performance may not solely depend on capacity limits viewed as a storage space, but also on how efficient brains are able to protect representations through attentional processes that filter out distractors (Awh and Vogel, 2008, Vogel et al., 2005). As a neural index, we therefore choose to focus on posterior alpha power because EEG oscillations in alpha frequency bands (9–14 Hz) are considered to play a key role in gating and protecting internal representations from distraction (Bonnefond and Jensen, 2012, Jensen and Mazaheri, 2010, Palva et al., 2011, Sauseng et al., 2009). Alpha power is considered to be particularly important for the sustained maintenance of working memory item information (Kundu et al., 2015, Hsieh et al., 2011). Previous research has shown that increased alpha in parietal–occipital regions during a delayed match-to-sample task was associated with better working memory performance (Jensen et al., 2002). Moreover, previous EEG studies using an identical delayed match-to-sample task (Haenschel et al., 2009) and studies using other tasks manipulating working memory (Crespo-Garcia et al., 2013, Michels et al., 2008) have found that alpha power increased after the offset of the stimulus during working memory maintenance. Our previous study using the same ADHD post-secondary education (PSE) population found strong evidence for reduced alpha power in the ADHD compared to a comparison group during a resting state (Woltering et al., 2012). These effects were interpreted as the ADHD group experiencing a lack of inhibition over sensory stimuli. As such, alpha power may be involved in the active inhibition of external/distracting stimuli, which could, if desynchronized, explain some of the attentional problems experienced by ADHD.

We choose postsecondary education (PSE) students with ADHD as they constitute a relatively understudied subgroup of ADHD. They are unique in that they are relatively high-functioning despite ongoing impairments (DuPaul et al., 2009). In our previous study involving participants in the present sample, standardized and normed scores did not show clinical levels of impairment on executive function and neuropsychological tasks, but their scores were lower compared to their PSE peers. Moreover, strong evidence for impairment was found in their qualitative, self-report, and measures specific to functional impairment in PSE (see Gray et al., 2015).

Studies investigating alpha power during a delayed match-to-sample working memory tasks in ADHD populations, however, are scarce and have mostly been conducted in child populations. Group difference studies in child populations thus far have been ambiguous; one study found no difference with peers (Gomarus et al., 2009; note this was a verbal working memory) whereas another found higher alpha for the ADHD group (Lenartowicz et al., 2014), which was explained as overcompensation. Therefore, the first goal (Study 1) was to compare PSE students with and without ADHD in terms of posterior alpha power during working memory maintenance to determine whether it would differentiate the two groups. Our second goal (Study 2) was to ascertain whether CWMT would influence (increase) posterior alpha power in these PSE students with ADHD. To do so we conducted a randomized controlled trial (RCT) in which participants with ADHD were randomly assigned into three treatment arms: a standard-length CWMT group (45 min/day), a shortened-length CWMT group which trained at one third of the standard-length intensity level (i.e., 15 min/day; see also Mawjee et al., 2015); and a waitlist group that received the same amount of contact with a training coach but did not receive CWMT. This design has a distinct advantage of controlling for engagement and ‘expectancy for improvement’ (Melby-Lervåg and Hulme, 2013). Furthermore, the presence of a dose effect would be strong evidence for the notion that the training caused any observed changes (e.g., the standard-length group would do better than the shortened-length group).

We formulated two main hypotheses:

Hypothesis study 1

Consistent with our previous findings in PSE students with ADHD (Kim et al., 2014, Woltering et al., 2012), we predicted lower posterior alpha and worse performance in our ADHD group compared to our Comparison group; and

Hypothesis study 2

We predicted posterior alpha power to increase (move in the direction of normalization) in ADHD in the training groups compared to the waitlist control group. More specifically, we expected a dose effect, whereby the increase in posterior alpha would be greater in the standard-length compared to the shortened-length training group, which in turn would show a greater increase compared to the waitlist who would show no change in posterior alpha power.

Section snippets

Participants

Participants with ADHD were recruited to participate in Study 1 and 2 through listserv emails sent from Student Disability Services. Registration with Student Disability Services in Canada requires comprehensive documentation or a new assessment to confirm their diagnosis. In addition, semi-structured telephone interviews were conducted to assess current eligibility and validate current ADHD symptomatology. Inclusion criteria were as follows: (1) registered with Student Accessibility/Disability

Behavioral data

To calculate each participant’s accuracy in the working memory task, we subtracted the false alarm rate from the hit rate. Using a two-way ANOVA (Group × Load) with age and sex as covariates, we found marginally significant main effects of Load, F(1, 173) = 3.24, p = .074, partial Eta square = .018 (90% CI = [0 .064]), and Group, F(1, 173) = 3.22, p = .074, partial Eta square = .018 (90% CI = [0 .064]). No significant interactions were found between the two factors (p = .229). It is worth mentioning that without

Summary of results

We set out to investigate whether alpha oscillation power during working memory maintenance was different between college students with ADHD and their healthy peers (Study 1), and whether this neural response would change after working memory training (Study 2). Our results showed suggestive evidence for small effects showing lower accuracy and lower alpha power for the ADHD compared to the comparison group during working memory maintenance. Within the ADHD sample, no training effects were

Trial registration

www.clinicaltrials.gov ‘Working Memory Training in ADHD (The Engage Study)’ # NCT01657721.

Acknowledgements

We thank Dr. Corinna Haenschel and Nikolaus Kriegeskorte for letting us use the task stimuli. This research was supported financially in part by a CIHR Operating Grant (# 245899, Tannock & Lewis) and by the Canada Research Chair program (Rosemary Tannock). We also want to acknowledge the Jewish Vocational Services for their openness in collaborating with us.

Conflict of interest: Authors report no conflict of interest.

References (95)

  • T. Klingberg

    Training and plasticity of working memory

    Trends Cogn Sci

    (2010)
  • T. Klingberg et al.

    Computerized training of working memory in children with ADHD-a randomized, controlled trial

    J Am Acad Child Adolesc Psychiatry

    (2005)
  • A.J. Krafnick et al.

    Gray matter volume changes following reading intervention in dyslexic children

    NeuroImage

    (2011)
  • A. Krishnan et al.

    Partial Least Squares (PLS) methods for neuroimaging: a tutorial and review

    NeuroImage

    (2011)
  • B. Kundu et al.

    Context-specific differences in fronto-parieto-occipital effective connectivity during short-term memory maintenance

    NeuroImage

    (2015)
  • S.M. Landau et al.

    A functional MRI study of the influence of practice on component processes of working memory

    NeuroImage

    (2004)
  • M. Lövdén et al.

    Experience-dependent plasticity of white-matter microstructure extends into old age

    Neuropsychologia

    (2010)
  • R. Martinussen et al.

    A meta-analysis of working memory impairments in children with attention-deficit/hyperactivity disorder

    J Am Acad Child Adolesc Psychiatry

    (2005)
  • A. May

    Experience-dependent structural plasticity in the adult human brain

    Trends Cogn Sci

    (2011)
  • A.R. McIntosh et al.

    Partial least squares analysis of neuroimaging data: applications and advances

    NeuroImage

    (2004)
  • L. Michels et al.

    EEG alpha distinguishes between cuneal and precuneal activation in working memory

    NeuroImage

    (2008)
  • M.D. Rapport et al.

    Do programs designed to train working memory, other executive functions, and attention benefit children with ADHD? A meta-analytic review of cognitive, academic, and behavioral outcomes

    Clin Psychol Rev

    (2013)
  • P. Sauseng et al.

    Brain oscillatory substrates of visual short-term memory capacity

    Curr Biol

    (2009)
  • Z. Shipstead et al.

    Cogmed working memory training: does the evidence support the claims?

    J Appl Res Mem Cogn

    (2012)
  • S.S. Simon et al.

    Cognitive intervention in amnestic Mild Cognitive Impairment: a systematic review

    Neurosci Biobehav Rev

    (2012)
  • C. Spironelli et al.

    Functional plasticity in Alzheimer’s disease: effect of cognitive training on language-related ERP components

    Neuropsychologia

    (2013)
  • K. Subramaniam et al.

    Computerized cognitive training restores neural activity within the reality monitoring network in schizophrenia

    Neuron

    (2012)
  • Y. Watanabe et al.

    Thalamic mediodorsal nucleus and working memory

    Neurosci Biobehav Rev

    (2012)
  • S. Woltering et al.

    Neural changes associated with treatment outcome in children with externalizing problems

    Biol Psychiatry

    (2011)
  • S. Woltering et al.

    Neurophysiological differences in inhibitory control between adults with ADHD and their peers

    Neuropsychologia

    (2013)
  • L.A. Adler et al.

    Preliminary examination of the reliability and concurrent validity of the Attention-Deficit/Hyperactivity Disorder Self-Report Scale v1.1 Symptom checklist to rate symptoms of Attention-Deficit/Hyperactivity Disorder in adolescents

    J Child Adolesc Psychopharmacol

    (2012)
  • D.E. Anderson et al.

    Induced alpha rhythms track the content and quality of visual working memory representations with high temporal precision

    J Neurosci

    (2014)
  • E. Awh et al.

    The bouncer in the brain

    Nat Neurosci

    (2008)
  • R.A. Barkley

    Behavioral inhibition, sustained attention, and executive functions: constructing a unifying theory of ADHD

    Psychol Bull

    (1997)
  • A.G. Barnett et al.

    Regression to the mean: what it is and how to deal with it

    Int J Epidemiol

    (2005)
  • A. Bollimunta et al.

    Neuronal mechanisms and attentional modulation of corticothalamic alpha oscillations

    J Neurosci

    (2011)
  • D.E. Broadbent et al.

    The cognitive failures questionnaire (CFQ) and its correlates

    Br J Clin Psychol

    (1982)
  • J.M. Chein et al.

    Neuroimaging studies of practice-related change: fMRI and meta-analytic evidence of a domain-general control network for learning

    Cogn Brain Res

    (2005)
  • L. Clare et al.

    Cognitive training and cognitive rehabilitation for people with early-stage Alzheimer’s disease: a review

    Neuropsychol Rehabil

    (2004)
  • M. Crespo-Garcia et al.

    Working memory processes are mediated by local and long-range synchronization of alpha oscillations

    J Cogn Neurosci

    (2013)
  • K.N. Dalby et al.

    Self-paced learning in children with attention deficit disorder with hyperactivity

    J Abnorm Child Psychol

    (1989)
  • M. D’Esposito et al.

    The cognitive neuroscience of working memory

    Annu Rev Psychol

    (2015)
  • B. Draganski et al.

    Neuroplasticity: changes in grey matter induced by training – newly honed juggling skills show up as a transient feature on a brain-imaging scan

    Nature

    (2004)
  • B. Draganski et al.

    Temporal and spatial dynamics of brain structure changes during extensive learning

    J Neurosci

    (2006)
  • D.L. Dunning et al.

    Does working memory training lead to generalized improvements in children with low working memory? A randomized controlled trial

    Dev Sci

    (2013)
  • G.J. DuPaul et al.

    College students with ADHD: current status and future directions

    J Atten Disord

    (2009)
  • K.I. Erickson et al.

    Training-induced functional activation changes in dual-task processing: an fMRI study

    Cereb Cortex

    (2007)
  • Cited by (29)

    • Neural and cognitive function in a pediatric brain injury model: The impact of task complexity

      2022, Cortex
      Citation Excerpt :

      MEG measures magnetic fields of electrical currents generated by oscillatory activity of neural populations (Hari et al., 2015). Oscillatory activity in theta (4–7 Hz) and alpha (8–12 Hz) frequency bands form synchronized networks that support supervisory attention, action selection and inhibition (Canolty et al., 2006; Cavanagh et al., 2012; Dambacher et al., 2014; Doesburg et al., 2016; Liu et al., 2016; Sauseng et al., 2010; van Schouwenburg et al., 2017; Zanto et al., 2010). Communication in the high gamma band (60–100 Hz) facilitates information processing, including primary visual processing and controlled visual-motor attention (Dockstader et al., 2014; Gaetz et al., 2013; Mainy et al., 2007; Cheyne, 2013; Müller et al., 2000).

    • Event-related brain oscillations in attention-deficit/hyperactivity disorder (ADHD): A systematic review and meta-analysis

      2022, International Journal of Psychophysiology
      Citation Excerpt :

      Other reported findings in children were increased alpha increases during WM maintenance (Lenartowicz et al., 2014, 2019), atypical alpha lateralization (Ellis et al., 2017; McAuliffe et al., 2020; Vollebregt et al., 2016), and no difference in parieto-occipital alpha decreases (Vainieri et al., 2020) and central alpha (Yordanova et al., 2013). Adult ADHD studies similarly showed consistently weaker modulations of alpha power, either lower decreases (Bozhilova et al., 2020; Deiber et al., 2020; Hasler et al., 2016; Michelini et al., 2018b; Missonnier et al., 2013) or lower increases (Leroy et al., 2018; Liu et al., 2016), except for one study showing no group differences (Vainieri et al., 2020). With regard to beta power, children with ADHD showed significantly weaker beta decreases compared to control children in three studies (Mazaheri et al., 2014; McAuliffe et al., 2020; Zammit and Muscat, 2019), a significantly stronger decrease in one study (Baijot et al., 2017), and no differences in the study with the largest sample (Vainieri et al., 2020) (Table 1).

    • A controlled clinical crossover trial of exercise training to improve cognition and neural communication in pediatric brain tumor survivors

      2020, Clinical Neurophysiology
      Citation Excerpt :

      Finally, the study adds to the repertoire of potential interventions that can be offered to improve brain function and cognition in clinical pediatric populations (Liu et al., 2016; Olson et al., 2017).

    • Effects of working memory training on neural correlates of Go/Nogo response control in adults with ADHD: A randomized controlled trial

      2017, Neuropsychologia
      Citation Excerpt :

      However, because the difficulty level of working memory training remains significantly lower than that in the experimental condition, factors such as motivation, expectation of improvement, and arousal level may differ between the control and experimental groups. To address this issue, our research group has used a dose-effect design, in which the control group and experimental group receive identical training, except that the training time (i.e., dosage) differs between the two groups (Liu et al., 2016; Mawjee et al., 2015). Both treatments are likely to be seen as beneficial and would therefore equate effects of expectancy and motivation in comparison to, for example, waitlist or non-adaptive control group activities.

    View all citing articles on Scopus
    View full text