CoChat
Sign Up
Cited by 32Cognitive Abilities and TestingCognitive Functions and MemoryNeural and Behavioral Psychology Studies

Adaptive working memory training does not produce transfer effects in cognition and neuroimaging

Open access
Ripp, Isabelle, Emch, Mónica, Wu, Qiong, Lizarraga, Aldana, Udale, Robert, von Bastian, Claudia Christina, Koch, Kathrin, Yakushev, Igor·2022·Translational Psychiatry
DOI 10.1038/s41398-022-02272-7PMC PMC9747798
Open this paper in CoChat — free accountOpen PDF

Abstract

Despite growing interest in cognitive interventions from academia and industry, it remains unclear if working memory (WM) training, one of the most popular cognitive interventions, produces transfer effects. Transfer effects are training-induced gains in performance in untrained cognitive tasks, while practice effects are improvements in trained task. The goal of this study was to evaluate potential transfer effects by comprehensive cognitive testing and neuroimaging. In this prospective, randomized-controlled, and single-blind study, we administered an 8-week n-back training to 55 healthy middle-aged (50–64 years) participants. State-of-the-art multimodal neuroimaging was used to examine potential anatomic and functional changes. Relative to control subjects, who performed non-adaptive WM training, no near or far transfer effects were detected in experimental subjects, who performed adaptive WM training. Equivalently, no training-related changes were observed in white matter integrity, amplitude of low frequency fluctuations, glucose metabolism, functional and metabolic connectivity. Exploratory within-group comparisons revealed some gains in transfer tasks, which, however, cannot be attributed to an increased WM capacity. In conclusion, WM training produces transfer effects neither at the cognitive level nor in terms of neural structure or function. These results speak against a common view that training-related gains reflect an increase in underlying WM capacity. Instead, the presently observed practice effects may be a result of optimized task processing strategies, which do not necessarily engage neural plasticity.

AI snapshot

Written from the paper’s full text.

TL;DR

In an 8-week randomized, active-controlled, single-blind trial in 55 healthy middle-aged adults, adaptive n-back training produced practice effects but no near or far transfer relative to non-adaptive control training. Multimodal PET/MRI found no training-related changes in white matter integrity, ALFF, glucose metabolism, or functional/metabolic connectivity.

Key findings

  • The experimental group improved on the trained n-back tasks: verbal d prime (Z = 7, p = 8.07e−6), visual d prime (Z = 13, p = 1.51e−5), verbal n-back level (Z = 21, p = 3.34e−5), and visual n-back level (Z = 18.5, p = 6.51e−5); the active control group showed no significant d prime changes.
  • The only significant group × time interaction was for the nearest transfer category (F(4,50) = 5.3, p < 0.013), driven by Digit Span forward (F = 16.97, p < 0.0001, η² = 0.243; BF10 = 182.42); there was no corresponding improvement in Digit Span backward, and no near or far transfer effects survived correction.
  • Neuroimaging showed no significant group × time interactions for ALFF, FDG-PET, DTI maps (FA, MD, AD, RD), or fMRI/PET network integrity after Bonferroni correction; uncorrected fMRI changes in the salience network and anterior DMN (p = 0.03) did not survive correction or contrast against the active control group.
  • Bayesian analyses mostly provided ambiguous or weak evidence against transfer, e.g., Corsi BF01 = 3.77, VLMT BFs01 = 2.75–3.66, short-APM BF01 = 1.78, and Iowa Gambling Task BF01 = 3.99.

Methods

  • Prospective, randomized-controlled, single-blind study: 55 right-handed healthy participants aged 50–64 years (mean age 55.9 ± 4.2 years; 30 males), randomized to adaptive visual/verbal n-back training (n = 28) or non-adaptive active control training (X-back/1-back; n = 27) for 8 weeks, about 20 min per session, four sessions per week.
  • Cognitive test battery administered one week before and one week after training assessed nearest transfer (Digit Span, Simple Visual Reaction Time, Corsi Block-Tapping), near transfer (Visual Simon, Colour Word Stroop, VLMT, RVIP), and far transfer (short Raven’s Advanced Progressive Matrices, Iowa Gambling Task); analyses used frequentist ANOVA/t-tests/Wilcoxon tests plus Bayesian t-tests and ANOVAs.
  • Imaging used a 3T hybrid PET/MRI scanner with FDG-PET, resting-state fMRI (ALFF and ICA-based network integrity), and DTI (TBSS), analyzed at voxel and network levels.

Limitations

  • The paper states that patients with affected working memory and neuroimaging indices of brain health might have more room for gains, but co-existing attention deficits and disease-related deterioration of neural plasticity may limit WM training efficiency; it suggests future studies target multiple cognitive domains and use techniques beyond fMRI, DTI, and FDG-PET.
  • The paper notes that exploratory uncorrected neuroimaging findings did not survive multiple correction or active-control contrasts, so they are not robust.
  • Evident from the text: the sample was healthy middle-aged adults only, limiting generalizability to younger, older, or clinical populations; the active control was a non-adaptive version of the same tasks, which is a stringent control but may also reduce between-group differences.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41398-022-02272-7

Cites

79 of 79 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

34 of 56 cited sources checked · 22 supported · 4 flagged · accuracy not yet measured

  • Premisecitation

    Subsequent meta-analyses did not find any significant far transfer effects, especially when the effects of WM training were compared with an active control group.

    “However, subsequent meta-analyses did not find any significant far transfer effects [5, 13, 17], especially when the effects of WM training were compared with an active control group, that is, a group practicing tasks with little WM demands.”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [5] Soveri A, Antfolk J, Karlsson L, Salo B, Laine M. Working memory training revisited: A multi-level meta-analysis of n-back training studies. Psychon Bull Rev. 2017;24:1077–96. https://doi.org/10.3758/s13423-016-1217-0. source

      The Training effects by cognitive domain section reports statistically significant transfer to cognitive control and fluid intelligence, contrary to the unqualified claim of no significant far transfer. The Control analyses section does support the active-control qualification: after accounting for control-group type, neither effect was significant. The paper defines active controls as groups practicing tasks not intended to substantially tax working memory (Background for the present study; page numbers unavailable). (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [13] Melby-Lervåg M, Redick TS, Hulme C. Working memory training does not improve performance on measures of intelligence or other measures of “far transfer”. Perspect Psychol Sci. 2016;11:512–34. https://doi.org/10.1177/1745691616635612. source
    • cites [17] Dougherty MR, Hamovitz T, Tidwell JW. Reevaluating the effectiveness of n-back training on transfer through the Bayesian lens: support for the null. Psychon Bull Rev. 2016;23:306–16. https://doi.org/10.3758/s13423-015-0865-9. source

      The Meta Analysis Bayes Factor section reports that, across 12 active-control studies, the estimated effects of n-back training on fluid-intelligence transfer were not different from zero and the meta-analytic Bayes factor favored the null. The claim is broader than this analysis: it refers to WM training and far transfer generally, whereas this paper reanalyzes n-back training and transfer to fluid intelligence. The Abstract also notes that the earlier meta-analysis found an overall positive effect. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
  • Premisecitation

    No FDG-PET study exists in conjunction with WM training.

    “Finally, although some FDG-PET studies investigated neural correlates of mental exercise [34–36], no such PET study exists in conjunction with WM training.”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites [34] Förster S, Buschert VC, Teipel SJ, Friese U, Buchholz H-G, Drzezga A, et al. Effects of a 6-month cognitive intervention on brain metabolism in patients with amnestic MCI and mild Alzheimer’s disease. J Alzheimers Dis JAD. 2011;26:337–48. https://doi.org/10.3233/JAD-2011-0025. source
    • cites [35] Shah T, Verdile G, Sohrabi H, Campbell A, Putland E, Cheetham C, et al. A combination of physical activity and computerized brain training improves verbal memory and increases cerebral glucose metabolism in the elderly. Transl Psychiatry. 2014;4:e487. https://doi.org/10.1038/tp.2014.122. source
    • cites [36] Small GW, Silverman DHS, Siddarth P, Ercoli LM, Miller KJ, Lavretsky H, et al. Effects of a 14-day healthy longevity lifestyle program on cognition and brain function. Am J Geriatr Psychiatry J Am Assoc Geriatr Psychiatry. 2006;14:538–45. https://doi.org/10.1097/01.JGP.0000219279.72210.ca. source
  • Premisecitation

    WM capacity is closely correlated with other higher-order cognitive functions such as fluid intelligence, abstract reasoning, and reading comprehension.

    “WM capacity is closely correlated with other higher-order cognitive functions such as fluid intelligence, abstract reasoning, and reading comprehension [3].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [3] Barrett LF, Tugade MM, Engle RW. Individual differences in working memory capacity and dual-process theories of the mind. Psychol Bull. 2004;130:553–73. https://doi.org/10.1037/0033-2909.130.4.553. source
  • Findingown data

    We did not detect any changes in ALFF or glucose metabolism after WM training.

    “Different from the present study, in which we did not detect any changes in ALFF or glucose metabolism after WM training, Takeuchi et al. reported increased brain activity in the dorsal prefrontal cortex following WM training [27].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [27] Takeuchi H, Taki Y, Nouchi R, Sekiguchi A, Kotozaki Y, Nakagawa S, et al. Neural plasticity in amplitude of low frequency fluctuation, cortical hub construction, regional homogeneity resulting from working memory training. Sci Rep. 2017;7:1470. https://doi.org/10.1038/s41598-017-01460-6. source
    • Rests on an assumption: The neuroimaging analyses had sufficient sensitivity to detect training-induced changes if present.
  • Premisecitation

    Meta-analyses that explicitly distinguish between types of control group typically report a lack of far, sometimes also near, transfer effects.

    “Critically, meta-analyses that explicitly distinguish between the types of a control group typically report a lack of far, sometimes also near, transfer effects [5, 15, 17].”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [5] Soveri A, Antfolk J, Karlsson L, Salo B, Laine M. Working memory training revisited: A multi-level meta-analysis of n-back training studies. Psychon Bull Rev. 2017;24:1077–96. https://doi.org/10.3758/s13423-016-1217-0. source

      The Control analyses (page not available) distinguish active from passive controls and find no statistically significant transfer to cognitive control or fluid intelligence after accounting for control-group type. But they find significant transfer to working-memory tasks, and the Main analyses (page not available) describe other transfer effects as very small rather than absent. The paper therefore supports the claim about some far-transfer outcomes, not a general lack of transfer or a lack of near transfer in this analysis. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [15] Sala G, Aksayli ND, Tatlidil KS, Gondo Y, Gobet F. Working memory training does not enhance older adults’ cognitive skills: a comprehensive meta-analysis. Intelligence. 2019;77:101386. https://doi.org/10.1016/j.intell.2019.101386. source
    • cites [17] Dougherty MR, Hamovitz T, Tidwell JW. Reevaluating the effectiveness of n-back training on transfer through the Bayesian lens: support for the null. Psychon Bull Rev. 2016;23:306–16. https://doi.org/10.3758/s13423-015-0865-9. source
  • Classified as probably not a substantial claim. 28
  • May combine two claims. 28

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    A meta-analysis reported a trend for n-back being the most efficient WM training.

    “Importantly, a meta-analysis reported a trend for n-back being the most efficient WM training [13].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [13] Melby-Lervåg M, Redick TS, Hulme C. Working memory training does not improve performance on measures of intelligence or other measures of “far transfer”. Perspect Psychol Sci. 2016;11:512–34. https://doi.org/10.1177/1745691616635612. source

      The Results (Immediate Effects on Intermediate Transfer Measures; page unavailable) report that CogMed, not N-back, had a significantly larger effect on verbal working memory than the other training programs. The Results (Immediate Effects on Far Transfer Measures; page unavailable) attribute N-back’s apparent effect on nonverbal ability to problematic studies and say it became negligible when the most problematic study was removed. The paper therefore does not report a trend establishing N-back as the most efficient working-memory training. (LLM, conf=0.85)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationcitation

    As transfer effects within and between cognitive domains were shown to be of equivalent magnitude when comparing different WM training paradigms, it is reasonable to assume that our results are generalizable to other WM training paradigms.

    “As transfer effects within and between cognitive domains were shown to be of equivalent magnitude when comparing different WM training paradigms (e.g., n-back training and complex span training) [81], it is reasonable to assume that our results are generalizable to other WM training paradigms.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [81] Holmes J, Woolgar F, Hampshire A, Gathercole SE. Are working memory training effects paradigm-specific? Front Psychol. 2019;10:1103. https://doi.org/10.3389/fpsyg.2019.01103. source

      The Results > Transfer (page unavailable) report strong evidence of transfer to verbal and visuo-spatial n-back tasks after n-back training, but not comparable evidence of transfer after complex-span training. The Discussion (page unavailable) describes within-paradigm transfer to n-back but not complex span, and no cross-paradigm transfer. These findings contradict the attributed claim that transfer effects were equivalent across the training paradigms. (LLM, conf=0.88)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    No transfer effects were observed in 3 out of 8 studies, and only nearest transfer effects as measured with digit span forward and/or backward were observed by three other studies.

    “Critically, no transfer effects were observed in 3 out of 8 studies [72–74], and only nearest transfer effects as measured with digit span forward and/or backward were observed by three other studies [19, 75, 76].”
    Quote in the paperSubstantial claim4 of 6 sources checked
    • cites [72] Chan S-Y, Wu Q, Liang D, Yan J. Visuospatial working memory training facilitates visually-aided explicit sequence learning. Acta Psychol. 2015;161:145–53. https://doi.org/10.1016/j.actpsy.2015.09.008. source
    • cites [73] Cuenen A, Jongen EMM, Brijs T, Brijs K, Houben K, Wets G. Effect of a working memory training on aspects of cognitive ability and driving ability of older drivers: merits of an adaptive training over a non-adaptive training. Transp Res Part F Traffic Psychol Behav. 2016;42:15–27. https://doi.org/10.1016/j.trf.2016.06.012. source

      The Discussion reports near-transfer effects in working memory but no differential improvement in other cognitive functions. However, §2.4 says the working-memory outcome was measured with the Automated Operation Span task, not forward or backward digit span; §2.7 lists backward digit span as a training task. This paper also does not establish the sentence’s aggregate count of studies. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [74] Wayne RV, Hamilton C, Jones Huyck J, Johnsrude IS. Working memory training and speech in noise comprehension in older adults. Front Aging Neurosci. 2016;8:49. https://doi.org/10.3389/fnagi.2016.00049. source
    • cites [19] Brehmer Y, Westerberg H, Bäckman L. Working-memory training in younger and older adults: training gains, transfer, and maintenance. Front Hum Neurosci. 2012;6:63. https://doi.org/10.3389/fnhum.2012.00063. source

      The citation to this paper is attached to the claim that it found only near-transfer effects. The Results (Far-transfer tasks) instead report a significant intervention-by-time effect on PASAT, a far-transfer task: the adaptive-training groups improved more than the low-level practice groups. The paper therefore does not support the word “only.” (LLM, conf=0.90)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
    • cites [75] McAvinue LP, Golemme M, Castorina M, Tatti E, Pigni FM, Salomone S, et al. An evaluation of a working memory training scheme in older adults. Front Aging Neurosci. 2013;5:20. https://doi.org/10.3389/fnagi.2013.00020. source

      The Results (page not supplied) report improvement on Digit Span Forwards, but also significant post-training improvement in immediate and delayed word recall. The Discussion (page not supplied) explicitly calls the word-recall findings evidence of transfer to episodic memory, contrary to the claim that this study observed only nearest transfer. The word-recall gains were not maintained at follow-up. (LLM, conf=0.86)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
    • cites [76] Weicker J, Hudl N, Marichal E, Müller K, Lepsien J, Trapp S, et al. Training of working memory in healthy elderly subjects - A randomized controlled trial. 4th Joint Meeting of the European Societies of Neuropsycholgy (ESN) and the 28th Meeting of the German Society of Neuropsychology (GNP), 2013.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    More recent meta-analyses that distinguished between nearest and near transfer effects found significantly greater effect sizes for the former, suggesting that WM training may mainly yield task-specific transfer rather than a general improvement in WM.

    “However, more recent meta-analyses that distinguished between nearest and near transfer effects found significantly greater effect sizes for the former, suggesting that WM training may mainly yield task-specific transfer rather than a general improvement in WM [5, 15].”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [5] Soveri A, Antfolk J, Karlsson L, Salo B, Laine M. Working memory training revisited: A multi-level meta-analysis of n-back training studies. Psychon Bull Rev. 2017;24:1077–96. https://doi.org/10.3758/s13423-016-1217-0. source
    • cites [15] Sala G, Aksayli ND, Tatlidil KS, Gondo Y, Gobet F. Working memory training does not enhance older adults’ cognitive skills: a comprehensive meta-analysis. Intelligence. 2019;77:101386. https://doi.org/10.1016/j.intell.2019.101386. source

  • Classified as probably not a substantial claim.

    Premisecitation

    Neuroimaging might depict evidence of neural plasticity that precedes measurable changes in cognition.

    “Specifically, neuroimaging might depict evidence of neural plasticity that precedes measurable changes in cognition [21].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [21] Karuza EA, Emberson LL, Aslin RN. Combining fMRI and behavioral measures to examine the process of human learning. Neurobiol Learn Mem. 2014;109:193–206. https://doi.org/10.1016/j.nlm.2013.09.012. source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Results indicated training-associated changes in frontal and parietal white matter tracts or the corpus callosum.

    “Results indicated training-associated changes in frontal and parietal white matter tracts [30, 33] or the corpus callosum [31].”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites [30] Dziemian S, Appenzeller S, von Bastian CC, Jäncke L, Langer N. Working memory training effects on white matter integrity in young and older adults. Front Hum Neurosci. 2021. https://doi.org/10.3389/fnhum.2021.605213. source
    • cites [33] Takeuchi H, Sekiguchi A, Taki Y, Yokoyama S, Yomogida Y, Komuro N, et al. Training of working memory impacts structural connectivity. J Neurosci. 2010;30:3297–303. https://doi.org/10.1523/JNEUROSCI.4611-09.2010. source
    • cites [31] Lövdén M, Bodammer NC, Kühn S, Kaufmann J, Schütze H, Tempelmann C, et al. Experience-dependent plasticity of white-matter microstructure extends into old age. Neuropsychologia. 2010;48:3878–83. https://doi.org/10.1016/j.neuropsychologia.2010.08.026. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The authors of that study found increased activity in the right dorsolateral prefrontal cortex, frontopolar area, and medial prefrontal cortex.

    “The authors found increased activity in the right dorsolateral prefrontal cortex, frontopolar area, and medial prefrontal cortex.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [27] Takeuchi H, Taki Y, Nouchi R, Sekiguchi A, Kotozaki Y, Nakagawa S, et al. Neural plasticity in amplitude of low frequency fluctuation, cortical hub construction, regional homogeneity resulting from working memory training. Sci Rep. 2017;7:1470. https://doi.org/10.1038/s41598-017-01460-6. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    N-back training is one of the most extensively studied forms of WM training.

    “N-back training is one of the most extensively studied forms of WM training [5].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [5] Soveri A, Antfolk J, Karlsson L, Salo B, Laine M. Working memory training revisited: A multi-level meta-analysis of n-back training studies. Psychon Bull Rev. 2017;24:1077–96. https://doi.org/10.3758/s13423-016-1217-0. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The meta-analysis of Au et al. reported a small, but significant positive transfer effect on fluid intelligence in healthy young adults.

    “For example, the meta-analysis of Au et al. [16] reported a small, but significant positive transfer effect on fluid intelligence in healthy young adults.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [16] Au J, Sheehan E, Tsai N, Duncan GJ, Buschkuehl M, Jaeggi SM. Improving fluid intelligence with training on working memory: a meta-analysis. Psychon Bull Rev. 2015;22:366–77. https://doi.org/10.3758/s13423-014-0699-x. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Meta-analyses averaging effects of training on WM tasks with varying similarity have reported significant near transfer effects.

    “Meta-analyses averaging effects of training on WM tasks with varying similarity have reported significant near transfer effects, e.g., [13, 14].”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [13] Melby-Lervåg M, Redick TS, Hulme C. Working memory training does not improve performance on measures of intelligence or other measures of “far transfer”. Perspect Psychol Sci. 2016;11:512–34. https://doi.org/10.1177/1745691616635612. source
    • cites [14] Schwaighofer M, Fischer F, Bühner M. Does working memory training transfer? A meta-analysis including training conditions as moderators. Educ Psychol. 2015;50:138–66. https://doi.org/10.1080/00461520.2015.1036274. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Takeuchi et al. reported that WM training increased the mean diffusivity in regions of the dopaminergic system.

    “Takeuchi et al. [32] reported that WM training increased the mean diffusivity in regions of the dopaminergic system.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [32] Takeuchi H, Taki Y, Nouchi R, Hashizume H, Sekiguchi A, Kotozaki Y, et al. Working memory training impacts the mean diffusivity in the dopaminergic system. Brain Struct Funct. 2015;220:3101–11. https://doi.org/10.1007/s00429-014-0845-2. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Only one study has investigated effects of WM training on the magnitude of intrinsic neural activity at rest.

    “Moreover, to the best of our knowledge, only one study has investigated effects of WM training on the magnitude of intrinsic neural activity at rest [27].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [27] Takeuchi H, Taki Y, Nouchi R, Sekiguchi A, Kotozaki Y, Nakagawa S, et al. Neural plasticity in amplitude of low frequency fluctuation, cortical hub construction, regional homogeneity resulting from working memory training. Sci Rep. 2017;7:1470. https://doi.org/10.1038/s41598-017-01460-6. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Only a few studies have investigated changes in structural connectivity following WM training.

    “Similarly, only a few studies have investigated changes in structural connectivity, that is, neural tracts, following WM training [30–33].”
    Quote in the paperNot a substantial claim3 of 4 sources checked
    • cites [30] Dziemian S, Appenzeller S, von Bastian CC, Jäncke L, Langer N. Working memory training effects on white matter integrity in young and older adults. Front Hum Neurosci. 2021. https://doi.org/10.3389/fnhum.2021.605213. source
    • cites [31] Lövdén M, Bodammer NC, Kühn S, Kaufmann J, Schütze H, Tempelmann C, et al. Experience-dependent plasticity of white-matter microstructure extends into old age. Neuropsychologia. 2010;48:3878–83. https://doi.org/10.1016/j.neuropsychologia.2010.08.026. source
    • cites [32] Takeuchi H, Taki Y, Nouchi R, Hashizume H, Sekiguchi A, Kotozaki Y, et al. Working memory training impacts the mean diffusivity in the dopaminergic system. Brain Struct Funct. 2015;220:3101–11. https://doi.org/10.1007/s00429-014-0845-2. source

      The Results section (‘The effect of WMT on MD’; page not available) reports changes in mean diffusivity after working-memory training, so the paper is an example of research on training-related neural microstructure. It does not report changes in structural connectivity or neural tracts, and it does not establish that only a few studies have investigated them. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites [33] Takeuchi H, Sekiguchi A, Taki Y, Yokoyama S, Yomogida Y, Komuro N, et al. Training of working memory impacts structural connectivity. J Neurosci. 2010;30:3297–303. https://doi.org/10.1523/JNEUROSCI.4611-09.2010. source

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Practicing WM tasks can also trigger cognitive changes in domains different to WM capacity, such as attention and visual integration, or the development of task- or material-specific expertise.

    “Moreover, practicing WM tasks can also trigger cognitive changes in domains different to WM capacity, such as attention and visual integration [19], or the development of task- or material-specific expertise [20].”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [19] Brehmer Y, Westerberg H, Bäckman L. Working-memory training in younger and older adults: training gains, transfer, and maintenance. Front Hum Neurosci. 2012;6:63. https://doi.org/10.3389/fnhum.2012.00063. source

      The Discussion (page unavailable) reports that working-memory training improved performance on a sustained-attention test, supporting the attention example attributed to this source. Our copy does not report a change in visual integration. The task- or material-specific expertise example is attached to a different citation [20], not this paper. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites [20] De Simoni C, von Bastian CC. Working memory updating and binding training: Bayesian evidence supporting the absence of transfer. J Exp Psychol Gen. 2018;147:829–58. https://doi.org/10.1037/xge0000453. source

  • May combine two claims.

    Interpretationcitation

    Both training interventions, no matter how specific they have been designed, can trigger performance improvements which are not due to an increased WM capacity, but most likely due to non-specific cognitive training effects, such as improved attentional processing, or due to the acquisition of strategies such as chunking or visualization.

    “This implies that both training interventions, no matter how specific they have been designed, can trigger performance improvements which are not due to an increased WM capacity, but most likely due to non-specific cognitive training effects, such as improved attentional processing [19], or due to the acquisition of strategies such as chunking or visualization [80].”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [19] Brehmer Y, Westerberg H, Bäckman L. Working-memory training in younger and older adults: training gains, transfer, and maintenance. Front Hum Neurosci. 2012;6:63. https://doi.org/10.3389/fnhum.2012.00063. source

      Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured
    • cites [80] Laine M, Fellman D, Waris O, Nyman TJ. The early effects of external and internal strategies on working memory updating training. Sci Rep. 2018;8:4045. https://doi.org/10.1038/s41598-018-22396-5. source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Aksayli et al. reported a consistent lack of far transfer, small to medium effect sizes for near transfer, and that these transfer effects depended on the similarity of the transfer tasks to the WM training paradigm.

    “Similar findings were reported by Aksayli et al. [79], who evaluated the effects of adaptive Cogmed WM training, a commercial training programme, relative to non-adaptive Cogmed training. The authors reported a consistent lack of far transfer, small to medium effects sizes for near transfer and that these transfer effects depended on the similarity of the transfer tasks to the WM training paradigm”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [79] Aksayli ND, Sala G, Gobet F. The cognitive and academic benefits of Cogmed: a meta-analysis. Educ Res Rev. 2019;27:229–43. https://doi.org/10.1016/j.edurev.2019.04.003. source

      The Results (section 4.2.1) report a medium near-transfer effect moderated by task overlap, and the Discussion (section 5) reports consistently negligible far transfer. The comparison is narrower than the paper's design: the Inclusion criteria (section 3.2) and Moderators (section 3.4) show that the meta-analysis included several kinds of control group, not only non-adaptive Cogmed training. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    No training-related changes were observed in white matter integrity, amplitude of low frequency fluctuations, glucose metabolism, functional and metabolic connectivity.

    “Equivalently, no training-related changes were observed in white matter integrity, amplitude of low frequency fluctuations, glucose metabolism, functional and metabolic connectivity.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The experimental group improved in the incongruent trials of the Visual Simon task, whereas the control group improved both in congruent and incongruent trials.

    “Here, the experimental group improved in the incongruent trials (RT-incong), whereas the control group improved both in congruent and incongruent trials.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    Only one participant reached the 9-back level, indicating the absence of a ceiling effect at the group level.

    “Only one participant reached the 9-back level, indicating the absence of a ceiling effect at the group level.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    At an uncorrected threshold of p < 0.05, there was a change for the SN and aDMN in the fMRI data, but not in the PET data.

    “At an uncorrected threshold of p < 0.05, there was a change for the SN and aDMN in the fMRI data (Table 4 and Fig. 5), but not in the PET data (Table 5 and Supplemental Fig. S2).”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The neuroimaging analyses had sufficient sensitivity to detect training-induced changes if present.

  • May combine two claims.

    Findingown data

    The experimental group showed a significant improvement in the n-back level achieved in verbal n-back training and in visual n-back training.

    “It also showed a significant improvement in the n-back level achieved in verbal n-back training (Z = 21, p = 3.34e−05) and in visual n-back training (Z = 18.5, p = 6.51e−05), Fig. 3B.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Subjects in both groups showed quicker reaction times in the Visual Simon task at T2 compared to T1.

    “Subjects in both groups showed quicker reaction times in the Visual Simon task (VST) at T2 compared to T1.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In this prospective, randomized, actively controlled, and single-blind study, we provide strong and consistent evidence for the absence of near and far transfer effects following 8-weeks of adaptive WM training in healthy middle-aged adults, despite a pronounced improvement in the n-back training task.

    “In this prospective, randomized, actively controlled, and single-blind study, we provide strong and consistent evidence for the absence of near and far transfer effects following 8-weeks of adaptive WM training in healthy middle-aged adults, despite a pronounced improvement in the n-bask training task.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Only exploratory post hoc tests revealed reduced integrity of the aDMN in the experimental group, but this observation did not survive multiple correction or contrasting against the active control group.

    “Only exploratory post hoc tests revealed reduced integrity of the aDMN in the experimental group. However, this observation did not survive multiple correction or contrasting against the active control group.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The neuroimaging analyses had sufficient sensitivity to detect training-induced changes if present.

  • May combine two claims.

    Findingown data

    The experimental group showed significant practice effects both in verbal n-back training and visual n-back training.

    “The experimental group showed significant practice effects both in verbal n-back training (Z = 7, p = 8.07e−6) and visual n-back training (Z = 13, p = 1.51e−05), Fig. 3A.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Following 8-week n-back training, neither near nor far transfer effects were detected in middle-aged adults, despite the presence of significant practice effects.

    “Following 8-week n-back training, neither near nor far transfer effects were detected in middle-aged adults, despite the presence of significant practice effects.”
    • Also stated in: § Lack of evidence for transfer effects · plain
      “Neither near nor far transfer effects were detected.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Measurement noise and impurity of cognitive tasks may mimic or obscure true underlying changes in cognitive abilities.

    “Specifically, measurement noise and impurity of cognitive tasks may mimic or obscure true underlying changes in cognitive abilities.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [18] Shipstead Z, Redick TS, Engle RW. Is working memory training effective? Psychol Bull. 2012;138:628–54. https://doi.org/10.1037/a0027473. source

  • May combine two claims.

    Interpretationown data

    WM training produces transfer effects neither at the cognitive level nor in terms of neural structure or function.

    “In conclusion, WM training produces transfer effects neither at the cognitive level nor in terms of neural structure or function.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The X-back task demands primarily attentional processing but no working memory.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    The study was prospective, randomized-controlled, and single-blind.

    “In this prospective, randomized-controlled, and single-blind study, we administered an 8-week n-back training to 55 healthy middle-aged (50–64 years) participants.”
    • Also stated in: § Participants · plain
      “They were randomly assigned single-blinded to an experimental or an active control group.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Only two fMRI studies have investigated resting-state functional connectivity changes in relation to WM training.

    “Yet, only two fMRI studies have investigated resting-state functional connectivity changes in relation to WM training [28, 29].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [28] Jolles DD, van Buchem MA, Crone EA, Rombouts SARB. Functional brain connectivity at rest changes after working memory training. Hum Brain Mapp. 2013;34:396–406. https://doi.org/10.1002/hbm.21444. source
    • cites [29] Takeuchi H, Taki Y, Nouchi R, Hashizume H, Sekiguchi A, Kotozaki Y, et al. Effects of working memory training on functional connectivity and cerebral blood flow during rest. Cortex J Devoted Study Nerv Syst Behav. 2013;49:2106–25. https://doi.org/10.1016/j.cortex.2012.09.007. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The presence or absence of an active control group has been proposed as a key reason of the discrepant findings.

    “The presence or absence of an active control group has been proposed as a key reason of the discrepant findings [6].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [6] Teixeira-Santos AC, Moreira CS, Magalhães R, Magalhães C, Pereira DR, Leite J, et al. Reviewing working memory training gains in healthy older adults: a meta-analytic review of transfer for cognitive outcomes. Neurosci Biobehav Rev. 2019;103:163–77. https://doi.org/10.1016/j.neubiorev.2019.05.009. source

  • May combine two claims.

    Premisereasoning

    Neither of these studies used an active control group nor did their analyses cover the whole brain.

    “However, neither of these studies used an active control group nor did their analyses cover the whole brain.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    These two studies reported connectivity changes, both increases and decreases, within and between regions of the frontal parietal and the default mode network (DMN).

    “Using a priori defined ROIs or networks of interest, these two studies reported connectivity changes, both increases and decreases, within and between regions of the frontal parietal and the default mode network (DMN).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [28] Jolles DD, van Buchem MA, Crone EA, Rombouts SARB. Functional brain connectivity at rest changes after working memory training. Hum Brain Mapp. 2013;34:396–406. https://doi.org/10.1002/hbm.21444. source
    • cites [29] Takeuchi H, Taki Y, Nouchi R, Hashizume H, Sekiguchi A, Kotozaki Y, et al. Effects of working memory training on functional connectivity and cerebral blood flow during rest. Cortex J Devoted Study Nerv Syst Behav. 2013;49:2106–25. https://doi.org/10.1016/j.cortex.2012.09.007. source

  • May combine two claims.

    Interpretationreasoning

    These strategies are likely paradigm-specific and, thus, could be applicable to both the experimental and active control group, thereby also explaining the cognitive improvements in the control group.

    “These strategies are likely paradigm-specific and, thus, could be applicable to both the experimental and active control group, thereby also explaining the cognitive improvements in the control group.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The gains are generalizable neither within the same domain nor to other cognitive domains.

    “Moreover, the gains are generalizable neither within the same domain nor to other cognitive domains.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    The study was prospective, randomized, actively controlled, and single-blind.

    “In this prospective, randomized, actively controlled, and single-blind study”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    The study was approved by the Federal Office for Radiation Protection and the local ethics review board.

    “The study was approved by the Federal Office for Radiation Protection and the local ethics review board (project number 399/13).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    ANOVA is known to be robust to not normally distributed data with an equal sample size.

    “ANOVA is known to be robust to not normally distributed data with an equal sample size [63].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [63] Blanca MJ, Alarcón R, Arnau J, Bono R, Bendayan R. Non-normal data: is ANOVA still a valid option? Psicothema. 2017;29:552–7. https://doi.org/10.7334/psicothema2016.383. source

  • May combine two claims.

    Study design and conductcitation

    The study tested and closely supervised a larger-than-average sample of a typically neglected age group using an intensive, adaptive home-based intervention.

    “Third, we tested and closely supervised a larger-than average sample [69] of a typically neglected age group using an intensive, adaptive home-based intervention.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [69] von Bastian CC, Guye S, Simoni CD. How strong is the evidence for the effectiveness of working memory training? Oxford University Press; 2019. https://doi.org/10.1093/oso/9780199974467.003.0004. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    To reduce the likelihood of false positives, the study rigorously corrected for multiple testing.

    “Fifth, to reduce the likelihood of false positives, we rigorously corrected for multiple testing.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The order of tasks trained (visual or verbal) was counterbalanced between participants within each group.

    “The order of tasks trained (visual or verbal) was counterbalanced between participants within each group.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    No correction for age, sex, BMI, or years of education was applied.

    “Thus, no correction for these variables was applied [68].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [68] Spector PE, Brannick MT. Methodological urban legends: the misuse of statistical control variables. Organ Res Methods. 2011;14:287–305. https://doi.org/10.1177/1094428110369842. source

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Participants were randomly assigned single-blinded to an experimental or an active control group.

    “They were randomly assigned single-blinded to an experimental or an active control group.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    There was no significant difference for age, sex, BMI, or years of education between the experimental and the active control groups.

    “There was no significant difference for age (p = 0.92), sex (p = 0.70), BMI (p = 0.19), or years of education (p = 0.38) between the experimental and the active control groups.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    To control for expectancy and non-specific cognitive intervention effects, the study compared n-back training to an active control training with low cognitive demand.

    “To control for expectancy and non-specific cognitive intervention effects, we compared n-back training to an active control training with low cognitive demand.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The X-back task demands primarily attentional processing but no working memory.

  • May combine two claims.

    Premisecitation

    Near and far transfer effects were reported only by Brehmer et al. for the PASAT task and by Simon et al. for the Digit Symbol task.

    “Near and far transfer effects were reported only by Brehmer et al. [77] for the PASAT task requiring sustained attention and by Simon et al. [78] for the Digit Symbol task requiring psychomotor ability, sustained attention, processing speed and, to a lesser degree, WM.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [77] Brehmer Y, Rieckmann A, Bellander M, Westerberg H, Fischer H, Bäckman L. Neural correlates of training-related working-memory gains in old age. NeuroImage. 2011;58:1110–20. https://doi.org/10.1016/j.neuroimage.2011.06.079. source
    • cites [78] Simon SS, Tusch ES, Feng NC, Håkansson K, Mohammed AH, Daffner KR. Is computerized working memory training effective in healthy older adults? Evidence from a multi-site, randomized controlled trial. J Alzheimers Dis. 2018;65:931–49. https://doi.org/10.3233/JAD-180455. source
46 of 46

  • The authors state this as an assumption.

    The auditory network is unaffected by visual and verbal n-back training.

    “The auditory network was chosen as reference network, as it was assumed to be unaffected by visual and verbal n-back training.”

  • The authors state this as an assumption.

    Training-induced neural plasticity constitutes the basis of sustained cognitive improvement.

    “Neuroimaging complimentary to cognitive assessments has the potential to pick up training-induced neural plasticity, which can be assumed to constitute the basis of sustained cognitive improvement.”

  • The paper relies on this without support.

    The X-back task demands primarily attentional processing but no working memory.

    “Hence, the X-back task demanded primarily attentional processing but no working memory.”

  • The paper relies on this without support.

    The five neuroimaging markers assessed provide a comprehensive assessment of neural correlates of WM training and transfer effects.

    “To the best of our knowledge, the present work provides the most comprehensive assessment of neural correlates of WM training and transfer effects reported so far.”

  • The paper relies on this without support.

    The neuroimaging analyses had sufficient sensitivity to detect training-induced changes if present.

    “The group × time interaction was not significant in both analyses.”
5 of 5

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Not confirmedz = 7reported p = .00000807 · recomputed p < .001

    “(Z = 7, p = 8.07e-6”

  • Not confirmedz = 13reported p = .0000151 · recomputed p < .001

    “(Z = 13, p = 1.51e-05”

  • Not confirmedz = 21reported p = .0000334 · recomputed p < .001

    “(Z = 21, p = 3.34e-05”

  • Not confirmedz = 18.5reported p = .0000651 · recomputed p < .001

    “(Z = 18.5, p = 6.51e-05”

  • ConsistentF(4, 50) = 5.3reported p < .013 · recomputed p = .001

    “F(4,50) = 5.3, p < .013”

5 of 5

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • RVIP outcome is described as both a proportion and a count difference

    Minor

    The methods say the RVIP outcome was the proportion of correct target detections, whereas the table defines its accuracy measure as the number of correct hits minus false alarms. Those are different measures.

  • Stroop standard deviation is incompatible with its paired t statistic

    Minor

    Table 2 gives a post-test SD of 2643 for the experimental group's congruent Stroop reaction time. With 28 participants and a mean change of 91.2 ms, that SD and the pre-test SD of 272.5 imply a paired |t| of at most about 0.20, not the 1.80 reported in Table 3.

  • Different Bayes factors reported for the same within-group tests

    Minor

    The prose reports Bayes factors of 18.53 and 1.87 for the experimental and control groups' forward Digit Span changes, respectively. Table 3 reports 20.97 and 1.78 for those comparisons.

3 of 3

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Ripp, I., Emch, M., Wu, Q., Lizarraga, A., Udale, R., von Bastian, C. C., Koch, K., & Yakushev, I.
Affiliations
  • TUM Klinikum (DE)
  • University of Sheffield (GB)
Corresponding author
Yakushev, Igor
ORCID
Venue
Translational Psychiatry
Publisher
Springer Science and Business Media LLC
Published
2022-12-13
Volume
12
Issue
1
Pages
512-512
ISSN
2158-3188 (ISSN-L)
DOI
10.1038/s41398-022-02272-7
PMC
PMC9747798
OpenAlex
W4311316951
Semantic Scholar
a205feb7dfd040f08c439e96a602dcb3c5e33b89
Cited by
31
References
79
Citations by year
2026: 7 · 2025: 11 · 2024: 9 · 2023: 4
FWCI
5.31
Type
journal article
Language
en
Open access
Licence
Locations
  • Translational Psychiatry (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • White Rose Research Online (University of Leeds, The University of Sheffield, University of York) (acceptedVersion)
  • PubMed Central (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • mediaTUM – the media and publications repository of the Technical University Munich (Technical University Munich) (submittedVersion)
Funders
Deutsche Forschungsgemeinschaft (YA 373/3-1, KO 3744/8-1)
Topics
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Cognitive Abilities and Testing
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Cognitive Functions and Memory
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural and Behavioral Psychology Studies
Keywords
working memory training, transfer effects, adaptive training, practice effects, cognitive intervention, white matter integrity, amplitude of low-frequency fluctuation, glucose metabolism
MeSH terms
Cognitive Training, Cognition, Humans, Memory, Short-Term, Middle Aged, Prospective Studies, Single-Blind Method, Neuroimaging

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-08.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-08.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review