Dual N-Back research

  1. Improving fluid intelligence with training on working memory.

    Authors: Jaeggi, Buschkuehl, Jonides, and Perrig

    What they tested
    Four controlled experiments compared adaptive Dual N-Back training for 8 to 19 days with no-training controls. Tests measured Dual N-Back performance and matrix reasoning.
    What they found
    The study reported improved reasoning-test performance, with larger gains after more training sessions.
    Keep in mind
    This historically important positive result used no-training controls rather than a matched activity. Later active-control studies were mixed or null, so it does not establish a settled fluid-intelligence or everyday benefit.
  2. No evidence of intelligence improvement after working memory training: a randomized, placebo-controlled study

    Authors: Redick and colleagues

    What they tested
    Young adults completed 20 sessions of adaptive Dual N-Back, adaptive visual search, or no training. Researchers tested several cognitive abilities.
    What they found
    Participants improved at the practiced tasks. The study found no positive transfer to the cognitive-ability tests.
    Keep in mind
    One trial tests a particular training schedule and population. It cannot settle every possible protocol or outcome.
  3. Improving fluid intelligence with training on working memory: a meta-analysis

    Authors: Au and colleagues

    What they tested
    A meta-analysis examined 20 controlled N-Back studies in healthy adults aged 18 to 50.
    What they found
    The authors reported a small positive pooled effect on fluid-intelligence tests.
    Keep in mind
    This is a result from laboratory tests across selected studies. It does not establish benefits for everyday learning or this particular app.
  4. Reevaluating the effectiveness of n-back training on transfer through the Bayesian lens: Support for the null.

    Authors: Dougherty, Hamovitz, and Tidwell

    What they tested
    A Bayesian reanalysis revisited the Au et al. data and compared the evidence for transfer by the type of control group used.
    What they found
    Studies with passive controls favoured a transfer effect, while studies with active controls modestly favoured the null.
    Keep in mind
    This is a reanalysis of an existing study set, not a new training trial. Its conclusion depends on the controls and studies included in the Au et al. dataset.
  5. Working Memory Training Does Not Improve Performance on Measures of Intelligence or Other Measures of “Far Transfer”: Evidence From a Meta-Analytic Review

    Authors: Melby-Lervåg, Redick, and Hulme

    What they tested
    A review covered 87 publications and 145 experimental comparisons across working-memory training programs.
    What they found
    Short-term gains appeared on working-memory measures. Broader outcomes showed no convincing reliable improvement against treated control groups.
    Keep in mind
    This review includes several kinds of working-memory training, not just Dual N-Back. Its conclusions concern the studies and outcomes it included.
  6. Working memory training revisited: A multi-level meta-analysis of n-back training studies

    Authors: Soveri and colleagues

    What they tested
    A meta-analysis combined 203 effect sizes from 33 randomized controlled trials in healthy adults.
    What they found
    The pooled gain was medium-sized on untrained N-Back tasks. Gains on other working-memory, fluid-intelligence, and cognitive-control tasks were very small.
    Keep in mind
    The review combined different N-Back tasks and outcomes. Improvement on a similar task is not the same as a broad change in ability.
  7. Near transfer to an unrelated N-back task mediates the effect of N-back working memory training on matrix reasoning

    Authors: Pahor, Seitz, and Jaeggi

    What they tested
    Three randomized trials, totaling 460 people, tested how gains on untrained N-Back tasks related to gains on matrix reasoning.
    What they found
    The statistical models linked near transfer with matrix-reasoning transfer, even though two trials had no overall intervention effect.
    Keep in mind
    This suggests a possible pathway. It does not show that everyone who improves at N-Back will improve at reasoning.
  8. Adaptive working memory training does not produce transfer effects in cognition and neuroimaging.

    Authors: Ripp, Emch, Wu, Lizarraga, Udale, von Bastian, Koch, and Yakushev

    What they tested
    A prospective, randomized, single-blind study compared eight weeks of adaptive visual and verbal N-Back with non-adaptive active-control training in healthy adults aged 50 to 64. Researchers used cognitive tests and multimodal imaging.
    What they found
    Adaptive training improved the practised tasks but produced no near- or far-transfer advantage and no training-related changes in the measured imaging outcomes.
    Keep in mind
    This was a modest sample of healthy middle-aged adults using a specific N-Back protocol, not XNBack. It does not settle effects in other age groups or clinical populations.
  9. Boosting Working Memory in ADHD: Adaptive Dual N-Back Training Enhances WAIS-IV Performance, but Yields Mixed Corsi Outcomes.

    Authors: Lintas, Bader, and Villa

    What they tested
    Young adults with and without ADHD, aged 18 to 30, completed at least 18 daily sessions in one month. They trained with fixed dual 1-back or adaptive Dual N-Back and took WAIS-IV and Corsi tests before and after.
    What they found
    Adaptive training produced larger Dual N-Back gains and larger gains on WAIS-IV working-memory measures, especially Digit Span Backward. Corsi gains were small and mixed.
    Keep in mind
    The ADHD sample excluded major psychiatric comorbidities, and the study measured only WAIS-IV working memory and Corsi. It does not show ADHD treatment or broad everyday transfer.
  10. Near and far transfer effects of working memory training: A preregistered, double-blind, randomized-controlled study

    Authors: Zhong and colleagues

    What they tested
    A preregistered trial compared adaptive visuospatial N-Back with active controls across three training schedules, including a one-month follow-up.
    What they found
    Longer training improved trained N-level more, but the authors found no advantage on the untrained cognitive tasks.
    Keep in mind
    This was visuospatial N-Back, not the combined position-and-letter task used here. It cannot directly test XNBack or its app.

Ready for a challenge?

Practice Dual N-Back, track your progress, and compare with friends. A demanding memory game, free to play.