A 2010 study recruited 11,430 adults, viewers of a BBC science program, and asked them to practice reasoning, memory, planning, and attention games for six weeks. They improved on all the games they practiced. On tests they hadn’t practiced, even very similar ones, they didn’t improve [1]. That result sums up much of this post: the brain learns what it practices and transfers little to anything else. Here are nine attempts to improve it, from Mozart to modafinil, along with what’s been measured for each.
Remember that on Pensar es Gratis we only cite reliable data and sources that pass our content filter. At the end of each post there’s a practical guide on the topic if it makes sense for one to exist.
What we’re measuring when we measure mental training
Researchers call it transfer when what’s trained is useful for something else. It’s near transfer if the new task resembles the practiced one, and far transfer if it doesn’t — like practicing a memory game and getting better at reading. Measuring it requires an active control group: people doing another activity just as entertaining for the same amount of time. If the comparison group does nothing, any improvement could be due to the expectation of having trained — the placebo effect.
A 2016 review examined more than 130 papers, including those cited by companies in the industry. It found clear improvement on trained tasks and little evidence of improvement on far-transfer tasks or in daily life [2]. Several meta-analyses (analyses that pool data from many studies) reach the same conclusion: minimal effect on general cognition, and differences between studies are explained by design quality and statistical artifacts [3].
The Mozart effect
In 1993, researchers at the University of California, Irvine published an experiment in Nature with 36 university students. Each one listened to ten minutes of a Mozart sonata, ten minutes of a relaxation tape, and ten minutes of silence, then took spatial reasoning tests. After Mozart, the score corresponded to a spatial IQ of 119; after relaxation, 111; after silence, 110. The advantage lasted between 10 and 15 minutes [4]. The study only measured spatial reasoning, not general intelligence, but the news spread as “Mozart makes kids smarter.”
Perhaps the truth is that before a long task, music you enjoy and that energizes you can help for a few minutes. That’s a conjecture of mine based on [6] and [7]; those studies didn’t test any specific playlist, and replications barely showed improvement.
Does bilingualism improve mental capacity?
The hypothesis was that managing two languages trains executive control: inhibiting distractions, switching tasks, holding information in mind. A 2018 Finnish meta-analysis pooled 152 studies with adults. Before correcting for publication bias, a very small advantage appeared in inhibition, task switching, and working memory; after correcting for it, none remained. In verbal fluency, bilinguals actually did somewhat worse [8]. In 2020, a group from Western University evaluated 11,041 people from several countries on 12 executive tasks. Bilinguals won on one, monolinguals on four, and all the differences disappeared once the groups were matched on factors that could confound the result [9].
My own view is that, if an effect exists, it doesn’t come from stronger executive control but from decades of running two language systems in active use. That would delay the appearance of symptoms even as the underlying disease progresses — what’s called cognitive reserve. That’s a conjecture; the studies above don’t prove it. What isn’t in question is that two languages let you communicate with more people.
Brain-training apps
In January 2016, the US Federal Trade Commission (FTC) closed a case against Lumos Labs, the company behind Lumosity. Its 40 games were advertised as helping people perform better at work and in school and as delaying cognitive decline, including Alzheimer’s. The company paid $2 million and agreed to a $50 million judgment, suspended due to its financial situation [13]. A randomized trial from the University of Pennsylvania assigned 128 young adults to 10 weeks of Lumosity or of common internet video games. They improved on what they practiced. On standard tests they improved just as much as those who played other video games and as another group evaluated with no training at all [14].
The exception with real data is from 2026. The ACTIVE trial randomly assigned US adults aged 65 and over to memory training, reasoning training, or processing-speed training, the last of which involves quickly finding visual information in increasingly complex tasks. Twenty years later, cross-referencing the data with Medicare records (2,021 participants), those who did the speed training plus at least one booster session had a 25% lower risk of a dementia diagnosis (hazard ratio, HR, of 0.75; 95% CI: 0.59–0.95). Those who had no booster sessions showed no difference (HR 1.01) [15]. A trial designed more than twenty years ago is still yielding new data.
This is a recent result, from a single trial and a specific program, and it doesn’t extend to just any app.
Dual n-back
The n-back is a working memory task: you receive a series of stimuli and must say whether the current one matches the one from n steps back. In the dual version, a screen position and an audio letter are presented simultaneously. In 2008, Jaeggi and colleagues (University of Bern and University of Michigan) published that training with it improved fluid intelligence — that is, reasoning without relying on prior learning. They recruited 70 students: the more days they trained, the greater the improvement, and at 8 days it wasn’t significant [16]. The result spread widely.
Redick and colleagues, at Georgia Tech, compared 20 sessions of dual n-back with 20 sessions of visual search and with doing nothing. Participants improved on the two practiced tasks and on no test of intelligence, multitasking, or memory [17]. A meta-analysis of 33 randomized trials found a medium effect on untrained versions of the n-back itself and a very small effect on other working memory, fluid intelligence, and cognitive control tasks. There were no differences by dose or between the single and dual versions [18]. The n-back improves the n-back.
Video games
Here there are three meta-analyses, and they don’t fully agree. The 2018 one showed that habitual action-game players perform better on perception and attention (g = 0.55, where g measures effect size in standard deviations), and that the advantage drops to g = 0.34 when people are randomly assigned to play action games versus other commercial games [19]. Another, from the same journal and year, found no causal improvement in cognitive ability [20]. The most recent, from 2026, pooled 133 studies and 14,245 participants. The relationship between playing and performing better was small but significant across all three designs: r = 0.162 in correlational studies, 0.22 when comparing gamers with non-gamers, and 0.088 in the 22 controlled trials. After correcting for multiple comparisons, neither age, sex, game type, nor duration significantly changed the result [21]. The advantage observed in habitual players is considerably larger than what trials actually produce.
In people over 60, a meta-analysis from Soledad Ballesteros’s group at UNED (Madrid) pooled 20 studies and found improvements in reaction time, attention, memory, and global cognition. Short interventions worked better than long ones [22]. The most striking case is NeuroRacer, a driving game designed at the University of California, San Francisco. Adults aged 60 to 85 who played it in multitasking mode reduced their multitasking cost more than control groups did, to the point of outperforming untrained 20-year-olds, and maintained the gain for six months [23]. This is a result from a single research group.
Transcranial stimulation
Transcranial direct current stimulation (tDCS) applies between 1 and 2 milliamps for 10 to 30 minutes through electrodes on the scalp, intended to modify the activity of the area underneath.
For cognition in healthy adults, the most-cited analysis is Australian. Horvath and colleagues, at the University of Melbourne, pooled single-session studies with adults aged 18 to 50. Of 59 analyses, none showed a significant effect, either in working memory or in language production [25].
For depression there’s more data. In São Paulo, Brunoni and colleagues compared tDCS, escitalopram, and placebo in 245 patients. The depression score dropped 9.0 points with tDCS, 11.3 with the drug, and 5.8 with placebo. tDCS was better than placebo, but didn’t meet the criterion to be considered equivalent to the drug [26]. In December 2025, the FDA approved the first home-use tDCS device for moderate to severe depression, by prescription, and described its benefit as modest. A 2023 trial in The Lancet had found no difference from placebo, and several favorable trials are funded by the company. As of September 16, 2026, it’s available in the US [27].
The alternating-current variant, tACS, has recent results in older adults. In Boston, four days of stimulation in people aged 65 to 88 improved either working memory or long-term memory, depending on the frequency and area stimulated, and the effect persisted a month later [28]. A meta-analysis of 102 studies and 2,893 people concluded that tACS improves cognition [29]. Other researchers point out that it pools together experiments that are too dissimilar.
Modafinil
Modafinil is a medication prescribed for excessive sleepiness that’s also used off-label to boost performance. In 2008, an informal Nature survey of 1,400 readers from 60 countries showed that one in five had used drugs to improve concentration or memory without medical need. Among those users, 62% cited methylphenidate and 44% modafinil, and about half reported side effects such as headache or jitteriness [30].
In people without sleep deprivation, two independent meta-analyses agree. The first, covering 19 single-dose trials, gave g = 0.10, with no differences between cognitive domains or between 100 and 200 mg [31]. The second, covering 14 studies, gave 0.12, with somewhat more effect on memory updating (0.28) [32]. An effect of 0.1 standard deviations is equivalent to moving from the 50th to the 54th percentile. Practically null.
A review from Oxford and Harvard adds some nuance. On basic tests, most studies show better executive function and only half show better attention, learning, and memory, with a few reporting worse divergent creative thinking. On complex tests, the improvement in attention, executive functions, and learning is consistent. In those controlled trials there were no more side effects or mood changes than with placebo [33].
Neurofeedback
Neurofeedback records the brain’s electrical activity with electrodes (EEG) and gives a reward — a sound, or an image that advances — when that activity approaches a target pattern. The idea is that the person learns to produce that pattern. Its most-studied use is childhood ADHD. The most recent meta-analysis, from the European ADHD guidelines group, pooled 38 randomized trials with 2,472 participants aged 5 to 40. In assessments made by raters likely blind to treatment (20 trials, 1,214 participants), the difference from control was practically null (SMD 0.04; 95% CI: -0.10 to 0.18). A small improvement only appeared when the analysis was limited to standard protocols, and in processing speed [34].
In 2021, the first large double-blind trial was published: 144 children aged 7 to 10 at two US centers, designed together with neurofeedback experts. The control group received the signal from a pre-recorded EEG, with the child’s own artifacts superimposed so the session would feel real. Neither at the end nor at 13 months was there a specific effect of neurofeedback [35].
What to do with all this
Train what you want to improve. Near transfer exists and far transfer barely does [3]. If you want to read better, read; if you want to speak English, speak English.
If you’re over 65, the only method with a randomized trial and 20-year follow-up is processing-speed training with booster sessions [15]. There’s an updated commercial version of that exercise, but the trial tested the original program, not that app.
The broadest support for maintaining cognition in older adults comes from combining measures. The FINGER trial, in Finland, assigned 1,260 people aged 60 to 77 at risk of decline to two years of diet, physical exercise, cognitive training, and vascular risk-factor control, or to general health advice. The intervention group improved more in cognition [38].
Learning languages is useful for communicating and, regarding dementia, has evidence in its favor that’s still debated [10][12]. For a long task, music you enjoy can energize you for a few minutes [6][7]. Video games have a small but real effect in trials [21].
In a healthy person, tDCS, modafinil, and neurofeedback currently have no reliable result for improving cognition. tDCS has an approved device in the US, by prescription, for depression [26][27]. Modafinil is a medication with a medical indication. Neurofeedback for ADHD didn’t outperform control in blind assessments or in the best available trial [34][35]. If any of these interest you for a specific problem, that’s a conversation to have with a doctor.
References
[1] Owen AM, Hampshire A, Grahn JA, Stenton R, Dajani S, Burns AS, Howard RJ, Ballard CG. Putting brain training to the test. Nature 2010;465(7299):775-778.
[2] Simons DJ, Boot WR, Charness N, Gathercole SE, Chabris CF, Hambrick DZ, Stine-Morrow EAL. Do “brain-training” programs work? Psychological Science in the Public Interest 2016;17(3):103-186.
[3] Sala G, Gobet F. Cognitive training does not enhance general cognition. Trends in Cognitive Sciences 2019;23(1):9-20.
[4] Rauscher FH, Shaw GL, Ky KN. Music and spatial task performance. Nature 1993;365:611.
[6] Pietschnig J, Voracek M, Formann AK. Mozart effect–Shmozart effect: a meta-analysis. Intelligence 2010;38(3):314-323.
[7] Thompson WF, Schellenberg EG, Husain G. Arousal, mood, and the Mozart effect. Psychological Science 2001;12(3):248-251.
[8] Lehtonen M, Soveri A, Laine A, Järvenpää J, de Bruin A, Antfolk J. Is bilingualism associated with enhanced executive functioning in adults? A meta-analytic review. Psychological Bulletin 2018;144(4):394-425.
[9] Nichols ES, Wild CJ, Stojanoski B, Battista ME, Owen AM. Bilingualism affords no general cognitive advantages: a population study of executive function in 11,000 people. Psychological Science 2020;31(5).
[10] Alladi S, Bak TH, Duggirala V, et al. Bilingualism delays age at onset of dementia. Neurology 2013;81:1938-1944.
[12] Anderson JAE, Hawrylewicz K, Grundy JG. Does bilingualism protect against dementia? A meta-analysis. Psychonomic Bulletin & Review 2020;27:952-965.
[13] Federal Trade Commission. Lumosity to pay $2 million to settle FTC deceptive advertising charges. Press release, January 5, 2016.
[14] Kable JW, Caulfield MK, Falcone M, et al. No effect of commercial cognitive training on brain activity, choice behavior, or cognitive performance. Journal of Neuroscience 2017;37(31):7390-7402.
[15] Coe NB, Miller KEM, Sun C, et al. Impact of cognitive training on claims-based diagnosed dementia over 20 years: evidence from the ACTIVE study. Alzheimer’s & Dementia: Translational Research & Clinical Interventions 2026;12(1).
[16] Jaeggi SM, Buschkuehl M, Jonides J, Perrig WJ. Improving fluid intelligence with training on working memory. PNAS 2008;105(19):6829-6833.
[17] Redick TS, Shipstead Z, Harrison TL, et al. No evidence of intelligence improvement after working memory training. Journal of Experimental Psychology: General 2013;142(2):359-379.
[18] Soveri A, Antfolk J, Karlsson L, Salo B, Laine M. Working memory training revisited: a multi-level meta-analysis of n-back training studies. Psychonomic Bulletin & Review 2017;24(4):1077-1096.
[19] Bediou B, Adams DM, Mayer RE, Tipton E, Green CS, Bavelier D. Meta-analysis of action video game impact on perceptual, attentional, and cognitive skills. Psychological Bulletin 2018;144(1):77-110.
[20] Sala G, Tatlidil KS, Gobet F. Video game training does not enhance cognitive ability. Psychological Bulletin 2018;144(2):111-139.
[21] Zhao R, et al. The association between video game play and cognitive ability: a systematic review and meta-analysis. Acta Psychologica 2026;267:107110.
[22] Toril P, Reales JM, Ballesteros S. Video game training enhances cognition of older adults: a meta-analytic study. Psychology and Aging 2014;29(3):706-716.
[23] Anguera JA, Boccanfuso J, Rintoul JL, et al. Video game training enhances cognitive control in older adults. Nature 2013;501:97-101.
[25] Horvath JC, Forte JD, Carter O. Quantitative review finds no evidence of cognitive effects in healthy populations from single-session tDCS. Brain Stimulation 2015;8(3):535-550.
[26] Brunoni AR, Moffa AH, Sampaio-Junior B, et al. Trial of electrical direct-current therapy versus escitalopram for depression. NEJM 2017;376(26):2523-2533.
[27] Cameron C. U.S. approves first device to treat depression with brain stimulation at home. Scientific American, December 12, 2025.
[28] Grover S, Wen W, Viswanathan V, Gill CT, Reinhart RMG. Long-lasting, dissociable improvements in working memory and long-term memory in older adults with repetitive neuromodulation. Nature Neuroscience 2022;25(9):1237-1246.
[29] Grover S, Fayzullina R, Bullard BM, Levina V, Reinhart RMG. A meta-analysis suggests that tACS improves cognition in healthy, aging, and psychiatric populations. Science Translational Medicine 2023;15(697):eabo2044.
[30] Maher B. Poll results: look who’s doping. Nature 2008;452:674-675.
[31] Kredlow MA, Keshishian AC, Oppenheimer S, Otto MW. The efficacy of modafinil as a cognitive enhancer: a systematic review and meta-analysis. Journal of Clinical Psychopharmacology 2019;39(5):455-461.
[32] Roberts CA, Jones A, Sumnall H, Gage SH, Montgomery C. How effective are pharmaceuticals for cognitive enhancement in healthy adults? European Neuropsychopharmacology 2020;38:40-62.
[33] Battleday RM, Brem A-K. Modafinil for cognitive neuroenhancement in healthy non-sleep-deprived subjects: a systematic review. European Neuropsychopharmacology 2015;25(11):1865-1881.
[34] Westwood SJ, Aggensteiner PM, Kaiser A, et al. Neurofeedback for attention-deficit/hyperactivity disorder: a systematic review and meta-analysis. JAMA Psychiatry 2025;82(2):118-129.
[35] Neurofeedback Collaborative Group. Double-blind placebo-controlled randomized clinical trial of neurofeedback for ADHD with 13-month follow-up. JAACAP 2021;60(7):841-855.
[38] Ngandu T, Lehtisalo J, Solomon A, et al. A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER). The Lancet 2015;385(9984):2255-2263.