Vision training

Don't use a training app or game in place of reading glasses; if you enjoy one, treat it as a game. Eye exercises have a recognised use only for convergence insufficiency, a specific difficulty turning the eyes inwards for close work, and even there, in adults, trials have not shown symptoms improving more than with a placebo.

Apps that claim to replace reading glasses: inventor-run and uncontrolled

Perceptual-learning apps such as GlassesOff (developed by Ucansi) target presbyopia, the age-related loss of close-up focus. They have people spot faint striped patches (Gabor patches) among distractors, on the theory that the brain can learn to read a blurred image better. The eye's optics do not change.

• The founding study. Thirty people with fading close-up focus, average age 51, did about 37 half-hour sessions over three months. Their near sharpness improved by roughly two chart lines and reading speed at the smallest print by about 17 words a minute, with no change in focusing or pupil size. The "control group" was three untrained people, not randomised, and the lead author was paid by and owned stock in Ucansi.^[1]
• The larger cohort. A follow-up of 119 people aged 40–63 found a similar two-line gain, larger in people whose near vision was worse to start with. It mixed people with normal near vision and people who had lost some, its 30 Berkeley participants very likely overlap the founding study, and three authors were company employees.^[2]
• An earlier commercial programme (NeuroVision) reported about two lines of gain in 18 people in the early stage of focus loss, against roughly three untrained controls, not randomised.^[3]
• Independent work is thinner still. In a non-inventor study with eight people per group, both training groups improved their own near-chart scores, but the paper reports no acuity comparison against untrained controls and found no contrast gain over them; switching focus between near and far screens was no better than training at one distance.^[4] A randomised pilot of 12 people starting multifocal contact lenses found better middle-distance sharpness after 20 days of training, with no difference in contrast.^[5] Six weeks of virtual-reality binocular training left near sharpness unchanged in 20 people with age-related loss of near focus.^[6]

No trial has compared stand-alone training of this kind with a sham version, and none tracks whether gains last. An independent German review concluded the approach "should be judged with caution".^[7]

Training in healthy older adults: the trained task improves

Five days of contrast and orientation training brought older adults averaging 71 (13 analysed) up to the level of untrained young adults on the trained lab task. On clinical charts, contrast sensitivity did not improve significantly, near sharpness rose by two or three letters, and there was no untrained control group.^[8] A 2025 study reported lasting contrast gains in 29 older adults, but its only comparison was trained young adults, and repeated testing alone also improved scores.^[9] No study has shown such training changing falls, crashes or daily function.

Action video games

Fifty hours of action games improved contrast sensitivity more than a slower game in one study of young non-gamers, but that result rests on six action players against seven controls.^[10] Meta-analyses disagree on whether action games improve thinking at all: one from the leading research group found a small-to-moderate gain, mainly in attention, with only "encouraging signs" for perception and published effects about 30% larger than the full literature;^[11] a rival analysis found "small or null" effects and no evidence of a causal benefit.^[12] In 58 older adults, an action game did no better than Tetris on a visual-attention test, and direct speed training beat both.^[13] A 2026 meta-analysis reporting "large" gains in visual function leaned on before-and-after studies without controls, which inflate effects.^[14] No trial has measured vision after action-game training in 40–65-year-olds.

Speed-of-processing training and driving

The Useful Field of View test measures how quickly you can take in several things at once, including at the edge of vision. Drivers with large impairments on it had 2.2 times the crash risk over three years in a cohort of 294 adults aged 55–87.^[15] Training on the same kind of task is the one visual-cognitive training with a real-world outcome, and that outcome is fragile:

• Crashes. In the ACTIVE trial, a large US trial of brain training in older adults, speed training was followed by about 43% fewer at-fault crashes per mile over six years in drivers averaging 73. But reasoning training did as well, which argues against anything specific to vision, and the lead author holds stock in the company that sells the training.^[16]
• Giving up driving and falls. Across the whole trial, speed training did not delay giving up driving over ten years^[17] or reduce falls; a benefit appeared only in a high-risk subgroup, at the edge of statistical significance.^[18]
• Midlife. In a trial of 681 primary-care patients, people aged 50–64 improved on the test as much as over-65s, but no crash or fall outcomes were measured, and the software maker supplied the training free.^[19] In a non-randomised study of 53 older adults, home online training improved the test score while hazard perception actually got slower.^[20]

Sports-vision products

Strobe glasses, tracking games and perceptual-learning games show gains on sport or lab tasks in athletes, which mostly vanish when the test differs from the training. Across 33 randomised trials, visual-attention gains were large when the test resembled the training and close to zero when it did not.^[21]

• Strobe eyewear. Nike's Vapor Strobe improved two of several lab measures against clear glasses in a study funded by Nike, with no gain in peripheral vision or object tracking.^[22]
• NeuroTracker-type tracking. In a randomised trial of 62 young footballers, training improved the trained task with no transfer to the pitch; the null came from the inventor's own group.^[23]
• ULTIMEYES. A baseball study reported sharper vision and fewer strike-outs in 19 trained players against untrained pitchers, without randomisation.^[24] The US Federal Trade Commission later charged the app's company, co-owned by the senior author, over unsupported "improve vision" claims; it settled for US$150,000.^[25]

None of these has been shown to improve everyday eyesight in ordinary midlife adults. For general brain-training apps, see Brain-training apps.

Adult amblyopia: proof the adult cortex can change

Amblyopia ("lazy eye") was long thought untreatable after childhood. Across 24 mostly uncontrolled studies of adult training, sharpness improved by about 1.7 chart lines on average, and age did not predict the result.^[26] A meta-analysis of 22 randomised trials in 422 adults found a moderate gain over controls.^[27] But in the best-controlled test, a double-masked trial of 115 people aged 7 and over (average 21), a binocular training game was no better than a placebo game, though fewer people in the active arm played as much as prescribed.^[28] Cochrane, the independent evidence-review network, has so far published only a protocol for an adult review.^[29]

Convergence insufficiency: the recognised exception

Convergence insufficiency is difficulty turning the eyes inwards for close work, causing strain and double vision when reading. In a placebo-controlled trial of 46 adults aged 19–30, office-based vision therapy improved the clinical measures, but symptoms improved in every group, placebo included.^[30] The American Academy of Ophthalmology concluded that, in young adults, office therapy has not been shown to relieve symptoms better than placebo, and evidence for home exercises such as pencil push-ups is inconclusive.^[31] A trial in people with age-related focus loss is planned but has not reported.^[32]

For why focusing exercises cannot restore close-up vision, see Vision decline with age.

Further reading

• Polat U et al. Training the brain to overcome the effect of aging on the human eye. Scientific Reports 2012.^[33]
• Heinrich SP. [Improvement of vision through perceptual learning in the case of refractive errors and presbyopia : A critical valuation]. Ophthalmologe 2017.^[34]
• Ball K et al. Cognitive training decreases motor vehicle collision involvement of older drivers. Journal of the American Geriatrics Society 2010.^[35]
• Guo Y et al. Does the "learning effect" caused by digital devices exaggerate sports visual training outcomes? A systematic review and meta-analysis. Frontiers in Physiology 2025.^[36]
• Gao TY et al. Effectiveness of a Binocular Video Game vs Placebo Video Game for Improving Visual Functions in Older Children, Teenagers, and Adults With Amblyopia: A Randomized Clinical Trial. JAMA Ophthalmology 2018.^[37]
• Chang MY et al. Home- and Office-Based Vergence and Accommodative Therapies for Treatment of Convergence Insufficiency in Children and Young Adults: A Report by the American Academy of Ophthalmology. Ophthalmology 2021.^[38]