Hacker News new | past | comments | ask | show | jobs | submit login
Professors create free research-backed games to train your brain (nyu.edu)
156 points by hhs 7 days ago | hide | past | web | favorite | 57 comments

Game 3 (All you can E.T.) which I just installed is 267.7MB on iOS, for what looks and plays like a flash game. The Android version is 39MB, so they should probably compress their textures or something in the iOS build.

Looks like it was built with minimal playtesting. Took me a couple tries to figure out how to play, for which I had to read the article.

On initial launch the game asks if the user is OK with their data being collected for research, but the 'No thanks' button/label is too tiny and too close to the large 'YES' button. I had meant to choose 'No thanks' but I think I clicked 'YES.

> "On initial launch the game asks if the user is OK with their data being collected for research, but the 'No thanks' button/label is too tiny and too close to the large 'YES' button. I had meant to choose 'No thanks' but I think I clicked 'YES."

Perhaps with more traning, your brain will be better at helping you avoid aggressive 'YES' buttons :)

That sounds plausible, but a little speculative. Perhaps a follow-up study is in order?

...aggressive 'YES' buttons

The opposite of rage quitting.

So far, no solid research has been able to show that brain training exercises change your thinking or behavior in measureable ways once you leave the smartphone screen


Interesting, and I also remember similar results from other studies, where essentially playing the game would make the subjects better at the game with no other provable benefit.

Now, the control for these experiments was amusing:

for one of the sessions:

> which played regular, dumb computer games

What would be a 'dumb' game is fascinating.

You'd have to find something that provokes no brain stimulation at all, no skill evolution while still being a game. Then if the 'game' was really completely void of content but the players kept playing it, would they end up in a kind o meditative state, potentially causing their abilities to increase ?

Another control group were doing crosswords, I guess to get a "well known quantity", but aren't crosswords also pretty demanding games depending on the people doing it? I'd be worse at crossword than building raiding strategies in an RPG for instance.

Somewhere I have the feeling these studies are deeply flawed with a lot of unchecked assumptions. Not by laziness per se, but it just seems that cognitive fields are inherently hard for rigorous studies.

Right, but then if 'Brain training' games are shown to be no different from other games that dont have that label then the label is still revealed to be bullshit.

See also: https://americanaffairsjournal.org/2019/08/rotten-stem-how-t...

"Finally, the big hope had been to motivate students to learn through interactivity and “gamification.” What any given software vendor might tout as a stimulating lesson, however, is often just a frustrating distraction for students. The case of edustar basketball shows that even highly engaged students may not be learning any­thing at all: in a randomized controlled trial, more than five hundred students were assigned either to an ordinary module teaching how to divide fractions or to a module teaching the same topic using a basketball game–like computer program. Students assigned to the basketball game spent about five minutes longer on the lesson, but performed significantly worse than students following the regular method. Ultimately, tech that boosted student engagement just took a longer time to teach them less."

Simply put, I don't think the engagement is there with the entire genre of "brain training" games. They don't feel like "play" in the same way that one could easily slip into Stardew Valley and lose hours at a sitting. Forcing oneself to interact for 30 minutes a day is not feasible. Even if the benefits were are as palpable as actual exercise.

Contrast with the "Math AR" tool GeoGebra. The interface is not perfect yet. But it does give a glimpse at what a truly interactive geometry lesson would look like. As well as a medium that invites life-long learning ;)


I'm not sure about learning games, but video games in general have definitely improved my skills in life. Especially when it comes to problem solving and persistence. I tend to work at difficult problems the same way I work on tough levels in games.

I do think many games today don't teach this though. You need to be able to lose. Old games made you lose constantly, you had to learn the patterns and build up your reflexes and coordination. Much of life is actually like this. It's a valuable skill to be able to keep going at something until you get it. I've seen a lot of people just give up on things at the first hint of difficulty, video games taught me not to do that and because of this, i'll push and keep learning skills even when it gets frustrating and not so fun. There's a lot of things I would've given up on if I hadn't learned that the rewards for pushing through are usually worth it.

>Forcing oneself to interact for 30 minutes a day is not feasible. Even if the benefits were are as palpable as actual exercise.

This makes no sense. People exercise for 30 minutes a day, or read, or clean, or play video games, or watch TV, or tend to a garden, or do yoga.

Why are all these feasible but exercising your brain is not?

> Forcing oneself to interact for 30 minutes a day is not feasible.

There is this thing called self-discipline which many of us exercise. Last January I bought a guitar as a complete novice, and 30 minutes a day later I've written over a hundred songs and some of them are quite good.

Sometimes I didn't have 30 minutes a day. That meant an hour the next day, or a few hours over the weekend. It's all about what your goals are, not feasibility.

I can lose hours at very simple games as well, though. I put more hours into .io games than I'd like to admit.

Maybe these games are so far mediocre because there haven't been many made yet, and most creative endeavors are mediocre. It doesn't mean that it's not possible to make an awesome memory-training game.

There have been a few interesting arguments about dual-n-back games. They have been shown with very solid evidence to improve cognitive performance. They are interesting in the beginning but get frustrating/difficult very fast, essentially you always play at a frustrating level where you final frequently. It is somewhat exciting but certainly not addicting/motivating in the sense of Civ or other well-developed 'addicting' games.

One of the theories on why they are effective (doubting the main interpretation that they train your working memory/cognitive processing) is that the game trains you to endure frustration and at some point also boredom. Dealing with frustration and boredom are actually important success skills - so if there was a way to get the game more engaging there's a good chance that the positive impact would be reduced.

tl;dr: issue is still under debate, but it might well be that the need to endure boredom/frustration in at least the dual-n-back games might be a feature (that improves impact) rather than a bug.

They have wrong control.

The control they should have is a group that lift weights [1], walks [3], swims or runs [2] for two hours.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3448565/ [2] https://www.ncbi.nlm.nih.gov/pubmed/14628975 [3] https://www.nemours.org/content/dam/nemours/www/filebox/serv...

Also, I think they did not test transience of effect observed - how long will it last?

Interestingly, people have run wm training studies with these control groups!

(Historical caveats about trusting the results of a single training study apply)


Agreed. In reality our time is zero sum, and if you could be doing something better with your time you should. Exercise has a whole host of mental benefits.

> “We found replicated evidence across multiple experiments that playing our games for two hours causes improvements in executive function skills as compared to a control group that plays an unrelated game,” said Mayer.

Christ what a low bar. That's like having your treatment group do jazzercise while the control watches TV, and then concluding that jazzercise is the most effective form of physical activity.

Your control group should be performing some sort of non-gaming intellectual exercise: reading a passage and answering some questions, having a group discussion on a serious topic, etc.

I decided to expand upon this thought in a short blog post [0] because I think it's a good general lesson.

[0] https://davefernig.com/2020/01/17/experimental-controls-as-o...

Low bars get you positive results. Positive results get you published.

The joke when I was in academia was that there were two ways to get lots of citations: publish a state-of-the-art model that gets cited because it's state-of-the-art, or publish a bad model that gets cited a lot because everyone wants to show how much they beat it by.

Also worth questioning whether the test was sufficiently far removed from the game itself that improved scores mean anything. Unfortunately the article is paywalled, but here's a quote from the abstract:

> Students who played Alien Game at a high level of challenge (i.e., reaching a high level in the game) developed significantly better performance on cognitive shifting tests compared to controls when they played for 2 h (Experiment 1, d = 1.44), but not when they played for 1 h (Experiment 2).

Probably it was a card sorting test. This sounds to me like "people who practiced card sorting got better at card sorting."

I have access to the article [1] through my university, here is the description of the procedure:

The study took place in the students' classroom, once a week over an 8 week period. [...] On the first week, demographic questionnaires and the computerized DCCS and Flanker tasks were administered. The Alien Game was then introduced to the students [..] They then played through the first, tutorial level to learn the basics of the game. For the next 6 weeks, students were asked to sign in and play for 20 min per week. Each week, the next two levels of the game were unlocked. [...] On the final week, the DCCS and Flanker tasks were administered again, and students were given a posttest questionnaire asking about their experiences with the game.

So if I understand correctly, that was:

- 1st week: just the questionnaire and test, plus demo

- weeks 2 to 7: play the game 20 minutes

- week 8: test again

So it seems the performance improvement was measured one week after the last game session.

[1] https://www.sciencedirect.com/science/article/pii/S036013151...

> The first game, Gwakkamolé, was designed to train inhibitory control, a subskill of executive functions. Inhibitory control is the ability to control one’s attention, behavior, thoughts, and/or emotions.

A really simple way test/train your inhibitory control is play this "game" with partner. Have them hold up either 1 or 2 fingers. When they hold up 1 finger, you have to say "2" and when they hold up 2 fingers, you have to say "1". Try to do this as quickly and accurately as possible.

Their study of adolescents[1] did not use a control group. This is an issue, as adolescent executive performance is likely naturally improving over time, and said improvement may be non-linear.


I guess this is as research-backed as movies are based on a true story.

Or the same way cosmetics are science backed.

Backed by a single simple study with too small a sample size and little or no control group, paid for by the group currently making the claim (a group who may have commissioned several other studies but didn't have them published because the results were not as complimentary to their cause).

Could well be. The entry for CrushStations on the Play store says "Our research suggests that CrushStations is an effective way to train working memory. The study supporting this claim will be published soon."

This isn't new. Games are proven to help a lot of things in the brain. For example reaction time: https://www.futurity.org/video-games-speed-up-reaction-time/

Maybe it's just me, but this gives me an ad feeling.

The article sounds like an ad, and 2 out of 3 games listed doesn't sound like anything new.

That being said, I think what is being said is that these 3 mechanics in the games are particulary suited to improve your brain activities.

The 3rd game is reimplementation of one of the games that is proved to improve your cognitive abilities - and I'm sorry, I cannot remember the name - which was about forcing your brain to work on disonnanced tasks.

What he actually said is that for these games the effect on cognitive skills was measured, which has not been done for other games. It may quite well be the case that many other games have the same or even a stronger effect, but it hasn't been measured yet.

> I cannot remember the name

That's all the endorsement we need.

:) I would like to clarify that I didn't pursue using it for learning, so that's why. I believe the app was probably worth time investment if that's what you are looking for.

This is fascinating to me because I get the sense that this is the beginning of a movement that will fully legitimize video games.

I've played lots of games that require quick decision making skills and working memory, and perhaps it will become increasingly clear that when I played them, I wasn't just wasting my time.

As younger generations gain power, traditional wisdom will change and maybe people will start to see video games as productive. Really, the only downside I see with them is that they can displace exercise and face-to-face communication -- I know games can be social, but you really just need time talking to other people with no distractions. But if you balance your life appropriately, why not?

I am from a younger (ish) generation, and I think video games have been really destructive in some ways. They offer an incredibly seductive and easily accessible way to waste enormous amounts of time, and the worst aspect is that some of the brightest minds I know are the most affected. They, unlike most traditional hobbies, fill a human desire to exert control over their sphere of influence in a totally artificial way. In that way, I definitely think they've siphoned off some productive time from our most talented and otherwise ambitious.

I quit playing videogames when I was about 16 or 17 and I still look back on that as one of the best decisions I've ever made.

I'm a millennial, I've grown up with video games every bit as much as you have.

Are games not already legitimized? I feel like with the rise of casual mobile games, almost everyone plays them now.

> Really, the only downside I see with them is that they can displace exercise and face-to-face communication -- I know games can be social, but you really just need time talking to other people with no distractions. But if you balance your life appropriately, why not?

My issue with video games is that playing them too much, especially when young, increases your dopamine tolerance too much. Many of them are designed to be addicting.

I have mostly stopped playing video games and feel mental well being has significantly improved.

Of course, that's not to say I think they should be banned or anything but in my humble opinion you are ignoring the significant downsides of playing video games. That being said, I have many fond memories and friendships created through video games too! Like most things there's good and bad.

It sounds like they've literally just reimplemented Whack-a-mole, Simon Says (or something similar), and... OK, the third one sounds newer, although it reminds me of the pizza trolls in that Zoombinis game.

I have a hard time believing an app is more beneficial to cognitive health or motor skills, especially for kids, than actively exploring the world outside.

Where does it say these improve motor skills? I saw no such claim being made.

Also I don’t really get your point anyways. I’d think expanding one’s mind would be accomplished by routines with good variety, and see no reason why there needs to be a choice between play outside and play on a tablet. Especially when a child is too young to really do much without supervision...

Cool! I always wished somebody would create something like BrainWorkshop but not boring. In fact I myself implemented the BrainWorkshop-like 2n-back game (and it helped me a lot) immediately as the paper behind it got published, when no publicly available implementations existed but I'm not good at graphics and game design so a non-boring version was not something I could make.

It is still controversial whether n-back games improve working memory outside of the game context.

From https://en.wikipedia.org/wiki/N-back:

> Two studies published in 2012 failed to reproduce the effect of dual n-back training on fluid intelligence. These studies found that the effects of training did not transfer to any other cognitive ability tests.[12][13] In 2014, a meta-analysis of twenty studies showed that n-back training has small but significant effect on Gf and improve it on average for an equivalent of 3-4 points of IQ.[14] In January 2015, this meta-analysis was the subject of a critical review due to small-study effects.[15] The question of whether n-back training produces real-world improvements to working memory remains controversial.


> I meta-analyze the >19 studies up to 2016 which measure IQ after an n-back intervention, finding (over all studies) a net gain (medium-sized) on the post-training IQ tests. The size of this increase on IQ test score correlates highly with the methodological concern of whether a study used active or passive control groups. This indicates that the medium effect size is due to methodological problems and that n-back training does not increase subjects’ underlying fluid intelligence but the gains are due to the motivational effect of passive control groups (who did not train on anything) not trying as hard as the n-back-trained experimental groups on the post-tests. The remaining studies using active control groups find a small positive effect (but this may be due to matrix-test-specific training, undetected publication bias, smaller motivational effects, etc.) I also investigate several other n-back claims, criticisms, and indicators of bias, finding: payment reducing performance claim: possible dose-response relationship of n-back training time & IQ gains claim: not found kind of n-back matters: not found publication bias criticism: not found speeding of IQ tests criticism: not found

That's curious to know, so I thank you for the reference, yet I don't care much as it made and makes a huge difference for me personally. Perhaps that's because it may be really effective for just a portion of people with specific conditions in specific circumstances (e.g. I have ADHD and I also take nootropics which are meant to increase neuroplasiticy, neurogenesis and cerebral blood flow so the result is synergetic, nootropics alone don't give a boost so profound) or maybe what gets improved is not the working memory but something else playing important role in cognition. By the way it seemingly has also improved my emotional intelligence, not just analytical one - I didn't expect that.

Fair point, but may I ask if you feel there is an improvement intuitively, or if you also measure it in some way (other than the game score itself).

I don't really measure but I also wouldn't label it "an intuitive feeling" - it's an observation (not just a feeling like when you're stoned and feel genius), although not scientifically credible. I have a well-developed self-observation skill and can always tell when my cognition is less (for an extreme example - avoid sleeping for some 36 hours, drink some booze and try to be smart - you'll notice it's hard and you don't really do well, no need to measure digitally) or more efficient. I believe I could measure if I knew a good way [to measure fluid intelligence] and was interested enough but that's not the case.

So, I recommend everybody to try n-back exercises for some time [and see if it helps them] and I usually mention that's kind of proven to help scientifically (at least in some papers) and empirically but avoid saying its efficiency is anything close to an unquestionable fact. As a result people (including very smart and educated already) immediately report feelings like stains have been removed from their brains and their mental gears got greased.

> I usually mention that's kind of proven to help scientifically (at least in some papers) and empirically but avoid saying its efficiency is a strict fact.

All good but I must disagree with “kind of proven”. Things can be either proven or not proven, and this is not proven.

> Things can be either proven or not proven, and this is not proven.

On the contrary, nothing in science is ever proven, only disproven. We can have less or more confidence in a particular model; and whether you should use a particular model depends on the cost/benefit analysis.

In this case, there is some evidence that it might be helpful, although somewhat disputed; but playing the game certainly doesn't hurt, and doesn't cost much, so it might make sense to give it a try.

> On the contrary, nothing in science is ever proven, only disproven.

Sure, but science is a specific case and I was talking about the all-or-nothing nature of the meaning of 'proven' itself.

I've read the original paper by Jaeggi et. al., one paper replicating it successfully and one paper proving the results transfer to different tasks. At this point wasn't it logically correct to say that's scientifically proven? Now we know there are papers which prove those proofs questionable. That's what I mean.

If only we had a bullet-proof test to measure fluid intelligence (classic IQ tests have quite a well-known number of problems, solving math problems and puzzles is known to be much more of a skill, working memory is working memory - not intelligence itself) objectively...

> At this point wasn't it logically correct to say that's scientifically proven?

Ah, I see. Yes, at that point it would have been correct to claim that it was proven.

Any game where history matters (past information is relevant but removed from view, so not chess or Go, but yes Blackjack or Go Fish) is like an n-back game.

A number of studies have demonstrated cognitive improvement from playing video games: https://www.psychologytoday.com/us/blog/freedom-learn/201502...

> Improved visual contrast sensitivity. Successful treatment of amblyopia. Improved spatial attention. Improved ability to track moving objects in a field of distractors. Reduced impulsiveness. Overcoming dyslexia. Improved ability to engage in multiple tasks simultaneously. Increased mental flexibility. Reduction of mental decline that accompanies aging.

Mainstream culture doesn't really want to accept this, for...a variety of reasons. And I still limit my son's screen time, even with games, because too much turns him into a dickhead and other activities are also important. But I don't doubt that for challenging games, there are cognitive benefits.

There doesn't appear to be a link to the "eight research articles reporting on the effectiveness of these games" and google scholar doesn't show any relevant publications by these authors - has anyone come the actual data?

Are those games FOSS?

so play more memory games? gotcha

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact