The Marshmallow Test: What Does It Really Measure? • Hacker News Board

Re: The Marshmallow Test: What Does It Really Measure?

#51

Post by sonnyblarney » Sun, Jun 03, 2018, 9:35 AM UTC

orbifold wrote: post #22

I've come across variations of this conclusion many times now: That it is really your parents affluence and educational background that predicts your success in school and how much you will earn in the future. This is often portrayed as being unjust. To me it always seemed obvious that children with affluent and well educated parents are more likely to be intelligent, have good self-control and other traits that are…

" simply because those traits are hereditary"

I think by far the most important traits are cultural and behavioural.

'Intelligence' is a small thing.

Grit, self-discipline, conscientiousness, curiosity, disposition, manners, curiosity, ethics, integrity, attitude towards education, planning for the future, views towards family etc. etc. I think are far more important than raw intelligence.

Of course if your parents are CEO's in industry, they're going to expose you to that and the frame of mind necessary, much like so many actors parents are also actors/directors/creatives etc..

Re: The Marshmallow Test: What Does It Really Measure?

#52

Post by Smaug123 » Sun, Jun 03, 2018, 9:37 AM UTC

skookumchuck wrote: post #42

Smaug123 wrote: post #20

I found the following post [1] to make a very compelling argument that the Marshmallow test isn't really testing for the ability to defer gratification, but for a "generalised ability/desire to take and pass tests" (manifesting itself here as the desire to pass the Marshmallow test). The post argues that this is an underrecognised character trait which can predict success in today's Western test-oriented world. Certa…

Do you feel that testing is irrelevant to mastery of a subject? That testing does not test for useful skills?

That's too absolute a statement. If you're really good at a subject but terrible at taking tests, you can still do really well in a test by pure brute ability. But one of the money quotes from the blog post is:

> When my friends said I was a “genius,” I would sort-of-joke that I didn’t know anything but could score 75% on any test with a Scantron.

Test-taking is an ability that you can train (and, like any ability, some people enjoy exercising it and some do not), which in a select few situations can substitute for mastery of many other skills. Those select few situations also happen to be pretty common in the Western world.

Re: The Marshmallow Test: What Does It Really Measure?

#53

Post by popnroll » Sun, Jun 03, 2018, 9:41 AM UTC

yasp wrote: post #38

popnroll wrote: post #28

Earlier quoted context omitted.

It can be argued that gratification delay is more psychology related than genetics. And while it can be found a pattern on genes, probably it will be falsified like happened before with these kind of experiments.

How could you possibly argue that without studying the question?

I never said to ignore the question.

Experiments require resources which are limited. And after the falsification of intelligence genes, it makes sense to find a psychological cause for delay of gratification first, and test other logical explanations second.

Edit: I was wrong. Reading Wikipedia article (https://en.wikipedia.org/wiki/Heritability_of_IQ), it seems intelligence is related to both genes and enviroment.

Re: The Marshmallow Test: What Does It Really Measure?

#54

Post by ACow_Adonis » Sun, Jun 03, 2018, 9:43 AM UTC

thaumasiotes wrote: post #34

Smaug123 wrote: post #20

I found the following post [1] to make a very compelling argument that the Marshmallow test isn't really testing for the ability to defer gratification, but for a "generalised ability/desire to take and pass tests" (manifesting itself here as the desire to pass the Marshmallow test). The post argues that this is an underrecognised character trait which can predict success in today's Western test-oriented world. Certa…

> Certainly I personally possess this trait in spades, and my personal history is full of instances where this trait was extremely useful to me. I love taking tests. Sadly, this has essentially never been any use.

I, on the other hand, am completely devoid of any desire to take or pass tests or perform for grader's/test takers, yet subjectively, I'd rate myself out to the right on the delayed gratification ability.

The first created/creates significant problems in engaging with formal schooling: when I was younger I began simply not turning up to class or participating in assessment. It created the awkward situation where I did do a national standardised test one day and topped the year group for maths, but received a failing grade for maths class. In uni, the pattern essentially repeated itself. I (correctly) percieved the test taking and assessment as pedagogically irrelevant, and deviated from the course curriculums where my interests and own study took me.

Delayed gratification however IS an important skill. One that, almost mathematically by definition, must give an advantage since it has an absolute advantage over the alternative. A person capable of delayed gratification has a greater range of possible options and strategies to achieve their goals, and can always opt for immediate gratification where it is appropriate, whereas a person incapable of delayed gratification can partake in no such choice.

If empirical attempts to measure the effect of delayed gratification show no advantage therefore (assuming successful experiment design), there's actually another troubling possibility: society is (now) basing the distribution rewards and outcomes on those behaviours that are not determined by delayed gratification.

Re: The Marshmallow Test: What Does It Really Measure?

#55

Post by TomMckenny » Sun, Jun 03, 2018, 9:44 AM UTC

orbifold wrote: post #22

I've come across variations of this conclusion many times now: That it is really your parents affluence and educational background that predicts your success in school and how much you will earn in the future. This is often portrayed as being unjust. To me it always seemed obvious that children with affluent and well educated parents are more likely to be intelligent, have good self-control and other traits that are…

The study and article literally say that kids in a given socio-economic group have the same life outcomes whether the grab the marshmallow or not.

And not just here but the notion that hereditary wealth correlates to intelligence has been endlessly refuted.

But if it were true, wouldn't it mean that blue counties, being richer than red counties, have a much higher average IQ?

Re: The Marshmallow Test: What Does It Really Measure?

#56

Post by gaius » Sun, Jun 03, 2018, 9:46 AM UTC

popnroll wrote: post #10

gaius wrote: post #5

Our cats have never in their lives had to worry about food, yet they still become anxious around mealtimes. That suggests that it is trait common to all mammals regardless of their "wealth".

You think the behavior of your cats represent the mammal population?

Yes, it’s called sampling

Re: The Marshmallow Test: What Does It Really Measure?

#57

Post by scns » Sun, Jun 03, 2018, 9:46 AM UTC

skookumchuck wrote: post #42

Smaug123 wrote: post #20

I found the following post [1] to make a very compelling argument that the Marshmallow test isn't really testing for the ability to defer gratification, but for a "generalised ability/desire to take and pass tests" (manifesting itself here as the desire to pass the Marshmallow test). The post argues that this is an underrecognised character trait which can predict success in today's Western test-oriented world. Certa…

Do you feel that testing is irrelevant to mastery of a subject? That testing does not test for useful skills?

You need to selftest your knowledge to master a subject

Re: The Marshmallow Test: What Does It Really Measure?

#58

Post by fvdessen » Sun, Jun 03, 2018, 9:49 AM UTC

TheOtherHobbes wrote: post #27

gcthomas wrote: post #7

I see so many forum discussions where posters criticise the poor for their 'weak' decisions and effectively blame the poor for their own plight. Why don't they just work harder and earn the money to pay for their healthcare and education? they say. It is easy to withhold funding from deprived areas if you convince yourself that it is their own fault. These results should be on the front pages of newspapers and on TV…

Scarcity is an outstanding book about this, and makes the point that severe constraints on anything - money, food, free time - can have psychologically damaging effects. In fact it’s extremely toxic - far more damaging than manynthings that have been banned now.

Is poverty really linked to food scarcity in western countries anymore ? The statistics show that overall, the poor are fatter than the rich.

Re: The Marshmallow Test: What Does It Really Measure?

#59

Post by wruza » Sun, Jun 03, 2018, 9:53 AM UTC

finolex1 wrote: post #16

yasp wrote: post #2

Has anyone studied whether time preference might be genetically heritable? Edit: why am I being downvoted for this?

Because it's closely related to research on the genetic basis of intelligence. Poor research on these topics in the past have been used to justify rather egregious social policies (eugenics, birth control, racial discrimination, etc). Even if it is heritable, making the knowledge public can only lead to a sense of fatalism.

Maybe key thing is to do a good research and to not use it as social policies rather than doing no research at all?

If I knew that my natural behavior is disadvantageous (given that time preference is an advantage), I could behave more consciously and do more analysis than intuition. Rather than just killing everyone suspected to be more succesful than me, if you know what I’m saying.

Re: The Marshmallow Test: What Does It Really Measure?

#60

Post by michaelt » Sun, Jun 03, 2018, 9:56 AM UTC

orbifold wrote: post #22

I've come across variations of this conclusion many times now: That it is really your parents affluence and educational background that predicts your success in school and how much you will earn in the future. This is often portrayed as being unjust. To me it always seemed obvious that children with affluent and well educated parents are more likely to be intelligent, have good self-control and other traits that are…

The problem I have with this is it’s always tempting to believe the aristocratic class deserve their riches when I’m one of them.

Perhaps the idea only “feels right” to me because I like the conclusion it leads to.