When Does Intelligence Peak? | Hacker News Reader

When Does Intelligence Peak?

blogs.scientificamerican.com

103 pointsheadalgorithm49 comments

This[1] is the paper they quote. The paper is pretty confused, mostly by their own fault. For example:

Recent evidence shows that whereas short-term memory for names and inverted faces peaks around the age of 22 years, neither short-term memory for faces nor quantity discrimination peaks until around the age of 30, a fact difficult to assimilate into the fluid-/crystalized-intelligence dichotomy (Germine, Duchaine, & Nakayama, 2011; Halberda, Ly, Wilmer, Naiman, & Germine, 2012). Whether face memory and quantity discrimination are exceptions to the fluid/crystalized rule or represent more systematic and previously unrecognized patterns of age-related difference is an open question

First, there's no "fluid-/crystalized-intelligence dichotomy", since the two factors are highly correlated with each other, to the point where psychometricians prefer to derive the higher-order order factor, g.

Second, clearly the authors will be very confused by results like this if they don't actually calculate g-loadings, and at the very least correlate the vector of g-loadings with the vector of age declines.

Now compare the above paper with a paper studying basically same question, but using proper factor analysis[2]. In particular, if you're low on time or background, look at the graphs in Fig 2. Clearly, being able to separate g-loading from group factors and test-specific measurement error gives us much clearer picture of age-related decline.

[1] - https://sci-hub.tw/10.1177/0956797614567339 [2] - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3637652/

FWIW I turned 55 this week and feel smarter and more capable than any time before in my life.

I feel I write better code, and faster, than before. I understand articles and books better.

I can't ski, bike, run, or backpack like I used to in my 20s. This is not a huge surprise.

Similarly, I turned 55 last month, but I have noticed a definite decline in some aspects of my intelligence with regard to software development.

I work in a company that's committed to pair programming: two of us working together on one computer to develop. In some ways it's like a footrace, me against a twenty-five-year-old, except it's software development, not a footrace.

I marvel at my twenty-something pair's ability to hold a complex model in their head. I can do it, but it takes me longer, and usually I need to scratch notes on a piece of paper to help.

Their memory has better retention than mine. "Remember we ran into that problem in November?" I'll have absolutely no recollection.

And sometimes it takes me longer to figure something out. I remember last week waking up and realizing the reason why we hadn't been able to recreate a certain failure was that we were using a layer 4 load balancer instead of a layer 7 with a redirect — something I was confident I would have figured out sooner ten years ago.

I also notice that, in respect to programming languages that I've picked up in the last ten years, I tend to rely on stackoverflow for syntax, even simple things like, "How do a write a for loop in {Ruby,Python,Golang,JS}?" whereas I know the syntax cold for my earlier languages (Perl, C++, C, Pascal).

On the plus side I have 35 years of experience to draw upon, both from a development side as well as from a managing-expectations side. When I'm on a project with little oversight the project managers don't worry about it going off the rails. Still, I envy the elasticity of the minds of my younger fellow developers.

I can't say about others but since I know a bit about programming I think programming is more about experience than intelligence or iq. I think it's like any other skill that we hone with time and at least in my case we subconsciously pick up things while debugging and looking at other people's code, snippets and examples that makes us better as we get older.

FWIW, I am 35 and I feel like my memory has declined from when I was 10 - 20. I used to be able to remember a lot of content for exams, song lyrics, and general knowledge.

Nowadays, I sometimes have this occur to me: I am thinking about something that I know that I know, but I cannot recall it.

How much sleep are you getting? I found that was the case for me, but once I started sleeping and exercising more regularly, I feel like that dissipated.

I wonder though how much of that has to do with the wealth and ease of information that is available now vs. 30 years ago? Not that I am saying that you feel smart because you have read more. But the fact that there is so much information that would not have been as easily attainable 20 or 30 years ago as there is now. At the very least the fact that I can read and comment here definitely impacts my 'intelligence'. Plus all the information that I am exposed to and so easily with no effort other than reading. It's simply not the same as back in the library or 'read books' days. Here I can be doing one thing and then easily learn (to take a break) about something entirely different. With almost zero overhead to do so. Once again I am talking about the potential mental sharpness that produces independent of the actual information that may or may not be retained.

Lucky you!

I feel like my coding peak was at 35 and is slowly declining ever since.

I'm slowly getting better in dealing with people though.

I don't believe that "coding" is a single axis of quality, either. I'm now better at some types, and worse at others.

When I was young, I was much better at reading or writing complex functions that worked correctly. As I get older, I'm much less able to, but I also wouldn't want to even if I could.

I don't think that's just sour grapes. I don't want to inflict complex code on my coworkers, either. And there's many times (when I'm tired, when I'm fixing an emergency bug, etc) that I'm not at 100% mental capacity and I value not having to work right at my brain's complexity limit.

16 year old me would probably not have believed that.

At 50, I have noticeable fast and slow periods. But when everything is firing correctly the correlations and ability to quickly see the forest for the trees is better than at any other time in my life. People are also more interesting and I tend to enjoy each nice day a little more than the last. Most importantly for professional work I now know when to give up and move on or dig in and finish.

It is orders of magnitude harder to write a “simple” solution. Developers should not be proud of complexity at any age.

Are you suggesting one should be proud of something just because it's harder?

No, one should be proud of creating an elegant solution.

At 42 I feel like my coding gets better as my context increases. Getting worse with people though.

Man, that gives me a lot of hope :)

I feel the same way at the same age. I've been feeling that I've been at my intellectual peak for many years now, and keeping my fingers crossed for the years to come.

I think it was Michael Jordan who said he was at his physical best at 25 but 35 was when he was playing his best basketball. Though he declined physically, he made up for it with his knowledge of the game and experience.

Though at 55, you are not at your intellectual peak of the your 20s, you have gained knowledge and experience that can more than compensate for loss of brain power.

I like to think of it in terms of hardware and software. While the hardware may stagnate and decline, you can always update the software to get improved performance.

> Though at 55, you are not at your intellectual peak of the your 20s...

I am not convinced I was at a peak 30 years ago.

From a narrow physical aspect, pretty much everyone peaks sometime in their 20s. Your brain is a physical organ and it declines along with the rest of your body. Like your eyesight, endurance, strength, etc. It's the inevitable biological aspect of aging.

But from a holistic wider view, you don't necessarily peak in your 20s as your gains in knowledge, experience and wisdom compensates for the decline in pure physical performance.

Basically, you know lots of shortcuts.

One thing should be clarified... Everyone is not at their physical best around 25. They may, on average, have the -potential- to be at their physical best around age 25, but excercise and training has a lot to do with it.

35 year old me could beat 25 year old me in almost any physical contest. Because I am in much, much better shape now. I do regular cardio and I train and lift weights. I was chubby and out of shape at 25.

I suspect the same could easily be true for mental prowess.

That's true. However, if we look at one's peak, your 25 yr old self had a higher peak than your 35 yr self. If you had exercised as a 25 yr old self and reached your potential, you would be physically better than your 35 yr old self.

It's like driving 25 mph in a porsche vs 50 mph in a SUV. In this instance, the SUV is faster because you chose to go 25 mph in a porsche. You didn't fully explore the full potential of the porsch.

Your physical peak could be at 25, but you also could choose not to exploit it fully.

That's the "natural potential plateau" question - if we are to believe this theory, then it means that you can reach your absolute maximum potential in 10-15 years.

I am not so sure - I wonder if there are ways, in some fields, to keep getting better well into your 50s. For example, I don't really know of the ability to gain strength decreasing. So it seems like you could lift well into your 50s and end up stronger than a peak 25 year old.

I.e. an elite 25 year old wouldn't drop off in strength.

...I just answered my own question: https://torokhtiy.com/blogs/warm-body-cold-mind/weightliftin...

Seems like 35 is where Olympic weight lifters stop winning medals. There is also this: "Harvard University says the average man loses up to 5 percent of muscle mass every decade, starting at age 30; most men lose about 30 percent of muscle mass as they age."

Well, I guess I am going to have to stick with the guy who posted "older people just stop giving a " to comfort myself about getting old.

> processing speed and short-term memory for family pictures and stories peak and begin to decline around high school graduation

I am skeptical that "short term memory for family pictures" is a discrete cogntive function.

I'm even more skeptical that you can extrapolate anything generally useful about cognitive decline, since this result is exactly what I'd expect if there were no cognitive decline -- people are better at remembering things they can contextualize to the familiar, so of course they'll be able to remember new information about their families better when they still live with them than after they move out.

agreed. I'd even be inclined to take it further and suggest that once leaving high school, pressures and stresses related to self sustainability (being employable, financial concerns, housing, etc etc) would become more real than any previous family memory.

Also, wouldnt an understanding that materializes as one gets older of the back story into each memory story (with the frame that memory stories here as having a positive impact) have the potential to change the frame of the memory, which decreases the likelihood of it being remembered? As in, certain memories are pretty great until ypu understand how it came to be that way, in which case either the back story becomes more signigicant or it becomes more mediocore than remembered. And mediocrity is easily forgotten...

I'd also suppose that as people move on with their lives, people relate to each other via shared stories... family stories have little meaning to those not involved, and so have little impact in "fitting in" in the rest of society. As that dawns on individuals leaving home, so are such stories repressed until they more actively need to be remembered.

Come now,this is silly.

There are people in their 60s that are so physically fit they can win a fair fist fight against a 20 something year old who ever so rarely works on his/her muscles.

Why is there a presumption that intellectual capacity is any different? A person that trains their mind will do better than those that don't,that's obvious.but how many people reach anywhere near their intellectual maximum capacity?

My theory is that as people get older the items of life take up much more of their time than in their earlier years which steals their potential to maintain or exceed their earlier intellectual capacity. Those that manage to make time for a mental workout in latter years would likely feel as competent as when they were young if not more.

Just a theory based on observation.

I hear you. But consider this. So you're making a case that you can exercise muscles and cardiovascular system, so therefore you can exercise the brain.

But what about other organs? Can you keep skin youthful through exercising it? What about hair? Eyes?

I think it's equally as plausible that no, you can't exercise the brain just by using it well. (though I don't personally believe this)

There is also clearly a peak and decline, even with exercise. The article does seem to cover this given the wide error bars for some of the cognitive tasks.

Intellectual capacity increases when people study intellectually demanding subjects and decreases when neglected,this is similar to the muscular and cardiovascular systems and afaik undisputed.

> Can you keep skin youthful through exercising it

No but you certainly can by avoiding excessive sunlight exposure and staying hydrated!

PSA: Being determined is a superior predictor of success than intelligence. With age, you have a better idea of what you feel is worth doing, and why, than the easy enthusiasms of youth.

Of course, those with wisdom, intelligence and youth will beat you. It's rare.

You are probably the smartest at around 5 to 10 years old. At least when it comes to creativity and learning. Peak performance of brain capacity is probably around 20 to 25. After that you aren't getting particular less intelligent per see but stress and lack of sleep are deterrent. The risk of stroke also increases as you get older. There are probably a peak of intelligence at around age 50 when your brain is still healthy and you have acquired experience and knowledge.

Degree of change would be nice. These traits could all be very flat over time and therefore small peaks make things statistically significant.

Disappointed by the tasks they chose. This is because the data is collected in a lab but we have so much more data these days.

Does anyone work for Udacity/EdX/Coursera? I would be interested to know if/how performance on those courses decline/improve with age.

It would also be interesting to see if performance is correlated because of lurking third factor. (My money is on exercise, or time).

Performance in MOOCS (or any educational setting) != Intelligence or ability to perform

> performance in educational setting != ability or intelligence

especially when the educational setting has an incentive to increase their performance metrics (not the same as standards)

I guess I'm more interested in whether my ability to learn machine learning (or whatever the equivalent is in 10 years) will decrease over whether I'll be able to remember a 7 digit number.

Call that what you will.

I am confused about what you consider "ability to perform". Education is a context for performance.

The theory is that your ability to memorize a 7 digit number (working memory) IS fluid intelligence.

The problem is that memorizing an x digit number is proven to be trainable, without much carry over to g or even other working memory tasks. I.e. - it's hard to train fluid intelligence rather than train the specific task which it is applied to.

I feel my ability to think quickly and see intuitive connections has decreased, but on the other hand I've become more logical and rational in my thinking. I think the latter has become more valuable to me, because it is key for communicating with others. And clear communication is much more useful right now than rapid thought.

Fluff article with misleading quote: "In this view, knowledge does not compensate for a declining adult intelligence; it is intelligence!"

This doesn't introduce anything of interest to people who already know that fluid intelligence decreases and knowledge can continue increasing.

Depressing.

Indeed. Besides, and anecdotally, I think it’s moreso give-a-f%%k that decreases rather than intelligence....

Anything that purports to measure “intelligence” will be confounded by give-a-f%%k. You can’t separate the two.

What if the article is aimed at people who didn't know that?

Then you got a decent introduction sprinkled with misleading commentary.

Humans have an odd focus on individual intelligence, even though most of the achievements that we have accomplished as a species have been through collective intelligence – rarely would I say that people function and create alone.

Our obsession with individual intelligence is usually about selection for (leadership of) elite teams.

For a brief moment I thought this article will talk about the point where humanity itself peaks in intelligence and knowledge.

On the other hand looking at the USA, I'd definitely say it already happened.

And when does the definition of intelligence stabilize?

This is a classic example of Betteridge's law.

Edit: QED

>It seems that the question "When does intelligence peak?" is actually a rather meaningless question. Not only do our various cognitive functions peak at different times, but past a certain age it might make more sense to view adult intelligence not through the lens of youthful general processing speed and reasoning, but through the lens of expertise, wisdom, and purpose.

Reply on news.ycombinator.com