Hacker News Reader

Comment by tabtab | Hacker News Reader

For most of human history, men were pretty much forced to be the warriors and primary defenders of a village. Thus, there was strong genetic pressure to react quick (or die). But women seem better at "find Waldo" kind of tasks, so perhaps there's a tradeoff. We are tuned for different specialties.

View on HN

It's worth mentioning that the cognitive science community generally shuns evolutionary arguments, as they tend to be "just so" assertions.

It's not that it's necessarily wrong, but rather unfalsifiable as such.

Who is “the cognitive community”, and does it include the evolutionary psychologists?

That's the parent posters point. There's no such thing as "evolutionary psychologists". At least so far as academics within psychology and neurology recognize it.

These disciplines examine behavior from two different angles, one looking at more proximate causes while the other looks at ultimate causes.

You can dismiss evolutionary psychology as irrelevant to other social or neurological cognitive sciences (which I would argue is also wrong), but it doesn't mean interesting and useful research is not being done.

Paleontologists and archaeologists can do research by looking at old things and comparing them to other old things and also to current things. They can be very precise. "This process of this bone was so many millimeters wider, probably indicating a stronger attached muscle." "The clay in these pots comes from this particular remote river valley, indicating a trade network between these locations." Etc.

How does a psychologist do anything like this? Mental processes of living things and how those respond to environment can be observed. How can mental processes of millennia ago be observed?

Not sure about psychology, but the comparative method is also used in linguistics to reconstruct dead languages.

That's true, and this linguistic work is in some sense "between" the hypothesized evopsych work and the rigorous work we see with genetics. When the geneticists tell us that a particular population underwent a bottleneck at a particular point in time, they're probably correct about that. Linguistics is like genetics in that, because of writing, experts have something physical they can study to infer how languages have evolved over time. Perhaps something similar might someday be used to see how non-linguistic mental processes have evolved over time, but at this point it's difficult to imagine what that might be. Psychology itself is on a fairly tenuous footing as a science, so it's hardly a sufficient foundation for whatever has been envisioned by evopsych enthusiasts.

The way you do evolutionary psychology scientifically is that you think about which physiological traits ought to have been produced by selective pressure and then check your predictions. Coming up with a post hoc justification after the evidence is in just isn't science.

So the distinction is post hoc reasoning, not evolutionary explanations? I interpreted the OP’s comment as condemning the latter.

The vast majority of people making evolutionary psychology arguments in Hacker News threads or newspaper op-eds or such are also doing the later so the bad associations are understandable, even if they're not really the fault of evolutionary psychology the academic field.

My point (albeit awkwardly stated) was that evolutionary psychology is fraught with post hoc reasoning. The post to which I was replying is one such example.

The result is that neighboring academic fields are (rightly, IMHO) skeptical of EP theories.

Evolutionary arguments are complementary to work done in other cognitive sciences. Both are important to understand the complete picture of how we behave.

Many evolutionary explanations begin as 'just so' assertions, but so does anything in science, really. Evolutionary arguments can be very difficult to test because our knowledge is incomplete. There are many interacting variables at play and we can't go back in time. It's doesn't mean we can't test hypothesis, though, and there is a huge body of evolutionary behavior research that shows how much we can do.

A "just so" story like this is certainly unfounded, I think, but I wouldn't call it unfalsifiable. You could search for supporting evidence by investigating, for instance,

- Do we see this effect in species other than humans?

- Is this effect more pronounced in individuals descending from ancestors who were more prone to hunting?

>Do we see this effect in species other than humans?

You can always find a difference between humans an the other species that explains the absence in the latter.

>Is this effect more pronounced in individuals descending from ancestors who were more prone to hunting?

Same.

Evolutionary psychology is ... not popular with cogsci depts...

Good evolutionary hypotheses generate testable predictions about the trait being examined. This can lead to other predicted (testable) effects, and so on, building evidence (or lack thereof) for an idea. Just because we can't draw definitive conclusions doesn't mean there is not strong evidence for many evolutionary theories.

> Evolutionary psychology is ... not popular with cogsci depts...

That's a shame. It seems there is a lot to be gained from a comprehensive approach to understanding.

You may have heard the phrase “nothing in biology makes sense except in the light of evolution" (Dobzhansky). Well, this is true for the human brain and psychology, too. We can't correctly address major behavioral issues without a deeper understanding of the evolved functions of our brain.

>Good evolutionary hypotheses generate testable predictions about the trait being examined.

To clarify: yes, absolutely!

The problem (and the reason for it's unpopularity), is that the EP community, in general, produces many bad hypotheses.

But authorities closed down my human breeding lab.

The important part is how well tuned.

In formal terms, how big is the overlap and effect size.

If overlap is big or effect size is small, the difference will be irrelevant for most tasks.

The question on thresholds for these values requires further major study. Such as devising a wide set of tasks and comparing performance in real tasks that putatively use it vs these measured motion detection differences. I'd start with catching a thrown ball. Avoiding getting touched by a moving object. Matching speed by movement. Etc.

Additionally, it has to be seen if specific training can change those values and to what extent.

Women¹ are also much better at seeing slight color variances. E.g. differentiate between two similar red berries.

¹humans whose eyes are using cells that are filled female hormones, to be precise

It's nothing to do with men being "forced" to do anything, it's simply that in every sexual species the selective pressures on males and females is very different. In mammals the females have to be strong enough to carry a baby until birth. While the males have to be stronger than all other males; long term health is much less important.

Reply on news.ycombinator.com