Disagree and false opposition. More later :)
-
-
Sometimes exp training can even hinder theoretical development. Not saying no empirical expertise is useful for theoretician, but also see how drilling of exp training can kill all conceptual creativity & make people conflate statistical hypothesis with substantive theory.
-
If you're analyzing data, I think that's essentially empirical experience. I think the important aspect is just to know that data can be messy so that you don't get too attached to a particular interpretation of it.
-
This is a good point, e.g. was at meeting with logicians & a cognitive scientist showed a plot of human data, with 3 noisy lines. I and other empirically trained ppl understood the general pattern. A logician asked "why do the lines zig-zag & how does your theory explain that?"
-
Just saying: that some experience with the variability in data / phenomena of interest can give one a better sense of what pattern in the observations is in need of substantive theoretical explanation.
-
Yeah, I've been where that logician was.
The only stats-related thing I knew before I moved to Psychology were Bayesian graphical models. I can't believe after more than a decade the module still exists and is taught by the same person! https://www.york.ac.uk/students/studying/manage/programmes/module-catalogue/module/COM00032H/2018-19 … -
But those are very useful too! Especially for better understanding Bayesian models in cognitive science and/or designing them oneself. -
Yes. Absolutely! But it made learning frequentist stuff very confusing! Especially at the start... I was like "why are we learning the equation of a line?" when we were taught linear regression for example. Also, oops. Sorry for breaking the thread.https://twitter.com/o_guest/status/1049777498056220673?s=19 …
-
I wasn't used to how stats is taught in psychology at all.
End of conversation
New conversation -
-
-
But maybe modellers are not who you are talking about, in which case I guess my points are moot.

-
Of course we also are speaking of modellers. How would you position yourself? As a (theoretical) computational modeler?
-
I've models where I capture the data, reproduce the data others have collected. I also have ones where I predict data that doesn't exist while I did the modelling. I also do qualitative modelling and/or theoretical modelling. But like I said I usually work with others' data.
-
I also do experiments on models to further understand the models themselves. Like science on a deep network.
-
This is why I am not sure in all this where I would fit in.
-
Above I made a typo. I meant to link to http://redistrict.science — what kind of modelling is that? It uses data but it doesn't really predict anything in and of itself.
-
I'd be say/guess in cognitive (neuro)science you span the part of the spectrum ranging from theoretical modeling to data modeling. (the 1 dimension is a simplification ofc). And your gerrymandering work is an example of mathematical/computational modeling for social engineering?
-
Yeah, exactly. BTW TIL because I have no idea social engineering meant what you just used it as! I only knew it as: https://en.wikipedia.org/wiki/Social_engineering_(security) …
- 2 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
