natehaines.bsky.social
Paid to do
p(b | a) p(a)
p(a | b) = —————————
p(b)
Data Scientist | Computational Psychologist | Devout Bayesian
https://bayesianbeginnings.com/
133 posts
784 followers
354 following
Regular Contributor
Active Commenter
comment in response to
post
yeah this is a great set
comment in response to
post
Wish I could have made it in person!
comment in response to
post
i have now vented on reddit and bsky so maybe i will feel better about it now
comment in response to
post
tariffs amirite
comment in response to
post
Thanks for the endorsement! Awesome to hear it's been influential to your work 😁
And yes the KL finding is super interesting by itself, happy to hear someone found it buried in the supplement ha
comment in response to
post
Thank you! Glad you have found it useful 🤓
comment in response to
post
it was horrible, there were 10+ papers based on the initial preprint/blog that were in print before this one eventually made it there 😅
comment in response to
post
Yeah I've found LLM tech good for stuff that is mostly boilerplate (e.g. some general software development stuff), but when it comes to modeling work that is necessarily bespoke, they are quite bad
comment in response to
post
That's called MDMA
comment in response to
post
But I have not read into this lit much tbh, so I have a biased sampling of what people think 😁
comment in response to
post
"Regressions to the mean = statistical artifact = the effect is not real" is my read based on papers like this
comment in response to
post
My read is quite different, e.g. www.sciencedirect.com/science/arti...
comment in response to
post
hey we grow good bourbon and that's food
comment in response to
post
Yes that's exactly what I mean—the claim that, because regression to the mean captures the effect, it is a statistical artifact and not something with a psychological explanation
comment in response to
post
I think they could potentially be good items for the classroom! It is a fun item set at least, although not sure about calibration as I really only looked at quartiles
comment in response to
post
Yeah IMO, I think the ideal design would be something within-person such that we could sample people across their individual skill distributions so that we could understand if the effect arises at the person level
comment in response to
post
"That explanation ... would not be able to explain self estimated performance going up again at very low ability"
I still think it is a better explanation than the "stats artifact" explanation. The models I used are too simple on purpose—I didn't have data for fitting anything more complicated
comment in response to
post
I feel there is some talking in circles going on here lol character limits be damned.. my stance is that the mechanism generating the noise is not well understood, but that psych-motivated models help provide an explanation. To get more fine grained (empirically) we need better datasets
comment in response to
post
Ah yeah this form of probability distortion is generally attributed to Goldstein & Einhorn (1987). e.g. see eq ~23
sci-hub.se/https://psyc...
comment in response to
post
I mean snipping the quote like that is a bit misleading ..
there are of course many different ways that noise could arise. the two models have different implications re mechanisms, but the data I had available were not able to distinguish between them
comment in response to
post
If you build a model with these assumptions based in, you will get the DK effect. In the same way, if you simulate multivariate normal data with one mean higher than the other and a moderate covariance, you get the DK effect. IMO the former is a plausible explanation, the latter just a description
comment in response to
post
The two models are essentially different parameterizations of the same idea. The assumptions being:
1) some latent skill gives rise to obj performance
2) people have some understanding of what their latent skill is, but there's noise in estimating it
3) people generally over-estimate their skill
comment in response to
post
My goal in writing the blog was mostly to point out that:
1) if obj and perceived skill are moderately correlated, you get something akin to the DK effect
2) you need to explain how the correlation arises
3) models based on confidence lit show how the cor/effect can arise per psych mechanisms
comment in response to
post
If proposing theoretically informed models that can explain things we see in the world is an "off strategy", I'm curious what you think an "on strategy" looks like 😁
comment in response to
post
I appreciate the shout out! It was a fun post to work through for sure. I wish there were more well sampled datasets to use so that the curve could be estimated within-people. Maybe one day!
comment in response to
post
Ha thank you 🤓 I think most study designs make it impossible to know for sure, but the confidence literature has analogous results that make me believe something real is going on there
comment in response to
post
lmao just the concept of a WMD being used in hospitals regularly is so funny
comment in response to
post
Also weirdly slow pacing given how season 1 ended with such a climax
comment in response to
post
Like every episode introduces new characters and plot lines, then only gives you a tidbit on past plot lines and characters. Too much novelty without enough closure
comment in response to
post
I'm so glad someone else feels this way
comment in response to
post
Very nice blog @hochmanshachar.bsky.social ! The dot probe is an interesting task. Lots of room for modeling, and I think researchers also need more trials if they want certain inferences. e.g., some work I did showing that the condition effect is best captured w/out indiv diffs: bit.ly/4ky2Oli
comment in response to
post
like a more techno, but equally melancholic, version of The Trilogy
comment in response to
post
See I must not experience the mood that induces a preference for Canadian whiskey
comment in response to
post
this is so incorrect man, but I suppose that is what makes this controversial
comment in response to
post
def true believers
comment in response to
post
hmm good q.. both ;)
comment in response to
post
i have always known