Anyone with a lot of experience in social psychology lit?

[sciatrix]

This book I'm reading is giving me a lot of interesting thoughts, but I've also seen a lot of yelling about the replicability of social psychology results (like the marshmallow experiment) in the years since I've finished my bachelor's. Since I've more or less ignored the human psych literature except as it relates to my field, I'm a little bit at a loss when it comes to evaluating that research as a whole.

How much of the irreproducibility discussion undermines this body of work? How much smoke is there in that fire? Anyone know?

Comments

[zenolalia]

Basically anything done before about 2000 needs to be checked from reproducibility and confounding factors.

Which is really unfortunate, because in a lot of ways the field is only about 110 years old and the first 90 of those years are as good as useless.

That said, anything in the post-2000 range is a little more trustworthy, as long as its research on the general population and not on various mentally ill groups.

Reliable, reproducible, non-confounded research on to social behaviours and underlying causes of those behaviours in autism, cluster-B PDs, and a host of other mental illnesses and neurotypes is extremely new, within the last 10 years in most cases.

It's also worth noting that a lot of the field is pretty dedicated to re-performing and re-analyzing these older, more widely accepted studies and figuring out the legitimate from the trash. Almost any major study (like the marshmallow study) from the "bad old days," has been re-created and re-analyzed to find out the confounders and the outliers already.

The real problem is that a lot of the methodologies developed and popularized in preceding generations of research in the field are themselves both biased and widely taught, because of course 2000-2010 was both extremely recent and a major upheaval in the field.

That said, as a research professional in a related field, you can probably trust your own judgement in the end. You'll get a quick sense for when confounders such as poverty, race, and trauma are ignored or suspiciously absent from the methodology.

[siderea]

Hmmmmm, this sounds wildly optimistic to me - like, the replicability crisis didn't start until the late 00s, so, no, I wouldn't assume that research 2000-2010 is a-okay.

But it also sounds like you have information I don't? I don't know anything about a revolution in the study of what is classically called "psychopathology", i.e. "social behaviours and underlying causes of those behaviours in autism, cluster-B PDs, and a host of other mental illnesses and neurotypes". This is right on my turf, so I'm happy to do my homework. Can you throw me a search string or other pointer so I know what I'm looking for?

[staranise]

I'm not sure anybody actually knows what's signal and what's noise. I'm interested in what you're reading, partly because I know better how to evaluate something when I know what it is.

[sciatrix]

Right, so this book is Secrets From the Eating Lab, Traci Mann. It's largely about motivation-to-eat kinds of stuff, and how to make healthy choices but also that losing significant weight is basically impossible etc etc etc. Most of that I know pretty well how to assess, but I know that she cites the marshmallow stuff in a couple of places and there's a lot here about distraction and eating quantity/choices. There's a piece

One of the big things making me go "hmmmm" is her work on food labels and their influence on food choices, which I know there was a recent scandal about re: actively fudging data and deliberately p-hacking with no plausible deniability in, but am not sure if her lab was involved or she knows people who are. (Upon a quick google, I'm thinking here of Brian Wansink's group; haven't looked into Mann's background to see if she's ever worked with Wansink.)

On the other hand, the book is sparking a lot of thoughts I've been having about food-seeking motivation and stress, especially long-term and unpredictable stress, and about interactions between leptin and cortisol maybe integrating internal and external cues about what's going on as an organism tries to respond. There's a particular study involving telling people that a milkshake is either a diet "healthy" milkshake or a full-fat "tasty" one shaping the level of ghrelin released in respose to that milkshake that I want to track down, and also some discussions of how people eat when stressed--she first argues that most people eat more when stressed but then discusses astronauts eating less, typically, and I've been thinking along the lines of different responses to food and stress based on different modes of comfort seeking in different people, and also whether or not food itself has become imbued with stress in some people.

And then she went into a discussion debunking the entire concept of comfort food, and now I'm just thinking about how mood, stress, and hunger interact with one another in an organismal context.