Yep, any other conclusion from the mentioned study would be suspect unless the asymmetric cost of going backwards was somehow controlled for.

Otherwise this is 100% explainable as a hoarding instinct because there is uncertainty about what else is available. Depending on a given person’s risk aversion, it’s entirely rational to take this view of the opportunity cost of passing on early choices you can’t revisit later.

Not to mention that there are people who just want to eat something and will pick the first item that's above said threshold because they just don't care.

That is me. I fill the plate with stuff that I want to eat and stop filling it when it is enough.

I do look in advance whether there is something awesome worth longer wait, typically it is just more of the same.

I like the buffet analogy. It is a good example of "choice architecture" a la Cass and Sunstein from "Nudge". Nudge is one of a number of works arguing that we factor in more than economically rational criteria--such as marginal cost of going backwards in the buffet line--when making decisions.

While Cass and Sunstein argue for government action in framing healthier choices, this post puts the onus on us to a.) recognize that information consumption involves choices, and b.) be more active thinkers consumers of information.

The funny thing about governmental "nudges". The vast majority of them - and there are a lot of them - are already used a lot, to meet unwritten administrative goals.

A small example is the DMV person failing to tell you about an option you have that costs them more work. A big one is the Rick Scott administration's design of the unemployment system in Florida:

https://nymag.com/intelligencer/2020/04/florida-unemployment...

I applaud the idea of deliberately thinking about this stuff; one reason why governmental programs go off the rails is that programs are implemented in bad faith (or just "lazy faith"). So insofar as you actually want competent government, bringing attention to this stuff is positive. I'm a bit less optimistic about improving individual behavior that way, but then I'm a pessimist about that topic in general.

The buffet line is a simplified test, meant to isolate an effect. Compare going to the grocery store with making a grocery delivery shopping list. You will find that end caps, promotions, coupons, shelf position, etc. all have an effect on you.

You will also find a lot of literature saying these things don't work. What they mean is "they don't work in the way they were sold to the people buying the end cap." But they sure do have an effect.

With a buffet (and when browsing the Internet), you can absolutely go backwards. Just get back in line. Yet we act as if there is a shortage of $resource.

That takes effort, which is its own cost.

I -could- go back and forth 30 times and end up with a satisfaction-maximizing plate.

But I'm not gonna do that. I'm going to stack some lettuce because I know it's a good base for whatever I get next (though cheap for the restaurant). I'm going to pick a quantity that leaves room for unexpected yummies ahead out of sight, but also is sufficient that if I find few things I like ahead I'll still be happy with my salad.

I don't even manage to maximize the outcome at buffets that I've been at before with mostly static contents.

Going back has an extra cost -- time to stand in a line again. So unless one enjoys standing in lines, there is absolutely an extra cost.

> Just get back in line.

Which might take a long time, and by the time you're back at $item, it may be out.

I wonder how long the lines were in the study, or if this was controlled for at all.