This is (and this is no hyperbole) fucking nuts

“Detrimental effects of same-sex co-housing on mice”

Took 2+ years to work it out. Well worth it. I wish Howie was around to see it.

doi.org/10.1016/j.ne...

Then we show it solves a lot of other problems.

I didn't think this was going to work so well. I meant it as a critical review of the work of my late advisor, Howie Rachlin, on maximization. His work turned out to be the foundation for something bigger.

👇

It tackles the question of how to measure how much an animal wants something. For decades we've been deluded into thinking that this is trivial (choice? response rate? PR breakpoint?). It is not. Simple solutions imply fallacious assumptions. We show this. Then we show a solution.

👇

I'd have to claim infantile amnesia

The making of controlled experiments is at the same time constitutive for the knowledge that can be gained in the experiment, contingent on a research situation, and historically shaped.

To understand the nature of controlled experimentation, one needs to consider the making– the design phase– of controlled experiments, particularly the conceptualization and treatment of background factors. 👇

It argues that controlled experimentation has a material-technical and a conceptual side. It shifts the focus from control experiments, comparisons with a control, to the broader issue of controlling for background factors as the epistemologically fundamental issue in experimentation. 👇