(Interesting. So the formal notions drove theoretical innovation such as pursued by Simon, Newell and others, while the technologies seemed most relevant for *acceptance* of those notions and theoretical constructs by the psychology community. What can we learn from this?) 9/n
(After "Tools-to-theories", Gigerenzer turns to "As if explanations". Did some research in this area myself as well: https://twitter.com/IrisVanRooij/status/955100914934800384?s=20) 10/n
"One of the most influential explanations of behavior is the theory of expected utility maximization. ...[It] addresses the question: How should one choose an action from a set of possible actions? The action with the highest expected utility represents the rational choice." 11/n
"Despite widespread use, theories of expected utility maximization have been criticized as being computationally intractable, assuming perfect knowledge about the exhaustive set of actions & future consequences & lacking in empirical evidence for stable utility functions" 12/n
"In his classic defense of expected utility theory, the economist Friedman (1953) countered that the psychological realism of a theory does not matter. According to Friedman, an explanation is only *as‐if*: People behave as if they maximized expected utility." 13/n
"In psych, distinction between as‐if and process models is often expressed in terms of Marr's (1982) distinction between a computational & an algorithmic level of explanation. (...) Like an as‐if theory, a computational analysis is mute about nature of the cognitive process" 14/n
(For those familiar with my work, my own thoughts ofc diverge from Gigerenzer on this point. I think it's a mistake to have non-realist interpretations of the comp level or otherwise allow for arbitrarily loose relations between levels of explanation https://twitter.com/IrisVanRooij/status/955101237673897984?s=20) 15/n
“Inspired by economic theory, cog processes—& their neural underpinnings—have been modeled as Bayesian inference (...) The problem of Bayesian computations & their approximations being intractable is irrelevant because there is no commitment to an algorithmic explanation” 17/n 👀
(Some commonly accepted but mistaken claims seem to persist by virtue of repetition despite careful and extensive arguments that show their flaws. And here Gigerenzer seems to help reify this point he contested himself in the past 🤔) 18/n
“Although researchers do not always agree whether a theory is as‐if or a process model, two distinguishing criteria exist. One was already mentioned: If the theory implies computations that are intractable, it is as‐if.” 19/n
(This definition of as-if surprises me, because on that definition even Gigerenzers Adaptitive Toolbox has as-if components: https://twitter.com/irisvanrooij/status/1193659636366368768?s=21 I doubt this would be his intent) 20/n
(After As-if theories, Gigerenzer turns to Adaptive Toolbox theories)

“Simon proposed that under uncertainty, cognitive processes involve heuristics that can guide behavior quickly and accurately.” 22/n
“Although Gestalt psychologists had used the term heuristics earlier to describe search for information, Simon insisted on formal models of heuristics and on studying how these are adapted to the structure of the environment.” 23/n
“The adaptive use of heuristics became systematically studied in the work on the adaptive decision maker (Payne, Bettman, & Johnson, 1993) and on the adaptive toolbox (Gigerenzer, Todd, & the ABC Research Group, 1999).” 24/n
“What the tools‐to‐theories framework and the study of formal theories of heuristic have in common is that they model cognitive processes. However, the study of heuristics does not assume that these processes correspond to the statistical tools used in a research community.” 25/n
On Marr's levels, Gigerenzer writes the following "the claim that there is a computational theory for all problems, Bayesian or otherwise, was rejected by Marr himself (see Brighton, in press)." This is true, see: https://twitter.com/IrisVanRooij/status/964589565402075138?s=20 26/n
"Marr (1977) distinguished between “Type 1” theories that yield to a computational analysis & “Type 2” theories that do not. According to Marr, Type 2 theories are the majority in AI." 27/n

This seems not consistent with what Marr wrote: https://twitter.com/IrisVanRooij/status/964589568828825601?s=20
"As I have argued, and as emphasized by Marr (1977), that distinction [between computational & algorithmic explanations] is irrelevant when cognition has to deal with uncertainty that cannot be measured by probability." 28/n

Marr said nothing about this AFAIK. I'm confused ...
(I sincerely do not understand this viewpoint. I mean, there are evidently computational-level theories that do not specify probabilities (but deal with uncertainty); e.g., models of analogical problem solving or similarity based categorization.) 29/n
“I think that psychology needs two agendas that editors of journals should consider promoting. First, we need to think more about the limits of theories. No theory should be published without specification of the domain where it does not work or apply.” 30/n
(This is a very good idea, insofar as we are able to know the domain and the limits. We cannot always know it a priori. But I agree, where we do, we would do well to explicate these limitations in our papers.) 31/n
“The program of ecological rationality is an example of such program, which states the conditions under which a given heuristic will succeed or fail. “

(That doesn’t strike me as stating limits of the *theory*of ecological rationality. This does: https://twitter.com/irisvanrooij/status/930557140485406726?s=21) 32/n
“Second, we need to think more about the integration of existing theories. Theory integration is an alternative and complementary route to Popper's program of theory elimination and, IMO, one of the most vital challenges to strengthening theoretical fundament of psychology.” 33/n
These were my highlights. The thread seems to be longer than this (short!) paper. Check out the full paper here: https://twitter.com/irisvanrooij/status/1195758273628196865?s=21

I enjoyed the paper but remain puzzled by some of the claims. /fin
You can follow @IrisVanRooij.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: