Rob Boyd & Pete Richerson - How Ice Age Climate Chaos Made Humans Cultural Animals

[@josephnoelwalker]

Was shocked to learn this was Pete and Rob's first ever joint interview, and glad I could bring them together for it (all of us had to travel to make it happen!). Their work is super underrated outside of their field, yet it provides one of the most powerful frameworks I've yet found for explaining so much of humanity's trajectory. Hope you all enjoy! Joe

[@bhastro9959]

at 23:30 Boyd explains that the crucial difference between "cultural evolution" and "cumulative cultural evolution" is in the fidelity of social learning, so that when an individual learns a behavior from another individual the learning is sufficiently accurate that he is able to exactly copy that behavior and then modify (and perhaps improve) it. I found this somewhat obscure. First, it's almost counterintuitive: the more accurate the social learning the less the learned behavior will deviate from the model. Second, and related, if a behavior is learned from another, there is no "evolution" - there is "culture" but not change. Thirdly, the way that an individual could modify a socially learned behavior isn't specified. One possibility is that he simply fails to accurately learn the behavior, without realizing it; the question then becomes is this accidental variant more successful, either easier to learn, or more effective in reaching a goal, so it spreads more rapidly than the origiinal version. Or instead he is able to deliberately improve the behavior (based on experiment, insight etc). All of this requires very precise definition of "culture", "evolution", "learning", "social learning" , "change", "fidelity" etc. In my view one needs to construct a detailed mathematical/computational model of all these things, so there is no handwaving. Inevitably such an attempt would be highly simplified, and omit most of the real-world complexities, in an effort to describe exactly what "cumulative cultural evolution" is. Just as the neodarwinian synthesis/population genetics successfully blended Darwin and Mendel, one needs to blend neuroscience of learning and psychology/culture, starting first with "beanbag" models.

[@RyanFaulkner-Hogg]

Epic mate, great job.

[@josephnoelwalker]

🙏

[@yarrowflower]

Where is the interview recorded? It looks like the Internet Archive headquarters in San Francisco.

[@josephnoelwalker]

correct! mentioned in the text overlay at the very start :)

[@yarrowflower]

Cool! How did you end up there?