Andrew Buskell just published in Acta Biotheoretica an insightful combined review of Cultural evolution in the digital age and Hugo Mercier’s Not Born Yesterday. There are many things to appreciate in the review. First - but I may be biased here… - it is a positive one. Andrew concludes by writing:
Both Mercier and Acerbi’s volumes are enjoyable, story-driven journeys. They guide the reader through large and expanding literatures in an entertaining and illuminating way. They are upbeat about human capabilities and provide a useful optimistic counterpoint to the recent spate of epistemological doomsayer.
I could not have said it better.
Second, it is pleasing to be in good company. Not born yesterday is an excellent book. In fact, while writing, my own views about how social information is evaluated have somehow glided from a more standard cultural evolution framework to the epistemic (or “open”) vigilance perspective presented in Hugo’s volume. If I would start working on Cultural evolution in the digital age today it would probably be a quite different book.
Third, and last, the “concerns” are also perceptive. One of those is that Cultural evolution in the digital age did not discuss enough the importance of the integration between the literature on social network models and the largely individual-level perspective endorsed in the book. I completely agree. The main theoretical perspectives I drew upon are indeed not particularly sensitive to the topic. Evolutionary psychology theories are mostly theories about individual minds, and while social interactions and communications are certainly important for these theories, they are important with respect to how they shaped those individual minds. (Incidentally, I always thought the evolutionary psychology is one of the approaches that would have more to gain by adopting a modelling perspective, but only very few evolutionary psychology works use it.) Cultural evolution seems more receptive: virtually all theories in cultural evolution take into account population-level dynamics, still populations are generally considered as uniform or - I think - fully-connected networks, i.e. networks were all individuals have the same probability to interact with any other individual in the network.
Of course, there are exceptions. At least a few papers adopt an explicit “cultural evolution on networks” perspective and there is an enormous grey area (the “grey” is just because I am not familiar myself) of works that are relevant from both perspectives. In my folder “archive/projects_discussed/interesting_to_do/ok_lets_see/but_probably_not” there is an old modelling project aimed at comparing the diffusion of cultural traits under different selective biases (neutral, model-based, frequency-based, and content-based) in different population structures: fully-connected (the control condition, standard in cultural evolution), Erdos-Reyni random networks and Scale-free networks. As the name of the folder suggests, the project is still there.
As a matter of fact, one of the reasons that motivated me working on Cultural evolution in the digital age was exactly the lack of integration between the two literatures. However, my unsatisfaction was going in the opposite direction. Research on online cultural dynamics was (and still is) dominated by the network/complexity science perspective. Whereas these works are computationally sophisticated, and some of them are illuminating, I have been sometimes frustrated by the lack (if not, worst, the implausibility) of the assumptions about individual psychology. My feeling is that many questionable ideas circulating about the negative effects of digital online media, e.g. the overestimation of “echo chambers” effects or of the influence of bots and misinformation, may have found support in the scientific literature because of this problem.
Unfortunately, I have been not explicit about that in the book and, as Andrew’s review rightly criticises, not particularly constructive, but the hope (implausible, I admit) was that researchers from the network/complexity science perspective would integrate some of the ideas in their models. But perhaps it is up to us. Interdisciplinarity is hard.