On February 14th, 2022, Dr. Igor Grossmann from the University of Waterloo gave us an interesting talk about scientists ability - or, rather, inability - to predict cultural change. This talk was a part of our weekly Brownbag colloquium series. Click here to learn more about his research!
See below for the title of the talk and abstract!
Explanation and prediction of cultural change
It is self-evident that human cultures are not static—consider how art, fashion, child rearing, and the workplace have changed in recent decades and especially in the last two years. Yet, how well do psychological scientists capture the dynamic nature of psychological and societal processes? How accurate are they at estimating and predicting cultural change and what strategies are they using to arrive at their conclusions? In the first part of my talk, I will illustrate a few examples of long- and short-term societal changes in human psychology, with implications for inferences psychologists tend to draw about human nature. Next, using COVID-19 pandemic as a lens, I will turn to behavioral and social scientists’ intuitions about cultural change. I will show that at the onset of the pandemic, scientists made a range of predictions about how societies will evolve, and every fourth case (out of 700 analyzed media interviews with scientists) they did so outside their domain-specific expertise, chiefly basing their judgments on intuitions and analogies. Moreover, large-scale prediction surveys conducted in spring and fall of 2020 (N = 717) show that behavioral scientists’ intuition-based judgments about cultural change were largely inaccurate: they were no more accurate than the average American, and that they did not update their inaccurate expectations in the face of six months of experience. Similarly, in the Behavioral and Social Science Forecasting Collaborate tournament (predictions.uwaterloo.ca) with over 120 teams of scientists from around the world, data point predictions about cultural change in well-being, prejudice, and political attitudes for 2020-2021 were not better than naïve statistical benchmarks (e.g., linear regression, average of historical trends) or wisdom of naïve crowds. These insights matter, because lay public generally expected psychological scientists to make accurate predictions, and preferred their recommendations to base COVID-related policy responses on societal change. In the final part of the talk, I will discuss ways to increase intuitions and reasoning about cultural change in our field, calling for graduate training in methods helping to quantitatively model prediction (vs. chiefly reply on qualitative post-hoc explanations).