Psychohistory

Psychohistory is the study of human behavior throughout history by applying psychology, especially unconscious motives. It combines psychology, history, and related social sciences to explain why individuals and groups act the way they do. In Asimov’s Foundation stories, Psychohistory is a mathematical technique used to predict the behavior of large populations.

Humans are not entirely predictable. Small events, leadership shifts, or technological shocks can upend long-term forecasts. If enough people act in statistically consistent ways, large-scale trends like collapse, unrest, or recovery might become predictable.

By examining history, demographics, economics, and social behavior, it is possible to identify key turning points and trends. Since it’s impossible to predict exact individual actions or very distant futures with certainty, researchers inspired by psychohistory typically present their work as probabilistic analysis rather than prophecy.

In Asimov’s books, psychohistory works like a statistical physics of societies. It doesn’t predict individual actions but the behavior of large populations using overall laws. Asimov’s model compares people to many particles in a gas. Each person is unpredictable, but the crowd as a whole can follow steady statistical patterns. The approach depends on large numbers. So, it only works when the population is very large, with few outliers or surprises.

Psychohistory can forecast broad outcomes like empire collapse, migration waves, instability, and recovery paths over decades or centuries. It is probabilistic and macro-level, more like forecasting weather or markets than giving a precise day-by-day script.

Psychohistory forecasts become unreliable when the population is too small, when individuals are aware of the predictions, or when exceptional actors cause major disruptions. The math cannot reliably account for single exceptional individuals, highly reactive small groups, or people who already know the prediction.

In Asimov’s logic, psychohistory works best when individual choices average out across millions or billions of people. Another problem is foreknowledge. If people know a prediction about the future, they may change their behavior in response, which can invalidate the prediction.