Figure 4.

In humans and chimpanzees, the recency of past encounters predicts the probability of future encounters consistent with a power-law function. (a) The recency of a word appearing in The New York Times headlines predicts the probability of the word appearing again. Recency is measured as the number of other words since the last occurrence of the target word. Adapted with permission from Anderson & Schooler [4]. (b) The recency of a human participant encountering a social partner predicts the future probability of contact with that partner. Adapted with permission from Pachur et al. [24]. (c) The recency of a chimpanzee encountering a social partner predicts future contact with that partner. Error bars with binomial 95% CIs are too small to plot. Dashed line represents overall mean probability of contact in dataset. Curves illustrate best fitting power functions.