Fig. 4: The effects of different forms of social learning on diet-repertoire development and size.

a, Diet-repertoire development over simulated time. The horizontal lines indicate the total number of food items in the simulated environment, the estimated size of the adult diet repertoire at Suaq and the adult-like repertoire (90% of the estimated adult diet repertoire). The minimum and maximum ages for the onset of independence are indicated by vertical dashed purple lines. The Zone of No Exposure (contained between grey dashed lines) indicates all possible outcomes if we also modified the rate at which simulated immature orangutans were exposed to different food items (thus, a theoretical baseline control for the removal of peering, enhancement and exposure; see Extended Data Fig. 1 for simulations of development within this zone). For each treatment, a weighted curve indicates the mean repertoire size across development, and the shaded region around each curve indicates ±1 standard deviation of the mean. b, Diet-repertoire size of simulated immatures at the maximum age for the onset of independence (nine years). c, Diet-repertoire size at the end of immaturity (15 years). For b and c, N = 250 simulations per treatment. Experimental treatments are indicated in different colours. Three asterisks (***) indicate a significant difference between treatments of P < 0.001, with significance determined by a Poisson GLM (two-sided tests; all means compared in the same model; see ‘Simulated diet-repertoire development’). In the box plots, the centre lines indicate the medians, the top and bottom boundaries of boxes indicate the upper and lower quartiles, the whiskers extend to 1.5× the interquartile range, and the points indicate outliers.