Figure 1.
Effect of selection pressure, group size, interaction between groups and learning costs on the mean number of cultural traits per individual.
a, b: no learning costs; c, d: learning costs 1 energy unit per learning event; a, c: isolated groups; b, d: interaction between neighboring groups. Legend indicates number of resource units per square. Results are grouped according to selection differential (x-axis). a, b: Higher selection pressure can increase the number of traits per individual up to a certain point (selection differential 0.5), but can lead to a decrease if it is too high (1.0). Higher resource availability significantly increases trait number for isolated groups and for interacting groups with intermediate selection differentials. Interacting groups always have a higher trait number than isolated groups. c, d: Learning costs can significantly decrease the number of traits per individual. If there are learning costs higher selection pressure always increases trait number (see text for details).
Table 1.
Mean number of cultural traits ± standard deviation and mean group sizes in populations with isolated or interacting groups, with different selection differentials and resource availabilities.
Figure 2.
Effect of innovation costs on number of cultural traits per individual. Higher innovation costs can significantly reduce the number of cultural traits per individual, especially in isolated groups with low selection differentials. Maximum energy score of individuals capped at 50.
a, d: inventing a new trait costs 10 energy units; b, e: inventing a new trait costs 20 energy units; c, f: inventing a new trait costs: 40 energy units. Upper row: isolated groups, lower row: individuals could learn from or choose partners from neighboring groups; error bars indicate standard deviation; results were compared using Wilcoxon rank sum tests (see Table S7); legend shows max resource value per square.