Fig. 4 | Scientific Reports

Fig. 4

From: A neural network model for the evolution of reconstructive social learning

Fig. 4

Joint evolution of self-guided individual learning and socially guided learning. For three frequencies (f) and magnitudes (m) of environmental change, the panels depict the characteristics of the evolved learning strategy after 20 K generations when self-guided learning and socially guided learning could jointly evolve. Each panel shows the outcome of 60 replicate simulations. Each point represents the population mean number of learning episodes (x-axis) and mean proportion of social learning (y-axis), both averaged over the last 2 K generations of the simulation. The colour of each point represents the average performance of the evolved learning strategy. A small number of learning episodes indicates that learning did not evolve. If learning evolved, a low proportion of social learning indicates the evolution of pure individual learning (here: self-guided individual learning), a high proportion indicates the evolution of pure social learning (here: socially guided learning), and an intermediate proportion indicates the evolution of a combined learning strategy. In all simulations, the width of the environmental profile was σ = 0.25.

Back to article page