Successful Social Learners Can Choose the Better of Two Options: An Agent-Based Simulation

Successful social learning strategy

Abstract

It has been argued that the presence of social learning does not increase population-level fitness because its advantage is frequency-dependent: Social learners cannot flourish unless a sufficient number of individual learners make valid information available. However, recent studies suggest that social learners can achieve higher fitness than individual learners even when individual learners are rare. This study investigated the effects of two mechanisms that confer an advantage on social learners over individual learners: information filtering by social learners and information aggregation through a preference for popular options. Using Smolla et al.’s (2015) model of social learning, which incorporated these two mechanisms, we systematically removed either or both of these mechanisms, rendering the study comparable to a 2 (information filtering: absent vs. present) × 2 (information aggregation: absent vs. present) factorial design. In addition, we tested these effects under the conditions with and without resource competition among agents, which was the original variable of interest in Smolla et al.’s study. The results showed that information filtering conferred a fitness advantage on social learners over individual learners especially when no resource competition was involved. In contrast, information aggregation alone was insufficient to help social learners outcompete individual learners.