Several forms of social learning rely on the direct or indirect evaluation of the fitness of cultural traits. Here we argue, via a simple agent-based model, that payoff uncertainty, that is the correlation between a trait and the signal used to evaluate its fitness, plays a pivotal role in the spread of beneficial innovation. More specifically, we examine how this correlation affects the evolutionary dynamics of different forms of social learning and how each can generate divergent historical trajectories depending on the size of the sample pool.In particular, we demonstrate that social learning by copying the best model is particularly susceptible to a sampling effect caused by the interaction of payoff uncertainty, the number of models sampled (the sample pool) and the frequency with which a trait is present in the population. As a result, we identify circumstances in which smaller sample pools can act as 'cultural incubators' that promote the spread of innovations, while more widespread sampling of the population actually retards the rate of cultural evolution. 3 the explicit use of replicator dynamics to explore optimal strategies for copying with payoff uncertainty (Schlag 1998). In this paper we cross-cut some of these established themes by exploring the impact of population size/sampling and payoff uncertainty on the early stages in the spread of innovations. Our finding augments the results of some existing studies (e.g., Baldini 2012Baldini , 2013 Lake and Crema 2012), but also identifies circumstances in which the positive correlation between population size and the rate of cultural evolution (Henrich 2004;Powell et al. 2009Powell et al. , 2010Shennan 2001) is significantly modified by the joint effect of payoff uncertainty and sample pool size. We begin with a brief review of findings from the most immediately relevant recent literature on the value of social learning, the nature of payoff and success-biased social learning and how it is mediated by payoff uncertainty and population size and/or the way in which the population is sampled.