Mirror neurons fire both when executing actions and observing others perform similar actions. Their sensorimotor matching properties have generally been considered a genetic adaptation for social cognition; however, in the present chapter we argue that the evidence in favor of this account is not compelling. Instead we present evidence supporting an alternative account: that mirror neurons’ matching properties arise from associative learning during individual development. Notably, this process was not ‘designed’ by genetic evolution specifically to produce mirror neurons, but just happens to produce them when the developing system receives correlated experience of observing and executing similar actions. Sensorimotor experience with non-matching actions, or with objects and actions, is hypothesized to generate other cell types in the same regions through the same process. The associative account has major implications for research into mirror neuron function and suggests several important lines of future research.