This paper addresses the question of the domain-specificity of learning biases for phonological processes. In two artificial grammar learning experiments we explore the role of learning biases in shaping the distribution of phonological patterns across the world’s languages. In Experiment 1, we demonstrate that learners are biased toward phonological patterns that occur in natural language, as opposed to patterns that are not found across the world’s languages. Specifically, learners are biased towards directional vowel harmony spreading processes. In Experiment 2, we exposed learners to a non-linguistic analogue to vowel harmony. Learners processed spreading such that learners favored the cross-linguistically valid pattern only when the first item of the series underwent spreading. This set of similarities and differences in learning may provide some insight into the origin of learning biases for spoken languages.