Understanding how and why pronunciations vary and change has been a dominant theme in variationist sociolinguistics (Labov 1966, 2001). Linguistic variability has also been an area of focus in psychology and cognitive science. Work from these two fields has shown that where variation exists in language, an alternative form, once used, persists in working memory and has a greater chance of reuse (Bock 1986; Bock & Loebell 1990; Branigan et al 2000). While there have been efforts to connect priming research with sociolinguistics at the level of grammar (Poplack 1980; Travis 2007), there has been less work which explicitly considers the potential role of priming as a motivating factor in accent variation and change. This paper explores the role of priming in a socially-conditioned sound change. There are two main findings: (1) phonetic variants with the same voicing tend to cluster together in naturally occurring speech and (2) repetition of phonetic form interacts with widely attested sociolinguistic predictors of variation. I argue that there are benefits to both cognitive science and sociolinguistics from this synergy: incorporating research from cognitive science into sociolinguistics provides us with a better understanding of the factors underpinning a sound change in progress; incorporating insights from sociolinguistics into cognitive science shows that priming does not always operate in the same way for all speakers.