Weak links between speech perception and speech production at the subphonemic level

Recent research has shown individual differences in gradiency, defined as sensitivity to fine-grained acoustic differences at the subphonemic level. Consistency also varies, with some listeners providing more stable responses than others to repeated presentations of the same stimuli. However, it remains unclear whether gradiency and consistency are linked between perception and production. Across two experiments, 160 Spanish speakers completed a Visual Analogue Scale (VAS) task measuring perceptual gradiency and consistency along a Spanish /ba/–/pa/ continuum manipulated in 7–8 ms steps of voice onset time (VOT). The same participants also performed an explicit imitation task using the same stimuli. Results revealed only weak relationships between perceptual gradiency and imitation accuracy, as well as between perceptual consistency and imitation consistency, even after accounting for other factors related to speech perception and imitation. These findings suggest that models proposing strong perception–production coupling may not extend to the subphonemic level.