Training type and composition moderate efficacy of perceptual learning for speech

Perceptual learning for novel speech sounds can occur via active training, in which a discrimination task is coupled with feedback. Although effective, active training is significantly more effortful than passive exposure to the target sounds. While passive exposure alone is insufficient for learning, a combination of active and passive training has been shown to be just as beneficial for perceptual learning gains (Wright et al., 2015). However, the underlying mechanisms at play, and therefore the ratio and order of active and passive training required to achieve learning, remain unclear. In the current study, participants completed a training paradigm to discriminate a dental–retroflex stop contrast. Participants were assigned to one of 8 conditions (n = 20 per condition) varying by the amount of training type, whether mixed training was blocked or interleaved, and training type order, and whether passive training was a nonlinguistic task or task-free. Preliminary results demonstrate learning in both mixed training conditions (blocked and interleaved) but only when active training came before passive, and opposite effects of passive training type (nonlinguistic task versus task-free) depending on mixed training organization. While data analysis is ongoing, results suggest a complex interplay between attention to target stimuli and the organization of other elements of training.