Cents and Shenshibility: The role of reward in talker-specific phonetic recalibration

Abstract

We, as listeners, encounter many challenges during speech comprehension that extend far beyond resolving ambiguities in the acoustic phonetic signal. An understudied component of naturalistic speech comprehension is how motivated a listener is to attend to various aspects of the speech signal. This includes, when listening to multiple novel talkers, whether we give talkers equal attention when adapting to their idiosyncratic productions. We asked if listener adaptation is influenced by an external reward that is intended to bias attention towards one talker versus the other. In a lexically guided perceptual learning paradigm, participants heard two talkers of different genders—“Jane” and “Austin”—with idiosyncratic productions of the /s/ and /ʃ/ fricatives. Participants were more likely to receive a small monetary reward for one talker than the other following correct responses during a phoneme-monitoring cover task. We hypothesized that participants would show greater phonetic recalibration for the more-rewarded talker than the less-rewarded talker; however, our results indicate that external reward may undermine learning for both talkers. The role of reward and motivation in talker-specific phonetic recalibration raises interesting questions for how these domain-general factors influence language processing more broadly.