Effect of Speech-Shaped Noise on Pseudoword Learning in Young Adults

This study examined the effect of background speech-shaped noise (SSN) on novel word learning. We asked whether training outcomes differ when pseudowords are learned in Quiet, +10 dB SNR, or 0 dB SNR conditions.

Sixty young adults (ages 18-30 years) completed a 5-day training paradigm in which novel pseudowords were paired with pictures of familiar objects. Participants were assigned in a counterbalanced manner to one of three listening conditions (Quiet, +10 dB SNR, 0 dB SNR). Recognition accuracy was tested after each daily session. Mixed-effects models evaluated baseline group equivalence, session-by-session changes, and final performance outcomes.

Groups did not differ significantly at baseline, and performance improved across all five sessions. Importantly, background noise did not disrupt word learning: by the final session, all three groups performed comparably. At intermediate sessions, participants trained at 0 dB SNR showed steeper improvements, although this did not translate into lasting group differences.

Findings suggest that moderate levels of SSN do not impair novel word learning in young adults. These results are partially consistent with theories of stochastic resonance, which predict performance maintenance from low-level noise. Although preliminary, this work indicates that SSN may not be uniformly detrimental to learning and raises the possibility that controlled noise exposure could be leveraged in future training paradigms. Further research should test generalizability across populations and task types.

https://doi.org/10.23641/asha.31079449.