Auditory learning of time-compressed speech reveals acoustic–phonetic constraints on generalization across age groups

Understanding how listeners adapt to temporally altered speech and the extent to which training effects generalize across different speech materials and forms remains an important open question in auditory perception research. This study examined these issues using a multi-session time-compressed speech (TCS) training protocol in Arabic. We assessed whether adaptive training in Modern Standard Arabic improves TCS perception and whether training-related gains extend to novel items and to a related but acoustically distinct speech form (spoken Palestinian Arabic). Fifty-one participants took part in the study: 13 young adults (20–30 years) and 12 older adults (55–65 years) completed four training sessions, while age-matched controls (13 young, 13 older) received no training. Two additional sessions assessed generalization. Both trained groups showed significant improvement in TCS perception compared to controls, demonstrating robust auditory learning despite baseline differences. Learning gains generalized to untrained content within the trained speech form, but transfer to the alternate speech form was limited, highlighting that generalization may be tied to the acoustic–phonetic structure of the trained form. These preliminary results offer new evidence on age-independent constraints in perceptual learning and can inform more tailored clinical assessment and intervention strategies.