Knowledge-based and signal-based cues are weighted flexibly during spoken language comprehension.

Area of Science:

• Psycholinguistics
• Cognitive Science
• Speech Processing

Background:

• Listeners utilize both knowledge-based and signal-based information for spoken language comprehension.
• The integration and weighting of cues from different representational levels remain underexplored.

Purpose of the Study:

• To investigate the flexible weighting and integration of morphosyntactic gender (knowledge-based) and speech rate (signal-based) cues.
• To determine if these cues are integrated online during spoken language comprehension.

Main Methods:

• An eye-tracking experiment utilizing the visual world paradigm.
• Manipulation of morphosyntactic gender marking and contextual speech rate.

Main Results:

• Participants immediately used morphosyntactic gender cues for referent prediction, even with cue uncertainty.
• Speech rate normalization effects were observed in gaze patterns, irrespective of preceding morphosyntactic information.
• Rate normalization effects persisted even in participants favoring the morphosyntactic gender cue, suggesting robustness.

Conclusions:

• Cues are weighted and integrated flexibly online, challenging strict hierarchical models.
• Speech rate normalization appears robust and potentially automatic during spoken language processing.
• Findings inform theories of cue integration and acoustic context effects in language comprehension.