Prediction in a visual language: real-time sentence processing in American Sign Language across development

Area of Science:

• Cognitive Science
• Linguistics
• Developmental Psychology

Background:

• Sign language comprehension involves integrating visual linguistic and non-linguistic information.
• Predictive processing is a key aspect of language understanding across modalities.

Purpose of the Study:

• To investigate semantic prediction in American Sign Language (ASL) comprehension in deaf adults and children.
• To explore how visual attention is allocated during predictive sign language processing.

Main Methods:

• Two eyetracking experiments using a visual world paradigm.
• Participants viewed ASL sentences with sentence-initial verbs that were either neutral or constrained relative to a target noun.
• Gaze patterns were analyzed to identify anticipatory looks and visual attention shifts.

Main Results:

• Adults and children exhibited anticipatory looks to target pictures in the constrained condition, indicating semantic prediction.
• Gaze alternated between the sign stimulus and target picture specifically when the object was predictable from the verb.
• Signers optimized visual attention between linguistic and referential signals during prediction.

Conclusions:

• Semantic prediction is evident in ASL comprehension among deaf adults and children.
• Predictive processing in sign language involves strategic visual attention allocation.
• Findings support the hypothesis that prediction is a modality-independent cognitive function.