Measuring Collaboration Load With Pupillary Responses - Implications for the Design of Instructions in Task-Oriented...

Area of Science:

• Cognitive Science
• Human-Computer Interaction
• Linguistics

Background:

• Speakers build common ground incrementally through fragmented utterances, not complete ones.
• Understanding the cognitive load implications of fragmented speech is crucial for effective communication.

Purpose of the Study:

• To investigate how fragmented utterance construction impacts cognitive load during production and comprehension.
• To assess the effectiveness of fragmented speech in collaborative tasks.

Main Methods:

• Collaborative furniture assembly task involving participants instructing each other.
• Measurement of pupil diameter (pupillometry) to assess cognitive effort.
• Analysis of eye-gaze behavior.

Main Results:

• Fragmented utterance construction increased cognitive resources for speakers.
• Listeners' cognitive resources decreased with each new fragmented utterance, indicating successful ambiguity resolution.
• Adapting speech for listeners via fragmentation is cognitively demanding for speakers.

Conclusions:

• Fragmented speaking minimizes overall collaboration load but increases speaker effort.
• Audience design through fragmented utterances aids comprehension by reducing listener ambiguity.
• Findings have implications for designing task-oriented human-robot interactions and assistive robots.