What's in a Name? Interlocutors Dynamically Update Expectations about Shared Names.

Area of Science:

• Cognitive Science
• Linguistics
• Psychology

Background:

• Effective communication relies on shared knowledge, or common ground (CG).
• Speakers typically use referring expressions (e.g., names) that they believe their addressee recognizes.
• Shared learning experiences facilitate the use of common names, but this is not always the case in conversations.

Purpose of the Study:

• To investigate how speakers choose referring expressions when common ground is uncertain.
• To determine if interlocutors can infer a partner's knowledge of names from their referring expression choices.
• To explore the role of the name and description (N+D) form in updating beliefs about common ground.

Main Methods:

• Utilized Bayesian approaches to language processing as a theoretical framework.
• Conducted two experiments involving participants' use and interpretation of referring expressions.
• Analyzed participants' ability to estimate common ground and infer knowledge based on referring expression choices.

Main Results:

• Participants successfully used referring expression choices to estimate common ground more accurately.
• Interlocutors could infer a partner's knowledge of other names based on their use of the N+D form.
• The N+D form provides valuable evidence about the speaker's knowledge and potential shared understanding.

Conclusions:

• Interlocutors can generalize from referring expression choices to update beliefs about shared knowledge.
• Models of common ground need to consider the status of referring means, not just referents.
• Referring expression choice is a dynamic signal that shapes conversational common ground.