Sampling Assumptions Affect Use of Indirect Negative Evidence in Language Learning

Area of Science:

• Cognitive Science
• Psycholinguistics
• Computational Linguistics

Background:

• Children acquire complex language rules without explicit negative evidence.
• Probabilistic models frame language learning as statistical inference from sentence samples.
• Learners' sensitivity to input sampling is a key aspect of language acquisition.

Purpose of the Study:

• Investigate how learners utilize sampling assumptions in language acquisition.
• Determine if learners can adapt to different sampling distributions (strong vs. weak).
• Explore the role of input presentation in guiding learning strategies.

Main Methods:

• Adult participants completed artificial language learning experiments.
• Exposure to controlled linguistic input under varying sampling conditions.
• Behavioral analysis to assess learning outcomes and strategy use.

Main Results:

• Participants' behavior aligned with strong sampling assumptions when input reflected underlying distributions.
• Participants' behavior aligned with weak sampling assumptions when input was less constrained.
• Learning strategies were adaptable based on the presented sampling method.

Conclusions:

• Human language learners are sensitive to the statistical properties of their input.
• Learners can flexibly employ different sampling assumptions (strong and weak) to guide acquisition.
• Understanding input sampling is crucial for explaining how complex linguistic knowledge is attained.