Language play facilitates language learning: Optimizing the input for gender-like category induction

Area of Science:

• Psycholinguistics
• Cognitive Science
• Language Acquisition

Background:

• Gender assignment in language is challenging without semantic cues.
• Learners may use syntactic cues, like articles, to infer noun gender.
• Children's songs often present syntactic cues in a structured manner.

Purpose of the Study:

• To investigate gender-like category induction using artificial language.
• To determine the impact of input structure and modality on category learning.
• To assess the efficacy of structured, rhyming, and melodic input for language acquisition.

Main Methods:

• Participants learned an artificial language with syntactic gender cues.
• Training involved structured or unstructured input in prose, rhyme, melody, or rhyme-and-melody modes.
• Category induction was measured to assess learning effectiveness.

Main Results:

• Structured input significantly improved gender-like category induction compared to unstructured input.
• The Rhyme-and-Melody mode showed the highest learning gains, particularly with structured input.
• Rhyme-only and Melody-only modes yielded intermediate results, outperforming prose.

Conclusions:

• Highly structured, rhyming, and melodic input substantially facilitates gender-like category induction.
• This structured input approach supports the use of children's songs in language education.
• Syntactic cues, when presented effectively, are crucial for learning grammatical gender.