Lexical adoption shows no significant condition effect after controlling for input volume, but AI-first participants adopted more words at the second dialogue position¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2 · q3 · i2 medium
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? After accounting for input volume and learner output, condition does not reliably predict lexical adoption (β=−1.77, p=.062). → Russell Scheinberg 2026 q3 i2 At the second dialogue position, AI participants adopted significantly more words (M=4.6 vs. 2.6, d=0.72, p=.003); at position 1 the direction was consistent but nonsignificant (d=0.42, p=.079). → Russell Scheinberg 2026 (2)

Evidence¶

Russell Scheinberg 2026¶

Russell Scheinberg, Ameeta Agrawal, Tetyana Sydorenko, Kalab Kahsay, Nina Vyatkina, Griet Boone. (2026). What Changes When the Interlocutor Is an AI? Interactional Fluency and Linguistic Uptake in L2 Spoken Dialogue. https://arxiv.org/abs/2606.22225 doi:10.48550/arxiv.2606.22225

q3 · i? · causal · r2

Mixed-effects regression on lexical adoption from the within-participant study; the AI produced over twice as many novel lemmas (M=79.6 vs. 36.4). "Pool size (β= 0.10,p < .001) and learner word count (β= 0.005,p <.001) were the significant predictors".

"However, a mixed-effects regression con- trolling for pool size and learner word count found no sig- nificant condition effect (β=−1.77,SE= 0.95,z=−1.87, p=.062)."

Russell Scheinberg 2026 (2)¶

Russell Scheinberg, Ameeta Agrawal, Tetyana Sydorenko, Kalab Kahsay, Nina Vyatkina, Griet Boone. (2026). What Changes When the Interlocutor Is an AI? Interactional Fluency and Linguistic Uptake in L2 Spoken Dialogue. https://arxiv.org/abs/2606.22225

q3 · i2 · causal · r2

Position-matched Welch's t-tests comparing adoption at the same sequential position to disentangle partner type from task order (Group 1 n=46, Group 2 n=32). The effect held at position 2 but not position 1.

"In position 2—where the depleted vocabulary pool provides a more sensitive test—AI participants adopted significantly more words (M= 4.6 vs. 2.6;d= 0.72,p=.003). In position 1, the direction is consistent but nonsignificant (d= 0.42,p=.079)."

Discussion¶

Related Claims¶

• Learners show stronger short-term syntactic priming with the AI partner, significant across all three recency-window operationalizations including one equalizing input volume — related
• Interactive learning tasks produce larger incidental spoken word-learning gains than noninteractive tasks — related
• Attitudes toward AI improved after direct experience, and satisfaction was predicted by within-turn production fluency rather than linguistic uptake — related
• L2 proficiency positively predicts L2 incidental vocabulary acquisition through reading — related
• L2 learners pick up vocabulary incidentally from meaning-focused input, with reading the most-studied mode and listening yielding similar proportions of words learned — related