Next-turn recovery from impasses declines steadily as impasse depth increases, with help-seeking impasses declining most sharply¶

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

Subclaims¶

q3 i? Each additional impasse turn lowered the odds of next-turn recovery by about 13% (AOR = 0.87, 95% CI [0.81, 0.94], p<.001), and raw recovery fell from 39.7% at onset to 18.8% at depth 6 or more. → Ahtisham 2026

Evidence¶

Ahtisham 2026¶

Ahtisham, B., Vanacore, K., Napoli, A., Arens, J., Ionova, K., Cohn, C., Salehi, S., & Kizilcec, R. (2026). Examining Variation in How Guided AI Tutors Resolve Student Impasses. arXiv. https://arxiv.org/abs/2609.38346 doi:10.48550/arxiv.2609.38346

q3 · i? · associational · r2

Descriptive next-turn recovery rates by depth from 6,630 impasse turns, with 95% CIs from a session-level cluster bootstrap (Figure 1). A multivariable logistic model on N=6,163 turns found "each additional turn at an impasse was associated with 13% lower odds of recovery (AOR = 0.87, 95% CI [0.81, 0.94], p<.001)". No standardized effect size beyond the odds ratio is printed.

"Across all impasse types, next-turn recovery fell steadily as impasses persisted, from 39.7% at onset to 32.6% at depth 2, 26.2% at depth 3, and 18.8% at depth 6 or more, less than half the onset rate"

Ahtisham 2026 (2)¶

Ahtisham, B., Vanacore, K., Napoli, A., Arens, J., Ionova, K., Cohn, C., Salehi, S., & Kizilcec, R. (2026). Examining Variation in How Guided AI Tutors Resolve Student Impasses. arXiv. https://arxiv.org/abs/2609.38346. doi:10.48550/arxiv.2609.38346

q3 · i? · associational · r2

Type-specific recovery trajectories from the same corpus show help-seeking impasses were most recoverable at onset but declined most sharply, while uncertainty remained comparatively recoverable through depth 4 (33.3% at depth 4).

"Help-seeking was initially the most recoverable state (47.0% at onset) but declined most sharply once initial assistance had failed (25.3% at depth 3; 12.5% at depth≥6)."

Discussion¶

Related Claims¶

• Student impasses in LLM chemistry tutoring concentrate in the relevant-concepts planning phase, and impasse type depends strongly on problem-solving practice — related
• Prompt specificity changes AI tutor pedagogy: a no-direct-answer guardrail alone produces uniform questioning, while a detailed guided protocol produces varied, impasse-sensitive responses — related
• The association between tutor questioning and impasse recovery decays as impasses persist, while addressing the student's error becomes more beneficial — related
• After a failed scripted question, repeating it is followed by lower recovery than switching to addressing the student's error — related
• Early dropouts are trapped in recursive concept-elicitation loops before reaching problem execution — related