A design-enactment gap in cognitive demand: 37.7% of conversations under-reached the teacher-targeted DOK level, and under-targeting increased with target ambition¶

Claim · All claims
Evidence · 1 study (2 entries) · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? 37.7% of conversations showed students performing below target DOK, with a mean DOK gap of +0.39 (SD = 0.81). → Alex Liu 2026 q2 i? When teachers targeted DOK 3, 46.0% of conversations failed to reach that level, whereas targeting DOK 1 produced a negative mean gap (−0.58). → Alex Liu 2026 (2)

Evidence¶

Alex Liu 2026¶

Alex Liu, Min Sun, Lief Esbenshade, Victor Tian, Zachary Zhang, and Kevin He. (2026). Teacher-Authored Prompts for Configuring StudentAI Dialogue: K12 Classroom Implementation. https://arxiv.org/abs/2604.16738 doi:10.48550/arxiv.2604.16738

q2 · i? · design · r2

Analysis of demonstrated versus targeted DOK across 1,362 linked conversations coded by LLM with human validation. The article reports "37.7% of conversations showed students performing below target DOK" and a mean gap of +0.39 (SD = 0.81).

"37.7% of conversations showed students performing below target DOK, while 52.9% achieved alignment and 9.4% exceeded targets (Appendix Figure 7). The mean DOK gap was +0.39 (SD = 0.81)"

Alex Liu 2026 (2)¶

Alex Liu, Min Sun, Lief Esbenshade, Victor Tian, Zachary Zhang, and Kevin He. (2026). Teacher-Authored Prompts for Configuring StudentAI Dialogue: K12 Classroom Implementation. https://arxiv.org/abs/2604.16738

q2 · i? · design · r2

Stratified analysis by teacher target DOK level (Table 4). The article reports 46.0% under-target for DOK 3 targets (N=824) and a negative mean gap of −0.58 for DOK 1 targets (N=67), indicating students exceeded recall targets.

"However, when teachers targeted DOK 3 (strategic thinking), nearly half of conversations (46.0%) failed to reach that level with many students demonstrating only DOK 2 engagement (Appendix Figure 8)."

Discussion¶

Related Claims¶

• In an OLS regression, explicit finish lines in teacher prompts were associated with smaller DOK gaps, while explain-reasoning epistemic framing was associated with larger gaps — related
• Teachers authoring student-AI dialogue prompts predominantly write highly specific tasks targeting higher-order thinking (92% at DOK 2–3) — related
• LLMs show a directional asymmetry in educational control: they reliably increase cognitive demand but struggle to lower it — related