Knowing What You Know: A Trait Theory of Reviewer Value in Human-AI Cognitive Systems
Organizations: Docyt Inc., San Jose, CA 95110 USA
Abstract
When a capable model and a human reviewer cooperate on the same decision, which human still adds value? We argue this is a question about a single cognitive trait, metacognitive calibration: the alignment between a reviewer's behavioral confidence and the actual correctness of their judgments. We develop a trait theory of reviewer value to make it precise. The theory makes five claims: value is governed by calibration rather than by activity, speed, agreement, or tenure; calibration leaves observable traces in behavioral telemetry; those traces are non-diagnostic on their own, because the fast, low-revision signature of a calibrated expert is indistinguishable from that of a confident novice until it is gated by outcomes; human calibration and a model's self-knowledge are the same construct, so the joint system should defer to whichever agent is better calibrated on each item; and individual calibration is causally identifiable only above the level of the single judgment. We ground each claim in a six-month production window of an AI-assisted accounting system (541621 reviewer events, 483 reviewers), and find that a purely behavioral expertise profile inverts the ground truth, scoring an overconfident cohort (5.2 percent correction survival) higher than genuine experts (99.2 percent); that reviewer quality is flat across tenure, so the trait is not learned on the job; and that the same calibration logic that separates good from bad reviewers, applied to the model, concentrates 92.5 percent of human effort on genuine errors. We close with the boundary the theory must respect: under majority-vote aggregation, individual causal value is unidentifiable.
Figures & tables
| Cohort (low self-revision) | Behavioral profile | Correction survival | |
|---|---|---|---|
| Expert (survival ) | 1,211 | 70.1 | 99.2% |
| Overconfident (survival ) | 577 | 76.2 | 5.2% |
| Tenure (weeks) | 0-1 | 1-2 | 2-4 | 4-8 | 8-16 | 16+ |
|---|---|---|---|---|---|---|
| (events) | 26,195 | 5,547 | 7,977 | 10,528 | 12,303 | 44,897 |
| Survival | 98.6% | 99.1% | 98.7% | 99.2% | 98.5% | 99.4% |
| Proxy | Computation | Range / weight |
|---|---|---|
| review-time score | step function of mean review seconds: s ; 30-300s ; 300-600s ; s | , weight 0.4 |
| self-correction rate | events the pair later unverified itself, over total verifications; enters as | weight 0.3 |
| consistency | over the run-length encoding of the per-event confirm/correct sequence | , weight 0.3 |
Human-AI Decision Making
Human-AI decision making studies how people and AI systems share, support, or delegate decisions across tasks and institutions.
See all 199 papers