POSTER: Overclaiming as Convenient Proxy for Confidence and Overconfidence
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Introduction Ideally, measuring over confidence (specifically, overestimation) requires an objective measure of ability to contrast with self-estimates. Administering such parallel tests can costly. Furthermore, if overconfidence is measured as a difference, it remains confounded by the reference ability and self-estimates. Overclaiming is an efficient, unobtrusive technique that compares claiming familiarity of genuine items (Reals) against claiming familiarity with fake items (Foils). Objectives To explore how the overclaiming technique distinguishes between respondents’ actual knowledge and their perceived knowledge as predictors of academic performance. Methodology Undergraduate students were given both a vocabulary test and then an overclaiming measure of vocabulary, as well as other knowledge tests which included confidence ratings for general knowledge items. Their overall grades in an introductory psychology course were also collected (not self-report). Overconfidence was calculated as the difference between a respondent’s standardized confidence ratings and their standardized correct knowledge score. Results Vocabulary ability measured by Overclaiming (Reals-claiming minus Foils-claiming) strongly correlated with a conventional vocabulary ability measure ( r (163) = .77***, CI .95 = [.69, .82]), and moderately with course grade, r (163) = .37***, CI .95 = [.23, .50]. Overconfidence was necessarily confounded by confidence and general knowledge measures (e.g. r = .55), yet negatively predicted course grade, r (162) = -.30***, CI .95 = [-.43, -.15]. Reals-claiming captured confidence ( r (162) = .33***, CI .95 = [.19, .46]) and Foils-claiming captured overconfidence, r (162) = .24**, CI .95 = [.08, .37]. Regression models showed no overlap between the two associations. Conclusions Assessing overconfidence via combined ability and reported confidence is onerous and yields confounded measures which can’t be combined in predictive regression models. Our vocabulary overclaiming technique, despite covering a different knowledge area, captures general confidence and overconfidence in a way that can be combined for predicting academic outcomes. NOTE: * p < .05, ** p < .01, *** p < .001
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.007 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it