Modeling juror decisions: a comparison of perceptions of innocence and guilt
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
The research in this dissertation investigates the consistency of juror decision models when evaluating incriminating and exonerating evidence. Many stochastic and psychological models indicate that an interaction between a person’s prior beliefs and their evaluation of the evidence contribute to their verdict decision. However, less is known regarding how this interaction occurs for different forms of evidence. In particular, a pervasive assumption is that jurors use the same or similar models to evaluate exonerating and incriminating evidence. The data from this dissertation indicates that this may not be the case. Participants adjust estimates of probability of guilt in a Bayesian inference problem more when case specific evidence is incriminating versus exonerating. Further, their response patterns and reported and observed measures of the type and variety of information they are using to produce probability of guilt estimates indicate that they engage in a process of mental estimation more often than they report that they do. The findings indicate that jurors may potentially use different decision models to evaluate different forms of evidence. Further, the framing of the search for culpability provides a plausible explanation for differences in the decision models that are used. Specifically, a juror’s selection criteria and perceived importance of a given piece of evidence will vary depending on its relevance to their decision task. Thus, asking jurors to estimate likelihoods of guilt may lead to their underutilization of evidence implying innocence.
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 | 0.006 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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