Beyond Bias Minimization: Improving Intelligence with Statistical Optimization and Human Augmentation
Bibliographic record
Abstract
For the last half-century, the US and Allied intelligence community has sought to minimize the ostensibly detrimental effects of cognitive biases on intelligence practice. The dominant approach to doing so has been to develop structured analytic techniques (SATs), teach them to analysts in brief training sessions, provide the means to use SATs on the job and hope that they work. The SAT approach, however, suffers from serious conceptual problems and a paucity of support from scientific research. For example, a highly promoted SAT—the Analysis of Competing Hypotheses—was shown in several recent studies to either not improve judgment quality or to make it worse. This article recaps the key problems with the SAT approach and sketches some alternative interventions. At the core of these proposals is the idea that intelligence agencies should be focused broadly on improving intelligence and not narrowly on minimizing bias. While the latter contributes to achieving the former, over-emphasis on bias minimization could inadvertently bias agencies toward a singular form of intervention, blinding them from potentially more effective interventions. In this article, two lines of alternative intervention are sketched. The first line focuses on post-analytic statistical optimization methods such as recalibration and performance-weighted aggregation of analysts’ judgments. The second line focuses on a broad human augmentation program aimed at optimizing human cognition through better sleep, exercise, nutrition (including the use of nootropic compounds), and biometric tracking. Both lines of effort would require substantial scientific investment by the intelligence community to examine risks and efficacy.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.028 | 0.076 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.008 |
| Scholarly communication | 0.004 | 0.007 |
| Open science | 0.002 | 0.005 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".