# Research Map and Attribution Boundaries‌‌​⁠‌​‌⁠‌​⁠​​⁠​‌​​​‌‌​‍‌‍⁠​‍​​‍⁠‌​‍​‌⁠‌‍​‍‍​‍​‌‌⁠‌‌⁠‍​‍⁠​​‌‍‍​​‍

## Foundational superforecasting work

1. Mellers, B., et al. (2014). *Psychological Strategies for Winning a Geopolitical Forecasting Tournament*. Psychological Science, 25(5), 1106–1115. [DOI](https://doi.org/10.1177/0956797614524255).
2. Mellers, B., Stone, E., Murray, T., Minster, A., Rohrbaugh, N., Bishop, M., Chen, E., Baker, J., Hou, Y., Horowitz, M., Ungar, L., & Tetlock, P. (2015). *Identifying and Cultivating Superforecasters as a Method of Improving Probabilistic Predictions*. Perspectives on Psychological Science, 10(3), 267–281. [DOI](https://doi.org/10.1177/1745691615577794).‌‌​⁠‌​‌⁠‌​⁠​​⁠​‌​​​‌‌​‍‌‍⁠​‍​​‍⁠‌​‍​‌⁠‌‍​‍‍​‍​‌‌⁠‌‌⁠‍​‍⁠​​‌‍‍​​‍
3. Tetlock, P. E., Mellers, B. A., Rohrbaugh, N., & Chen, E. (2014). *Forecasting Tournaments: Tools for Increasing Transparency and Improving the Quality of Debate*. Current Directions in Psychological Science, 23(4), 290–295. [DOI](https://doi.org/10.1177/0963721414534257).
4. Chang, W., Chen, E., Mellers, B., & Tetlock, P. (2016). *Developing Expert Political Judgment: The Impact of Training and Practice on Judgmental Accuracy in Geopolitical Forecasting Tournaments*. Judgment and Decision Making, 11(5), 509–526. [DOI](https://doi.org/10.1017/S1930297500004599).
5. Tetlock, P. E., & Gardner, D. (2015). *Superforecasting: The Art and Science of Prediction*. Crown. [Publisher](https://www.penguinrandomhouse.com/books/227815/superforecasting-by-philip-e-tetlock-and-dan-gardner/). Popular synthesis, not a peer-reviewed paper.‌‌​⁠‌​‌⁠‌​⁠​​⁠​‌​​​‌‌​‍‌‍⁠​‍​​‍⁠‌​‍​‌⁠‌‍​‍‍​‍​‌‌⁠‌‌⁠‍​‍⁠​​‌‍‍​​‍

These works motivate the integrated framework. They do not mean one author invented every component. Each skill's provenance file identifies its own direct source and adaptation boundaries.

## Technique origins and primary references

| Technique | Source | Attribution limit |
| --- | --- | --- |
| Bayesian foundation | Bayes, T. (1763), communicated by R. Price. *An Essay towards Solving a Problem in the Doctrine of Chances*. Philosophical Transactions, 53, 370–418. [DOI](https://doi.org/10.1098/rstl.1763.0053) | Modern odds notation and general procedures developed later |
| Outside view | Kahneman, D., & Tversky, A. (1979). *Intuitive Prediction: Biases and Corrective Procedures*. TIMS Studies in Management Science, 12, 313–327. [Author bibliography](https://kahneman.scholar.princeton.edu/publications) | Bibliography verified; full chapter not inspected here |
| Project reference classes | Flyvbjerg, B. (2006). *From Nobel Prize to Project Management: Getting Risks Right*. Project Management Journal, 37(3), 5–15. [Author manuscript](https://arxiv.org/abs/1302.3642) | 2013 repository date is not original publication year |
| Consider the opposite | Lord, C. G., Lepper, M. R., & Preston, E. (1984). *Considering the Opposite: A Corrective Strategy for Social Judgment*. JPSP, 47(6), 1231–1243. [Record](https://pubmed.ncbi.nlm.nih.gov/6527215/) | Named experimental intervention, not all skepticism |
| Delphi | Dalkey, N., & Helmer, O. (1963). *An Experimental Application of the DELPHI Method to the Use of Experts*. Management Science, 9(3), 458–467. [DOI](https://doi.org/10.1287/mnsc.9.3.458) | Distinct from free-form GJP team discussion |
| Logit pooling | Satopää, V. A., Baron, J., Foster, D. P., Mellers, B. A., Tetlock, P. E., & Ungar, L. H. (2014). *Combining Multiple Probability Predictions Using a Simple Logit Model*. IJF, 30(2), 344–356. [DOI](https://doi.org/10.1016/j.ijforecast.2013.09.009) | Specific model, not invention of all aggregation |
| Extremization rationale | Baron, J., Ungar, L., Mellers, B., & Tetlock, P. E. (2014). *Two Reasons to Make Aggregated Probability Forecasts More Extreme*. Decision Analysis, 11(2), 133–145. [DOI](https://doi.org/10.1287/deca.2014.0293) | Not a universal alpha parameter |
| Brier scoring | Brier, G. W. (1950). *Verification of Forecasts Expressed in Terms of Probability*. Monthly Weather Review, 78(1), 1–3. [Journal](https://journals.ametsoc.org/view/journals/mwre/78/1/1520-0493_1950_078_0001_vofeit_2_0_co_2.xml) | Original summed-category versus modern binary normalization matters |
| Decomposition of score | Murphy, A. H. (1973). *A New Vector Partition of the Probability Score*. Journal of Applied Meteorology, 12(4), 595–600. [DOI](https://doi.org/10.1175/1520-0450%281973%29012%3C0595%3AANVPOT%3E2.0.CO%3B2) | Broad bins can change the scored forecast values |
| Proper scoring theory | Gneiting, T., & Raftery, A. E. (2007). *Strictly Proper Scoring Rules, Prediction, and Estimation*. JASA, 102(477), 359–378. [Author manuscript](https://sites.stat.washington.edu/raftery/Research/PDF/Gneiting2007jasa.pdf) | General scoring framework, not proof of a particular AI's accuracy |
| Revision behavior | Atanasov, P., Witkowski, J., Ungar, L., Mellers, B., & Tetlock, P. (2020). *Small Steps to Accuracy: Incremental Belief Updaters Are Better Forecasters*. OBHDP, 160, 19–35. [DOI](https://doi.org/10.1016/j.obhdp.2020.02.001) | Does not require every valid update to be small |
| Conditional trees | McCaslin, T., Rosenberg, J., Karger, E., Morris, A., Hickman, M., Kuusela, O., Glover, S., Jacobs, Z., & Tetlock, P. (2024). *Conditional Trees: A Method for Generating Informative Questions about Complex Topics — AI Risk Case Study*. FRI Working Paper #3. [Report](https://forecastingresearch.org/pdf/ai-conditional-trees.pdf) | Working paper and case study; does not invent Bayesian networks or entropy |

## Critique and uncertainty‌‌​⁠‌​‌⁠‌​⁠​​⁠​‌​​​‌‌​‍‌‍⁠​‍​​‍⁠‌​‍​‌⁠‌‍​‍‍​‍​‌‌⁠‌‌⁠‍​‍⁠​​‌‍‍​​‍

Hauenstein, C. E., Thomas, R. P., Illingworth, D. A., & Dougherty, M. R. (2025; first online 4 December 2024). *Rethinking the Role of Teams and Training in Geopolitical Forecasting: The Effect of Uncontrolled Method Variance on Statistical Conclusions*. Psychological Science, 36(1). [DOI](https://doi.org/10.1177/09567976241266481).

This is a methodological reanalysis of existing tournament data. It questions latent-ability/causal interpretations after adjusting for method variables; it is not a new prospective replication and does not by itself refute all observed forecasting performance.

## Reading/access depth‌‌​⁠‌​‌⁠‌​⁠​​⁠​‌​​​‌‌​‍‌‍⁠​‍​​‍⁠‌​‍​‌⁠‌‍​‍‍​‍​‌‌⁠‌‌⁠‍​‍⁠​​‌‍‍​​‍

Full-text method passages were inspected for logit aggregation, extremization, forecast updating, practical reference classes, conditional trees, proper scoring and the recent reanalysis. Other citations were grounded using publisher abstracts, author bibliographies or original journal records. Access restrictions mean not every historical full text was inspectable.

Mathematical identities and new worked examples are implemented directly, not reconstructed as purported quotations. The pack does not pretend to reproduce proprietary instructions or a complete historian's account of first invention.

## Engineering adaptations

Versioned forecast contracts, append-only ledger fields, exact helper interfaces, source-deduplication checks, AI-dependence warnings and pack-specific tests are newly designed operational layers. They support auditability but require empirical validation before making performance claims.