A Magic Act in Causal Reasoning: Making Markov Violations Disappear
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
A desirable property of any theory of causal reasoning is to explain not only why people make causal reasoning errors but also when they make them. The mutation sampler is a rational process model of human causal reasoning that yields normatively correct inferences when sufficient cognitive resources are available but introduces systematic errors when they are not. The mutation sampler has been shown to account for a number of causal reasoning errors, including Markov violations, the phenomenon in which human reasoners treat causally related variables as statistically dependent when they are normatively independent. A Markov violation arises, for example, when an individual reasoning about a causal chain X→Y→Z treats X as informative about the state of Z even when the state of Y is known. Recently, the mutation sampler was used to predict the existence of previously untested experimental conditions in which the sign of Markov violations would switch from positive to negative. Here, it was used to predict the existence of conditions in which Markov violations should disappear entirely. In fact, asking subjects to reason about a novel causal structure with nothing but generative causal relations (a cause makes its effect more likely) resulted in Markov violations in the usual positive direction. But simply describing one of four causal relations as inhibitory (the cause makes its effect less likely) resulted in the elimination of those violations. Theoretical model fitting confirmed how this novel result is predicted by the mutation sampler.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| 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) | 0.000 | 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