Two tests, same data, opposite conclusions. Wei, Nathoo & Masson show how model misspecification makes Bayes factors contradict ANOVA in repeated-measures designs, and random-slope models resolve the paradox! 🔗sciencedirect.com/science/ar…

MathPsych/ICCM 2026 starts this Friday! Join us July 17–21 at Le Centre Sheraton Montreal for workshops, keynotes, and the latest in mathematical & computational cognitive modeling. We can't wait to see everyone there! mathpsych.org/conference/23/

If we only see the choices people make, can we mathematically deduce their hidden internal rankings? Applying Shannon’s Maximum Entropy principle, this paper derives an exact algebraic formula for hidden ranking probabilities in random utility models. 🔗sciencedirect.com/science/ar…

How does the brain tell correlation from causation across time? Niels Verosky shows one small tweak to a classic associative learning rule lets the same machinery behind temporal difference learning compute causal contingencies, not just co-occurrences. sciencedirect.com/science/ar…

The connectivity hypothesis (stronger symptom interactions increase vulnerability to mental disorders) has been influential but never formally proven. Van der Mass & Waldop provide the first analytic derivation using mean-field Ising models of depression. sciencedirect.com/science/ar…

Standard correlation only captures linear relationships. A new tutorial introduces mutual information and higher-order measures that reveal what pairwise methods leave behind. 🔗sciencedirect.com/science/ar…

Can two completely different statistical models produce identical predictions? For knowledge assessment models that use multi-answer items, the answer is yes! New theoretical work maps out when,why and what it means for measuring what people truly know. sciencedirect.com/science/ar…