Bayesian inference, reformulated as Q-learning, has symmetric decreasing learning rates, and fitting constant-rate Q-learning to such optimal behavior can spuriously produce positivity and confirmation biases.
Information about action outcomes differentially affects learning from self-determined versus imposed choices
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Bias or Optimality? Disentangling Bayesian Inference and Learning Biases in Human Decision-Making
Bayesian inference, reformulated as Q-learning, has symmetric decreasing learning rates, and fitting constant-rate Q-learning to such optimal behavior can spuriously produce positivity and confirmation biases.