Building worlds of your own: the functional role of metacognitive bias in metacognition.

Metacognition, the capacity to monitor and evaluate one's own decisions, has become a central topic in psychological and neuroscientific research. While research has largely focused on metacognitive sensitivity, defined as the ability to discriminate between correct and incorrect decisions, considerably less attention has been devoted to metacognitive bias, defined as the systematic tendency to report higher or lower confidence irrespective of objective accuracy. Here, we argue that this imbalance has limited current theories of metacognition. Instead, greater focus on metacognitive bias can deepen our understanding of internal models of confidence, revealing how inter-individual differences give rise to distinct internal representations of the world, even when external information is processed in comparable ways. This perspective reframes metacognitive bias as a regulatory mechanism that balances flexibility and stability in self-evaluation, rather than a failure of calibration. Given its theoretical and empirical relevance, we emphasize the importance of estimating this construct within a Signal Detection Theory (SDT) framework. We propose the meta-criterion (meta-c), derived from the model introduced by Maniscalco and Lau, as a principled and quantitative index of metacognitive bias. Unlike purely descriptive measures, meta-c captures fine-grained individual differences along a continuous scale, opening to the possibility of defining more conservative versus more liberal confidence policies. We foresee metacognitive bias as one of the central axes along which the next generation of metacognitive research will develop.