Anomalous Evidence, Confidence Change, and Theory Change

Area of Science:

• Cognitive Psychology
• Philosophy of Science

Background:

• Understanding how individuals update their beliefs when confronted with contradictory information is crucial.
• Previous research has explored factors influencing belief revision, but a nuanced understanding of confidence shifts and theory change remains.
• An experimental paradigm measuring theory change and confidence offers a novel approach to investigate these cognitive processes.

Purpose of the Study:

• To experimentally investigate the effects of anomalous evidence on theory change and confidence.
• To examine how the amount, type (convergent vs. replicating), and explanatory context (alternative theory) of anomalous evidence influence confidence and theory revision.

Main Methods:

• Three experiments utilized a novel paradigm to measure participants' theory change and confidence levels.
• Experiment 1 manipulated the amount of anomalous evidence.
• Experiments 2 and 3 manipulated the nature of anomalous evidence (convergent vs. replicating) and the availability of an alternative theory, respectively.

Main Results:

• Confidence changes were directly proportional to the amount of anomalous evidence presented.
• Larger decreases in confidence significantly predicted theory changes.
• Convergent anomalous evidence and the provision of an alternative theory resulted in greater confidence shifts and more theory changes.
• Even without complete theory change, anomalous evidence and alternative theories reduced confidence in existing beliefs, facilitating incremental shifts.

Conclusions:

• The amount and nature of anomalous evidence, along with the presence of alternative explanations, play a critical role in belief revision.
• Confidence reduction is a key mediator in the process of theory change.
• These findings have implications for understanding scientific progress and individual learning when encountering challenging data.