Cognitive dissonance reduction as constraint satisfaction

Area of Science:

• Cognitive Psychology
• Computational Neuroscience
• Artificial Intelligence

Background:

• Cognitive dissonance theory explains discomfort from conflicting beliefs or behaviors.
• Existing models struggle to fully capture nuances in dissonance paradigms like insufficient justification and free choice.

Purpose of the Study:

• To introduce and evaluate a constraint satisfaction neural network model (consonance model) for simulating cognitive dissonance.
• To compare the consonance model's performance against traditional cognitive dissonance theory.
• To explore the relationship between dissonance reduction and other psychological processes.

Main Methods:

• Simulated data from insufficient justification and free choice dissonance paradigms using the consonance model.
• Compared model fits to human data against predictions from cognitive dissonance theory.
• Conducted a new psychological experiment to test model predictions for free choice scenarios.

Main Results:

• The consonance model demonstrated superior fits to human data in several cases compared to cognitive dissonance theory.
• Model's enhanced precision and inclusion of additional constraints accounted for improved performance.
• Experimental validation confirmed the model's predictions for choices between undesirable alternatives.

Conclusions:

• The consonance model offers a more precise computational account of cognitive dissonance.
• Unforeseen similarities exist between dissonance reduction and other psychological processes constrained by beliefs and attitudes.
• This framework advances understanding of decision-making and belief consistency.