Computational model for behavior shaping as an adaptive health intervention strategy

Area of Science:

• Behavioral Science
• Computational Psychology
• Digital Health

Background:

• Adaptive behavioral interventions (JITAI) leverage real-time sensing but often lack theoretical grounding.
• Traditional interventions face limitations due to imprecise procedures and transient effects.
• Enhancing theoretical fidelity in JITAI is crucial for improving intervention effectiveness.

Purpose of the Study:

• To explore the use of shaping, a behavioral theory principle, as a JITAI adaptation strategy.
• To computationally model and investigate the optimal parameters for behavior shaping in adaptive interventions.

Main Methods:

• A computational model of behavior dynamics and operant conditioning was modified to include behavior shaping.
• Digital experiments were conducted using the updated model to test various shaping parameters.
• The model simulated varying the range of reinforced behaviors over time to identify optimal strategies.

Main Results:

• Continuous narrowing of the reinforcement window over time yielded better results than discrete narrowing.
• A rapid initial narrowing followed by slower decreases in reinforcement window size was most effective.
• The computational shaping model demonstrated efficacy in generating target behaviors.

Conclusions:

• Shaping can be effectively implemented as a JITAI adaptation strategy.
• Computational modeling provides a valuable tool for investigating and optimizing JITAI adaptation.
• Further research is needed to validate these computational findings in real-world experimental settings.