Adaptive purchase tasks in the operant demand framework.

Area of Science:

• Behavioral economics
• Operant conditioning
• Quantitative psychology

Background:

• Quantifying reinforcer effects on behavior is crucial.
• Traditional operant demand models often use nonlinear regression (e.g., Hursh & Silberberg, 2008).
• These models have strengths but also trade-offs in application.

Purpose of the Study:

• Introduce and demonstrate an adaptive task for operant demand assessment.
• Elucidate key features of operant demand without relying on nonlinear regression.
• Estimate participant-level peak work (P_MAX) using a novel approach.

Main Methods:

• Developed an adaptive task employing reinforcement learning.
• Systematically guided questioning to estimate P_MAX.
• Evaluated the algorithm across four iteration lengths (5, 10, 15, 20 questions).
• Used equivalence testing with simulated agent responses.

Main Results:

• Tasks with five or more sequentially updated questions yielded P_MAX values statistically equivalent to seeded values.
• Demonstrated that nonlinear regression may not be essential for analyzing reinforcer consumption.
• Showcased the ability to reveal how consumption scales with price.

Conclusions:

• An adaptive, reinforcement learning-based task can effectively estimate P_MAX.
• This approach offers a viable alternative to complex nonlinear regression for operant demand analysis.
• The findings support simpler methods for understanding consumer demand and reinforcer scaling.