In search of the feedback function for variable-interval schedules

Area of Science:

• Behavioral psychology
• Operant conditioning

Background:

• Variable-interval schedules are crucial for modeling reinforcement dependent on external factors.
• Existing feedback functions for these schedules fail to meet all theoretical and empirical criteria.
• A key theoretical requirement is that the function's slope must approach or equal 1.0 at the origin.

Purpose of the Study:

• To evaluate proposed feedback functions for variable-interval schedules.
• To address the unsolved problem of finding a theoretically sound and empirically valid feedback function.
• To investigate the empirical puzzle of high response rates at low reinforcement rates.

Main Methods:

• Evaluation of three feedback functions against empirical data sets.
• Comparison of a new, theoretically unjustified function with functions proposed by Prelec and Herrnstein (1978) and Nevin and Baum (1980).

Main Results:

• All three evaluated functions demonstrated good data-fitting capabilities.
• Theoretically incorrect functions (Prelec and Herrnstein, Nevin and Baum) fit the data well.
• An empirical puzzle emerged: low reinforcement rates correlated with unexpectedly high response rates.

Conclusions:

• The search for a theoretically sound and empirically correct feedback function for variable-interval schedules continues.
• The success of theoretically flawed models highlights a gap in understanding operant behavior at low reinforcement rates.
• Further research is needed to derive a function from fundamental principles of operant behavior.