Computational behavior dynamics: an alternative description of Nevin (1969)

Area of Science:

• Behavioral science
• Computational modeling
• Psychology

Background:

• Nevin's (1969) discrete-trials experiment is a foundational study in behavioral research.
• Existing algebraic behavior theories have limitations in explaining complex response dynamics.

Purpose of the Study:

• To develop and instantiate a computational processing behavior-dynamic model.
• To evaluate the model's ability to reproduce key findings from Nevin's task.
• To explore the model as a stochastic alternative to algebraic behavior theory.

Main Methods:

• A computer program was created to simulate a behavior-dynamic model.
• The model's performance was assessed against established results from Nevin's (1969) discrete-trials experiment.
• Qualitative reproduction of experimental outcomes was the primary evaluation criterion.

Main Results:

• The model qualitatively reproduced three key findings from Nevin's task: matching of response and reinforcement frequencies, inverse relationship between changeover probability and run length, and independent variation of response rates.
• The model did not quantitatively replicate the results in specific detail.
• The model generated novel, testable predictions for future research.

Conclusions:

• The developed behavior-dynamic model offers a viable stochastic alternative to traditional algebraic behavior theories.
• Computational models can capture complex behavioral dynamics observed in classic experiments.
• Further research is needed to quantitatively validate the model and its predictions.