Reanalysis of the equation for simple action

Area of Science:

• Behavioral science
• Psychology
• Animal behavior

Background:

• The matching law and its derived simple action equation have been influential in predicting behavioral output.
• Previous work by De Villiers and Herrnstein (1976) reported high variance accommodation for this equation across various experiments.

Purpose of the Study:

• To re-evaluate the predictive accuracy of the simple action equation.
• To investigate the impact of data averaging on the equation's fit.
• To explore potential improvements or limitations of the simple action model.

Main Methods:

• Reanalyzed data from positive-reinforcement studies cited by De Villiers and Herrnstein (1976).
• Minimized averaging of behavioral output rates before applying least-squares fitting.
• Introduced parameters to allow the equation's scaling constant to vary with reinforcement value.

Main Results:

• Minimizing rate averaging reduced data variance accommodation from 94% to 78%.
• Adding parameters improved data variance accommodation but lacked clear psychological interpretation.
• The simple action equation showed diminished predictive power without rate averaging.

Conclusions:

• The simple action equation, as applied, may be an inadequate model for behavioral output.
• Averaging behavioral rates inflates the apparent fit of the simple action equation.
• Further research is needed to develop more robust models of behavioral output.