Punishment in human choice: direct or competitive suppression?

Area of Science:

• Behavioral psychology
• Operant conditioning
• Decision-making models

Background:

• Two models, Deluty's competitive-suppression and de Villiers' direct-suppression, explain reinforcement and punishment integration in operant choice.
• Previous research tested these models in nonhumans, but not at the individual human level.

Purpose of the Study:

• To compare the predictive accuracy of the competitive-suppression and direct-suppression models for human operant choice.
• To provide the first individual-subject evidence on human punishment integration models.

Main Methods:

• College students' mouse clicking was reinforced with money under a two-alternative concurrent schedule.
• Punishment was introduced as variable-interval money losses, with qualitative and quantitative model comparisons.
• Generalized matching law and least-squares regression were used for quantitative analysis.

Main Results:

• The direct-suppression model was supported by 12 of 14 individual-subject functions in qualitative comparisons.
• Quantitative analyses using the generalized matching law showed superior fits for the direct-suppression model in 6 of 7 subjects.
• Further quantitative tests in Experiment 3 favored the direct-suppression model in 11 of 13 cases.

Conclusions:

• The direct-suppression model provides a better account of human punishment in operant choice than the competitive-suppression model.
• Results align with nonhuman investigations, offering robust individual-subject evidence for the direct-suppression model.
• Findings have implications for refining punishment models and future human choice research.