Preference for fixed-interval schedules of reinforcement

Area of Science:

• Behavioral psychology
• Animal cognition
• Operant conditioning

Background:

• The matching hypothesis predicts response distribution based on reinforcement rates.
• Concurrent chain schedules are used to study choice behavior and interval timing.
• Pigeons are a common model organism for studying learning and decision-making.

Purpose of the Study:

• To investigate pigeon choice behavior on a concurrent chain schedule with varying fixed-interval (FI) schedules.
• To examine deviations from the matching hypothesis in initial link responding.
• To assess the influence of reinforcement history (recency effect) on choice.

Main Methods:

• Pigeons were trained on a two-link concurrent chain schedule.
• Initial link choices led to terminal links with food reinforcement on fixed-interval (FI) schedules (20s vs. 5, 14, 30, or 60s).
• Response rates in the initial link were recorded and analyzed.

Main Results:

• Pigeons consistently chose the key associated with the shorter fixed interval (FI) in the initial link, over-selecting it compared to matching predictions.
• A significant negative recency effect was observed: pigeons responded less on the key that last delivered reinforcement.
• Deviations from the matching hypothesis were pronounced, particularly with shorter FIs.

Conclusions:

• Pigeon choice behavior in concurrent chain schedules is influenced by factors beyond simple reinforcement rate matching, including interval length.
• Reinforcement history, specifically recent reinforcement, significantly impacts immediate choice behavior (negative recency).
• These findings highlight the complexity of decision-making processes and limitations of basic matching models in explaining animal behavior.