Learning variable and stereotypical sequences of responses: Some data and a new model.

Area of Science:

• Behavioral neuroscience
• Animal behavior studies
• Operant conditioning

Background:

• Frequency-dependent reinforcement schedules shape animal behavior.
• Understanding response patterns is key to reinforcement learning models.

Purpose of the Study:

• To investigate pigeon behavior under novel frequency-dependent reinforcement schedules.
• To test a linear operator model's ability to predict response patterns.

Main Methods:

• Pigeons were trained in a left-right choice task.
• Schedules selectively reinforced least-frequent response pairs or triplets.
• Behavioral patterns were analyzed and compared to a computational model.

Main Results:

• Reinforcing least-frequent pairs induced double alternation (e.g., RRLL).
• Reinforcing least-frequent triplets led to random-like behavior in most pigeons.
• A linear operator model accurately predicted these emergent patterns.

Conclusions:

• Pigeon behavior adapts predictably to frequency-dependent reinforcement.
• The findings support the utility of mathematical models in explaining complex learned behaviors.
• Response complexity influences the resulting behavioral patterns.