Transitive responding in animals and humans: Exaptation rather than adaptation?

Area of Science:

• Cognitive Science
• Animal Behavior
• Neuroscience

Background:

• Animals must navigate complex environments for survival and reproduction.
• Cognitive abilities, such as transitive inference, aid in this navigation.
• Transitive inference is the ability to deduce relationships, e.g., if A>B and B>C, then A>C.

Purpose of the Study:

• To investigate transitive inference abilities in non-human animals.
• To review and evaluate theories explaining transitive behavior.
• To present a learning model that simulates transitive responding.

Main Methods:

• A non-verbal multiple instrumental discrimination task was used for animals.
• Premise pairs (e.g., A+B-, B+C-) were used for training.
• Transitive responding was tested using novel pairs (e.g., B∘D∘).

Main Results:

• Pigeons, rats, monkeys, apes, and humans commonly exhibit transitive preferences.
• A simple algebraic learning model effectively simulates observed transitive behavior.
• Value transfer, while present, plays a minor role; classical conditioning is key.
• A modified model accounts for individuals failing to respond transitively.

Conclusions:

• Transitive responding can be explained by learning mechanisms, not necessarily complex logic.
• An algebraic learning model, convertible to a neural network, robustly simulates results.
• The evolution of transitive responding is linked to environmental demands and biological constraints.