Spontaneous decisions and operant conditioning in fruit flies.

Area of Science:

• Neuroscience
• Behavioral Biology
• Learning Systems

Background:

• Historical debates in the 1930s by Skinner and Konorski explored similarities and differences between classical (Pavlovian) and operant (instrumental) conditioning.
• Operant conditioning involves complex interactions between multiple learning systems that are challenging to separate.
• Modern neurobiology offers tools to dissect and study the biological underpinnings of operant conditioning.

Purpose of the Study:

• To investigate the distinct biological processes involved in operant conditioning, including behavioral variability, world-learning, and self-learning.
• To understand the interactions between these learning systems.
• To elucidate the mechanisms of habit formation in operant learning.

Main Methods:

• Utilized neurobiological tools to dissect the biological processes of operant conditioning.
• Investigated behavioral variability generation and its role in environmental interaction detection.
• Examined the processes of world-learning (stimulus valuation) and self-learning (action valuation) in flies using visual stimuli.
• Analyzed the inhibitory role of world-learning on self-learning in the mushroom bodies and the conditions for habit formation.

Main Results:

• Behavioral variability is actively generated by the brain, serving to probe environmental responses.
• World-learning, the process of assigning value to environmental stimuli, inhibits self-learning, which assigns value to actions.
• This inhibition, mediated by the mushroom bodies, requires extended training to be overcome, leading to habit formation.
• Self-learning transforms flexible, spontaneous actions into rigid, habitual responses.

Conclusions:

• Operant conditioning involves a complex interplay between active behavioral variability generation, world-learning, and self-learning.
• The brain actively generates variability to explore environmental contingencies.
• Habit formation is achieved through overcoming the initial inhibition of action-specific learning (self-learning) by stimulus-specific learning (world-learning) via extended training.