A behavioral analysis of force-controlled operant tasks in American lobster

Operant conditioning is a common tool for studying cognitive aspects of brain functions. As the first step toward understanding those functions in simple invertebrate microbrains, we tested whether operant conditioning could be applied to train American lobster Homarus americanus that has been extensively adopted as an animal model for neurophysiological analyses of nervous system functions and behavioral control. The animal was trained by food rewarding for gripping of a sensor bar as the operant behavior. Lobsters were first reinforced when they acted on the bar with a stronger grip than a pre-set value. After this reinforcement, the animal learnt to grip the bar for food pellets. The yoked control experiment in which the animal received action-independent reinforcement excluded the possibility of pseudoconditioning that the food simply drove the animal to frequent gripping of the sensor bar. The association of the bar-grip with food was extinguished by rewarding nothing to the operant behavior, and was restored by repeating the reinforcement process as before. In addition, lobsters successfully carried out differential reinforcement regarding the gripping force: their gripping force changed depending on the increased force threshold for food reward. These data demonstrate that lobsters can be trained by operant conditioning paradigms involving acquisition and extinction procedures with the precise claw gripping even under the force control.