Measuring Social Motivation In Mice Using Novel Operant Conditioning Paradigms

Background: Much advancement has been made in the understanding and treatment of human disease through studies on animal models. For human disorders involving abnormal social behavior such as autism, research on mouse models will benefit from the development of novel assays of complex social behavior including social motivation.

Objectives: The goal of this research is to develop and validate new tasks of social motivation and social escape for mouse models of autism. The proposed tasks involve the use of original operant conditioning paradigms programmed through a computer system that will allow a test mouse to control access to another mouse within an operant box. The access to the stimulus mouse will serve as a social reward for mice with prosocial tendencies but may serve as an aversive stimulus for mice with nonsocial tendencies.

Methods:  Initial research has been carried out through two experiments comparing individually-housed versus group-housed C57BL/6J mice, a prosocial inbred mouse strain. In the first experiment, each test mouse was trained to press a lever to seek a social reward in the form of access to an unfamiliar stimulus mouse. The social reward was set on a progressive ratio schedule with a step size of three. The number of lever presses achieved in the final trial of a testing session (breakpoint) was used as an index of social motivation. For the second experiment, motivation for a food reward was compared to a social reward. The mice were conditioned to associate one lever consistently with a food reward and another consistently with the same social reward described in the previous experiment. After this conditioning period, a series of 1 hr testing sessions assessed reward preference by allowing test mice to freely choose between the rewards by pressing the associated lever. 

Results: In experiment 1, all mice demonstrated social motivation as determined by an increase in the number of lever presses when a stimulus mouse was present. In addition, group-housed mice consistently demonstrated higher breakpoints than individually-housed mice. In experiment 2, all mice showed a preference for the food reward over the social reward as demonstrated by a higher number of food lever presses than social lever presses across testing sessions. While performance was similar during the first half of the testing sessions, individually-housed mice demonstrated a heightened food to social reward ratio over group-housed mice in the latter half of the testing sessions.

Conclusions: The results suggest that these operant conditioning paradigms may be valuable tools to assess social motivation in various mouse strains including models of autism. The validation of these paradigms is ongoing through the testing of additional group-housed and individually-housed mice, as well as mouse strains that have previously demonstrated social deficits.  Preliminary data suggests that individually-housing mice long-term may decrease social motivation perhaps due to their lack of social experience. Future experiments will test mice lacking social motivation in an operant paradigm in which they can choose to escape a forced social interaction.