An operant social self-administration and choice model in mice.

Little is known about how social factors contribute to neurobiology or neuropsychiatric disorders. The use of mice allows one to probe the neurobiological bases of social interaction, offering the genetic diversity and versatility to identify cell types and neural circuits of social behavior. However, mice typically show lower social motivation compared with rats, leading to the question of whether mice should be used to model complex social behaviors displayed by humans. Studies on mouse social behavior often rely on measures such as time spent in contact with a social partner or preference for a social-paired context, but fail to assess volitional (subject-controlled) rewarding social interaction. Here, we describe a volitional social self-administration and choice model that is an extension of our previous work on rats. Using mice, we systematically compared female adolescent and adult C57BL/6 mice and outbred CD1 mice, showing that operant social self-administration, social seeking during periods of isolation and choice of social interaction over palatable food is significantly stronger in female CD1 mice than in female C57BL/6J mice, independently of age. We describe the requirements for building the social self-administration and choice apparatus and we provide guidance for studying the role of operant social reward in mice. We also discuss its use to study brain mechanisms of operant social reward, potentially extending its application to mouse models of neuropsychiatric disorders. The training commonly requires ~4 weeks for stable social self-administration and 3-4 additional weeks for tests, including social seeking and choice.

Waving Through the Window: A Model of Volitional Social Interaction in Female Mice.

BACKGROUND: Mouse models of social behavior fail to account for volitional aspects of social interaction, and current neurobiological investigation of social behavior is performed almost exclusively using C57BL/6J mice, the background strain of most transgenic mice. Here, we introduce a mouse model of operant social self-administration and choice, using a custom-made apparatus. METHODS: First, we trained adolescent and adult female C57BL/6J and CD1 mice to self-administer palatable food pellets and then to lever press under increasing fixed-ratio response requirements for access to an age-matched female social partner. Next, we tested their motivation to seek social interaction using a progressive ratio reinforcement schedule, relapse to social seeking after social isolation, and choice between palatable food versus social interaction. We also tested social conditioned place preference in adult female CD1 and C57BL/6J mice. RESULTS: Adolescent and adult female mice of both strains showed similar rates of food self-administration. In contrast, CD1 mice demonstrated significantly stronger social self-administration than C57BL/6J mice under both reinforcement schedules. CD1 but not C57BL/6J mice demonstrated robust social seeking after social isolation. In the choice task, CD1 mice preferred social interaction, whereas C57BL/6J mice preferred food. CD1 but not C57BL/6J mice demonstrated robust social conditioned place preference. The strain differences were age independent. CONCLUSIONS: Our data show that CD1 mice are a better strain for studying female social reward learning. Our mouse operant social model provides a tool for research on neurobiological substrates of female social reward and disruption of social reward in psychiatric disorders using mouse-specific genetic tools.