SIGNIFICANCE: Humans exhibit broad heterogeneity in social behavior from infancy to adulthood. However, there are vast knowledge gaps in understanding neurobiological mechanisms that contribute to this heterogeneity. For neurodevelopmental disorders, such as autism spectrum disorders (ASDs), social dysfunction is even more heterogeneous, affecting clinical treatment efficacy. Neurodevelopmental disorders also disrupt neural circuitry involved in mental processes, including cognition and affect. Testing cognitive functions is relatively straightforward in children, but affective processing—which influences attention, motivation/reward, and emotional regulation—is a more difficult construct to test. The challenge is even greater when assessing these functions in a nonverbal child or a child with language impairments. Tools that probe complex internal responses—such as feelings, drives, and motivations independent from language—become necessary for these populations and even for typically developing (TD) children.

INNOVATION: We propose a unique translation of a research strategy from animals to children, utilizing conditioned place preference (CPP)—a time-honored paradigm—to probe differences in motivation, reward, and aversion. This is the first attempt to establish CPP in children and the first to establish a social CPP paradigm in humans. These studies will allow us to study social motivation in children with disrupted social behaviors, specifically ASD.

APPROACH: This study addresses two questions: (1) Can children learn a CPP for toys? and (2) Can children learn a social CPP? In addition, we examine heterogeneity in CPP and explore child characteristics associated with preference scores. Demonstrating CPP in TD children allows us to describe heterogeneity in associative learning, and then establishing social CPP provides us with a means for describing social behavior heterogeneity in TD children. This will guide future studies in which we examine social deficits in children with ASD.

METHOD: Participants are 75 TD children aged 36 to 60 mo (to first establish CPP) and an additional 75 TD children (to establish social CPP). We use a child-friendly, custom-designed castle with two unique training rooms connected to a neutral room. Each room is uniquely decorated by color and other visual cues and is outfitted with age-appropriate toys. Conditioned stimuli (castle rooms) are repeatedly paired with unconditioned stimuli (toys or social experimenter), which elicits an unconditioned response (positive affect), and after successful conditioning, the conditioned stimulus elicits a conditioned response (choosing to spend time in the room) similar to the unconditioned response. Ceiling mounted video cameras and microphones allow for continuous monitoring. Noldus Observer^® XT software is used to analyze time spent in each room, language elicitations, time spent playing with toys, and locomotion. Correlations compare performance on the Mullen Scale of Early Learning with behaviors elicited during CPP. We explore whether child characteristics are associated with preference scores using logistic regression.

RESULTS: Pilot data indicate that we can reliably establish a place preference in young TD children. These results strongly support the use of CPP in children to understand heterogeneity of both associative learning and social behaviors.