Brain Mechanisms of Play Behavior

Play behavior is most common in young mammals and birds, two taxonomic groups that are also characterized by relatively large brains (see Chapter 5). Although the implication that the occurrence of play may be correlated with brain size is tempting, this relationship does not hold when comparisons are made at the level of species within specific orders of mammals or birds (Pellis & Iwaniuk, 2004). Furthermore, both “brain size” and “play” are

Early Social Learning and Behavior 387 complex concepts that include several components. For example, different types of play do not correlate with each other, suggesting independent brain control. There is no evidence of a positive correlation between time allocated to locomotor (solitary) play and social play in 3- to 5-year-old children. One approach is to focus in a restricted taxonomic group and determine what aspects of brain function are related to specific categories of play behavior. A comparative survey of play fighting and locomotor play in rodent species from 25 genera shows that, in fact, there is a negative correlation across species in the complexity of these two forms of play. Thus, an evolutionary increase in the complexity of social play is accompanied by a tendency to reduce the complexity of locomotor play. Evolutionary analysis of two types of fighting play (agonistic and sexual) in 24 rodent genera suggests that the likely ancestral condition involved a mixture of these two types of play behavior. This condition is prevalent in sciurid rodents, a family that includes ground squirrels. By contrast, murid rodents (the mammalian family with the largest number of species, including laboratory rats and mice) are characterized by specialization in sexual fighting play, in which the targets and tactics resemble adult precopulatory behavior.

In rats, complete or partial decortication does not seem to affect locomotor play and, in some cases (e.g., with lesion of the prefrontal cortex), it may actually disinhibit play behavior (Panksepp, Burgdorf, Turner, & Gordon, 2003). However, neurochemical inhibition of some prefrontal areas has subtle effects on play in rats. For example, activating inhibitory GABA receptors in the medial prefrontal cortex and orbital cortex (equivalent to a reversible lesion) reduces the frequency of locomotor play without affecting general activity, and it enhances social exploration (van Kerkhof, Damsteegt, Trezza, Voorn, & Vanderschuren, 2013). Rats also display social play, especially during juvenile development (between weaning and sexual maturity), characterized by interactions between peers—rough-and-tumble play. Neural activity after a session of social play was assessed by measuring the expression of c-fos, a gene expressed when neurons depolarize (van Kerkhof, Trezza, Mulder, Gao, Voorn, & Vanderschuren, 2014). Social play enhanced activity in several areas, including the medial prefrontal cortex, the dorsal striatum (linked to habits), the nucleus accumbens (part of the reward system), and the amygdala (involved in emotion), among other regions. A tentative circuitry underlying social play in rats can be sketched by correlating c-fos expression among several brain areas. This indicated, for example, that activity in the amygdala was correlated with prefrontal cortex and nucleus accumbens expression, both pointing to the rewarding nature of play behavior in rats.

Thus, play behavior in rodents reflects the operation of different brain mechanisms, evolving at different rates, and in different lineages. Among these brain mechanisms, the neocortex is especially important for the plasticity associated with play behavior (Pellis & Pellis, 2016). Decorticated rats engage in social play but cannot readily modify their play behavior as a function of the social context.