**2. Neural circuits involved in social cognition**

In humans, brain regions implicated in social cognition have been identified mainly by lesion studies or by functional magnetic resonance imaging (fMRI) detecting differential activation in response to social versus nonsocial cues. Accordingly, a network of key brain structures involved in the perception, integration, and coding of social cues have been identified, receiving the denomination of "the social brain." These structures comprise brain areas traditionally involved in *cognitive processes*, such as frontal and temporal cortices; *motivation and reward*, such as the basal ganglia; and *processing of emotions*, such as the amygdala [8–10]. Proper function of the social brain should be considered in terms of the coordinated activity of the neural network involving these structures [11]. Accordingly, individuals with autism have been reported to present structural and/or functional alterations in these areas. Aberrant cortical organization is a pathological observation commonly seen in postmortem brain tissue of individuals with ASD [12]. Connectivity studies using fMRI indicate alterations in the PFC, with increased local and decreased long-range connectivity [13], which might be accounted for by the observed disorganized cortical structure. Structural MRI studies report

#### *The Cerebellum and Autism: More than Motor Control DOI: http://dx.doi.org/10.5772/intechopen.85897*

increased amygdala size in children with autism, and the extent of amygdala enlargement is correlated with social deficits [14]. However, fMRI studies conversely report both hyperactivation [15] and hypoactivation [8] of the amygdala in response to social stimuli. These contrasting results have been suggested to indicate either a failure of the amygdala to engage upon social stimuli or, alternatively, an overreaction to such stimuli, coding them as threatening. Both cases would in turn result in social withdrawal. Similar contrasting results have been found when studying the reward system in ASD, as some studies observe a generalized decreased activation of the NAcc independent of stimuli [16], and others indicate that the hypoactivation is stimulus-dependent as individuals with ASD react to cues that are salient to them but not necessarily to neurotypical individuals [17], indicating that in autism a different set of cues might be coded as salient.

The cerebellum, conventionally associated with motor functions, is lately being considered a key structure within the social circuitry [18]. In fact, cerebellar neuroanatomical alterations, including the reduced size and number of Purkinje cells, are among the most replicated findings in postmortem brain samples of individuals with autism [19]. In addition, cerebellar defects have been proposed to affect the functioning of distal brain areas to which the cerebellum projects [20]. For example, the PFC long-range connection deficits observed in ASD mentioned above include the cerebellum [13]. The *developmental disconnection hypothesis* of autism suggests that certain areas of the brain that normally connect to the frontal lobe become disconnected during development. Thus, a change in connectivity in a certain area could affect the functioning of other brain regions either through compensation or adaptation of nearby circuitries [21]. Along these lines, deficits in connectivity of the cerebellum could account for dysfunction in connected areas, being possible to affect some of the social brain structures previously mentioned.

Although human neuroanatomical and functional studies have been very useful in the identification of the brain regions involved in social cognition, animal studies are critical to understand how information is processed at the circuit and molecular level, from the perception of the stimulus to the expression of a behavioral response. The mouse (*Mus musculus*) is currently the most widely studied, mainly for practical reasons and technical amenability. In addition, as mammals, mice present the same key brain structures and express a wide range of social behaviors that can be easily measured in the lab [22]. A schematic representation of the brain structures and circuits implicated in social cognition processes in rodents is presented in **Figure 1**. Animal studies have corroborated the role of previously described structures in social behavior and have given insight into circuit function. For example, disruption of the ratio between cortical excitation and inhibition (E/I) has been extensively hypothesized to be a causal mechanism in autism [23]. Again, one must consider that the cortical E/I balance is a complex process controlled both locally and distally by neuromodulation from connecting circuits, arguing against the specificity of a certain structure as preferentially involved in ASD since alterations in distally connected regions could also lead to an altered cortical E/I ratio. Nevertheless, in 2011, the application of the recently developed optogenetic techniques allowed to test for the first time this hypothesis in vivo [24]. The authors found that increasing, but not reducing, the E/I balance in the PFC would lead to social dysfunction in mice. Similarly, recent studies have indicated the role of the reward system and, specifically, dopaminergic projections from the ventral tegmental area to the NAcc, in the modulation of social interaction. Dopamine (DA) is a neurotransmitter traditionally linked to movement control and reward processing, including social reward and domains of behavior that are modulated by two separated DA pathways. DA-producing neurons are located in two main brain areas, the ventral tegmental area (VTA) and the substantia nigra (SN). The *nigrostriatal pathway* originates in the

#### **Figure 1.**

*Brain circuits linked to social cognition in mice. Olfactory signals from a social stimulus are perceived through the olfactory bulb and transferred to the amygdala (AMYG) and probably other structures to be processed. This social cue will modulate the activity of several structures. One of such, the ventral tegmental area (VTA), mainly composed of dopaminergic neurons, projects to the PFC (executive function) and NAcc (reward system). Of note, the cerebellum modulates VTA activity implying a role in social behavior and reward. Adapted from [28].*

SN and projects to the striatum, modulating movement; and the *mesocorticolimbic pathway*, involved in cognitive processes including social cognition, originates in the VTA and is further subdivided into two pathways: the *mesocortical pathway*, which projects to the cortex, and the *mesolimbic pathway* which projects to limbic areas such as the NAcc [25]. Recently, it has been found that in mice, optogenetic stimulation of dopaminergic VTA-NAcc projections increased, while inhibition decreased during the time the animals were engaged in social exploration [26]. Interestingly, very recently Carta et al. [27] have demonstrated the role of the cerebellum in modulation of the reward pathway through direct control of the activity of the VTA, which could have profound implications for social behavior. The authors found that the cerebellum sends excitatory projections to the VTA and that optogenetic modulation of the cerebellum-VTA pathway bidirectionally modulated social behavior and reward. Considering previous studies where stimulation of VTA-NAcc DA projections modulates social behavior, it is highly likely that the cerebellum indirectly controls the activity of these projections. A cartoon deciphering the main brain structures and connections involved in social cognition in mice is presented in **Figure 1**.
