Schizophrenia and other psychiatric disorders are generally diagnosed based on a collection of symptoms defined by a combination of an individual’s feelings, perceptions, and behaviors. Many of these disorders are characterized by specific cognitive and social deficits. Although it is nearly impossible to recapitulate the full phenotypic spectrum of schizophrenia in mice, mouse models play an indispensable role in understanding the pathogenesis of this disorder and the development of new therapeutics. Genetic mouse models of schizophrenia and mouse behavioral tests provide a feasible approach for elucidating causal relationships between susceptibility gene(s) and schizophrenia-related symptoms. There has been a proliferation of studies characterizing basic behavioral phenotypes in mice. Since there is no way to completely model human psychiatric symptoms in mice, the major role of behavioral tests is to provide insights into underlying affected circuitry and pathophysiology. Given that the recovery of cognitive and social abilities significantly benefits functional outcomes, there has been an increasing interest in characterizing cognitive and social functions in mutant mice; however, these functions are not easy to measure. In this review, a selection of conventional behavioral tasks was briefly described and three specific behavioral tasks aimed at characterizing social communication, attentional function, and choice behavior in mice were highlighted. The choice of specific behavioral tasks during experimental planning should take into consideration a variety of factors, including their validity, reliability, sensitivity, utility, and specificity. Based upon the hypothetical hypofunction of N-methyl-D-aspartate receptor (NMDAR)-mediated signaling pathways in the involvement of cognitive and social impairments in schizophrenia, three NMDAR-related compounds/drugs, D-serine, sarcosine, and D-cycloserine, are discussed as an example.