Discovering the hormonal and neural mechanisms that promote affiliative social behavior is a high priority in behavioral neuroscience. Although studies with standard laboratory rodents have afforded many important insights, exciting advances are also occurring through comparative research with nonstandard species that vary in sociality or form socially monogamous pair bonds, work that is often informed by an explicitly evolutionary perspective. Research with prairie voles has examined the roles of sex steroid hormones, adrenal glucocorticoids, oxytocin family peptides, and dopamine in the formation of monogamous pairs. Corticosterone facilitates pairing by males but inhibits it in females, vasopressin (acting via the V1a receptor) and oxytocin facilitate pairing, and dopamine in the nucleus accumbens also facilitates pairing. Research with zebra finches is testing the limits of generality of these mechanisms, and has shown how sex steroid effects early in life along with social experience lead to an adult's sex preference in a pairing partner. Estrogen manipulations during the embryonic or nestling periods result in females that prefer to pair with other females. An all-female social environment can reinforce such effects, and can also produce males and females that will pair with either sex. Research with multiple species of estrildid finches is revealing the contributions of peptidergic and dopaminergic mechanisms to the evolution of species differences in whether animals are gregarious or territorial. Mechanisms for and responses to vasotocin (avian vasopressin) in the septal region of the brain are predicted by sociality in this group of birds.