Wireless Ad Hoc Networks (MANETs) can to provide first responders and disaster management agencies with a reliable communication network in the event of a large-scale natural disaster that devastates majority of the existing communication infrastructure. Without requiring a fixed infrastructure, MANETs can be quickly deployed after a large-scale natural disaster or a terrorist attack. On the other hand, MANETs have dynamic topologies which could be disconnected because of the mobility of nodes. This paper presents a decentralized approach to maintain the connectivity of a MANET using autonomous, intelligent agents. Concepts from the social network analysis along with flocking algorithms are utilized to guide the deployment decision of agents. Unlike a basic flocking algorithm where all nodes have the same importance, network metrics are used to quantify the relative importance of nodes. Computational results are presented to demonstrate the effect of various local agent behaviors on the global network connectivity.