The application of natural language to improve the interaction of human users with information systems is a growing trend in the recent years. Advances in cognitive computing enable a new way of interaction that accelerates insight from existing information sources. In this paper, we propose a modular cognitive agent architecture for question answering featuring social dialogue improved for a specific knowledge domain. The proposed system has been implemented as a personal agent to assist students learning Java programming language. The developed prototype has been evaluated to analyze how users perceive the interaction with the system. We claim that including social dialogue in QA systems increases users satisfaction and makes them easily engage with the system. Finally, we present the evaluation results that support our hypotheses