The principal challenge in designing intelligent agents (IAs) is to make their interactions with humans feel natural and acceptable. This review focuses on the employed social cues in intelligent agents that can enhance the quality of Human-IA interactions. We conducted a literature review and classified social cues in IAs in two main categories of behavioral and appearantal anthropomorphism. Although these social cues are broadly investigated in literature, their influence on each other and their simultaneous impact on the formed impression is still ambiguous. Using the obtained information, we proposed a research model that can unveil the influence of behavioral anthropomorphism on both appearantal anthropomorphism and positive impressions. This paper contributes to research and practice by providing insights into human-IA interactions.