Recent research into ‘The Social Brain Hypothesis’ has found variation in human social group size to be explained by differing levels of cognitive ability (Stiller and Dunbar 2007). The Current study aimed to develop a measure of social complexity that took into account the depth of our relationships with others as well as simpler measures of social group size, and emphasises the difficulties associated with measuring social networks in humans based around self report measures. We assessed one complete social network by gathering two-way reports of relationship closeness from a university hall of residence. Social complexity measures were then related to a number of factors believed to influence an individual’s social ability in order to assess how variation in social complexity is best explained. These included measures of empathy, systemising, altruism and personality, as well as general intelligence in line with Dunbar’s research. Personality was found to be the key factor influencing social complexity at every level of relationship closeness. This offers greater insight into what factors are most influential in the early stages of social network development, and calls for further research into social networks at varying stages of development.