We present here the analysis of multimodal data gathered during realistic face-to-face interaction of a target speaker with a number of interlocutors. Videos and gaze of both interlocutors were monitored with an experimental setup using coupled cameras and screens equipped with eye trackers. With the aim to understand the functions of gaze in social interaction and to develop a gaze control model for our talking heads we investigate the influence of cognitive state and social role on the observed gaze behaviour.