In this paper, we discuss personality from the viewpoint of training personal assistance in the 3D virtual worlds. Personality is a set of psychological traits that characterize the individual human. It is hypothesized that we have a set of innate factors from which social cognition or social intelligence emerges. Throughout this emergence, conditioning will play a primary role in allowing individuals to become attached to their social surroundings, especially to their parents or care-takers. A fundamental core personality is gradually developed in the early stages of life in the form of the referential associative knowledge required in social life. Given this hypothesis, we have developed a self-biased conditioning model composed of two basic units: a set of primary response networks and an associative memory in which a set of sensors are associated with motor centers and give reinforcement feedback to these centers. We have conducted an initial experiment to use this mechanism to train a virtual world agent through interactions with its user. In this paper, we will report some observations obtained through this experiment.