The purpose of the present thesis was to characterize the cognitive mechanism that underlies aesthetic pleasure. This problem was attacked using a bottom-up and a top-down approach. The concern of the bottom-up approach was to elucidate the mechanism responsible for a number of well- established historical (i.e., learning-based) preference effects - the mere-exposure effect, preference for prototypicality and structural-relation preferences. Critical literature reviews and empirical results from a number of experiments that employed a category learning procedure established that these preference effects are best conceptualized as stemming from conscious and inferential processes, that is, from the strategic use of either processing fluency, feelings of familiarity or explicit recognition to determine liking. The concern of the top-down approach was to empirically examine the often articulated notion that aesthetic pleasure may be traced to a psychological process involving 'the striving for and the acquisition of meaning'. Using a procedure in which participants were first given an impoverished passage or picture to make sense of before being given the full passage or picture to evaluate, the results of a number of experiments showed that the more participants had to strive after meaning (the more impoverished the preceding passage or picture became) the more pleasure that was engendered upon the acquisition of that meaning (the more favorably they subsequently evaluated the full passage or picture). The findings from both approaches led to the following cognitive characterization of aesthetic pleasure : it consists of a 'meaning making' process in which difficult and uncertain processing (as one strives for meaning) gives way to easy and certain processing (when one acquires meaning) to produce feelings of relief (because one's exertions have abated) and joy (because of the challenge that was successfully overcome in acquiring meaning). The relevance of this m̀eaning making' process for our understanding of a number of preference effects is discussed