With the growing usage of mobile devices, manufacturers have to distinguish themselves in their devices to keep their share of the market. Most important is that the consumer is satisfied with a given brand, since then the chances are high that he will stay loyal to that brand when it is time for a new device. A consumer buys a mobile device based on several factors. The user interface is one of these factors, and it is an easy aspect for the consumer to evaluate. It is what he sees and uses, and therefore, has to be made easy to use, intuitive and attractive. One way of making a user interface attractive is by using a new technology, such as an auto-stereoscopic display. This type of display makes use of binocular depth, also called stereopsis, which allows users to see depth without the necessity of using viewing aids. Nowadays, static images on a standard 2D display get an impression of depth by using monocular depth cues, e.g. occlusion, shadows, relative size, and linear perspective. However, an auto-stereoscopic display provides a real sense of depth with objects popping out of the display screen. It is interesting to investigate the effect on the overall experience of a user interface of a mobile device with an auto-stereoscopic display when combining stereopsis with these monocular depth cues. To answer this research question three experiments have been performed. The goal of the first experiment – the pilot experiment – was to find which attributes were appropriate for evaluating the overall experience of a graphical user interface of a mobile device. In this experiment subjects saw stimuli based on a simplified menu of a mobile device, consisting of a background image with five background icons, and one ‘selected icon’ on the foreground. The stimuli varied in the monocular depth cue used, being shadowing, relative size, and luminance differences, and in the background image and depth mode used. The subjects were asked to score these stimuli on three predefined attributes, which were affordance, aesthetics, and preference. From this experiment the attributes affordance and aesthetics were found to represent different assessment criteria, which implied that these two attributes were useful for the assessment of a graphical user interface of a mobile device. The goal of the second experiment – the tuning experiment – was to find the preferred setting of the subjects for the disparity difference between foreground and background items in the graphical user interface, for the size increase of the ‘selected icon’ and for the luminance increase of the ‘selected icon’. Subjects were requested to indicate their preferred setting via a tuning for each of the variables separately. The results showed that there was quite some spread in preferred disparity among the subjects. The preferred size of the ‘selected icon’ was on average about 130% of its original size. The increase in luminance for the ‘selected icon’ was found to be too small to be detected in the absence of reference material. Therefore, it was decided to leave this variable out of the rest of the study. The goal of the third experiment, being the main experiment, was to find the added value of the various depth cues for the user interface of a mobile device with an auto-stereoscopic display. In this experiment the subjects were selected based on having about the same preferred disparity. These subjects were asked to score the stimuli on the attributes perceived amount of depth, image quality, affordance, and aesthetics. The stimuli consisted of combinations of the same variables as in the first experiment, with the exception of the luminance cue. The results showed that the bigger size of the ‘selected icon’ and the addition of a shadow both contributed to higher scores for all four attributes. Additionally, the depth mode ‘sinking background’, in which the ‘selected icon’ was displayed at the screen and the background behind the screen, gained the highest scores for the perceived amount of depth, image quality and affordance. The depth mode ‘none’, representing the background image that already contained depth had the highest score on aesthetics. The natural background image had the highest score on perceived depth and affordance, while the uniform background had the highest score on image quality. Overall it can be concluded that an auto-stereoscopic display has clearly added value for the design of a graphical user interface on a mobile device. The various variables studied for the graphical user interface affected different aspects of its overall experience. Clearly, a ‘sinking background’ depth mode is more appreciated than a ‘floating icon’ depth mode. A background containing depth is clearly appreciated from an aesthetic point of view. Additionally, the user interface should be designed such that no artifacts affecting the overall image quality are introduced. As the amount of disparity that has to be used is very dependent on the viewer, it may be a good option to allow the user of a mobile device to personalize the amount of disparity to his or her own preference.