Contents I. Introduction
The visual information about the 3D world utilized by humans is provided by a set of 2D images on the eye retina. It leads to uncertainty and/or discrepancy in image interpretations because the same projections could belong to different 3D objects. As an additional complexity, the visual system has to recover the information about objects' depth (i.e. the mutual disposition of objects) with respect to the observer. To overcome these difficulties one needs to utilize and combine in an effective way a variety of visual characteristics (or so called cues) in order to achieve inferences, more informative and potentially more accurate than if they were obtained by means of a single cue. The process of combining, manipulating and interpreting information in stimulus integration problem is beneficial because it allows the human visual system to estimate and perceive more accurately the objects' properties and to take appropriate actions, leading to improved reasoning (judgment) under uncertainty or/and possible conflicts between different visual stimulus. The uncertainty, associated with the utilized visual cues and the possible conflicts between them influences the decision making and action control in the process of human aging due to the increased level of internal noise in the visual system [1]. If the visual system neglects some of the available information [2], the visual signal/noise ratio will additionally deteriorate. Throughout the life cycle many aspects of vision and visual information processing decline and affect everyday task performance. 3D shape of objects and their spatial layout are specified on the base of both: static and dynamic visual cues. Age-related impairments in visual processing and perception are observed for both of them [3]. Therefore the task of vision inherently requires the integration of all available visual cue information to determine 3D object's shape. This paper focuses on human way of integration of motion and texture information in the process of object's slant estimation. Our goal is to reveal not only the age-related changes in the process of visual information assessment, but also the plasticity of the visual system to best adapt to these changes and to efficiently exploit all the available information in the visual scene in order to provide the visual system with a meaningful output, concerning more accurate and robust spatial information about the 3D objects. We will present and compare the performance of three fusion rules to model human way of visual cue integration: Normalized Conjunctive Consensus (NCC), Averaging (AVE), and the probabilistic Proportional Conflict Redistribution rule no.5 (pPCR5) defined recently within Dezert-Smarandache Theory (DSmT) for plausible and paradoxical reasoning. In section II we present briefly the visual cue integration problem and recall the principles of NCC, AVE and pPCR5 fusion rules. In section III, we present the experimental strategy and procedure, methods, subjects, involved in the experiments, stimulus and used apparatus. In section IV, the research reasoning logic is presented as well the results, obtained on the base of applied fusion rules. Concluding remarks are given in section V.
