Humans perform numerous interactions with objects in the environment, like reaching for a mug on the breakfast table, in everyday live. Even though these interactions can be done without bigger cognitive effort, the underlying mechanisms are fairly complex. One crucial mechanism is to code the location of the object that we want to act on (target), which is done by using so called reference frames. These reference frames provide coordinate systems and origins to which targets can be related to. In the egocentric reference frames, any body part can serve as the origin of the coordinate system whereas in the allocentric reference frame, any object in the environment could serve as the origin. The use of these reference frames for coding a target location to perform an action was research question of several studies in the past. However, these studies manly used rather artificial stimuli like dots and bars on black screens, lacking ecological validity. Moreover, in some studies participants were biased to prefer one reference frame over the other.In the present thesis, I aimed to overcome these limitations by using naturalistic and complex stimuli which provide multiple potential allocentric cues for coding a target location without biasing participants coding behavior. To this end, I applied a paradigm in which participants had to perform a memory-guided reaching movement to a target location that was represented by an object in a breakfast scene that was presented either on a computer screen or in virtual reality. By conducting four studies, I was able to provide novel insights into the integration of the two reference frames in complex environments. First, I confirmed that both reference frames are used for coding a target location for memory-guided reaching. Second, I revealed several factors that determine the use of objects as allocentric cues in a complex environment, such as the task-relevance of objects, the distance between a target and task-relevant objects and the reliability of the allocentric information. Third, I have shown that the use of the allocentric information is higher for coding target locations in depth compared to coding a target location in the horizontal plane. And fourth, I found that object size is an important depth cue for allocentric coding of object locations in depth.
Verknüpfung zu Publikationen oder weiteren Datensätzen
