UC Berkeley

It can be argued that individuals are most often concerned with the objects that exist in an environment, as it is these objects that are recognized, localized and acted upon. However, the space between objects plays as critical a role in activities of daily living as the objects themselves because without the perception of an underlying spatial structure, accurate localization, navigation, and manipulations of objects would not be possible. To date, questions still exist as to how space is perceived across the visual field as a function of eccentricity (i.e., where an object is located within the visual field) and the factors that influence space perception in the visual periphery. In particular, conflicting reports have been found in the literature regarding inherent biases in peripheral localization and the modulating affect of attention on the perceived structure of visual space. The aim of these studies was to help resolve some of these issues. In the first two chapters it is argued that perception of both the locations and shapes of objects presented in the periphery are influenced by voluntary and involuntary attention. In Chapter 2, the results of three experiments demonstrate that changes in the distribution of sustained voluntary attention alter the perceived locations of target dots presented at different eccentricities along the cardinal axes. Specifically, the results show that when attention is focused on a region of space, targets appear to be located more peripherally than when attention is distributed across the visual field. The next three experiments in Chapter 3 show that rapid shifts in the location of involuntary attention can distort the perceived shape of an oval. These studies support the assertion that changes in attention alter the underlying structure of visual space, and thus, alter the perceived locations and/or shape of any object presented in that space. The experiments in Chapter 4 investigate the role of visual boundaries in spatial localization and argue that different classes of borders are associated with the different reference frames and metrics used in defining the underlying perceptual structure of visual space. The results of these experiments show that when spatial localization occurs relative to the intrinsic borders of the visual field, participants show a peripheral bias and a non-linear scaling of target locations across eccentricity. The introduction of external boundaries first leads to a linear scaling of target locations and can change the pattern of mislocalization from a peripheral to a foveal bias. In Chapter 5, it is argued that the borders of the visual field are used to compute location in a more natural metric (percentage of visual field extent) than degrees of visual angle for determining spatial location within a retinotopic reference frame, and that this metric may determine the allocation of processing resources across the visual field. Using a crowding paradigm in which participants perform significantly better when target gratings are presented along the lower vertical meridian than the upper vertical meridian, the results of the final experiment show that both inter-subject and intra-subject variability can be accounted for by this new metric. Collectively, these experiments highlight the fact that visual space is not a stable mapping of the external environment. Rather, the perceived structure of visual space is flexible and can be altered by both the borders that define a space as well as the attentional state of an observer.