Perceptual issues of visual attention and depth perception in augmented reality

Long, M.L. (2019) Perceptual issues of visual attention and depth perception in augmented reality. PhD thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (2MB)


There has been a recent development and growth of augmented reality devices across the world. The technology has the potential to enhance a user’s perception and experience of the environment which surrounds them. For designers and engineers of 3D displays (e.g. vehicle HUDs) the technology offers an opportunity to form a more enriched and exciting experience for drivers. A key anticipation of AR is the capability to present virtual information at depth, as well as on displays with a wide field of view. Applying psychological theories and models of visual attention (e.g. perceptual load theory) and depth perception (e.g. modified weak fusion), the aim of the study was to establish human abilities with AR advancements in depth and useful field of view (UFOV). Conclusions were applied to the development of psychological theory and helpful design guidelines for AR designers and engineers. This has been done by creating two sets of experimental paradigms. The first set of experiments presented peripheral Landolt C target stimuli on a large display (120o), simultaneously whilst participants controlled a central tracking task (CTT). Over two experiments workload was manipulated on both central and peripheral tasks, including presentation time, contrast, tracking speed, and size. Results indicated that peripheral performance did not decrease as a function eccentricity. Conclusions suggested a demonstration of attentional selection dependent on stimulus parameters, with performance potentially dependent on more temporal characteristics than eccentricity. ‘Best-practice’ design guidelines are presented for AR HUDs with large display size. The second design paradigm implemented a two-alternative forced-choice psychophysical depth judgment task, assessing thresholds for correctly determining the depth of a virtual diamond in reference to a real-world object. Over a series of experiments a number of parameters were assessed including virtual image height in the visual field, virtual image cue properties (relative size and brightness), and background scene manipulations. Depth thresholds for the AR image were tested with the pedestrian target at 5m, 10m, 20m and 25m. Results demonstrated that depth resolution is poor in augmented reality compared to real-world environments, with conclusions indicating a potential difference in how relative cues of depth interact with each other in AR environments, i.e. in AR cues from virtual imagery and real-world background may not be effectively combined together. Design guidelines regarding where in depth a virtual image needs to be placed to ‘match’ that of a specific real-world object, and how relative cues to influence depth perception are presented.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Allen, H.A.
Burnett, G.
Keywords: augmented reality; visual perception; depth perception
Subjects: B Philosophy. Psychology. Religion > BF Psychology
Q Science > QA Mathematics > QA 75 Electronic computers. Computer science
Faculties/Schools: UK Campuses > Faculty of Science > School of Psychology
Item ID: 56110
Depositing User: Long, Michael
Date Deposited: 13 Dec 2019 04:40
Last Modified: 06 May 2020 12:32

Actions (Archive Staff Only)

Edit View Edit View