Academic journal article Perception and Psychophysics

The Visual Perception of Lines on the Road

Academic journal article Perception and Psychophysics

The Visual Perception of Lines on the Road

Article excerpt

The present work demonstrates that observers grossly underestimate the length of lines parallel to their line of sight. In Experiment 1, observers, working from memory, estimated the length of a dashed line on the road to be 0.61 m. This result is consistent with observers' using an average visual angle converted to the physical length of visible lines on the road to estimate their length. In Experiment 2, observers gave verbal and matching estimates that significantly underestimated the length of a 3.05-m line on the ground that was parallel to their line of sight. In Experiment 3, observers significantly underestimated the length of dashed lines on the road while in a moving car. The results of Experiments 1 and 3 are described well by Euclidean geometry, whereas the tangle model that utilizes an increasing function of the visual angle to describe perceived extent best describes the results of Experiment 2.

(ProQuest: ... denotes formulae omitted.)

For years, psychophysicists have investigated the relationship between physical extent and observers' visual perception of extent. Although some of this work has focused on the metrics of visual space in general and has shown that visual space is curved (i.e., hyperbolic or elliptical; Battro, Netto, & Rozestraten, 1976; Blank, 1958; Koenderink, van Doorn, & Lappin, 2000; Luneburg, 1950), an abundance of other work has focused on how visual space is distorted in an affine manner, so that perceived intervals in depth (or parallel to the observer's line of sight) appear to become systematically compressed with depth, relative to perceived intervals in the horizontal or vertical directions (i.e., the frontoparallel plane perpendicular to the observer's line of sight; Beusmans, 1998; Foley, Ribeiro-Filho, & Da Silva, 2004; Gilinsky, 1951; Gogel, 1964; Haber, 1985; Harway, 1963; Hecht, van Doorn, & Koenderink, 1999; Kudoh, 2005; Loomis, Da Silva, Fujita, & Fukusima, 1992; Loomis & Philbeck, 1999; Matsushima, de Oliveira, Ribeiro-Filho, & Da Silva, 2005; Norman, Crabtree, Clayton, & Norman, 2005; Norman, Todd, Perotti, & Tittle, 1996; Todd & Norman, 2003; Todd, Tittle, & Norman, 1995; Toye, 1986; Wagner, 1985). Some of this work has shown that when extents are viewed in near space, depth intervals must be made approximately 1.5-2 times as large to be perceived to be the same as horizontal extents (Norman et al., 1996; Todd & Norman, 2003). Many of these studies have been conducted in outdoor environments with several stakes or L-shapes placed on an outdoor grassy field at different distances. They typically have observers either estimate distances between stakes or adjust a stake or one part of the L-shape so that a depth interval is perceived to be equivalent to a horizontal interval. This work has shown that in cases in which stakes are ~4-40 m away from the observer, depth intervals are adjusted to be between 1.5 and 2 times as large, relative to the horizontal interval of the same physical size (Foley et al., 2004; Hecht et al., 1999; Kudoh, 2005; Loomis et al., 1992; Loomis & Philbeck, 1999; Matsushima et al., 2005; Norman et al., 2005; Toye, 1986; Wagner, 1985). Hecht et al. concluded that many findings from other studies could be rephrased in terms of depth compression and suggested that there is a general mechanism for sagittal compression for far space.

The affine distortion of perceptual space shown in the aforementioned work has been applied mainly to unfilled distances between stakes or with L-shape stakes connected to one another (or touching one another) on the ground. To our knowledge, less work has been done considering whether an object or a filled area, such as a stake placed on the ground plane parallel to the line of sight of the observer, would also be subject to the large affine distortions that distances between stakes are. Previous work has shown that unfilled space is generally seen as much smaller than filled space (Luria, Kinney, & Weissman, 1967). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.