Is the depth cue of Familiar Size computed using the biological equivalent of a trigonometric table?

Undergraduate Just-In-Time Abstract

Poster Presentation 26.363: Saturday, May 18, 2024, 2:45 – 6:45 pm, Banyan Breezeway
Session: Undergraduate Just-In-Time 1

Jhera Darefsky1 (), Amoolya Chengalasetty1; 1Arizona State University

Familiar size is a pictorial depth cue in which the distance to an object could be computed from an object's known physical size divided by the tangent of its visual angle. If the visual system uses this type of computation, distances, and visual angles that observers have not experienced would be as informative as those experienced. A second approach, not involving trigonometry, is one in which connections are formed between the distance of an object and the visual angle it subtends at a particular distance. Distance is recovered using a look-up table. This study aimed to explore these two explanations by varying the distance and visual angles subtended by an object during the familiarization phase. One hundred thirty-six college students estimated the distance of an identical object that varied in height from 7 to 14 cm. Using a between-subjects design, eliminating relative size, the distance to the display during training was varied. Participants were asked to study an unfamiliar object that was 10 cm tall. The test object was then placed 150 cm from the observer and viewed by one eye. The apparent distance was measured by having observers position a rod, visible to both eyes, to match the distance. Regardless of whether they had previously experienced the test display at its distance and visual angle, a depth illusion was evident. The results suggest that familiar size can be generalized to previously unseen distances, but the process may not necessarily require knowledge of the geometry of triangles. What happens may be a generalization gradient around a stored look-up value. We will have to present differences between experience and testing using more considerable distances and shifts in visual angle to see if this pattern of results occurs in more demanding conditions. Unfortunately, we are testing a null hypothesis.