Purchase this article with an account.
Suzanne Khalil, Michael McBeath; Canonical representaion: An examination of preferences for viewing and depicting 3-dimensional objects. Journal of Vision 2006;6(6):267. doi: https://doi.org/10.1167/6.6.267.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
This study examines angular vantages of canonical views of three-dimensional objects, and finds that angular variance is consistent with principles of information optimization. We tested if canonical views systematically vary across (1)Domain: Photography versus line-depictions, and (2)Action: Choosing a preferred view versus generating it. The principal statistical hypothesis is that canonical views remain consistent, independent of action in the photographic domain, but vary with action in the drawing domain. The secondary hypothesis is that canonical views vary with object elongation. 144 adults specified best views of objects in a 2(DOMAIN:photographic,line-depiction) by 2(ACTION:choose,generate) by 3(TYPE:geometric, automotive,animal) by 3(ELONGATION:short,medium,long) factorial design. The expected three main-effects (Action, Domain, and Elongation) and the interaction of the three significantly varied as predicted. In the photographic domain, the mean preferred-view was relatively constant across action, while in the linear-depiction domain, the preferred-view angled more toward a profile when drawn than when chosen. There was also a systematic decrease in preferred angle with object elongation. Findings are consistent with the idea that people prefer skewed vantages that optimally provide more overall information regarding both front and side features. The increased information also makes these vantages potentially more difficult to draw; so people prefer such views when allowed to choose or take a photograph, but not when required to draw. In summary, canonical preferences systematically vary depending on domain of representation, action-task, and object feature of elongation, and are consistent with an information optimizing strategy.
This PDF is available to Subscribers Only