The Assignment of Absolute and Relative Depth in Constructing a 3-D Mental Model from Viewed Orthographic Displays

Abstract

Constructing the 3-D mental structure from viewed 2-D orthographic displays may require establishing both the relative depth relation (RD) between adjacent areas in the orthographic displays and the exact magnitude of absolute depth (AD) for a given area. We examined the processes of depth assignment involved in 3-D model construction by asking subjects to mentally imagine the 3-D object portrayed by a pair of orthographic projections, and then choose among four alternative isometrics the one that matched the imagined 3-D model. One alternative corresponded to the previously shown orthographic views (ie, target). The other alternatives were distractors that differed from the target in terms of RD, AD, or both. Results from four experiments show that (a) when subjects mistook a distractor for target, they were more likely to misidentify the distractor sharing the same RD with target but with different AD, than to misidentify the distractor sharing the same AD but with different RD; (b) subjects could shorten their solution time by 40% without greatly compromising their performances when times available for imagining 3-D models were reduced in proportion to those consumed in a self-pacing condition. The implications of these results are discussed in terms of a model in which an observer may try to decide the relative depth prior to figuring out the exact absolute depth, although it is necessary to consider both aspects of depth assignment in the construction of 3-D mental models.