Purchase this article with an account.
Zoe J. Oliver, Mark V. Roberts, Alan J. Pegna, Charles Leek; The time course of three-dimensional object recognition in human vision: An ERP study. Journal of Vision 2014;14(10):908. doi: 10.1167/14.10.908.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
One of the great mysteries of the human brain is how we are able to perceive and identify three-dimensional (3D) objects rapidly and accurately despite variability in viewing conditions and sensory input. While some kinds of image classification have been reported to occur as quickly as 100-150ms post-stimulus onset (Thorpe et al., 1996, Nature), we know relatively little about the time course of different perceptual processes involved in object recognition. We investigated these issues using 1000 Hz, 64 channel event-related potentials (ERPs). Participants completed a perceptual matching task in which they had to make shape equivalence judgments about two sequentially presented 3D surface rendered multi-part novel objects. Trials could either contain two identical objects (differing only in scale to eliminate pixel overlap), or two different objects. There were three types of different object trials: Stimulus pairs could share volumetric parts but differ in spatial configuration (Same parts/Different configuration), contain different volumetric parts but share spatial configuration (Different parts/Same configuration) or share neither parts or spatial configuration. Analyses of the ERP waveforms showed no differences between conditions on the early P1 component – indicating that the conditions were well-matched in terms of low-level image properties. In contrast, we found evidence of differential sensitivity to volumetric part and spatial configuration overlap in the different response trials on posterior electrodes. This was reflected in amplitude modulations of negative deflections on the N1 component between different object conditions approximately 170ms post-stimulus onset. These differences suggest an early perceptual sensitivity to shared spatial configuration but not local part structure. We propose that this pattern of results is consistent with the operation of two parallel processes during shape perception involving a fast, low-spatial frequency, analysis of global shape configuration, and a slow, high spatial frequency, analysis of local part structure.
Meeting abstract presented at VSS 2014
This PDF is available to Subscribers Only