Cortical responses to layout change specified by two pictorial cues: An fMRI study

Abstract

Humans perceive depth from various types of visual information: binocular cues; motion carried cues; and static monocular or pictorial depth cues. Although the neurological substrates of these processes may differ, it is reasonable to hypothesize that at higher levels of processing the perception of depth converges to common cortical locations. Our study used functional magnetic resonance imaging (fMRI) to investigate the neural correlates of the perception of depth specified by two pictorial depth cues: the T-junction cue (interposition), which provides information for depth order; and linear perspective, which provides information for surface slant. These cues were used to generate two displays that varied in spatial layout: T-junctions specified two sets of cubes that differed in their depth order. Linear perspective specified two different surface slants corresponding to a floor and a ceiling. Within 30s blocks, displays employing the same depth cue alternated with each other every two seconds. Blocks of the T-junction figures and blocks of the linear perspective figures were randomly presented in the same 8 minute scan along with control displays. Control displays were matched to each individual depth cue display. That is, they contained the same number of lines, matched in orientation and length, with the lines repositioned to eliminate depth information. Statistical parameter maps were generated using the general linear model. Cube displays containing T-junctions consistently showed more activation bilaterally in the occipital and parietal regions than their controls, including areas MT+. However, subtracting control condition from displays based on linear perspective showed less robust activation in the occipital and parietal areas.