Elsevier

Physica Medica

Volume 32, Issue 10, October 2016, Pages 1308-1313
Physica Medica

Original paper
Brain activation difference evoked by different binocular disparities of stereograms: An fMRI study

https://doi.org/10.1016/j.ejmp.2016.07.007Get rights and content

Highlights

  • Functional location of stereoscopic vision evoked by binocular disparity.

  • Relationship between brain blood response and binocular disparity in an extended range.

  • An advanced Random Dot Stereogram (RDS) eliminating shape factor from common used RDS.

  • Functional division between dorsal and ventral vision pathway.

Abstract

The binocular disparity of two retina images is a main cue of stereoscopic vision. However, the global dependency between brain response and binocular disparity still remains unclear. Here, we used functional Magnetic Resonance Imaging (fMRI) to identify stereopsis-related brain regions with a modified Random Dot Stereogram (RDS) and plotted the activation variation curves under different disparity size. In order to eliminate the confounding shape difference between the stereogram and the plane, commonly seen in RDS, we modified the RDS to a checkerboard version. We found that V3A, V7 and MT+/V5 in dorsal visual stream were activated in stereoscopic experiment, while little activation was found in ventral visual regions. According to the activation trends, 13 subjects were divided into three groups: 5 subjects with turning points (a shift from increased to decreased activation), 5 subjects without turning points and 3 subjects with activation unrelated to disparity. We inferred that the dorsal visual stream primarily processes spatial depth information, rather than shape information.

Introduction

Binocular disparity, one of the most important cues of human stereopsis, refers to the difference between two retinal images seen by the left and right eyes [1]. It is important to understand the neural mechanisms of disparity-evoked stereopsis in both psychological research and ophthalmology clinical application [1], [2]. Disparity size, which determines apparent depth, is an important entry point to reveal the mechanisms of stereoscopic vision.

Neuron-based studies on monkeys and cats provide some fundamental proofs that a large number of disparity-tuned cells exist in the visual cortex [3], [4], [5], [6], [7], [8], [9], [10], [11]. Their firing rates vary with disparities and different neurons generally have different preferred disparity ranges [5], [12], [13], [14], [15], [16], [17]. Some studies also found that regions such as the inferior temporal cortex (LOC) include neurons that respond to concave or convex surfaces selectively [18], [19], which indicates a functional division for stereoscopic vision processing in the cortices exists.

Further, functional Magnetic Resonance Imaging (fMRI) studies have shown that dorsal visual cortices, particularly V3A, V5/MT and V7 [20], [21], [22], [23], [24], [25], [26], [27], and some ventral visual regions, such as V4 and LOC [24], [28], [29], are important regions in stereoscopic vision processing. The dorsal and ventral visual streams are believed to play different roles in stereopsis processing. The dorsal visual regions process absolute disparity and stereoscopic depth identification, while the ventral visual regions process relative disparity and spatial shape recognition [28], [30], [31], [32], [33], [34].

Several fMRI-based studies have specialized in the relationship between brain response and disparity size. Preston et al. and Minini et al., in 2008 and 2010, investigated this question respectively [24], [25]. They found that the response of dorsal visual regions increased with the binocular disparity, while the response of ventral visual regions remained approximately unchanged. They inferred that stereoscopic vision was processed mainly in the dorsal visual stream, while the ventral visual stream extracts shape information from disparity images. However, the range of disparity they selected was both relatively small (smaller than 0.25° in Preston’s experiment [24], smaller than 0.7° in Minini’s experiment [25]). Therefore, the disparity-response curve in larger disparity remains unclear.

In light of this, we designed an fMRI-based experiment that included a more extensive binocular disparity range. Since there were additional shape differences between the Random Dots Stereograms (RDS) with zero disparity and nonzero disparity, cortical activations were not only from the designed stereoscopic depth differences but also from the confounding shape differences when comparing the two types of stimuli (Fig. 1). In order to verify further the functions of different regions, we modified the general RDS to eliminate confusion of shape (see Section 2). We hypothesized that cortical activity would not always increase with binocular disparity, and further, that the main cortices for processing stereoscopic depth were dorsal visual regions, rather than ventral visual regions consistent with Preston and Minini [24], [25], so there would be no activation in the ventral visual stream. The results showed that most activated regions were localized in the dorsal visual stream and intensity of dorsal activations did not always increase with disparities.

Section snippets

Subjects

Thirteen subjects (6 males and 7 females, 23–29 years) participated in the study. Experiments were approved by the local ethics committee and were in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki). All subjects provided informed consent and received compensation for their participation. All subjects have normal or corrected-to-normal vision. No participant revealed any brain tissue abnormality on anatomical MRI. Before the main experiment all

Activation contrast map

The activation contrast map was derived by comparing zero disparity condition and nonzero disparity conditions, and the average activation contrast map among 13 subjects were presented in Fig. 3. Comparing the activation distribution, there were some differences in the location of activation clusters between the left and right hemispheres. For the left hemisphere, the main clusters were localized at the end of the dorsal visual stream (V7). There were several smaller fragments that appeared

Discussion

Our study focuses on the relationship between visual cortex activation and binocular disparity size with a more extensive disparity range, which provides further evidence for the functions of dorsal and ventral visual streams in stereoscopic vision. The visual cortex is structurally divided into dorsal and ventral visual streams, which have been proved to be different in function [31], [35]. However, the two streams are believed to be both related with stereoscopic vision [5], [10], [15], [19],

Conclusions

The results of the present study indicate that the primary regions related to stereoscopic vision are several dorsal visual regions (V3, V3A, v7 and MT+/V5). They mainly perceive disparity-evoked spatial depth information, rather than shape information. The activation intensity of these regions is tuned by binocular disparity. When binocular disparity increases, activation intensity enhances firstly, and then weakens. The turning points are not always consistent between individuals.

Acknowledgements

This work was supported by Beijing Natural Science Foundation (7162112) and the National Natural Science Foundation of China (81171330). The authors thank the reviewers for their comments and suggestions.

References (39)

  • P.O. Bishop et al.

    Binocular interaction fields of single units in the cat striate cortex

    J Physiol

    (1971)
  • D. Ferster

    A comparison of binocular depth mechanisms in areas 17 and 18 of the cat visual cortex

    J Physiol

    (1981)
  • J.H. Maunsell et al.

    Functional properties of neurons in middle temporal visual area of the macaque monkey. II. Binocular interactions and sensitivity to binocular disparity

    J Neurophysiol

    (1983)
  • D.J. Felleman et al.

    Receptive field properties of neurons in area V3 of macaque monkey extrastriate cortex

    J Neurophysiol

    (1987)
  • J.W. Gnadt et al.

    Neurons in monkey parietal area LIP are tuned for eye-movement parameters in three-dimensional space

    J Neurophysiol

    (1995)
  • B.G. Cumming et al.

    Disparity detecting neurons in visual cortical area V1 give selective responses to anticorrelated random dot patterns that are impossible to fuse perceptually

    Soc Neurosci Abs

    (1996)
  • G.F. Poggio et al.

    Stereoscopic mechanisms in monkey visual cortex: binocular correlation and disparity selectivity

    J Neurosci

    (1988)
  • I. Ohzawa et al.

    Stereoscopic depth discrimination in the visual cortex: neurons ideally suited as disparity detectors

    Science

    (1990)
  • G.C. DeAngelis et al.

    Organization of disparity-selective neurons in macaque area MT

    J Neurosci

    (1999)
  • Cited by (0)

    1

    Present address: Medical Physics, Department of Radiology, University Medical Center Freiburg, Freiburg 79106, Germany.

    View full text