Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Mar 2;22(4):12.
doi: 10.1167/jov.22.4.12.

What are the visuo-motor tendencies of omnidirectional scene free-viewing in virtual reality?

Affiliations

What are the visuo-motor tendencies of omnidirectional scene free-viewing in virtual reality?

Erwan Joël David et al. J Vis. .

Abstract

Central and peripheral vision during visual tasks have been extensively studied on two-dimensional screens, highlighting their perceptual and functional disparities. This study has two objectives: replicating on-screen gaze-contingent experiments removing central or peripheral field of view in virtual reality, and identifying visuo-motor biases specific to the exploration of 360 scenes with a wide field of view. Our results are useful for vision modelling, with applications in gaze position prediction (e.g., content compression and streaming). We ask how previous on-screen findings translate to conditions where observers can use their head to explore stimuli. We implemented a gaze-contingent paradigm to simulate loss of vision in virtual reality, participants could freely view omnidirectional natural scenes. This protocol allows the simulation of vision loss with an extended field of view (\(\gt \)80°) and studying the head's contributions to visual attention. The time-course of visuo-motor variables in our pure free-viewing task reveals long fixations and short saccades during first seconds of exploration, contrary to literature in visual tasks guided by instructions. We show that the effect of vision loss is reflected primarily on eye movements, in a manner consistent with two-dimensional screens literature. We hypothesize that head movements mainly serve to explore the scenes during free-viewing, the presence of masks did not significantly impact head scanning behaviours. We present new fixational and saccadic visuo-motor tendencies in a 360° context that we hope will help in the creation of gaze prediction models dedicated to virtual reality.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
The 29 omnidirectional images were used in this experiment, the first one (red border) was part of the training phase.
Figure 2.
Figure 2.
Masking conditions are presented here in a viewport measuring 90° by 90° of field of view. Radii are proportionally accurate. From left to right: central masks of 6° and 8° of radius, peripheral masks of 4° and 6° of radius.
Figure 3.
Figure 3.
(a) Absolute angles appear in green between a saccade vector (black arrows) and the horizontal axis (orange dashed lines). The horizontal saccade directionality (HSD) reports the proportion of left-directed saccades within horizontal saccades; the horizontal saccade percentage (HSP) measures the proportion of horizontal saccades among all saccades observed. (b) Relative angles (green arcs) are angles between two saccades vectors (black arrows and orange dashed lines). The saccadic reversal rate (SRR; Asfaw et al., 2018) measures the number of backward saccades falling between 170 and 190 as a proportion of the total amount of saccades. The forward saccade segment length (FSSL) reports on the length distribution of consecutive saccades directed approximately in the same relative forward direction.
Figure 4.
Figure 4.
Average and 95% CI of fixation durations calculated across subjects and stimuli (on a log-scale). The X axis labels have been replaced with icons representing mask types and radii, from left to right: central mask 6°, central 8°, peripheral 6°, peripheral 4°. Mask conditions are ordered by increasing surface masked. The control condition is present as a black line crossing the plots horizontally (mean is shown as a solid black line, dashed lines report 95% CI).
Figure 5.
Figure 5.
Average and 95% CI of the head, eye and combined movement amplitude during saccades calculated across subjects and stimuli (on a log-scale). For a more detailed view, we present in Figure A.7 density distributions of gaze, eye and head movements as a function of the amplitude of the motion during saccades. The X axis labels have been replaced with icons representing mask types and radii, from left to right: central mask 6°, central 8°, peripheral 6°, peripheral 4°. The control condition is present as black lines crossing the plots horizontally (mean is shown as a solid black line, dashed lines report 95% CI).
Figure 6.
Figure 6.
Joint distribution of saccade amplitude and absolute direction as a function of masking condition. The red circles represent the mask radius.
Figure 7.
Figure 7.
Joint distribution of saccade amplitude and relative direction as a function of masking condition. The red circles represent the mask radius.
Figure 8.
Figure 8.
Mean and 95% CI of forward saccade segment length (FSSL) as a function of viewing time. A higher FSSL value means that a saccade at that point in time was on average part of a longer sequence of saccades travelling in the same direction. Colour legend: formula image no-mask, formula image central 6°, formula image central 8°, formula image peripheral 6°, formula image peripheral 4°.
Figure 9.
Figure 9.
Mean and 95% CI of visuo-motor variables as a function of viewing time. Colour legend: formula image no-mask, formula image central 6°, formula image central 8°, formula image peripheral 6°, formula image peripheral 4°. Fixation durations and saccade amplitudes are displayed on a log-scale.
Figure A.1.
Figure A.1.
Average amplitudes and confidence intervals (95%) of gaze saccades as a function of mask types and radii. “C-4” refers to a central mask of 4° of radius, “P-6” to a peripheral mask of 6°.
Figure A.2.
Figure A.2.
Time-course average of oculo-motor variables gathered on-screen in David et al. (2019). We report time-dependent means (coloured lines) and 95% CI (grey ribbons) for control (Ctrl), central-masking (C-*), and peripheral-masking (P-*) conditions. The numbers after C- and P- refer to the gaze-contingent mask's radius. Fixation durations and saccade amplitudes were log-transformed to better estimate distribution means. Fixation durations and saccade amplitudes are displayed on a log-scale.
Figure A.3.
Figure A.3.
Fixation position count (blue) by latitude (top, 0 to 180) and longitude (bottom, -180 to 180). PDF curves (red) are fitted with a Gaussian kernel for latitudes and a von Mises kernel for longitudes. The first second of the dataset was removed to decrease an effect of the starting position.
Figure A.4.
Figure A.4.
Density distribution estimation (by Gaussian kernel) of fixation positions in the viewport as a function of the headset's latitudinal rotation (pitch). Below each subfigure is reported the number of fixations sampled in the latitudinal range divided by the total number of fixations (RatioN). Medianδy informs about the median distance of the fixation positions on the vertical axis to the center of the viewport.
Figure A.5.
Figure A.5.
Average and 95% CI of HSD and HSP measures of absolute saccade direction calculated across subjects and stimuli. The X axis labels have been replaced with icons representing mask types and radii, from left to right: central mask 6°, central 8°, peripheral 6°, and peripheral 4°. The control condition is present as black lines crossing the plots horizontally (mean is shown as a solid black line, dashed lines report 95% CI).
Figure A.6.
Figure A.6.
Average and 95% CI of saccadic reversal rates (SRR) of relative saccade direction calculated across subjects and stimuli. The X axis labels have been replaced with icons representing mask types and radii, from left to right: central mask 6°, central 8°, peripheral 6°, peripheral 4°. The control condition is present as black lines crossing the plots horizontally (mean is shown as a solid black line, dashed lines report 95% CI).
Figure A.7.
Figure A.7.
Figures A.7a, A.7b and A.7c show density estimations of saccade distributions as a function of gaze, eye and head movement amplitudes. Gaze movement amplitudes (Figure A.7a) are presented with a maxima of 100°. Colour legend: formula image no-mask, formula image central 6°, formula image central 8°, formula image peripheral 6°, and formula image peripheral 4°.

Similar articles

Cited by

References

    1. Aguilar, C., & Castet, E. (2011). Gaze-contingent simulation of retinopathy: Some potential pitfalls and remedies. Vision Research, 51(9), 997–1012. - PubMed
    1. Albert, R., Patney, A., Luebke, D., & Kim, J. (2017). Latency requirements for foveated rendering in virtual reality. ACM Transactions on Applied Perception (TAP), 14(4), 25.
    1. Anderson, N. C., Bischof, W. F., Foulsham, T., & Kingstone, A. (2020). Turning the (virtual) world around: patterns in saccade direction vary with picture orientation and shape in virtual reality. Journal of Vision, 20(8), 1–19. - PMC - PubMed
    1. Antes, J. R. (1974). The time course of picture viewing. Journal of Experimental Psychology, 103(1), 62. - PubMed
    1. Asfaw, D. S., Jones, P. R., Mönter, V. M., Smith, N. D., & Crabb, D. P. (2018). Does glaucoma alter eye movements when viewing images of natural scenes? a between-eye study. Investigative Ophthalmology & Visual Science, 59(8), 3189–3198. - PubMed

Publication types