A paradoxical misperception of relative motion

Abstract

Motion perception is considered a hyperacuity. The presence of a visual frame of reference to compute relative motion is necessary to achieve this sensitivity [Legge, Gordon E., and F. W. Campbell. "Displacement detection in human vision." Vision Research 21.2 (1981): 205-213.]. However, there is a special condition where humans are unable to accurately detect relative motion: images moving in a direction consistent with retinal slip where the motion is unnaturally amplified can, under some conditions, appear stable [Arathorn, David W., et al. "How the unstable eye sees a stable and moving world." Journal of Vision 13.10.22 (2013)]. In this study, we asked: Is world-fixed retinal image background content necessary for the visual system to compute the direction of eye motion to render in the percept images moving with amplified slip as stable? Or, are non-visual cues sufficient? Subjects adjusted the parameters of a stimulus moving in a random trajectory to match the perceived motion of images moving contingent to the retina. Experiments were done with and without retinal image background content. The perceived motion of stimuli moving with amplified retinal slip was suppressed in the presence of visual content; however, higher magnitudes of motion were perceived under conditions with no visual cues. Our results demonstrate that the presence of retinal image background content is essential for the visual system to compute its direction of motion. The visual content that might be thought to provide a strong frame of reference to detect amplified retinal slips, instead paradoxically drives the misperception of relative motion.

Fig. 1.

Display configuration. (a) The projector drew a fixation target and the surrounding 17° field displayed either white light or a binarized noise pattern. The AOSLO generated the stimuli positioned 2° temporally. These images were simultaneously projected onto the retina through a beamsplitter. (b) Three stimuli conditions: a Gain −1.5 stimulus moves with increased retinal slip, a Gain +1.5 stimulus moves in advance of retinal motion, and a Gain 0 stimulus is world-fixed. Undergoing the same retinal motion, a Gain −1.5 stimulus moves 2.5x more on the retina than a Gain 0 stimulus, and a Gain 0 stimulus moves 2x more on the retina than a Gain +1.5 stimulus. (c) Simulated trajectories showing uncorrelated, Brownian motion (left) and positively correlated, persistent motion (right). The red dot indicates the starting position for each trace. Both paths were generated from the same average step length, L = 0.6 arcminutes, and have similar speeds (S), yet have different diffusion constants (D) due to different α values. Full descriptions of these parameters are in Materials and Methods. (d-e) Experiment sequence. Subjects fixated on the target and attended to the stimuli positioned 2° temporally. The retina-contingent stimulus moved contingent to each subject’s idiosyncratic fixational eye motion. The random walk stimulus moved independent to eye motion, with pre-generated random offsets. Under background-present conditions (d), the projector field displayed a binarized noise pattern which changed after each presentation and the fixation target remained on for the entire duration. Under background-absent conditions (e), the “Ganzfeld effect” was achieved by setting up a white paper with an aperture in front of the display permitting only the AOSLO and projector beams to enter the eye, shown in (a). LEDs were taped around the eye to illuminate the paper and the luminance was adjusted so that the subject saw only the stimuli in a white full-field surround. Owing to its proximity to the eye, the natural blur of the aperture rendered the transition between the display and luminance-matched paper invisible. The fixation target was timed to turned off when the stimuli were presented.

Fig. 2.

Individual results from six subjects tested under background-present and background-absent conditions. The conditions are indicated by labels on the top left corner of each graph. The small circles are the six perceptual matches for Gain −1.5 stimuli (blue open symbols), Gain +1.5 stimuli (green filled symbols), and Gain 0 stimuli (yellow filled symbols). The large circles are the averages of the matches with standard error of the mean bars. The red arrows show the extent to which the ${\alpha }_{WM}$ deviated from Brownian motion. Arrows pointing right indicate ${\alpha }_{WM}>1$, arrows pointing left indicate ${\alpha }_{WM}<1$, and no arrow means that ${\alpha }_{WM}=1+/-0.02$. Longer arrows correspond to higher deviations from Brownian motion. The arrow length in the legend indicates pure persistence (${\alpha }_{WM}=2$) if pointing right or pure antipersistence (${\alpha }_{WM}=0$) if pointing left.

Fig. 3.

Computed ratios from six subjects tested under background-present and background-absent conditions. The red arrows show the extent to which the ${\alpha }_{WM}$ deviated from Brownian motion. Arrows pointing up and right indicate ${\alpha }_{WM}>1$, arrows pointing left indicate ${\alpha }_{WM}<1$, and no arrow means that ${\alpha }_{WM}=1+/-0.02$. Longer arrows correspond to higher deviations from Brownian motion. The arrow length in the legend indicates pure persistence $\left({\alpha }_{WM}=2\right.$) if pointing right or pure antipersistence (${\alpha }_{WM}=0)$ if pointing up/right or pure antipersistence if pointing down/left (${\alpha }_{WM}=0$).

Fig. 4.

a-c Violin plots (9) showing the distribution of the (a) $D_{EM}$, the (b) ${\alpha }_{EM}$, and (c) $S_{EM}$ for three Gains (−1.5, +1.5, and 0) under two conditions (background-present and background-absent). In the center of the violin, the white circle represents the median and the dark bars represent the interquartile range. The black points are the (a) mean $D_{EM}$, (b) mean ${\alpha }_{EM}$, and (c) mean $S_{EM}$ for each subject across their six respective trials for each Gain. The asterisks indicate statistical significance (p < 0.05 and p < 0.01, respectively) from a post hoc Tukey-Kramer test following a two-factor repeated-measures ANOVA.

Fig. 5.

Example gaze traces for subject 10003L during Gain −1.5 retina-contingent presentations under (c) background-present (1500-ms duration and ${\alpha }_{EM}=0.90$) and (d) background-absent (750-ms duration and ${\alpha }_{EM}=1.29$) conditions. The red dot indicates the starting position for each trace. The gaze directions are labeled: left (L), right (R), up (U), and down (D).

Fig. 6.

Computed ratios from six subjects plotted as a function of each subject’s ${\alpha }_{WM}$. Data from two experiments are shown: Gain −1.5 stimuli tested under background-absent conditions and Gain +1.5 stimuli tested under background-present conditions. The black line is the simulated $D_{PM}/D_{WM}$ versus ${\alpha }_{WM}$, where the $MS{D}_{PM}=MS{D}_{WM}$ but the random walk stimulus’s $\alpha$ is equal to one (Brownian motion) while the retina-contingent stimulus’s ${\alpha }_{WM}$ ranged from 0.9 to 1.8. Both the model and data points show exponential decay.