Depth cue combination in spontaneous eye movements

Visual responses and eye movement (EM) -related activities were studied in single units of the superior colliculus (SC) of alert cats. Spontaneous EMs were encouraged by training. Throughout the SC (i.e., in intermediate and deep layers as well as in superficial layers), units were found to respond well to visual stimuli. Strong and consistent responses could be elicited by very dim, low-contrast stationary stimuli. Visual responses varied from phasic to tonic; some units responded tonically to stationary stimuli in the center of the receptive field, and phasically to peripheral stimuli. Many cells responded more vigorously to moving than to stationary stimuli, but very few responded exclusively to stimulus movement. The vast majority of cells were directionally selective. A small number of units were sensitive to the absolute, as well as the retinal, position of visual stimuli. These cells were activated by visual stimuli which fell in the receptive field only if the cat's gaze was fixated on one half of the screen. It seems that these cells must receive information about both eye position and the retinal (receptive field) position of the stimulus. It is possible that they reflect coding of target location within a head (or body) frame of reference. EM-related units were of two types: (1) about 20% of the sample responded prior to spontaneous or visually-triggered EMs, and (2) another 10% (or more) responded with, but not before, EMs. Some cells in the second group discharge almost synchronously with EMs and, thus, cannot plausibly be said to respond to the movement of images across the retina. All cells in the first group were directionally selective. The percentage of EM-related cells in the deep layers of SC is lower in cat than in monkey. Possible reasons for such differences are discussed.

Spatial computations underlying the coordination of the hand and eye present formidable geometric challenges. One way for the nervous system to simplify these computations is to directly encode the relative position of the hand and the center of gaze. Neurons in the dorsal premotor cortex (PMd), which is critical for the guidance of arm-reaching movements, encode the relative position of the hand, gaze and goal of reaching movements. This suggests that PMd can coordinate reaching movements with eye movements. Here, we examine saccade-related signals in PMd to determine whether they also point to a role for PMd in coordinating visual-motor behavior. We first compared the activity of a population of PMd neurons with a population of parietal reach region (PRR) neurons. During center-out reaching and saccade tasks, PMd neurons responded more strongly before saccades than PRR neurons and PMd contained a larger proportion of exclusively saccade-tuned cells than PRR. During a saccade relative position coding task, PMd neurons encoded saccade targets in a relative position code that depended on the relative position of gaze, the hand, and the goal of a saccadic eye movement. This relative position code for saccades is similar to the way that PMd neurons encode reach targets. We propose that eye movement and eye position signals in PMd do not drive eye movements, but rather provide spatial information that links the control of eye and arm movements in order to support coordinated visual-motor behavior.

In many situations, we perceive 3-D depth feeling from 2-D images. Among various depth cues, monocular depth cues make us perceive depth from single-view images. In this paper, we propose several methods to enhance monocular depth cues and improve depth perception of single-view images. The proposed methods deal with four depth cues: linear perspective, aerial perspective, focus, and shadow effects. Once we have a 2-D colour image and its corresponding depth information, we can modify its depth cues and enhance 3-D depth perception in the 2-D image. Experimental results show that the proposed methods can enhance depth perception of single-view images without any considerable visual distortions.

1. The spike activity of deep cerebellar nuclear neurons was recorded in the alert cat during spontaneous and during vestibularly and visually induced eye movements. 2. Neurons were classified according to their location in the nuclei, their antidromic activation from projection sites, their sensitivity to eye position and velocity during spontaneous eye movements, and their responses to vestibular and optokinetic stimuli. 3. Type I EPV (eye position and velocity) neurons were located mainly in the posterior part of the fastigial nucleus and activated antidromically almost exclusively from the medial longitudinal fasciculus close to the oculomotor complex. These neurons, reported here for the first time, increased their firing rate during saccades and eye fixations towards the contralateral hemifield. Their position sensitivity to eye fixations in the horizontal plane was 5.3 +/- 2.6 spikes s-1 deg-1 (mean +/- S.D.). Eye velocity sensitivity during horizontal saccades was 0.71 +/- 0.52 spikes s-1 deg-1 s-1. Variability of their firing rate during a given eye fixation was higher than that shown by abducens motoneurons. 4. Type I EPV neurons increased their firing rate during ipsilateral head rotations at 0.5 Hz with a mean phase lead over eye position of 95.3 +/- 9.5 deg. They were also activated by contralateral optokinetic stimulation at 30 deg s-1. Their sensitivity to eye position and velocity in the horizontal plane during vestibular and optokinetic stimuli yielded values similar to those obtained for spontaneous eye movements. 5. Type II neurons were located in both fastigial and dentate nuclei and were activated antidromically from the restiform body, the medial longitudinal fasciculus close to the oculomotor complex, the red nucleus and the pontine nuclei. Type II neurons were not related to spontaneous eye movements. These neurons increased their firing rate in response to contralateral head rotation and during ipsilateral optokinetic stimulation, and decreased it with the oppositely directed movements. 6. Saccade-related neurons were located mostly in the fastigial and dentate nuclei. Fastigial neurons were activated antidromically from the medial longitudinal fasciculus, while dentate neurons were activated from the red nucleus. These neurons fired a burst of spikes whose duration was significantly related to saccade duration. Dentate neurons responded during the omni-directional saccades, while some fastigial neurons fired more actively during contralateral saccades. 7. These three types of neuron represent the output channel for oculomotor signals of the posterior vermis and paravermis. It is proposed that type I EPV neurons correspond to a group of premotor neurons directly involved in oculomotor control.(ABSTRACT TRUNCATED AT 400 WORDS)

The role of visual salience in directing eye movements in visual object agnosia

Microsaccades are tiny eye movements typically occurring spontaneously and without awareness but can also be intentionally controlled with high precision. We used these tiny visual actions to investigate how intention modulates sensorimotor awareness by directly comparing intended (upon instruction), unintended (occurring despite instruction to fixate), and spontaneous microsaccades. In addition, we dissociated the effects of action intention and the actions' visual consequences on awareness. To achieve this, we presented a stimulus at high temporal frequency rendering it invisible during stable fixation. Critically, this stimulus became visible when it slowed down on the retina, either incidentally, due to a microsaccade with comparable direction and speed, or physically, when replaying the retinal consequence of previous microsaccades. Trials without a stimulus were included as control. Participants reported whether they perceived the stimulus (visual sensitivity), whether they believed they had made a microsaccade (microsaccade sensitivity), and their level of confidence that their eye movement behavior was linked to their perception (causality assignment). Visual sensitivity was high for both generated and replayed microsaccades and comparable for intended, unintended, and spontaneous eye movements. Microsaccade sensitivity, however, was low for spontaneous microsaccades, but heightened for both intended and unintended movements. Thus, the intention to saccade or fixate enhances awareness of otherwise undetected eye movements. Visual consequences failed to aid eye movement awareness, and confidence ratings revealed a poor understanding of a causal relationship between eye movement and sensory consequence. These findings highlight the functional relevance of intention in sensorimotor awareness at the smallest scale of visual actions.

Scanning eye movements in schizophrenic patients: Relationship to clinical symptoms and regional cerebral blood flow using 123I-IMP SPECT

Reaching movements towards stereoscopically presented virtual objects have been reported to be imprecise. This might be a problem for touch interaction with virtual environments. Estimating the distance to an object in personal space relies on binocular disparity and other depth cues but previous work on the influence of familiar size for reaching and grasping has produced conflicting results. We presented a virtual tennis ball and manipulated binocular disparity as well as the size of the tennis ball. The results suggest that depth information from binocular disparity and from familiar size is combined for reaching movements towards virtual objects. However, subjects differed in the weight they assigned to each depth cue.

Saccadic localisation of targets of various properties has been extensively studied, but rarely for texture-defined figures. In this paper, three experiments that investigate the way information from a texture target is processed in order to provide a signal for eye movement control are presented. Participants made saccades to target regions embedded in a background structure, and the saccade landing position and latency were measured. The textures comprised line elements, with orientations of the lines configured to form the figure and ground. Various orientation profile configurations (Block, Blur, and Cornsweet), were used in order to measure the role of edge profiles in driving eye movements and producing salience. We found that in all cases the visual system is in fact able to effectively segregate a texture figure from the ground in order to accurately plan a saccade to the target-figure. While saccadic latency was the highest for the Blur profile, the mean saccadic landing position was mostly unaffected by the various profiles (Experiment 1). More specifically, we showed that saccades were directed to the centre-of-gravity of the target (Experiment 2). We also found that figures with information of orientation contrast at both the edge and centre of figure (i.e. Block) produced the highest level of saliency in attracting eye movements (Experiment 3). Overall, the results show that saccades are planned on the representation of the whole target shape rather than a local salient region based on orientation contrast cues, and that the various texture profiles were important only to the extent that they affected the time to programme a saccade.

Chapter 12 - Modeling the effects of lexical ambiguity on eye movements during reading

Mixed reality technology can be applied to simulation training, improve surgical performance, enhance 3D game experience, and so on, attracting extensive attention of researchers. The perception of the user about head‐mounted display MR using HoloLens is critical, especially for precision applications such as virtual hoisting. Designing and adding appropriate depth cues in MR scenes is an effective way to improve users' depth perception. In this study, taking virtual hoisting training system as an example, the depth perception strategy of multi‐cue fusion is proposed to improve the perception effect. Based on the mechanism of human depth perception, five kinds of depth cues are designed. The depth perception effect of adding single cue is studied by perceptual matching experiment. Based on the principle of fuzzy clustering, a multiple‐cue comprehensive depth optimization strategy on viewing distance scale is proposed. Finally, the perceptual matching results demonstrate the effectiveness of the multi‐cue fusion strategy, and the average error is reduced by 20.68% compared with the single‐cue strategy, which can significantly improve the spatial depth perception. This research can provide a reference for improving users' depth perception in interactive MR simulation systems.

The integration of disparity, shading and motion parallax cues for depth perception in humans and monkeys

Spontaneous vertical eye movements in the critical care setting are often a source of confusion and alarm; while their origin remains at least partly theoretical, understanding their classification and associated clinical implications can inform the diagnostic workup and further clinical management. This case describes a patient who demonstrated ocular dipping: slow conjugate downward eye movements with a quick return to primary gaze. Ocular dipping is a rare phenomenon that was initially described in patients with hypoxic brain injury and has since been described in only a handful of cases. Dipping resides in a spectrum of spontaneous vertical eye movements, with ocular bobbing being the first of these described eye movements. Ocular bobbing is characterized by a fast downward movement followed by a slow return to the mid gaze position which is classically associated with pontine injury. Other vertical eye movements that can be seen in patients with a disorder of consciousness include other variations of ocular bobbing and dipping, vertical myoclonus, and small-amplitude mainly vertical movements.

Medial rectus motoneurons mediate nasally directed horizontal eye movements. These motoneurons receive two major excitatory inputs, from the abducens internuclear neurons (ABD Ints) and neurons of the lateral vestibular nucleus whose axons course through the ascending tract of Deiters (ATD). In the present work, we have recorded in the alert chronic cat preparation the discharge activity of these two premotor neurons simultaneously with eye movements, to discern their relative contribution to the firing pattern of medial rectus motoneurons. ABD Int discharge was accurately correlated with eye movements, displaying high sensitivities to eye position and eye velocity. ATD neurons also discharged in relation to spontaneous and vestibular eye movements but showed significantly lower eye position and eye velocity sensitivities. Outstandingly, ATD neurons presented a significantly lower eye position threshold for recruitment compared to both ABD Ints and medial rectus motoneurons. Therefore, ATD neurons exhibited eye position and velocity signals during spontaneous and vestibular eye movements, which were of lower magnitude than those of ABD Ints, but due to their low recruitment threshold, they could play a significant role in facilitating ABD Int signal transmission onto medial rectus motoneurons.

Polyphemus eye movements were recorded in both pitching and yawing planes, both in a static visual environment and with a sinusoidally moving stimulus. Spontaneous eye movements (average amplitude 1·7°) had different properties in the two planes, with trembling movements predominating in the pitching plane. A contour-sharpening function is proposed for these movements. An attempt to analyse the eye movement response system using a Bode diagram shows a very poor fit to the data, leading to the conclusion that a closed-loop control system is an inappropriate model in this case. The evoked eye movements are most convincingly represented by a model in which the time the stimulus takes to traverse a restricted sensitive zone in the central region of the eye controls the duration of a subsequent constant angular velocity saccade. The direction of the response movement follows the direction of the stimulus. A small-object tracking function is proposed for these movements.