An automated eye movement recording system for use with human infants

An infrared corneal reflection eye movement system modified for work with human infants is described. The system generates on-line digital information on eye position at a sampling rate that allows examination of the temporal and spatial characteristics of fixations and eye move-ments. An algorithm for defining fixation and eye movement episodes is described.     

On-line eye movement recorders: The good,the bad,and the ugly

The first section of this paper describes three types of problems inherent in the use of on-line eye movement recorders (EMRs), These types of problems are (1) that software is difficult to develop, (2) that an on-line EMR is subject to all difficulties that are normally inherent in computer use and, (3) that with an on-line EMR. it is difficult to detect when the gaze is falling on an object that has an irregularly shaped boundary. It is argued that a large class of eye movement experimentation could be done as easily or more easily with an off-line EMR. The second section of the paper describes an off-line EMR laboratory currently in use by the author.     

Focus identification during sentence comprehension: evidence from eye movements

Three eye movement experiments investigated focus identification during sentence comprehension. Participants read dative or double-object sentences (i.e., either the direct or indirect object occurred first), and a replacive continuation supplied a contrast that was congruous with either the direct or the indirect object. Experiments 1 and 2 manipulated focus by locating only adjacent to either the direct or indirect object of dative (Experiment 1) or double-object (Experiment 2) sentences. Reading-time effects indicated that the surface position of the focus particle influenced processing. In addition, Experiment 1 reading times were longer when the replacive was incongruous with the constituent that only adjoined, and particle position modulated a similar effect in Experiment 2. Experiment 3 showed that this effect was absent when only was omitted. We conclude that the surface position of a focus particle modulates focus identification during on-line sentence comprehension.     

Eye-hand coordination in goal-directed aiming.

In a number of studies, we have demonstrated that the spatial-temporal coupling of eye and hand movements is optimal for the pickup of visual information about the position of the hand and the target late in the hand's trajectory. Several experiments designed to examine temporal coupling have shown that the eyes arrive at the target area concurrently with the hand achieving peak acceleration. Between the time the hand reached peak velocity and the end of the movement, increased variability in the position of the shoulder and the elbow was accompanied by a decreased spatial variability in the hand. Presumably, this reduction in variability was due to the use of retinal and extra-retinal information about the relative positions of the eye, hand and target. However, the hand does not appear to be a slave to the eye. For example, we have been able to decouple eye movements and hand movements using Müller-Lyer configurations as targets. Predictable bias, found in primary and corrective saccadic eye movements, was not found for hand movements, if on-line visual information about the target was available during aiming. That is, the hand remained accurate even when the eye had a tendency to undershoot or overshoot the target position. However, biases of the hand were evident, at least in the initial portion of an aiming movement, when vision of the target was removed and vision of the hand remained. These findings accent the versatility of human motor control and have implications for current models of visual processing and limb control.     

The position of random autokinetic movement and the physiological position of rest are frequently stable and identical

Both the autokinetic illusion (AKI) and involuntary eye movements when a fixation point goes off have been attributed to unmonitored drift eye movements which result from constant features of the oculomotor system. Pilot studies confirmed that the visual directions where these effects had no directional bias, the position of random autokinetic movement (PRAKM) and the physiological position of rest (PPR), were highly correlated and reliable. The more precise main experiment showed that they were usually, but not always, stable and identical. Drifts are therefore not totally due to stable features of the oculomotor system but are a compound of several slow responses which can differ with both stimulus conditions and time. They are not all equally unmonitored.     

Calibration of an eye-movement system for use in reading

A method for measuring horizontal eye movements in the msec range is described. Accurate measurement of horizontal eye movement over a linear range of 12° is achieved by processing the image of the eye illuminated with infrared light and with the head position fixed. The system has given very reliable results, and a resolution of 6 min of visual angle can be achieved with a character space of 45 min of arc. We also describe efficient numerical-data processing which allows the precise determination of the absolute position of the eye.     

Performance of a computer system for recording eye fixations using limbus reflection

A computer system was developed to measure and record locus and duration of the eye axis of gaze using a limbus reflection technique. System hardware and operating procedures for on-line data collection and correction are described. Algorithms are presented that separate fixations from movements. Based on these algorithms, system performance was measured in terms of spatial accuracy of eye fixations and temporal accuracy of saccadic movements.     

Focus identification during sentence comprehension: Evidence from eye movements

Three eye movement experiments investigated focus identification during sentence comprehension. Participants read dative or double-object sentences (i.e., either the direct or indirect object occurred first), and a replacive continuation supplied a contrast that was congruous with either the direct or the indirect object. Experiments 1 and 2 manipulated focus by locating only adjacent to either the direct or indirect object of dative (Experiment 1) or double-object (Experiment 2) sentences. Reading-time effects indicated that the surface position of the focus particle influenced processing. In addition, Experiment 1 reading times were longer when the replacive was incongruous with the constituent that only adjoined, and particle position modulated a similar effect in Experiment 2. Experiment 3 showed that this effect was absent when only was omitted. We conclude that the surface position of a focus particle modulates focus identification during on-line sentence comprehension.     

Data acquisition system for precision eye-fixation studies

A system has been developed to sense and record eye point-of-gaze for on-line computer processing, using a Mackworth camera in conjunction with a closed-circuit television system. System optics and electronics are described, together with computer software for system calibration and automatic artifact removal.     

The GO model: A reconsideration of the role of structural units in guiding and organizing text on line

Healy (1994) and Koriat and Greenberg (1994) offered different theoretical accounts of the missingletter effect (MLE) in the letter-detection task, whereby a disproportionate number of letter-detection errors occur in frequent function words. Healy emphasized identification processes, whereas Koriat and Greenberg viewed the structural role of the embedding word to be crucial. Recent research suggests that neither position alone can account for the complete set of observations pertaining to the MLE. The present paper offers a theoretical integration of these competing explanations of letter detection in terms of a GO (guidance-organization) model of reading. This model specifies how structural processing of connected text helps guide eye movements to semantically informative parts of the text, enabling readers to achieve on-line fluency.     

High-speed data processing and unobtrusive monitoring of eye movements

A system has been developed that monitors eye movements without imposing any unnatural constraints or mechanical attachments on the subject. The system also features high-speed on-line data processing.     

On the role of refixations in letter strings: the influence of oculomotor factors

Recent studies of reading and word recognition have shown that eye-movement behavior depends strongly on the position in the word that the eye first fixates; the probability of refixating in a word is lowest with the eye near the middle of the word, and it increases as the eye fixates to either side. It has generally been assumed that the cause for this optimal landing position phenomenon lies in the very strong drop-off of visual acuity even within the fovea; refixation should be more likely when the eye starts from a noncentral position, because here less information can be extracted during one fixation. It may, however, be the case that the phenomenon is caused not by acuity drop-off, but by differences in within-word oculomotor scanning tactics as a function of the position that the eye initially fixates. To test this, in the present experiment we kept visual information constant while we varied the initial fixation position. We used homogeneous strings of letters of different length. One letter in each string was different from the rest (e.g., kkkkkok), and this was the letter that the subject initially fixated. This target letter had to be identified before saccading to a comparison string. The position of the target letter in the string was varied from trial to trial. If, owing to acuity limitations, refixations reflect insufficient information extraction, then, because the target letter is always directly fixated, the pattern of refixations in this condition should be independent of the first fixation position. However, the obtained refixation probability showed a strong dependence on the position of first fixation. The number of refixations was independent of the absolute length of the letter strings, but it seemed to be influenced by the proportion of the string over which the eye had to pass. The larger this proportion, the higher the probability of refixation. The results suggest that to a certain extent refixations in letter strings (or words) reflect properties of the oculomotor system rather than visual information extraction.     

Colour and spatial cueing in low-prevalence visual search

People are able to judge the current position of occluded moving objects. This operation is known as motion extrapolation. It has previously been suggested that motion extrapolation is independent of the oculomotor system. Here we revisited this question by measuring eye position while participants completed two types of motion extrapolation task. In one task, a moving visual target travelled rightwards, disappeared, then reappeared further along its trajectory. Participants discriminated correct reappearance times from incorrect (too early or too late) with a two-alternative forced-choice button press. In the second task, the target travelled rightwards behind a visible, rectangular occluder, and participants pressed a button at the time when they judged it should reappear. In both tasks, performance was significantly different under fixation as compared to free eye movement conditions. When eye movements were permitted, eye movements during occlusion were related to participants' judgements. Finally, even when participants were required to fixate, small changes in eye position around fixation (<2°) were influenced by occluded target motion. These results all indicate that overlapping systems control eye movements and judgements on motion extrapolation tasks. This has implications for understanding the mechanism underlying motion extrapolation.     

Extraretinal information about eye position during involuntary eye movement: optokinetic afternystagmus

Despite importance for theories of perception, controversy exists as to whether information is available to the perceptual system about involuntary as well as voluntary eye movements. We measured the perceived direction of targets flashed briefly in an otherwise dark field during the primary phase of optokinetic afternystagmus (OKAN), an involuntary eye movement that persists in darkness following optokinetic stimulation. Perceived direction was measured by unseen pointing in one experiment and by pointing made under visual control in a second experiment. Pointing was essentially veridical in both experiments, indicating that accurate extra-retinal information about eye position (presumably, as efference copy) exists for OKAN. Illusory motion of visual targets, which can occur during involuntary oculomotor responses, therefore cannot be attributed to a lack of efference-copy signals for such eye movements.     

A method for measuring eye movements using Hall-effect devices

A system for precise recording of eye position and movements in laboratory animals, by means of Hall-effect devices, is described. The system, useful in neurophysiological and neurobehavioral studies, allows the analysis of saccadic eye movements, optokinetic- and vestibular-induced nystagmus, slow tracking movements, eye vergences, and so forth. This small, light-weight, and inexpensive system uses a set of Hall-effect devices and associated electronics to sense variations in the position of high-power magnets fixed in the eye sclera or in scleral contact lenses. The output of the Hall-effect devices is amplified by operational amplifiers, collected through an A/D converter, and analyzed in a PC computer by specific software.     

An eye movement analysis system (EMAS) for the identification of cognitive processes on figural tasks

EMAS is a software system (written in VAX-11 FORTRAN) for the analysis of eye movement data recorded during the performance of figural tasks. Its main functions are: (1) Calibration of raw coordinates of eye movements to determine their actual position on the stimulus display. Different kinds of measurement distortion may be corrected. (2) Identification of eye fixations and the determination of their locations and durations. (3) Analysis of fixation sequences. The frequency of transitions of fixations among specified sectors of the stimulus display is computed. A sequential list is made of the successive fixations in which the fixated sector and the fixation duration are graphically indicated. (4) Plotting of raw or calibrated eye movement data and fixation points. The sequence of fixations in specified display sectors can also be plotted in real time. Applications of the programs to the Embedded Figures Test and the Hidden Figures Test are illustrated.     

Do common systems control eye movements and motion extrapolation?

People are able to judge the current position of occluded moving objects. This operation is known as motion extrapolation. It has previously been suggested that motion extrapolation is independent of the oculomotor system. Here we revisited this question by measuring eye position while participants completed two types of motion extrapolation task. In one task, a moving visual target travelled rightwards, disappeared, then reappeared further along its trajectory. Participants discriminated correct reappearance times from incorrect (too early or too late) with a two-alternative forced-choice button press. In the second task, the target travelled rightwards behind a visible, rectangular occluder, and participants pressed a button at the time when they judged it should reappear. In both tasks, performance was significantly different under fixation as compared to free eye movement conditions. When eye movements were permitted, eye movements during occlusion were related to participants' judgements. Finally, even when participants were required to fixate, small changes in eye position around fixation (<2°) were influenced by occluded target motion. These results all indicate that overlapping systems control eye movements and judgements on motion extrapolation tasks. This has implications for understanding the mechanism underlying motion extrapolation.     

Remembering a location makes the eyes curve away

Working memory is a system that keeps limited information on-line for immediate access by cognitive processes. This type of active maintenance is important for everyday life activities. The present study shows that maintaining a location in spatial working memory affects the trajectories of saccadic eye movements toward visual targets, as the eyes deviate away from the remembered location. This finding provides direct evidence for a strong overlap between spatial working memory and the eye movement system. We argue that curvature is the result of the need to inhibit memory-based eye movement activity in the superior colliculus, in order to allow an accurate saccade to the visual target. Whereas previous research has shown that the eyes may deviate away from visually presented stimuli that need to be ignored, we show that the eyes also curve away from remembered stimuli.