Horizontal Fusionol Responses to Stimuli Containing Artificial Scotomas

Horizontal Fusionol Responses to Stimuli Containing Artificial Scotomas Duane K. Boman* and Andrew E. Kerresz Horizontal fusional responses were studied with stimuli containing binocular or monocular, artificial, stabilized, scotomas. Binocular scotomas of 5-deg, 10-deg, and 15-deg diameters were utilized. The fusional responses to scotomatic stimuli were compared with full-field stimulus responses. AH responses contained significant motor and nonmotor (sensory) components. Overall motor compensation to stimuli with 10-deg and 15-deg scotomas was reduced, while the overall motor compensation to stimuli with 5-deg scotomas was not. With full-field stimuli and with stimuli containing binocular scotomas, the changes in the two eyes1 lines of sight were often asymmetric in response to symmetric disparity changes. This response asymmetry was exacerbated by the presence of monocular scotomas. Fixation was less steady with stimuli containing 10-deg or 15-deg binocular scotomas than it was in response to full-field stimulation. The fusional responses to annular stimuli were similar to those elicited by scotomatic stimuli. Invest Ophthalmol Vis Sci 26:1051-1056, 1985

The visual periphery plays an important role in both normal and abnormal human binocular vision and its influence on fusional response has come under close scrutiny.1"9 Burian1 was first to examine the effect of extrafoveal fusional stimulation on fusional responses. He found that the introduction of a vertical disparity between stimuli located up to 12 deg above the print of fixation could produce a displacement of centrally located nonius markers and could actually disrupt central fusion even though the subjects were unable to report whether the peripheral stimuli were fused. He later used this subjective technique to show that strabismic patients produced vergence responses to peripherally located fusional stimuli.2 Kertesz and Hampton 7 studied horizontal and vertical fusional responses to extrafoveal stimulation by stabilizing a 10-deg artificial scotoma over the central visual field of one eye. Using line stimuli that subtended 50 deg of arc and an objective eye movement measuring technique, they found the following: fusional responses to the introduction of step disparities included both motor compensation and a non-

motor (sensory) component that was limited to the extent of Panum's fusional areas; the overall motor compensation was reduced; and the monocular eye movements were markedly asymmetric. However, it is quite possible that some of the asymmetry was caused by the monocular nature of the artificial scotoma. The objectives of this report are to: characterize horizontal fusional response to stimuli containing binocular artificial scotomas; and compare this response to that elicited by stimuli containing monocular artificial scotomas or full-field stimuli that cover the central visual fields. Binocular artificial scotomas of 5-deg, 10-deg, and 15-deg diameters were utilized to study the effect of scotoma size on fusional response. Annular stimuli were also utilized to examine their adequacy to produce extrafoveal fusional stimulation.

Materials and Methods A computer-controlled, projection-type, video display device was used throughout these experiments. This device displays a 256 by 256 dot matrix to each eye allowing for dichoptic (red/green) presentations. The viewing distance was 115 cm, and the display screen subtended 40 deg of arc vertically and 50 deg horizontally. A detailed description of this device is provided elsewhere.10 The stimuli were based on the stimulus shown in Figure 1. It consisted of a 128 by 128 element random-dot (RD) stereogram upon which a 40 deg-long binocularly visible vertical line, composed of three segments, and two 25 deg-long horizontal nonius lines were superimposed. The intersec-

From the Biomedical Engineering Division and Department of Electrical Engineering and Computer Science, Northwestern University, Evanston, Illinois. Supported in part by research grant EY-1055 from the National Eye Institute. * Present address: Department of Neurology, Stanford University School of Medicine and Santa Clara Valley Medical Center, San Jose, CA 95128. Submitted for publication: October 22, 1984. Reprint requests: Dr. Andrew E. Kertesz, Biomedical Engineering Division, Northwestern University, 2145 Sheridan Road, Evanston, IL 60201.

1051

1052

INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / Augusr 1985

RIGHT EYE

LEFT EYE

y

/////////////A W///////////,

W////////////, W//////////A 25C

25°

Fig. 1. A random-dot stereogram subtending 50° by 40° was the basis for the stimuli used in the experiments. The horizontal and vertical coordinates of stimulus features are indicated relative to the fixation point. The random-dots within the crosshatched areas were presented at a 1 ° crossed disparity relative to the rest of the stimulus. A vertical line composed of three segments that was seen by both eyes and two horizontal nonius lines were superimposed on the random-dot stereogram.

tion of the vertical and horizontal lines provided a fixation point. The perceived appearance of the vertical line also provided a clue for the presence of fusion or diplopia while the nonius lines were used as suppression indicators. Two 4 deg-wide horizontal RD stripes, one 4 deg above the fixation point and the other 4 deg below it, were presented with a 1 deg crossed disparity with respect to the rest of the pattern, thus providing a depth cue. Horizontal eye movements were monitored by an infrared reflection method." This device has a resolution of 15 min of arc and is linear within a range of ±8 deg of arc. The eye positions were sampled at a rate of 20 Hz, digitized, and stored for later processing. In stabilization experiments, the eye positions were also directly fed back to the computer that controlled the stimulus device. This information was used to blank a circular area over the central visual field of one or both eyes. This artificial scotoma was centered and adjusted by a method similar to that of Kelly,12 and its position was updated at the rate of 20 Hz. Four stereonormal subjects participated in the experiments. They had stereoacuities of 1 min or better and no history of binocular problems. One wore contact lenses to achieve 20/20 acuity while the others were emmetropic. They were seated in a darkened room to eliminate conflicting visual cues. Chin and headrests were employed. Informed human consent was obtained prior to undertaking this study. Each experimental run began with the stimulus at the zero disparity position and the scotomas (if present) positioned in the visual periphery. The subjects were instructed to maintain fixation thoughout the run on either the intersection of the nonius lines and the vertical line or, in the runs with binocular scotomas, at the imaginary intersection of the vertical line and the nonius lines. The subjects were to report the

Vol. 2 6

presence of diplopia, suppression, loss of stereopsis, or if they "peeked" above or below the scotoma. In the rare occasions when either diplopia or "peeking" occurred, the run was discarded. Each run was replicated 12 times. The subjects initiated a run by pressing a button. The scotomas (if present) were then moved to and stabilized over the central visual fields. After 8 sec, a symmetric 5 deg convergent or divergent horizontal disparity was introduced in the form of a staircase with a 0.5 deg disparity step occurring each second. The 5 deg disparity position was maintained for 10 sec before the disparity was reduced to zero by another staircase. Each trial was 45 sec long. In each sitting, four or five stimulus conditions were presented three times in a random order along with two calibration runs. The eye movement records were analyzed to determine the overall motor compensation to the 5 deg disparity (disjunctive motor component) and the angular change in each eye's line of sight from the beginning of a run to the middle (during the presence of the 5 deg stimulus disparity). The disjunctive component of the eye movement recordings was extracted by subtracting the right eye's position from that of the left. Movements in the same direction as that of the stimulus were defined to be positive. Onetailed t-tests were employed at the 99% confidence level to compare: (1) the average overall motor compensation under different stimulus conditions; (2) the average change in the line of sight of the left eye to that of the right eye under each stimulus condition; and (3) the average number of saccades of greater than 0.5-deg amplitude under different stimulus conditions. Results Experiment 1 In the first experiment, vergence responses to fusional stimuli containing monocular and binocular artificial stabilized scotomas of various diameters were studied and compared to the responses to fullfield fusional stimuli that covered the central visual fields as well. Five stimuli were used: (1) full-field stimulus that was identical to that shown in Figure 1 and provided the baseline for comparison; (2) the same as (1) with a central 10 deg left monocular artificial scotoma; (3) the same as (1) with a central 5 deg binocular artificial scotoma; (4) the same as (1) with a central 10 deg binocular artificial scotoma; and (5) the same as (1) with a central 15 deg binocular artificial scotoma. Figures 2A, 2B, and 2C show typical responses by subject 2 to the introduction of a 5 deg convergent

No. 8

HORIZONTAL FUSIONAL RESPONSE TO SCOTOMATIC STIMULI / Bomon ond Kerresz

10* MONOCULAR SCOTOMA

0K8SCIRUN3

1053

DKBSC1 RUN 2

RIGHT EYE

Fig. 2. Responses by subject 2 to 5° convergent disparity presentations. The overall motor compensation and the change in each eye's line of sight are given in degrees of arc. A, Full-field stimulus. B, Stimulus with a central 10° scotoma stabilized on the left eye's visual field. C, Stimulus with a central binocular 10° scotoma. D, Stimulus with a 10° "annulus" presented to the left eye.

DISJUNCTIVE COMPONENT


STIMULUS DISPARITY

STIMULUS OlSPARrrY LEFT DIVERGENT

10° BINOCULAR SCOTOMA

DKBSCI RUN 1 10'MONOCULAR ANNULUS

D K 8 1 1 RUN 12


disparity while viewing the full-field stimulus, the stimulus with a monocular scotoma, and the stimulus with a 10-deg binocular scotoma, respectively. Substantial motor and nonmotor components are evident in each of these responses. The overall motor compensation to the scotomatic stimuli was less than that to the full-field stimulus, and the change in the line of sight of the left eye was much less than that of the right eye in the scotomatic responses. The subject's fixation was also less steady while viewing the stimulus with a binocular artificial scotoma than in the other responses. Table 1 presents each subject's average overall motor compensation (disjunctive) and the average change in each eye's line of sight under the different stimulus conditions. It can be seen that there was a substantial motor component in each case (table entry) as well as a substantial nonmotor component in many cases. Three of eight cases with monocular scotomas, three of eight cases with 10-deg binocular scotomas, and three of eight cases with 15-deg binocular scotomas showed significantly less overall motor compensation than was produced to the full-field stimulus. Most of these reductions occurred with convergent disparity presentations. Only one case with a 5-deg scotoma showed significantly different overall motor compensation than the full-field cases. Also, no significant differences in overall motor compensation were observed between the responses to the 10-deg and 15-deg scotomas. There was often asymmetry between the change in the left eye's line of sight and that of the right.

Significant asymmetry was found in four of eight cases with the full-field stimulus and in 11 of 24 cases with binocular scotomas. These asymmetries were evenly distributed between the convergent and divergent cases. All eight cases with monocular scotomas showed significant asymmetry. In all but one case, the scotomatic eye moved less than its fellow. Most subjects had less steady fixation when viewing stimuli with large binocular scotomas than they did when viewing the other stimuli. Three of four subjects produced significantly more saccades of greater than 0.5 deg amplitude when viewing stimuli with 10-deg or 15-deg binocular scotomas than when viewing fullfield stimuli. One subject showed this tendency while viewing stimuli with a 5-deg scotoma, but the number of large fixational saccades was significantly less than was produced while viewing stimuli with 10-deg or 15-deg scotomas. None of the subjects showed this tendency when viewing stimuli with monocular scotomas. Subject 4 had difficulty holding a steady vergence angle while divergent disparities were introduced in stimuli with monocular scotomas. One of these responses is shown in Figure 3. Fusion and stereopsis persisted throughout the trial, even though there were large vergence errors for periods of over 2 sec. This subject also had difficulty holding fusion when viewing divergent stimuli with 10-deg or 15-deg binocular scotomas when this experiment began but improved with practice. This subject had no difficulty holding a steady vergence angle when viewing any of the other stimuli.

1054

INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / August 1985

Vol. 26

Table 1. Average and standard deviations (in degrees) of eye movement responses to 5 deg horizontal disparities contained in fusional stimuli Stimulus type% Subject

Disparity direction

Motor component

1

Convergent

Disjunctive Left Right

4.7 ± 0.3 2.3 ± 0.3 2.4 ± 0.2

4.4 ± 0.4 1.9±0.5f 2.4 ± 0.2

4.7 ± 0.3 2.3 ± 0.3 2.4 ±0.1

4.4 ± 0.4 2.1 ±0.5 2.4 ± 0.3

4.3 ± 0.5 1.9±0.4t 2.5 ± 0.3

4.1 ±0.5* 1.7 ±0.4f 2.3 ± 0.2

4.0 ± 0.2* 1.9 ± 0.3 2.1 ±0.2

2

Convergent


4.6 ± 0.4 1.9±0.2f 2.6 ± 0.3

3.9 ± 0.5* 1.4±0.3t 2.5 ± 0.4

4.6 ± 0.3 1.9 ± 0 . 4 | 2.6 ± 0.2

4.0 ± 0.5* 1.6 ± 0 . 3 | 2.4 ± 0.4

4.0 ± 0.6* 1.8 ±0.4 2.2 ± 0.5

4.1 ±0.5* 1.8 ± 0.5 2.3 ± 0.5

4.5 ± 0.3 1.6±0.4f 2.9 ± 0.4

3

Convergent


4.9 ± 0.5 2.6 ± 0.2 2.4 ± 0.5

3.1 ±0.8* 0.5 ± 0.9f 2.7 ± 0.4

4.9 ± 0.4 2.4 ± 0.5 2.4 ± 0.4

2.7 ± 1.1* 1.2 ± 0.7 1.5 ±0.5

2.9 ± 0.6* 0.9 ± 0.8f 1.9 ±0.9

NR NR NR

3.2 ± 0.7* 1.7 ±0.5 1.5 ±0.6

4

Convergent


4.2 ± 0.6 1.8 ±0.6f 2.5 ± 0.3

4.2 ± 0.8 1.5 ±0.4f 2.7 ± 0.6

4.7 ± 0.8 2.1 ±0.7 2.5 ± 0.7

.4.2 ±0.8 2.1 ±0.7 2.1 ±0.7

4.3 ± 0.8 2.5 ± 0.4f 1.8 ±0.7

4.3 ± 1.1 1.7 ± 0.5f 2.7 ± 0.8

4.3 ± 0.6 2.1 ±0.9 2.2 ± 1.0

1

Divergent


4.4 ± 0.3 2.1 ±0.2 2.3 ± 0.2

4.1 ±0.4 1.9±0.2t 2.3 ± 0.2

4.4 ± 0.2 2.2 ± 0.2 2.2 ± 0.2

4.1 ±0.3 2.0 ± 0.2 2.1 ±0.3

4.1 ±0.2* 2.1 ±0.3 2.0 ± 0.2

4.2 ± 0.3 1.8 ±0.3f 2.4 ± 0.2

4.0 ± 0.3* 2.0 ± 0.3 2.0 ± 0.2

2

Divergent


4.1 ±0.3 2.4 ± 0.2f 1.8 ± 0.2

4.1 ±0.3 2.3 ± 0.3f 1.9 ± 0.3

4.3 ± 0.2 2.6 ± 0.2f 1.7 ±0.2

4.3 ± 0.3 2.3 ± 0.4 2.0 ± 0.5

4.3 ± 0.4 2.5 ± 0.4f 1.8 ± 0.3

4.1 ±0.3 2.0 ± 0.3 2.1 ±0.4

4.2 ± 0.3 2.6 ± 0.4f 1.6 ±0.4

3

Divergent


3.5 ± 0.5 1.0 ± 0.4f 2.5 ± 0.3

3.1 ±0.6 0.6 ± 0.7f 2.5 ± 0.5

3.5 ± 0.7 1.3 ±0.7f 2.3 ± 0.6

3.5 ± 0.7 1.6 ±0.7 1.9 ±0.4

3.2 ± 0.5 1.3 ± 0.8 1.9 ±0.6

3.7 ± 0.7 1.3 ± 0.3| 2.4 ± 0.6

3.7 ± 0.9 1.8 ±0.6 1.9 ± 1.0

4

Divergent


5.1 ±0.6 2.5 ± 0.6 2.5 ± 0.7

3.4 ± 1.3* -0.2 ± l.lf 3.6 ± 0.6

4.0 ± 0.9* 1.0 ± 1.2f 3.1 ±0.6

3.9 ± 1.3* 1.2 ± 1.4f 2.7 ± 0.9

4.1 ± 1.1 1.3 ± 1.2f 2.8 ± 0.3

2.9 ± 0.9* -0.2 ± 0.9f 3.3 ± 0.4

FF

LS10

BS5

BS10

BS15

LAW

BA10

2.9 ± 0.8* 0.9 ± 0.7| 2.0 ± 0.7

* Cases with significantly different overall motor compensation than the full-field case (P < 0.01). t Cases with significantly asymmetric monocular motor responses (P < 0.01).

t Stimulus types. FF: full-field stimulus; LS10: 10° left monocular scotoma; BS5: 5° binocular scotoma; BS10: 10° binocular scotoma; BS15: 15° binocular scotoma; LA 10: 10° left monocular annulus; BA10: 10° binocular annulus.

Experiment 2

fixation and central suppression. If annular stimuli could adequately provide extrafoveal stimulation, they would be useful in such an examination. Therefore, in this experiment, the central 10-deg portion of the stimulus shown to one or both eyes was blanked. The blanked portion, which was present throughout the entire trial, was fixed with respect to the pattern and not stabilized with respect to eye movements or the visual field. The fusional responses to these "annular" stimuli were compared to those to stimuli with monocular and binocular scotomas. Two stimuli based on the stimulus in Figure 1 were used:

Studies utilizing subjective methods have shown that the peripheral visual fields are important for fusion for strabismic patients.2-3 It is of interest to study the influence of the peripheral visual fields on the fusional responses of these patients by objective means, but it is difficult to implement artificial scotomas because these patients usually exhibit unstable 10° MONOCULAR SCOTOMA

JS9 RUN 11

Monocular annular stimulus: Full-field stimulus with the central 10 deg shown to the left eye blanked. Binocular annular stimulus: Full-field stimulus with the central 10 deg shown to both eyes blanked.

STIMULUS DISPARITY

Fig. 3. Response by subject 4 to a 5° divergent disparity presentation contained in a stimulus with a central 10° scotoma stabilized on the left eye's visual field.

Figure 2D shows a response by subject 2 to the introduction of a 5-deg convergent disparity while viewing the monocular annular stimulus. It can be seen that this response is quite similar to that evoked by the stimulus with a monocular scotoma of comparable size. Table 1 presents each subject's average overall motor compensation and average change in the line of sight of each eye in response to these

No.

HORIZONTAL FUSIONAL RESPONSE TO SCOTOMATIC STIMULI / Domon and Kerresz

stimuli (columns LA 10 and BA10). Subject 3 was unable to fuse the monocular annular stimulus when a convergent disparity was introduced. There was significantly less overall motor compensation in three of seven cases with monocular annuli and four of eight cases with binocular annuli than to the full-field cases. Five of seven cases with monocular annuli and three of eight cases with binocular annuli also showed significant asymmetry between the changes in two eyes' lines of sight. Subject 4 had difficulty holding a steady vergence angle when viewing the monocular annular stimulus with divergent disparity presentations but had no difficulty with any of the other stimuli. There were no significant differences between the subjects' average overall motor compensation to monocular annular stimuli and monocular scotomatic stimuli, while two of six cases showed significant differences between the overall motor compensation to 10 deg binocular annular stimuli and 10 deg binocular scotomatic stimuli (subjects 1 and 2, convergence). In general, the responses to these two classes of stimuli were very similar.

Discussion Fusional responses to stimuli that contain either 10-deg monocular or binocular artificial scotomas are alike in that they show less overall motor compensation than responses to stimuli that cover the central visual fields. On the other hand, they are different in that fixation is much less steady with binocular scotomas than with monocular scotomas or full-field stimuli, and the changes in the lines of sight of the two eyes are markedly asymmetric with monocular scotomas whereas there is much less asymmetry with binocular scotomas or full-field stimuli. This suggests that much of the response asymmetry that was found by Kertesz and Hampton 7 may have been due to the utilization of 10-deg monocular artificial scotomas. The fusional responses to full-field stimuli and stimuli with 5-deg binocular scotomas were similar in overall motor compensation, symmetry of the changes in the lines of sight of the two eyes, and steadiness of fixation. On the other hand, stimuli with 10-deg and 15-deg binocular scotomas evoked less overall motor compensation and less steady fixation than full-field stimuli, while the symmetry of the changes in the lines of sight of the two eyes was similar to that elicited with full-field stimuli. Overall motor compensation was reduced in 17 of 24 cases to stimuli with 10-deg or 15-deg artificial scotomas as compared to full-field stimulic that covered the central visual fields, but this reduction was significant in less than half the cases (9 of 24). The average overall compensation to full-field stimuli ranged from 70% to 102%, while for stimuli with 10-

1055

deg or 15-deg artificial scotomas the range was 54% to 88%. Although there was a reduction in this range, there was also much overlap. Previous tests using 10deg artificial scotomas and 1.5-deg horizontal disparity steps found the overall motor compensation to range from 40% to 77%.7 We observed fusional nonmotor components as large as 1.5 deg in response to full-field stimuli. Although Panum's fusional areas are known to be quite small in the central retina, nonmotor components of up to 4.9 deg have been found with stimuli that covered 57 deg of the visual field. It was suggested that interactions from peripheral retinal regions increase the extent of Panum's fusional areas.13 The results of this study support that suggestion. We frequently observed asymmetric changes in the lines of sight of the two eyes when viewing full-field stimuli (4 of 8 cases). The eye that made the larger change in response to convergent disparities often did not make the larger change in response to divergent disparities, so this was not a manifestation of ocular dominance. The changes in the lines of sight of the two eyes were always asymmetric in response to stimuli with monocular scotomas. In all but one case, the scotomatic (left) eye moved less than the other eye. The nonscotomatic (right) eye did not always adequately compensate for the disparity but would often undercompensate or overcompensate. The responses to stimuli that contained a monocular annulus also showed these characteristics. In a control experiment in which the central 10 deg of the right eye's stimulus was blanked, each subject produced asymmetric responses with the right eye moving less than the left. Two of the subjects had disparity direction dependent difficulties with their vergence responses. The fact that most of these difficulties only appeared in the responses to stimuli with monocular annuli or scotomas and rarely in response to stimuli with binocular annuli or scotomas suggest that this is not simply due to the extrafoveal nature of the fusional stimulus. One of the subjects produced large vergence errors that were present for up to 5 sec in response to stimuli with monocular annuli or scotomas. Fusion and stereopsis persisted throughout these periods. Vergence errors of this magnitude and duration have been reported by Hyson et al14 during fusion of misaligned random-dot stereograms. In their tests, the subjects corrected these vergence errors by producing convergent and divergent saccades. In these tests, the subject employed slow vergence movements to correct the vergence errors. These results are in agreement with the suggestion of Hyson et al that a neural memory compensates for variations of vergence over several seconds of time.

1056

INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / August 1985

The results of experiment 2 show that annular stimuli may be used to test the influence of extrafoveal retinal areas on fusional vergence response. We have successfully used this method to study the fusional response of patients with binocular anomalies. With this method, monocular fixation aids are not necessary but they help to steady fixation. In summary, this study extends the characterization of horizontal fusional responses to extrafoveal stimulation to include stimuli with binocular, artificial scotomas. Responses to these stimuli contain both motor and nonmotor components. Compared to responses to full-field stimulation, stimuli with binocular scotomas of 10 deg or 15 deg produce less steady fixation and evoke less overall motor compensation. Stimuli with monocular scotomas exacerbate the asymmetry of the changes in the lines of sight of the two eyes whereas stimuli with binocular scotomas do not. Key words: artificial scotomas, extrafoveal stimulation, vergence response, binocular fusion, binocular fixation

References 1. Burian HM: Fusional movements: role of peripheral retinal stimuli. Arch Ophthalmol 21:486, 1939. 2. Burian HM: Fusional movements in permanent strabismus: a study of the role of the central and peripheral retinal regions

3. 4. 5.

6.

7. 8.

9. 10.

11.

12. 13. 14.

Vol. 26

in the act of binocular vision in squint. Arch Ophthalmol 26: 626, 1941. Lyle TK and Foley J: Subnormal binocular vision with special reference to peripheral fusion. Br J Ophthalmol 39:474, 1955. Winkelman JE: Peripheral fusion. Arch Ophthalmol 45:425, 1951. Ludvigh E, McKinnon P, and Zaitzeff L: Relative effectivity of foveal and parafoveal stimuli in eliciting fusion movements. Arch Ophthalmol 73:115, 1965. Sullivan MJ and Kertesz AE: Peripheral stimulation and human cyclofusional response. Invest Ophthalmol Vis Sci 18: 1287, 1979. Kertesz AE and Hampton DR: Fusional response to extrafoveal stimulation. Invest Ophthalmol Vis Sci 21:600, 1981. Houtman WA and van der Pol BAE: Fusional movements by peripheral retinal stimulation. Graefes Arch Clin Exp Ophthalmol 218:218, 1982. Hampton DR and Kertesz AE: Fusional vergence response to local peripheral stimulation. J Opt Soc Am 73:7, 1983. Reuss JL and Kertesz AE: Microcomputer generation of dynamic stereo graphics for clinical use. IEEE Trans Biomed Eng 28:15, 1981. Stark L, Vossius G, and Young LR: Predictive control of eye tracking movements. IRE Transactions on Human Factors in Electronics HFE-3:52, 1962. Kelly DH: Photopic contrast sensitivity without foveal vision. Optics Letters 2:79, 1978. Kertesz AE: Effect of stimulus size on fusion and vergence. J Opt Soc Am 71:289, 1981. Hyson MT, Julesz B, and Fender DH: Eye movements and neural remapping during fusion of misaligned random-dot stereograms. J Opt Soc Am 73:1665, 1983.