Fusional response to extrafoveal stimulation.

Fusional response to extrafoveal
stimulation
Andrew E. Kertesz and David R. Hampton
Horizontal and vertical fusional responses to extrafoveal stimulation were examined. An objec-
tive binocular eye movement measuring technique was utilized to position an artificial, stabi-
lized scotoma over the central 10° diameter area of one eye. The responses contained both a
motor and a nonmotor, or sensory, component. The monocular eye movements were asym-
metrical. (INVEST OPHTHALMOL VIS SCI 21:600-605, 1981.)
Key words: binocular fusion, extrafoveal stimulation, vergence eye movements,
binocular disparity
I n man, the visual field is frequently consid-
ered to be divided into two regions: a center
or foveal area and a surround or visual pe-
riphery. These regions are distinguished on
the basis of different underlying anatomic
and physiologic features of the visual system.
The fusional response to stimulation within
the central visual field has received consider-
able attention, but no objective eye move-
ment measurements have been made to
explore the role played by more peripheral
regions in horizontal and vertical fusional re-
sponses. Burian1 was first to investigate the
particular role of peripheral retinal disparity
in the mechanism of binocular fusion. He
found, by means of a subjective measure of
vertical phoria, that strictly peripheral stim-
uli could elicit a fusional response. The most
peripheral of his fusional stimuli was located
From the Biomedical Engineering Center, Northwest-
ern University, and Division of Ophthalmology, Ev-
anston Hospital, Evanston, 111.
This research was supported in part by research grant
EY1055 from the National Eye Institute.
Submitted for publication Oct. 31, 1980.
Reprint requests: Dr. Andrew E. Kertesz, Biomedical
Engineering Center Technological Institute, North-
western University, Evanston, 111. 60201.
12° above each fovea. At this distance from
the fovea, the stimuli could not be identified
as actually fused because of their "indistinct"
appearance. Nevertheless, Burian estimated
the fusional response magnitude by measur-
ing the displacement of centrally located
nonius lines that were shifted by the re-
sponse to the peripheral stimuli. One of Bu-
rian's surprising results was that peripheral
fusion could actually disrupt central fusion,
causing the subject to experience diplopia in
his percept of a small fusional stimulus within
the central visual field. Winkelman2 con-
firmed Burian's results and extended them to
fusional stimuli 27° from the fovea, located in
the lateral periphery.
Only recently has the cyclofusional re-
sponse to strictly peripheral disparities been
measured by means of an objective binocular
eye movement measuring technique. It was
found that even torsional disparities that are
confined to strictly peripheral regions (annu-
lar stimuli of 30° to 50° in diameter) evoke a
cyclofusional response, with a substantial
motor component in the form of cyclovergent
eye movements and a significant nonmotor,
or sensory, component limited to the extent
of Panum's fusional areas.3
Thus it has been established that strictly
600 0146-0404/81/100600+06$00.60/0 © 1981 Assoc. for Res. in Vis. and Ophthal., Inc.

Volume 21
Number 4 Fusional response to extrafoveal stimulation 601
peripheral cyclofusional disparities evoke a
fusional response consisting of both motor
and sensory components3 and that peripheral
horizontal and vertical disparities also evoke
fusional responses.1' 2 Since neither Burian
nor Winkelman monitored eye positions dur-
ing their experiments, we do not know the
relative contributions of the motor and sen-
sory components to the overall fusional re-
sponse. The objective of this report is to
examine the way in which horizontal and
vertical fusional amplitudes are apportioned
between the motor and sensory components
during extrafoveal stimulation. We would
like to find out whether extrafoveal stimula-
tion with horizontal or vertical disparities
evokes fusional vergence eye movements as
does stimulation that includes foveal regions
as well.
Method
Two stereo-normal subjects participated in the
experiments. Their vision was corrected to 20/20
acuity, and measurements were made from their
phoria position. They were seated in a darkened
room. Their heads were held stationary by a
biteboard and a headrest. Eye movements were
monitored by an objective binocular technique.
Subjects wore tight-fitting scleral contact lenses
with embedded search coils for sensing eye posi-
tions within a magnetic field. Horizontal and ver-
tical movement of each eye was measured with a
resolution of 1 min arc. A detailed description
of our procedures and measurement technique
has been provided elsewhere.4' 5 The stimuli
were displayed dichoptically on oscilloscopes (with
short-persistence P15 phosphor) that were part of
a four-channel wide-angle (57° field of view) tachis-
toscope. The stimulus luminance level of 3.2
cd/m2 was within the mesopic range. The circular
disk stimuli subtended 57° and were composed of
11 equally spaced parallel horizontal or vertical
lines of type shown in Fig. 1. One of the monocu-
lar stimuli contained a fixation point (left eye view
in Fig. 1), whereas the central 10° portion of the
other monocular stimulus pattern was electron-
ically blanked (right eye view in Fig. 1). We will
refer to the central 10° blanked region as an arti-
ficial scotoma, the position of which was elec-
tronically controlled and stabilized with respect to
horizontal and vertical movmements of the eye
whose monocular image contained it. Thus, by
using the horizontal and vertical eye movement
57° STIMULUS, 10° SCOTOMA
A
\
\ A
\
\
LEFT EYE RIGHT EYE
Fig. 1
signals to negate the effect of eye movements, we
were able to stabilize the retinal position of the
scotoma.
Preliminary experiments with nonstabilized sco-
tomas revealed that after the presentation of the
stimulus disparity, the subject would sometimes
shift fixation to the edge of the blanked stimulus
region and thereby bring the disparity to foveal
regions. We therefore decided to stabilize the reti-
nal position of the scotoma in order to confine the
stimulus disparities to extrafoveal regions. After a
few experiments with stabilized scotomas the sub-
jects realized the futility of these disparity foveating
attempts and stopped making them. Our experi-
mental apparatus is only capable of stabilizing a
scotoma in one eye at a time, which is sufficient to
remove binocular stimulation from the foveas.
Each experiment began with a calibration run
for the horizontal and vertical eye movement
channels, which was followed by the gain adjust-
ments for the stabilization of the artificial scotoma
according to the method of Kelly.6 These calibra-
tions were followed by experimental runs in which
vertical parallel line stimuli were used for the
study of the horizontal fusional response, and
horizontal parallel line stimuli were used for mea-
surement of the vertical fusional response. In all
cases the subject was instructed to fixate the cen-
ter of the stimulus pattern even though only one of
the monocular images contained a fixation point.
Each run began with a 10 sec presentation of a
zero disparity stimulus followed by a 10 sec pre-
sentation of a symmetrical step disparity; that is,
the monocular stimulus images were displaced by
an equal amount but in opposite directions, after
which the stimulus disparity was once again re-
duced to zero for 10 sec. The magnitude of the
step disparity was chosen to be smaller than the
maximum fusible disparity, and the subjects were
able to fuse it without difficulty.

602 Kertesz and Hampton Invest. Ophthalnwl. Vis. Sci.October 1981
SCOT 12, 10°L, 1.5°H CONV, AEK, 10R
SCOT 16, 10°L, 1.5° H DIV, AEK, 10R
LEFT EYE HORIZONTAL
RIGHT EYE HORIZONTAL
1.5° CONV
RIGHT
n
TOTAL HORIZONTAL STIMULUS DISPARITY
(L+R)/2 5 SEC
DISJUNCTIVE (L-R)
Fig. 2. Average of 10 fusional responses to a 1.5°
horizontal, convergent step disparity presentation
(Subject A.E.K.). The 10° diameter artificial
scotoma was stabilized on the left retina. The per-
centage of the disparity compensated by the dis-
junctive motor component is provided on the right.
LEFT EYE HORIZONTAL
RIGHT EYE HORIZONTAL
1.5° DIV
TOTAL HORIZONTAL STIMULUS DISPARITY
(L+R)/2
DISJUNCTIVE (L-R)
Fig. 3. Average of 10 fusional responses to a 1.5°
horizontal, divergent step disparity presentation
(Subject A.E.K.). The 10° diameter artificial
scotoma was stabilized on the left retina. The per-
centage of the disparity compensated by the dis-
junctive motor component is provided on the right.
We presented left or right-hyper vertical dis-
parities in experiment 2. A left-hyper vertical dis-
parity was said to exist when the stimulus pre-
sented to the left eye was displaced in the upward
direction, and the stimulus seen by the right eye
was displaced in the downward direction from
zero disparity viewing conditions. A right-hyper
vertical disparity was said to exist when the stimu-
lus seen by the right eye was displaced in the
upward direction and the stimulus seen by the left
eye was displaced in the downward direction from
zero disparity viewing conditions.
Results
Experiment 1. The objective of this exper-
iment was to examine the horizontal fusional
response to extrafoveal stimulation. The stim-
ulus consisted of 11 parallel vertical lines as
shown in Fig. 1. The results are presented
in Figs. 2 and 3 and are summarized in
Table I.
Fig. 2 shows a response to a 1.5° horizontal
convergent step disparity presentation. Sub-
sequent to the introduction of the disparity,
both eyes moved in the direction required to
decrease the stimulus disparity. Thus the ex-
trafoveal convergent disparity presentation
elicited a vergence response. The total dis-
junctive motor response compensated for
70% of the disparity, allowing for the exis-
tence of a significant sensory component as
well. There was a difference, however, in the
net change in eye positions in response to the
stimulus. The eye whose retinal image con-
tained the scotoma (left eye) moved 0.45°,
whereas the fellow eye moved 0.6°. This dif-
ference may be explained by either of two
hypotheses. If one assumes a pure con-
vergence response, the differences in the
monocular eye movements were the result of
asymmetrical vergence contributions. In this
case the eye whose retinal image contained
the scotoma (left eye in Fig. 2) compensated
for only 30% of the stimulus disparity (or,
equivalently, followed 60% of its monocular
stimulus movement), whereas the other eye
eliminated 40% of the disparity (or followed
80% of its monocular stimulus movement).
Alternatively, the differences in the monocu-
lar eye movements may have been caused by
a combination of version and vergence re-

Volume 21
Number 4 Fusional response to extrafoveal stimulation 603
Table I. Horizontal eye movement response to the presentation of a 1.5° disparity
contained in the 11 parallel vertical line stimulus/
Subject Left eyeB Right eyeB
Total (L-R)
disjunctive movement
1.5° Convergent step disparity
1.5° Divergent step disparity
A. E. K. (R)
A. E. K. (L)
D. R. H. (R)
D. R. H. (L)
A. E. K. (R)
A. E. K. (L)
D. R. H. (L)
D. R. H. (L)
0.65°
0.45°
0.45°
0.1°
0.75°
0.4°
0.4°
0.3°
0.35°
0.06°
0.25°
0.6°
0.4°
0.7°
0.2°
0.55°
1.0°
1.05°
0.7°
0.7°
1.15°
1.1°
0.6°
0.85°
(67%)
(70%)
(47%)
(47%)
1
(77%)
(73%)
(40%)
1
(56%)
R or L denotes the eye whose retinal image contained the 10° stabilized artificial scotoma.
AEach entry represents an average of 10 responses.
BNet change in eye positions as discussed in the text.
cPercentage of stimulus disparity compensated by disjunctive eye movements.
Table II. Vertical eye movement response to the presentation of a 0.4° vertical right-hyper
step disparity contained in the 11 parallel horizontal line stimulus.A
Subject Left eye1 Right eyeB
Total (L-R)
disjunctive movement
0.4° Right-hyper step disparity A. E. K. (L)
A. E. K. (R)
D. R. H. (L)
D. R. H. (R)
0°
0.3°
0°
0.3°
0.3°
0°
0.3°
0°
0.3° (75%)c
0.3° (75%)
0.3° (75%)
0.3° (75%)
R or L denotes the eye whose retinal image contained the 10° stabilized artificial scotoma.
AEach entry represents an average of 10 responses.
BNet change in eye positions as discussed in the text.
•-Percentage of stimulus disparity compensated by disjunctive eye movements.
sponses. In this case a version movement
would have been elicited by the displace-
ment of the monocular fixation point in the
nonscotomatic eye, representing an attempt
by the eyes to maintain fixation. Symmetri-
cal vergence components would then have
served to minimize the extrafoveal stimulus
disparities. We were able to distinguish an
initial version component followed by a ver-
gence component in some of the responses. A
version component, however, was not distin-
guishable in the majority of the responses.
We therefore cannot rule out either hypoth-
esis on the basis of our results.
Fig. 3 shows a response to a 1.5° horizontal
divergent step disparity presentation. In this
case the introduction of the disparity was fol-
lowed initially by a version movement in
which the eye whose retinal image contained
the scotoma (left eye) was pulled by the fel-
low eye in a direction opposite to that re-
quired for the elimination of the disparity.
The initial version movement was followed
by a vergent movement, with the net result
of both eyes contributing to the minimization
of the stimulus disparity. There was a differ-
ence, once again, in the net change in eye
positions. The initial version component was
observed much more frequently in response
to stimuli containing divergent disparities
than it was for convergent disparity presen-
tations. Nevertheless, the version compo-
nent was not distinguishable in some of
the responses to divergent disparity presen-
tations.
Table I provides a summary of our obser-
vations for this experiment. The data show
that both eyes participated in the fusional
response. The motor compensation was in-
complete, but its overall amplitude remained
the same regardless of which monocular
image contained the scotoma. A substantial

604 Kertesz and Hampton Invest. Ophthalmol. Vis. Set.October 1981
SCOT 20, 10°L, 0.4° R HYPER, AEK, 10R
LEFT EYE VERTICAL
RIGHT EYE VERTICAL
0°
_ .
tup
0.4° R HYPER
TOTAL VERTICAL STIMULUS DISPARITY
(L+R)/2 5 SEC
I I
DISJUNCTIVE (L-R) 75%
Fig. 4. Average of 10 fusional responses to a 0.4°
vertical, right-hyper, step disparity presentation
(Subject A.E.K.). The 10° diameter artificial
scotoma was stabilized on the left retina. The per-
centage of the disparity compensated by the dis-
junctive motor component is provided on the right.
portion of the disparity, in some cases more
than half of it, was compensated for by the
sensory component.
Experiment 2. The objective of this experi-
ment was to examine the vertical fusional re-
sponse to extrafoveal stimulation. The stim-
ulus consisted of 11 parallel horizontal lines.
Since the vertical fusible disparity range is
much smaller than the horizontal one, and
we wanted to stay well within that range, we
used a 0.4° right-hyper step disparity in this
experiment.
Fig. 4 shows a response to a 0.4° right-
hyper vertical disparity presentation. The ex-
trafoveal fusional stimulus elicited a disjunc-
tive motor response that compensated for
75% of the stimulus disparity. The net
changes in eye positions in response to the
presentation of the stimulus were once again
unequal. The eye whose retinal image con-
tained the scotoma experienced no net
change in eye position, whereas the fellow
eye moved by 0.3°. In Fig. 4 a version com-
ponent preceded the vergence component,
but in many of the responses of Table II no
version was distinguishable. Once again, just
as in the horizontal case, we cannot rule out
either of the hypotheses on the basis of these
results. However, previously obtained re-
sults of vertical fusional stimulation that did
not involve the use of scotomas demonstrated
asymmetrical vergence responses to sym-
metrical disparity presentations.7
Discussion
Traditionally the fusional response has
been studied in terms of the three orthogonal
directions along which disparities are pre-
sented. Thus we speak of horizontal, vertical,
and cyclofusional responses. It was previ-
ously shown that torsional disparities that
were confined to strictly peripheral regions
evoked a cyclofusional response containing a
substantial motor component in the form of
cyclovergent eye movements.3 The present
results demonstrate that extrafoveal horizon-
tal or vertical disparities also elicit a disjunc-
tive motor response.
A comparison of full-field vs. extrafoveal
horizontal fusional stimulation indicates that
the presence of the monocular scotoma de-
creased the overall motor contribution and
caused an asymmetry between the two mon-
ocular motor components. In the case of ver-
tical fusional stimulation, the scotoma exac-
erbated the asymmetry between the two
monocular motor responses without appre-
ciably affecting the magnitude of the overall
disparity compensating motor response.
The fact that disparities positioned ex-
trafoveally are sufficient to evoke a fusional
response should not surprise clinicians, since
there are many examples of patients with
macular degeneration, often bilateral, who are
able to maintain ocular alignment. Clearly, in
these cases, it is the peripheral fusion lock that
maintains ocular alignment. Our results show a
way in which such an alignment may be ac-
complished via the motor and sensory compo-
nents of fusional response.
REFERENCES
1. Burian HM: Fusional movements: the role of periph-
eral retinal stimuli. Arch Ophthalmol 21:486, 1939.
2. Winkelman JE: Peripheral fusion. Arch Ophthalmol
45:425, 1951.