Preferred retinal loci and macular scotoma
characteristics in patients with age-related
Ronald A. Schuchard, PhD
ABSTRACT • RÉSUMÉ
Many patients with macular scotomas due to age-related macular degeneration do not
perceive black spots in the visual field where the scotomas are located. Rather, they
describe objects as “vanishing,” “jumping out of nowhere” or “having blurry parts,” or
a combination of features. In addition, when the macular scotoma affects the fovea, the
visual system uses 1 or more preferred retinal loci (PRLs) as a “pseudofovea” to
perform visual tasks. Visual function testing with the scanning laser ophthalmoscope
has provided a wealth of information regarding how patients perceive the visual world
and how the oculomotor system directs eye movements. This article describes
2 specific functions of the oculomotor system, fixation stability and refixation
precision, with data collected from normally sighted people and patients with visual
field loss. The implications of the characteristics of PRLs and macular scotomas for
clinical testing are discussed.
Plusieurs patients qui ont des scotomes attribuables à une dégénérescence maculaire
liée à l’âge ne perçoivent pas de points noirs du champ visuel là où se situent les
scotomes. Ils décrivent des objets « qui disparaissent », « venus de nulle part » ou
« partiellement flous » ou ayant un mélange de ces caractéristiques. De plus, lorsque
les scotomes maculaires affectent la fovéa, l’appareil visuel utilise une ou plusieurs
zones de rétine préférentielles comme « pseudofovéa » pour accomplir des tâches
visuelles. L’examen de la fonction visuelle avec l’ophtalmoscope laser à balayage a
donné une abondance de renseignements sur la perception qu’ont les patients du
monde visuel et sur la façon dont le système oculomoteur dirige les mouvements de
l’œil. Cet article décrit 2 fonctions particulières du système oculomoteur, la stabilité
de la fixation et la précision de la refixation, avec des données recueillies chez des
personnes qui ont une vue normale et des patients qui ont subi une perte du champ
visuel. Il traite aussi des implications des caractéristiques des zones de rétine
préférentielles et des scotomes maculaires pour les tests cliniques.
T he visual system with a central scotoma from age-
related macular degeneration (AMD) chooses
(often without the conscious knowledge of the
or more eccentric preferred retinal loci (PRLs) natu-
rally and reliably develop to perform the foveal visual
tasks such as recognizing objects and directing eye
patient) a preferred eccentric retinal area because the movements while reading or tracking objects.1–16 This
fovea can no longer perform visual tasks; that is, in an choosing by the visual system of an eccentric “pseu-
eye with a central scotoma affecting all of the fovea, 1 dofovea” for visual-performance functions has been
From the Department of Veterans Affairs, Rehabilitation Research & GA 30033-4004, USA; fax (404) 728-4837; firstname.lastname@example.org
Development, Center of Excellence for Aging with Vision Loss, Atlanta, Ga.
This article has been peer-reviewed.
Correspondence to: Dr. Ronald A. Schuchard, Atlanta VA Rehabilitation
Research & Development Center (151R), 1670 Clairmont Rd., Decatur, Can J Ophthalmol 2005;40:303–12
Preferred retinal loci—Schuchard 303
Preferred retinal loci—Schuchard
Fig. 1—Characteristics of preferred retinal loci (PRLs) and scotomas in the right eye (left image) and left eye (right image) of a
patient with exudative age-related macular degeneration (AMD).The green solid line describes the border of the dense scotoma
(DS), the yellow letters describe the area of the relative scotoma (RS), the red circle describes the border of the PRL, and the blue
F indicates the best estimate of the location of the nonfunctioning fovea.
noted primarily by the use of the scanning laser oph- mining where the eccentric PRL is located except by
thalmoscope (SLO). The SLO also allows accurate monitoring the location of visual-stimuli images on
characterization of the central visual field by macular the retina. Fig. 1 shows how visual stimuli can be
perimetry with direct retinal observation of fixation directly monitored with an SLO to determine scotoma
and perimetry targets.17,18 For example, fixation per- and PRL characteristics. Generally, in the low-vision
formance has 2 operational definitions: refixation population with central scotomas, there is no consis-
precision is defined as the retinal area that a patient tent retinal location relative to the scotoma for the
uses during repeated fixation (refixation); fixation sta- PRLs. In a study of 825 patients, 84.4% of low-vision
bility is defined as the retinal area that a patient uses eyes (1130 of 1339) demonstrated an established
while maintaining fixation. Refixation precision and PRL (foveal or eccentric) for fixation, which varied
fixation stability show variability, of course, and the from 1.0° to 9.0° in diameter.14 Only 4.4% of the
distribution of retinal positions used during fixation patients did not demonstrate a PRL in either eye.
performance is operationally defined as the retinal There was a dense scotoma within 2.5° of all of the
locus for fixation. eccentric PRLs, surprisingly, and it completely sur-
The concept of using a retinal locus for visual tasks rounded 17.4% of the PRLs (ring scotoma).
such as fixation is accepted in vision rehabilitation In an eye with a functioning fovea, fixation is the
but seldom used with normally sighted patients. The act of directing the eye toward the visual target of
term PRL has typically been reserved for patients regard, causing the image of the target to be within
whose visual system has chosen a preferred eccentric the fovea. The anatomic fovea is a retinal area about
retinal area for visual tasks because of a central 5° (1500 mm) in diameter in which the central
scotoma. However, the visual system of patients with 1.2°–1.7° diameter of photoreceptors (referred to as
the anatomic foveola or clinical fovea) is composed
paracentral scotomas or even normal sight does
entirely of cones.19,20 The traditional view is that
choose to use the fovea over other retinal areas; thus,
within the centre of the fovea, where the cone density
the fovea can be referred to as their PRL. is highest, is a small and spatially invariant retinal area
RETINAL CONSIDERATIONS referred to as the optimal locus.21 The optimal locus
has been hypothesized to direct the location of a fix-
For patients with a functioning fovea, visual tasks ation stimulus when fixation is initiated and to direct
are performed by aiming the eye such that the image drifts and microsaccades to maintain fixation.
of the visual target of regard is placed within the In an eye with a central scotoma affecting all of the
foveal area. For patients with a central scotoma from fovea, eccentric fixation is the act of directing the eye
AMD involving all of the fovea, visual tasks are per- toward the visual target of regard, causing the image
formed by aiming the eye such that the image of the of the target to be placed in 1 or more preferred
visual target of regard is placed within a PRL. eccentric retinal areas. Eccentric fixation refers to fix-
Unfortunately, there is no conclusive way of deter- ation performance in which the oculomotor system is
304 CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005
Preferred retinal loci—Schuchard
oriented to the eccentric retinal area, and the patient are more likely to use a “secondary” PRL when fixat-
has the sensation of looking directly at the visual ing on a target in a secondary position of gaze. The
target during fixation. Eccentric viewing, on the use of multiple PRLs has also been observed for dif-
other hand, refers to viewing performance in which ferent visual tasks13 and lighting conditions.15
the oculomotor system is not oriented to the target Patients are often completely unaware of when or
and the patient has the sensation of looking above, how they use multiple PRLs.
below or to either side of the target to see it. A shift
of the oculomotor reference from the fovea to a pre- PRL CHARACTERISTICS
ferred eccentric retinal area is possible in patients The ability to make efficient and effective eye
with bilateral macular disease.7,8 However, even movements with a PRL is often considered the most
though the saccades of patients with central scotomas
difficult and yet the most valuable component of
can consistently direct images to the PRL, they some-
visual tasks such as scanning and reading. The more
times have the latency and dynamic characteristics of
basic eye movements made by the visual system are
nonfoveal saccades.8 But the latency can be shortened
fixation, pursuit and saccadic movements. These
with practice, eventually tending asymptotically to
tasks are thought to possibly represent measurable
the normal saccadic duration.22 Eccentric viewing or
parameters of the PRL quality for visual performance;
fixation naturally and reliably occurs when the foveal
that is, to optimally function as a retinal locus for
areas in both eyes are no longer functional, such as
visual performance, the PRL needs to keep a visual
when both eyes have central scotomas.1–6,22
image in a discrete and stable retinal area (fixation
POSITION OF GAZE CONSIDERATIONS stability), to track moving objects through space
(pursuit) and to move rapidly to objects of interest
The primary position of gaze is defined as the posi- appreciated in the visual field away from the PRL
tion of the eye when a patient is looking at a visual (saccadic movements). These and other characteris-
target that is straight ahead. From this primary posi- tics of the PRL are thought to be important in activ-
tion all ocular movements are initiated. More specifi- ities of daily living for patients with AMD.
cally, the primary position of the eye is the position
against which all torsional, rotational and transla- Refixation precision
tional movements are measured. The position of the
Subjects were asked to fixate for 30 s on a target
eye at the primary position of gaze is not necessarily
randomly placed in 1 of 25 positions determined by
identical to an anatomically straightforward position
a 5 × 5 grid within the 17° × 12° SLO field of view.
of the eye (as in an AMD patient with a central
scotoma). A secondary position of gaze is defined as The centre of the grid was at the primary position of
any position of the eye represented by a vertical, hor- gaze, and the remaining grid locations provided 24
izontal or oblique (a combination of vertical and hor- secondary positions of gaze. Table 1 presents the char-
izontal) deviation from the primary position of gaze. acteristics of the PRL for fixation described by a
It is expected that patients with functioning foveal bivariate analysis that provides an elliptical fit of the
areas (e.g., patients with paracentral scotomas or 30-s retinal positions during fixation (unpublished
normal sight) will use the foveal area as their PRL for data, 2003). The characteristics of the ellipsis are ori-
visual tasks when visual targets are in either the entation of the major axis (± 90° from the positive
primary position of gaze or secondary positions of horizontal axis), eccentricity (the ratio of the major to
gaze. In addition, it is expected that patients with minor axes when a value of 1 indicates a circular
nonfunctioning foveal areas (owing to central sco- shape), major axis length (in minutes of arc) and area
tomas) will use a preferred eccentric retinal area as the (in minutes of arc squared).
PRL for fixation when targets are in either the An analysis of variance in the characteristics of the
primary position of gaze or secondary positions of PRL for fixation found by refixation precision was
gaze. However, patients with central scotomas use 1 first performed for 3 normally sighted subjects who
or more PRLs for visual tasks when visual targets are had extensive experience fixating. Effects with p <
in the secondary positions of gaze and almost always 0.05 are reported as significant in this and all further
use a “primary” PRL for visual tasks when the targets analyses of variance. The orientation of the major
are in the primary position of gaze;8 that is, patients axes of the PRLs was not affected by the positions of
CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005 305
Preferred retinal loci—Schuchard
Table 1—Characteristics of ellipses representing preferred retinal loci for fixation found by refixation
precision and studied by bivariate analysis
Primary position of gaze Secondary positions of gaze
Ma j o r a x i s M a j o r a xi s
Length Area (min Length Area (min
Angle (min of of arc Angle (min of of arc
Subject* Stimulus (degrees) arc) Eccentricity‡ squared) (degrees) arc) Eccentricity‡ squared)
EN-1 1° cross –14.1 26.5 0.78 434 43.8 42.2 0.79 1 105
12′ disc –23.6 30.5 0.61 443 –63.9 44.2 0.78 1 190
1° disc –46.0 31.9 0.56 452 –86.6 58.1 0.66 4 042
EN-2 1° cross –13.6 28.8 0.55 392 24.0 82.1 0.40 2 056
12′ disc –3.0 28.8 0.60 389 1.1 44.4 0.72 1 120
1° disc –20.6 27.9 0.78 476 –71.4 47.8 0.65 1 176
EN-3 1° cross 15.2 21.1 0.97 337 42.8 43.2 0.45 661
12′ disc –75.5 34.7 0.46 436 61.1 54.8 0.69 1 631
1° disc 65.5 33.0 0.83 608 11.4 30.6 0.68 3 007
IN-1 1° cross 42.8 59.8 0.77 2 172 14.2 100.5 0.56 4 443
IN-2 1° cross –38.6 57.9 0.88 2 317 37.7 199.2 0.44 13 769
IN-3 1° cross 78.3 76.9 0.44 2 050 –16.9 150.5 0.73 13 028
IN-4 1° cross 10.8 56.1 0.89 2 205 –81.2 88.5 0.61 3 741
PS-1 1° cross 47.9 81.9 0.60 3 163 53.7 87.2 0.61 3 660
PS-2 1° cross –63.5 48.4 0.48 890 –23.0 172.3 0.50 11 667
CS-1 1° cross –2.2 212.9 0.55 19 759 13.9 549.2 0.36 84 770
CS-2 1° cross 124.6 390.0 0.28 33 735 –64.4 555.0 0.64 155 790
*Subjects were experienced (E) or inexperienced (I) in fixation and had normal (N) sight or had scotomas (S), either paracentral (P) with a
functioning fovea or central (C). Subject PS-2 had a ring scotoma surrounding the fovea. All scotomas were due to age-related macular
Plus or minus 90° from the positive horizontal axis.
A value of 1 indicates a perfect circle.
gaze (primary or secondary), the different fixation position of gaze, additional studies were done with 4
stimuli or the different subjects. The eccentricity of normally sighted subjects who were inexperienced in
the PRLs also was not affected by the positions of fixation. These results and those for the experienced
gaze (primary or secondary), the different fixation subjects with the 1° cross were combined for analysis.
stimuli or the different subjects. However, the eccen- In all subjects, the major axis length and area of the
tricity values in Table 1 clearly indicate that the PRL PRLs were significantly larger with the primary posi-
for fixation is not a circle (eccentricity = 1) except in tion of gaze than with the secondary positions of
a few cases (e.g., subject EN-3 with 1° cross at the gaze. However, all the inexperienced subjects had
primary position of gaze). The major axis length and larger fixation areas than the experienced subjects.
the area of the PRLs were not affected by different The same effect of experience on fixation perform-
fixation targets or different subjects. Therefore, the ance has been reported for fixation stability.23
orientation and shape of the PRLs for fixation change Patients with central visual field loss not affecting
in apparently random and nonsystematic ways. How- the fovea (e.g., patients with ring scotomas) would be
ever, the major axis length and area were significantly expected to have fixation performance similar to that
larger for secondary positions of gaze than for of normally sighted subjects. The results for 2
primary positions of gaze with all 3 fixation targets patients with paracentral scotomas shown in Table 1
and all 3 subjects (the interaction terms were not sig- are typical of these types of field loss: the PRLs for fix-
nificant). ation found with primary and secondary positions of
To extend the finding that the major axis length gaze were similar to those of the normally sighted
and area of the PRL is affected by the fixation-target subjects. However, such patients sometimes place the
306 CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005
Preferred retinal loci—Schuchard
fixation target in an eccentric PRL for fixation. For area of 68′ in diameter. In addition, the patients did
example, patient PS-2 twice placed the fixation target not place the target in a small retinal area of 10.5′
in an eccentric retinal location outside the ring most of the time and very seldom placed the fixation
scotoma when the target was at a secondary position stimulus outside the small retinal area (a Gaussian
of gaze (these 2 eccentric fixation positions were not distribution). Rather, patients placed the target in a
included in the bivariate analysis). The patients are larger retinal area most of the time (a roughly flat dis-
often completely unaware of the change in the retinal tribution of 40′ or 50′) and outside this larger area
fixation position. The last 2 patients listed in Table 1 very seldom.
had central scotomas affecting the fovea. The increase
in size of the PRL for fixation is typical of this type of Comparison of results
field loss. Patient CS-1 used 2 eccentric retinal areas
The PRL sizes during a fixation-stability trial, at
to place the fixation target; patient CS-2 used only 1
both primary and secondary positions of gaze, were
PRL. Patients with central scotomas affecting the
very similar to the PRL sizes for the 24 repeated fixa-
fovea are also often completely unaware of using dif-
tions at the primary position of gaze during refixa-
ferent retinal positions for PRLs. It is not known why
tion-precision trials. Patients appear to use a larger
patients place fixation targets in locations less often
used for fixation. retinal area for fixation when asked to fixate on a
target that is away from the primary position of gaze.
Fixation stability However, when they have fixated on a target at a sec-
ondary position of gaze, they appear to be able to
The retinal loci for fixation found by fixation sta- maintain the image of the fixation target in a retinal
bility in 8 of the secondary positions of gaze (at the locus for fixation that is the same as when they are
corners and the horizontal/vertical axes of the 17° × asked to fixate and maintain fixation on a target at
12° SLO field) were found along with 8 primary- the primary position of gaze. Interestingly, even
position trials. An analysis of variance (repeated- though the patients used a larger retinal area for sec-
measures design) was performed on the characteris- ondary-position fixation, the diameters of the major
tics of the PRLs for fixation (again significance is axes of the PRLs were almost always 1° or less. It may
reported for p < 0.05). Like refixation precision, the not be a coincidence that the retinal positions for fix-
orientation of the major axis and the eccentricity of ation are typically within an area of about 1.2° to 1.7°
the PRLs were not affected by the position of gaze for most normally sighted patients. The central
(primary or secondary). Unlike refixation precision, 1.2°–1.7° area of the fovea, the foveola, is a retinal
however, major axis length was significantly affected area with roughly equal resolution capability.
by position of gaze. Previous studies of fixation sta-
bility in normally sighted subjects had shown PRL Binocular PRLs
and foveal areas for fixation of 31.6′ to 373′
squared.21–25 The PRL areas found with SLO were In previous investigations of PRLs each monocular
337′ to 443′ squared for small fixation targets with PRL was studied separately, but individuals typically
experienced patients. The main difference between perform activities of daily living with both eyes open.
the previously reported and the SLO values is the 3:1 Therefore, previous investigators have forced subjects
ratio between the 95% and 63% equal-frequency to perform visual tasks monocularly without fully
PRLs, that is, the difference in the retinal areas in understanding the consequences for the monocular
which one would expect to find 95% versus 63% of PRL characteristics. In a recent study of binocular
the retinal positions of fixation. PRLs,26 67% of the patients saw stimuli with only 1
To further investigate the concept of a larger retinal eye (monocular perception) while performing simple
locus for fixation, the 400 samples (25 samples free-viewing binocular fixation and perception tasks.
during 8 primary-position trials and 8 secondary- No single visual factor (acuity, contrast, threshold
positions trials) for each patient performing fixation sensitivity, fixation stability, saccadic ability, pursuit
stability in the SLO were analyzed to find the radius ability or scotoma characteristics) had a statistically
from the centroid for each sample. The normalized significant effect on perception of the stimulus,
cumulative frequency indicated that 95% of the time monocular or binocular. However, if the subjects had
the patients kept the fixation stimulus within a retinal a large difference in monocular visual function (e.g., a
CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005 307
Preferred retinal loci—Schuchard
Fig. 2—In a patient with AMD and monocular perception, the binocular PRLs are not in the same direction or at relatively the same
distance from the nonfunctioning fovea in the right eye (left image) and the left eye (right image).
logMAR [logarithm of the minimum angle of resolu- movement tasks; that is, the fellow eye is actually fol-
tion] difference greater than 0.66, or more than lowing the lead of the “dominant” eye. Therefore,
6 lines of letters on the ETDRS [Early Treatment determining the dominant PRL can be critical in
Diabetic Retinopathy Study] chart), the dominant guiding the prescription of monocular-vision-
PRL was the PRL with better visual-function capa- enhancing equipment, testing monocular visual func-
bility. But, in the large group of subjects (82%) who tion and other monocular issues.
did not have these large differences in visual-function
capability between the 2 eyes, the only way of IMPLICATIONS OF PRL AND SCOTOMA
knowing which PRL was dominant was by a binocu- CHARACTERISTICS IN AMD
lar perception test that directly evaluated which PRL
was dominant. The only other factor that seems to The idea of 1 or more PRLs instead of a single
contribute to monocular perception is noncorrespon- retinal point being used for visual tasks is an impor-
dence of binocular PRLs. Fig. 2 shows an example of tant concept for clinical vision testing and visual
binocular PRLs that are in different directions and at tasks. Many clinical tests or treatments presume that
different distances from the nonfunctioning foveas. the eye movement system is operating by referencing
Compare the PRLs in Fig. 2 with those in Fig. 1, a retinal point or at least a single retinal area. A few
where the binocular PRLs are in the same direction examples are presented to demonstrate the implica-
and at relatively the same distance from the nonfunc- tions of PRLs and scotomas for clinical testing and
tioning foveas. visual tasks.
The finding that 2 out of 3 patients with central Reading
scotomas see visual stimuli at fixation with only 1 eye
does not mean that the patients are seeing the entire The ability of the eye to make movements so that
visual world with only 1 eye. Patients reported seeing the visual-target image is placed or kept in the PRL
the world with both eyes, although not always appre- has been shown to be correlated to reading rate and
ciating that when looking at a target in the central reading errors.27–33 Previous reports on reading rate
visual field (e.g., a number, letter or word) they were had shown that the existence of a central scotoma was
seeing the target with only 1 eye. Previous studies of a much greater predictor of reading impairments than
PRL abilities in eye movements might be considered was reduced acuity.31,32 However, the ability of the
to have been testing similar types of tasks: typically PRL to direct eye movements, both saccadic ability
the subject was asked to fixate, saccade to a small (measured by the number of saccades and the saccade
target or follow a small target. It is possible that the characteristics) and fixation stability (measured by
eye that actually is used by the visual system in binoc- the retinal area of fixation), had a much higher corre-
ular tasks is a “dominant” eye that is better at eye- lation with reading speed and correct reading rate
308 CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005
Preferred retinal loci—Schuchard
than did either acuity or scotoma existence. Humphrey 10-2 test) could be the average of flashes
Furthermore, subjects with macular diseases who actually presented at multiple retinal locations. The
were undergoing vision rehabilitation clustered into 2 instability inherent in eccentric PRLs for fixation also
groups, 1 showing significant improvement and the makes accurate mapping of the scotoma boundaries
other showing no improvement in reading with very difficult.
vision-rehabilitation training.33,34 That research also Pericentral fixation targets are used when the
showed a strong association between saccade scores purpose of the testing is to measure visual function in
and reading rate or reading errors. Patients with the central macular retinal area (foveal area). These
AMD who present with low reading rates (fewer than targets are intended to guide fixation by stimulating
10 words/min and more than 2 errors with a short the perifovea (the outer limit of the macular area)
passage) and poor saccadic ability are less likely to while providing no visual stimulation of the inner
show improvement in reading with vision rehabilita- macular area, including the foveal area. When sub-
tion. jects are instructed to “fixate in the centre of the
target,” clinicians often assume that patients with
Visual search central scotomas place the centre of the pericentral
fixation target within the PRL for fixation. They also
Unlike reading, the ability of the eye to make often assume that, when instructed to “look directly
movements so that the visual-target image is placed at the centre of the target,” patients place the target
or kept in the PRL has been shown to not be highly centre in the foveal area, even though the fovea is no
correlated to visual search efficiency or accuracy. longer functioning owing to the central scotoma.
Saccades of patients with central scotomas were However, most patients with central scotomas appear
found to consistently direct fixation targets to the to orient images of targets to the PRL, and verbal
PRL even though the saccades sometimes had the instructions do not change the location of the peri-
latency and dynamic characteristics of saccades not central fixation target.7,8 Therefore, one cannot
directed by the fovea.12,22 The patients with AMD assume that using pericentral fixation targets in clini-
also had, on average, more saccades of shorter length cal tests (e.g., a big X in the tangent screen or Amsler
than normally sighted patients. However, the latency grid), with or without verbal instructions, provides
of the first saccade and the number of saccades to testing with the fovea directed at the centre of the test
move the search-target image into the PRL were not area.
correlated to the search time (together or sepa-
rately).12 Rather, the time after the search-target Patient awareness of macular scotomas
image was moved into the PRL was the greatest con-
tributor to the total visual search time; that is, The presence of scotomas in the macula hinders the
patients with macular scotomas deliberated longer performance of many daily activities, such as reading,
about the identity of the search target. Therefore, any but most patients with AMD are unaware of even
patient with a macular scotoma, regardless of the large defects in their central visual field. A valuable
retinal location of the PRL, will have impaired visual application of the information gained from PRL
search ability beyond the inability to see targets testing and macular perimetry is to educate the
located inside the scotoma. patient about the presence of the scotomas, especially
when they occur relative to fixation. Macular perime-
Perimetry testing try methods can be used to demonstrate to the
patient what a scotoma is and what compensatory eye
The size and location of the PRL have obvious movements can be used to “move the scotoma out of
implications for perimetric testing of patients with the way.” These compensatory techniques are espe-
AMD. For example, perimetric results can be trans- cially important for the nearly 1 out of 5 low-vision
lated in position, and if this translation is not recog- patients who have fairly good visual function (as good
nized the location of the macular scotoma will be as 20/40) but ring scotomas surrounding the PRL
inaccurately positioned in the visual field relative to (Fig. 3). The patient who has gained an awareness of
the fovea. If the patient has a very large PRL for fixa- the scotoma’s effects on visual tasks is more likely to
tion or multiple PRLs, the perimetry results reported make accurate compensatory eye movements to
at 1 point of the visual field (e.g., the grid point in a search for, find and identify pertinent visual informa-
CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005 309
Preferred retinal loci—Schuchard
Fig. 3—A ring scotoma, a scotomatous area completely surrounding the PRL, is a typical stage of the progression of vision loss in
geographic atrophy.This patient has AMD due to geographic atrophy in the right eye (left image) and the left eye (right image).
another example of when the concept of retinal locus
for fixation is important. Calibrating eye trackers
(e.g., the pupil eye tracker in the Humphrey Visual
Field Analyzer) is done by having patients fixate on
targets at known spatial locations. The recordings
from the trackers are transformed to real-world
dimensions by the relationships between these
recordings and the known spatial locations of the fix-
ation targets. The underlying assumption is that
patients place the fixation targets in exactly the same
retinal position each time they are instructed to look
at the target; that is, one must assume that the PRL
has no shift in retinal position when the fixation
target is placed in different spatial locations, even
though multiple PRLs occur. To make matters worse,
the SLO results show that patients use a larger retinal
Fig. 4—PRL shift in the left eye of a patient with AMD. The
locus for fixation when looking at fixation targets in
patient consistently used 2 PRLs in the same retinal locations secondary positions of gaze (see Table 1). To develop
and at approximately the same light levels during 1 year of the transformation algorithm for calibration, one
monitoring: the primary PRL (red circle) with typical bright must have patients look at fixation targets in many
indoor light (about 100 cd/m2 or more) and the secondary secondary positions of gaze.
PRL (purple circle) with less light (e.g., when watching televi-
sion).TS indicates the scotoma that exists at the threshold of Eccentric-viewing training
the secondary PRL.
The orientation of the oculocentric system (fovea
tion. The ability to be aware of the scotoma’s effects or eccentric PRL) affects eccentric-viewing training.
on visual tasks is even more challenging when the Many vision rehabilitation practitioners believe that
patient is using multiple PRLs (Fig. 4), often without patients with central scotomas benefit greatly from
conscious knowledge of changing retinal positions for such training.35,36 A goal of this training is to enable
different lighting conditions or visual tasks. the patient to use a specific eccentric area of the retina
Calibration of eye-position instruments as if it were a fovea. To the extent that the PRL can
play this role, this goal is reasonable. Results of fixa-
The calibration of eye trackers for vision testing is tion performance and visual-search performance,12 as
310 CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005
Preferred retinal loci—Schuchard
measured with an SLO, indicate that patients with 6. Whittaker SG, Budd JM, Cummings RW. Eccentric fixation
central scotomas are able, and indeed likely, to use a with macular scotoma. Invest Ophthalmol Vis Sci 1988;29:
PRL. The PRL appears to be developed by a patient
7. White JM, Bedell HE. The oculomotor reference in humans
simply through experience of a central scotoma, with bilateral macular disease. Invest Ophthalmol Vis Sci
without formal training. However, it is certainly pos- 1990;31:1149–61.
sible that eccentric-viewing training could facilitate 8. Schuchard RA, Raasch TW. Retinal locus for fixation: peri-
this behaviour, making fixation more stable and accu- central fixation targets. Clin Vis Sci 1992;7:511–620.
rate or influencing the patient to use a more optimal 9. Schuchard RA. Validity and interpretation of Amsler grid
reports. Arch Ophthalmol 1993;111:776–80.
eccentric retinal area for eccentric viewing. It is not 10. Guez JE, Le Gargasson JF, Rigaudiere F, O’Regan JK. Is there
known whether patients with central scotomas can be a systematic location for the pseudo-fovea in patients with
trained to develop a new, more optimal, PRL as the central scotoma? Vision Res 1993;33:1271–9.
reference for the oculocentric system. Further results 11. Culham LE, Fitzke FW, Timberlake GT, Marshall J.
are needed to determine what factors (e.g., resolution, Assessment of fixation stability in normal subjects and
patients using a scanning laser ophthalmoscope. Clin Vis Sci
field of view, proximity to the nonfunctioning fovea 1993;8:551–61.
and binocular conditions) the visual system uses 12. Schuchard RA, Fletcher DC. Preferred retinal locus: a review
when naturally choosing a retinal area as the PRL. with applications in low vision rehabilitation. Ophthalmol
Clin North Am 1994;7:243–55.
CONCLUSION 13. Sunness JS, Bressler NM, Maguire MG. Scanning laser oph-
thalmoscopic analysis of the pattern of visual loss in age-
Although there have been reports concerning some related geographic atrophy of the macula. Am J Ophthalmol
aspects of the PRL, the characteristics of the PRL are 1995;119:143–51.
still not well known and understood. For example, 14. Fletcher DC, Schuchard RA. Preferred retinal loci relation-
more results need to be collected to determine ship to macular scotomas in a low-vision population.
whether single or multiple PRLs are influenced by Ophthalmology 1997;104:632-8.
15. Lei H, Schuchard RA. Using two preferred retinal loci for dif-
retinal abnormalities. Also, what is known from previ-
ferent lighting conditions in patients with central scotomas.
ous work needs to be reevaluated, since it was never Invest Ophthalmol Vis Sci 1997;38:1812–8.
determined that the PRLs that were studied were actu- 16. Sunness JS, Applegate CA, Haselwood D, Rubin GS. Fixation
ally used by the visual system in binocular (natural) patterns and reading rates in eyes with central scotomas from
viewing tasks. However, the characteristics that are advanced atrophic age-related macular degeneration and
known about the PRL have been shown to be impor- Stargardt disease. Ophthalmology 1996;103:1458–66.
tant in their relationship to activities of daily living for 17. Fletcher DC, Schuchard RA, Livingstone CL, Crane WG,
Hu SY. Scanning laser ophthalmoscope macular perimetry
patients with AMD. Therefore, clinicians may want to
and applications for low vision rehabilitation clinicians.
consider assessing the PRLs and scotomas of their Ophthalmol Clin North Am 1994;7:257–65.
patients as part of a complete evaluation. 18. Sunness JS, Schuchard RA, Shen N, Rubin GS, Dagnelie G,
Haselwood DM. Landmark-driven fundus perimetry using
REFERENCES the scanning laser ophthalmoscope. Invest Ophthalmol Vis Sci
1. von Noorden GK, Mackensen G. Phenomenology of eccen-
19. McDonnell JM. Ocular embryology and anatomy. In: Ogden
tric fixation. Am J Ophthalmol 1962;53:642–61.
TE, editor. Retina. St. Louis: CV Mosby Company; 1989.
2. Dalgleish R, Naylor EJ. Bilateral eccentric fixation with no
20. Bishop PO. Binocular vision. In: Moses RA, Hart WM,
ocular deviation in a case of heredo-macular degeneration. editors. Adler’s physiology of the eye. St. Louis: CV Mosby
Br J Ophthalmol 1963;47:11–3. Company; 1987.
3. Cummings RW, Whittaker SG, Watson GR, Budd JM. 21. Kosnik W, Fikre J, Sekuler R. Visual fixation stability in older
Scanning characters and reading with a central scotoma. adults. Invest Ophthalmol Vis Sci 1986;27:1720–5.
Am J Optom Physiol Opt 1985;62:833–43. 22. Steinman RM. Effect of target size, luminance, and color on
4. Timberlake GT, Mainster MA, Peli E, Augliere RA, Essock monocular fixation. J Opt Soc Am 1965;55:1158–65.
EA, Arend LE. Reading with a macular scotoma. I. Retinal 23. Whittaker SG, Cummings RW, Swieson LR. Saccade control
location of scotoma and fixation area. Invest Ophthalmol Vis without a fovea. Vision Res 1991;31:2209–18.
Sci 1986;27:1137–47. 24. Steinman RM, Cunitz GT. Fixation of targets near the
5. Timberlake GT, Peli E, Essock EA, Augliere RA. Reading absolute foveal threshold. Vision Res 1968;8:277–86.
with a macular scotoma. II. Retinal locus for scanning text. 25. Skavenski AA, Hansen RM, Steinman RM, Winterson BJ.
Invest Ophthalmol Vis Sci 1987;28:1268–74. Quality of retinal image stabilization during small natural and
CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005 311
Preferred retinal loci—Schuchard
artificial body rotations in man. Vision Res 1979;19:675–83. 32. Legge GE, Ross JA, Isenberg LM, LaMay JM. Psycho-
26. Schuchard RA, Tekwani N, Hu SY. Binocular preferred physics of reading: clinical predictors of low-vision reading
retinal loci: relationship of visual factors to binocular percep- speed. Invest Ophthalmol Vis Sci 1992;33:677–87.
tion. OSA Tech Dig Ser 1995;1:332–5. 33. McMahon TT, Hansen M, Stelmack J, Oliver P, Viana MA.
27. Culham LE, Fitzke FW, Timberlake GT, Marshall J. Use of Saccadic eye movements as a measure of the effect of low
scrolled text in a scanning laser ophthalmoscope to assess read- vision rehabilitation on reading rate. Optom Vis Sci 1993;70:
ing performance at different retinal locations. Ophthalmic Phy- 506–10.
siol Opt 1992;12:281–6. 34. McMahon TT, Hansen M, Viana M. Fixation characteristics
28. Rubin GS, Turano K. Reading without saccadic eye move- in macular disease. Invest Ophthalmol Vis Sci 1991;32:
ments. Vision Res 1992;32:895–902. 567–74.
29. Rubin GS, Turano K. Low vision reading with sequential 35. Holcomb JG, Goodrich GL. Eccentric viewing training.
word presentation. Vision Res 1994;34:1723–33. J Am Optom Assoc 1976;47:1438–43.
30. Fletcher DC, Schuchard RA, Watson G. Relative locations of 36. Goodrich GL, Mehr EB. Eccentric viewing training and low
macular scotomas near the PRL: effect on low vision reading. vision aids: current practice and implications of peripheral
J Rehabil Res Dev 1999;36:356–64. retinal research. Am J Optom Physiol Opt 1986;63:119–26.
31. Legge GE, Rubin GS, Pelli DG, Schleske MM. Psycho-
physics of reading. II. Low vision. Vision Res 1985;25: Key words: age factor, macular degeneration, ocular fixation,
312 CAN J OPHTHALMOL—VOL. 40, NO. 3, 2005