Optometry (2009) 80, 630-634

Does the dynamic cross cylinder test measure

the accommodative response accurately?
Jaclyn A. Benzoni, O.D., M.S., Juanita D. Collier, O.D., M.S., Kimberley McHugh, O.D.,
Mark Rosenfield, M.C.Optom., Ph.D., and Joan K. Portello, O.D., M.P.H., M.S.
State University of New York State College of Optometry, New York, New York.

KEYWORDS
Accommodation;
Near testing;
Optometer;
Refraction

Abstract
BACKGROUND: The dynamic cross cylinder (DCC) test is a standard clinical procedure used to assess the
accommodative response (AR) subjectively. However, because of potential problems arising from the
ambiguous stimulus conditions, it is unclear whether this test provides an accurate measure of the AR.
The aim of this study was to compare clinical subjective findings with objective measurements of the AR.
METHODS: Subjective findings to a 2.50-diopter (D) accommodative stimulus obtained with the DCC
test (without fogging lenses) were compared with objective measurements of the AR obtained with a
Grand Seiko WAM 5500 optometer (RyuSyo Industrial Co. Ltd., Kagawa, Japan) in 25 young subjects.
As spherical lenses were introduced to quantify the subjective finding, objective measures of the AR were
also recorded through these lenses.
RESULTS: The mean AR recorded subjectively and objectively was 2.35 and 1.68 D, respectively
(P , 0.0001). Of the 10 subjects who demonstrated a lead of accommodation subjectively, only 1 had
a lead objectively. For the 8 subjects who showed a lag of accommodation subjectively, all had a lag objectively. Introducing lenses produced a significant change in the mean AR.
CONCLUSION: The subjective DCC test as performed here does not provide an accurate measurement
of the AR to a near target in a young population. We recommend that alternative techniques such as
using an objective, open-field optometer or Cross-Nott retinoscopy be adopted for determining the
within-task AR.
Optometry 2009;80:630-634

The assessment of the accommodative response (AR) to

a range of stimuli is an important part of the clinical
optometric examination.1 Patients symptoms frequently relate to near vision activities, and inappropriate responses,
whether under or over-accommodation relative to the object
of regard are a frequent cause of asthenopia.2 Accordingly,
it is essential for the clinician to determine the actual AR to
the stimulus condition for which the patient is reporting
Corresponding author: Mark Rosenfield, M.C.Optom., Ph.D, State
University of New York State College of Optometry, 33 West 42nd Street,
New York, New York 10036.
E-mail: Rosenfield@sunyopt.edu

difficulty. In some cases, a more complete examination of

a range of responses may be appropriate, which can be
achieved by plotting an accommodative stimulus-response
curve.3,4 An example of such a plot is illustrated in Figure 1.
Only for a single stimulus level (the so-called crossover
point) are the accommodative stimulus and response equal.
Whereas the plot illustrated in Figure 1 represents an
average finding, patients presenting with symptoms relating
to near vision activities may exhibit different results. For example, some patients show a tendency to overaccommodate
for near targets2 rather than exhibiting the more typical lag
of accommodation.3-6 A lead of accommodation can be a
primary problem resulting from trauma, systemic pathology,

1529-1839/09/$ -see front matter 2009 American Optometric Association. All rights reserved.
doi:10.1016/j.optm.2009.07.012

Benzoni et al

Clinical Research

or the administration of either ocular or systemic pharmacologic agents, or secondary to a vergence abnormality such as
convergence insufficiency.1,7 Although the object of regard
will remain clear as long as the accommodative error does
not exceed the depth-of-focus of the eye, such a patient
may still experience symptoms, possibly resulting from the
effect of excessive accommodative convergence. An excessive lag or underaccommodation relative to the accommodative stimulus may also produce asthenopia. This symptom is
most typically found in early presbyopia, but may also be associated with both systemic and ocular pathological conditions, pharmacologic therapy, neurologic abnormalities,
and functional disorders such as basic exophoria, divergence
excess, and accommodative infacility.7-9
One of the most commonly used clinical procedures to
assess the AR is the dynamic cross-cylinder (DCC) test.1
Here, subjects view a pattern of intersecting horizontal
and vertical lines through a cross cylinder (typically
60.50 diopters [D]) to create mixed astigmatism, with
the horizontal and vertical lines theoretically equidistant
in front of and behind the retina. If the patient is accommodating exactly in the plane of the target, then after introduction of the cross cylinder, the circle of least confusion
(COLC) will lie on the retina, and the patient will report
that both sets of lines (horizontal and vertical) appear
equally clear (or more accurately, equally blurred). However, if the patient has either a lag or lead of accommodation, then the horizontal or vertical lines, respectively, will
appear clearer. Spherical lenses can subsequently be introduced to move the COLC onto the retina, and the lens
power required to make the 2 sets of lines appear equally
clear provides a measure of the accommodative error.
A number of different procedures have been advocated
for performing the DCC test. In principle, the test may be
carried out under monocular, binocular fused, or binocular
unfused conditions.1 Although monocular testing should
assess the blur-driven and proximally induced AR only, under binocular fused conditions, convergent accommodation
will be added to the response. Binocular unfused testing
uses vertical prisms to dissociate the patient, thereby eliminating disparity-vergence. As might be expected, this
results in a response very similar to the monocular condition.10 Additionally, some techniques (e.g., the 14A and
14B procedures advocated by the Optometric Extension
Program2,11) recommend that additional plus lens power
should be added to the refractive correction so that the vertical lines appear clearer and then the plus power reduced
until the horizontal and vertical lines are equally clear
and dark. This is discussed later in this report.
A number of significant problems with the DCC test
make it of questionable value in prepresbyopic patients
with active accommodation, irrespective of the specific
technique being used. For example, it is assumed that a
patient who has minimal accommodative error under
naturalistic conditions will, during testing, produce an AR
to the dioptrically conflicting, rectilinear target that lies
exactly midway between the 2 foci (i.e., places the COLC

631

Figure 1

Static accommodative stimulus-response curve for a normal

subject. 1 5 initial nonlinear region, 2 5 linear region, 3 5 transitional
soft saturation region, 4 5 hard saturation presbyopic region. The diagonal line represents the unit ratio (or 1:1) line. Figure redrawn with permission from Ciuffreda and Kenyon4 (1983).

on the retina). Little evidence supports this proposal. For

example, Portello et al.12 used an infrared optometer to
measure the AR in subjects with uncorrected astigmatism.
They found that subjects generally exerted the minimum
accommodation necessary to place the anterior focal line
within the depth-of-focus of the eye. However, under
none of the conditions tested was the COLC positioned
on or close to the retina.
The observation that the AR changes after the introduction of lenses provides an additional difficulty with this test.
If a lag of accommodation is observed, plus lenses are
introduced to obtain the required endpoint. However, in a
young patient with active accommodation, the introduction
of additional plus power is likely to stimulate a reduction in
the blur-driven AR, provided the subject is able to detect
the change in accommodative stimulus in the presence of
the uncorrected astigmatism. If the reduction in accommodation is equal to the magnitude of the plus lens, then the
subjective response to the test will remain unchanged.13-16
Accordingly, the current study compared subjective
measurements of the AR obtained with the DCC procedure
with objective measurements taken using an open-field,
infrared optometer to assess the validity of this clinical
procedure.

Methods
The study was performed on 25 visually normal subjects
having a mean age of 23.4 years (range, 20 to 30 years). All
subjects were optometry students at the State University
of New York State College of Optometry, and had bestcorrected visual acuity of at least 6/6 (20/20) in each eye.
None had any manifest ocular disease or strabismus. The

Optometry, Vol 80, No 11, November 2009

Subj AR Obj AR (D)

632
2
1.5
1
0.5
0
-0.5
-1
0

2
Objective AR (D)

Figure 3

Difference between the subjective AR determined using the

DCC procedure and the objective AR measured with an infrared optometer
as a function of the objective response. The horizontal dashed line indicates the mean difference, and the area between the 2 solid horizontal
lines represents the 95% limits of agreement between the 2 values.

Figure 2

Photograph of the experimental setup. Subjects wore both

their habitual refractive correction and the 60.50 D cross cylinders in a
trial frame, and viewed a rectilinear target (not visible here as it was on
the side of the nearpoint card nearest the subject) at a viewing distance
of 40 cm. The subject was positioned at the autorefractor throughout,
which allowed objective measurements to be recorded in addition to the
subjective responses.

study followed the tenets of the Declaration of Helsinki,

and informed consent was obtained from all subjects after
an explanation of the nature and possible consequences of
the study. The protocol was approved by the Institutional
Review Board at the SUNY State College of Optometry.
All of the testing was completed in a single session.
The DCC test was performed without the use of additional
fogging lenses.1,17 Cross-cylinders of 60.50 D (minus axes
vertical) were introduced before each eye. Subjects were required to view a high-contrast, black on white, cross cylinder
target, mounted on a near point rod at a viewing distance of
40 cm. The target was viewed through the habitual refractive
correction, which was mounted in a trial frame. Subjects
were asked to indicate whether the horizontal or vertical lines
appeared either clearer or darker. If a preference was indicated, then spherical lenses were introduced before both
eyes until the 2 sets of rectilinear lines appeared equally
clear. This lens value was taken as the accommodative error,
i.e., the difference between the accommodative stimulus and
response. If no equal response was found, then the endpoint
was taken as the midpoint between the lenses that produced a
change in subjective response. For example, if the subject reported that the horizontal lines were clearer initially, but the
vertical lines appeared clearer through an additional 10.25 D
lens, this was recorded as a 0.12 D lag of accommodation. In
accordance with the conventional clinical procedure, the luminance of the target was maintained at a low level (approximately 10 cd/m2). Additionally, the AR to the DCC target
was measured objectively using a Grand Seiko WAM 5500
infra-red optometer (RyuSyo Industrial Co. Ltd., Kagawa,
Japan). This instrument has been described and evaluated
previously.18-20 Immediately after the subjective testing, objective measurement of the AR was performed with the

habitual refractive correction in place. If the subject indicated subsequently that one set of lines on the DCC target appeared clearer, then spherical lenses were introduced to
achieve the position of subjective equality, and the AR was
measured through each of these supplementary lenses using
the infrared optometer. All objective data were recorded
from the right eye only, and for each condition at least 10
readings of the refractive state of the eye were taken, converted into spherical equivalents, and averaged. Because
the instrument takes measurements at approximately 1- to
2-second intervals, the time to assess the AR for each stimulus level was approximately 15 seconds. The objective measurements typically took 1 to 2 minutes to complete.
However, subjects received a short break between measurements while the lenses were being changed.
The same experimental set up was used for both the
objective and subjective measurements. Subjects wore a
trial frame and lenses with their distance refractive correction, and additional lenses, when indicated by the subjective responses, were added to the trial frame as necessary.
For all testing, subjects viewed the nearpoint card mounted
on the infrared optometer, with their head against the
forehead rest and chin placed on the chinrest. The experimental setup is shown in Figure 2.

Results
The mean AR recorded subjectively and objectively was
2.35 D (SD 5 0.60) and 1.68 D (SD 5 0.49), respectively.
A paired t test indicated that this difference was significant
(t 5 7.13; df 5 24; P , 0.0001). Additionally, the difference between the subjective and objective findings was calculated for each individual, and the 95% limits of
agreement, calculated as 1.96 multiplied by the standard
deviation of the differences21 was 60.95 D. These differences are illustrated in Figure 3.
ARs for each individual are shown in Table 1. Linear regression analysis indicated a significant correlation between
the subjective and objective findings (r2 5 0.39; P 5 0.001),
with a regression line described by the equation: objective
AR 5 (0.51x subjective AR) 1 0.47. Ten subjects exhibited

Clinical Research

633

Table 1 Subjective and objective measurements of the AR

for each of the 25 subjects tested
Subjective AR, Objective AR, Difference
[D]
[D] (SD)
(subjective-objective)
2.75
3.00
3.25
2.62
3.00
3.25
2.50
2.50
1.75
2.25
2.75
2.62
2.62
1.75
2.50
1.75
1.25
1.75
1.75
2.50
2.50
2.50
2.50
2.50
0.75
Mean 2.35
SD
0.60

2.20
1.74
2.06
2.33
1.82
2.89
1.90
1.71
1.11
1.98
1.23
1.62
1.33
1.13
1.67
0.43
1.09
1.93
1.36
1.88
1.72
1.56
1.78
1.85
1.30
1.66
0.49

(0.87)
(0.13)
(0.33)
(0.12)
(0.35)
(0.20)
(0.19)
(0.24)
(0.27)
(0.30)
(0.28)
(0.73)
(0.24)
(0.16)
(0.18)
(0.57)
(0.17)
(0.25)
(0.15)
(0.17)
(0.47)
(0.12)
(0.14)
(0.24)
(0.19)

0.55
1.26
1.19
0.29
1.18
0.36
0.60
0.79
0.64
0.27
1.52
1.00
1.29
0.62
0.83
1.32
0.16
20.18
0.39
0.62
0.78
0.94
0.72
0.65
20.55
0.69
0.48

a lead of accommodation (AR . 2.50 D) based on the subjective findings. However, only one of these subjects demonstrated a lead of accommodation objectively. Of the 8
subjects who showed a lag of accommodation (AR , 2.50
D) subjectively, all also demonstrated a lag objectively.
For the 8 subjects with a lag of accommodation based on
the subjective response, plus lenses were introduced to
achieve subjective equality, i.e., so that the 2 sets of lines
appeared equally clear. Objective measurements using the
infrared optometer verified that the lenses produced a
decline in the AR, as shown in Figure 4. Friedmans nonparametric test indicated that the decline in AR after the introduction of the plus lenses was significant (c2 5 16.05;
P 5 0.000).

Discussion
The findings of the current study clearly indicate that the
subjective DCC test using the methodology adopted here
does not provide veridical measurements of the AR in young
subjects with active accommodation. The mean results
obtained with this test are significantly higher than those
obtained with an objective infrared optometer. In addition,
the range of differences between the 2 measurements of AR

Objective AR (D)

Benzoni et al

1.5
1
0.5
0

0.25

0.5

0.75

Lens power (D)

Figure 4

Mean AR measured objectively through plus lenses using the

infrared optometer in 8 subjects. These subjects all exhibited a lag of accommodation when tested subjectively with the DCC test (i.e., horizontal
lines appeared clearer) through their habitual refractive correction (0 D).
Accordingly, supplementary plus lenses were introduced to produce an
equal subjective response and the AR measured objectively though these
lenses. It is apparent that the AR changed significantly when the lens
power was modified. Error bars indicate 1 SEM.

exceeded 2 D, while the DCC test overestimated the number

of subjects having a lead of accommodation.
In comparing the findings of the subjective and objective
techniques for assessing the accommodative response, differences between the procedures should be noted. For
example, with the DCC technique adopted here, a single
measurement was taken. Had either a bracketing technique
been used, i.e., going beyond the first neutral response and
then changing lenses in the opposite direction, or a high
neutral procedure, whereby additional plus power is added
initially, and then reduced until an equal response is
obtained, then different (and possibly superior) findings
might have been found. In addition, 10 readings of the
accommodative state were obtained with the infrared optometer over approximately 15 seconds and subsequently
averaged. Accordingly, depending on the number of lens
changes required, the objective determination typically took
1 to 2 minutes to complete. In contrast, the single subjective
assessment generally took only 20 to 30 seconds. Further, it
is unclear whether subjects can be trained to respond better
with the DCC, i.e., to become more sensitive to the
differences in clarity of the rectilinear lines. However, our
personal experience when teaching this procedure to optometry students is that they quickly become capable of
achieving any response, i.e., altering their AR so that either
the horizontal lines appear clearer, the vertical lines appear
clearer, or both sets of lines appear equal. Finally, to our
knowledge, no studies have been performed to determine the
sensitivity of the DCC, i.e., its ability to identify accurately
those individuals with accommodative anomalies.
It is clear that the subjective procedure frequently
overestimates the AR in young subjects. One possible
explanation for this difference is that subjects are voluntarily changing their accommodation to the target. It is
presumed at the outset that a subject who is accommodating
accurately on the target (AR 5 2.50 D) will maintain the
COLC on the retina, with the horizontal and vertical lines
remaining equally blurred. However, subjects could increase their AR, thereby making the vertical lines appear
clearer, or alternatively reduce the AR to make the

634
horizontal lines clear. Interestingly, Bannon and Walsh22
stated that for the recognition of letters, the vertical strokes
are most important, so that a patient may prefer to keep the
vertical strokes clear. If a patient habitually accommodates
to make the vertical portion of the target clear, then this
would explain the overaccommodation seen with the subjective test. Certainly, use of a rectilinear target is not conducive to maintaining the COLC on the retina. It has also
been proposed that fogging lenses (e.g., an additional
11.00 sphere) be introduced over the distance refractive
correction so that both the horizontal and vertical focal
lines lie in front of the retina.17 This is also the high neutral procedure advocated by the Optometric Extension
Program and others.2,11,17 The term fogging is inappropriate
here, because it refers to making a patient myopic when determining the distance refractive error so as to minimize the
AR. During the DCC test, introduction of an additional
11.00 sphere will reduce the accommodative stimulus
from 2.50 D to 1.50 D but not make the patient myopic
or blur their visual acuity.
The introduction of any lens during the test procedure
will modify the accommodative stimulus and therefore alter
the resulting AR. If the goal is to determine the response for
a particular stimulus level (e.g., 2.50 D), then that stimulus
must be maintained throughout the test. Once plus lenses
are introduced, if the patient reduces the AR by an amount
equal to the magnitude of the lens power, then the
subjective response will remain unchanged. Under binocular test conditions, a change in subjective response may be
obtained only when the patient is no longer able to reduce
accommodation while maintaining accurate vergence on
the target, i.e., exert negative relative accommodation.1
Therefore, the DCC procedure is not a direct quantification
of the AR for a 2.50 D stimulus, but rather uses subjective
responses to determine the lens that yields zero accommodative error.
Accordingly, the DCC test as performed here does not
provide an accurate measure of the AR under normal
viewing conditions. When assessing the AR on a young
patient, an ideal test should use a naturalistic stimulus
placed at the patients habitual working distance to simulate
normal near-work conditions. Additionally, it should avoid
the use of supplementary lenses over the refractive correction because they will alter the accommodative stimulus
(and resulting AR) from the habitual state. The use of either
an objective, open-field optometer (if available) or CrossNott retinoscopy,1,23-25 whereby the point conjugate with
the retina during active accommodation is determined by
altering the retinoscopy working distance seem to be the
optimal procedures to determine the within-task AR.

Optometry, Vol 80, No 11, November 2009

References
1. Rosenfield M. Accommodation. In: Zadnik K, ed. The ocular examination. Measurement and findings. Philadelphia: Saunders; 1997:87-121.
2. Birnbaum MH. Optometric management of nearpoint vision disorders.
Boston: Butterworth-Heinemann; 1993:53-71.
3. Morgan MW. Accommodation and its relationship to convergence. Am
J Optom Arch Am Acad Optom 1944;21:183-95.
4. Ciuffreda KJ, Kenyon RV. Accommodative vergence and accommodation in normals, amblyopes and strabismics. In: Schor CM,
Ciuffreda KJ, eds. Vergence eye movements: basic and clinical aspects. Boston: Butterworths; 1983:101-73.
5. Heath GG. Components of accommodation. Am J Optom Arch Am
Acad Optom 1956;33:569-79.
6. Charman WN. The accommodative resting point and refractive error.
Ophthalmic Optician 1982;21:469-73.
7. Scheiman M, Wick B. Clinical management of binocular vision. Philadelphia, Lippincott; 1994.
8. Hofstetter HW. Factors involved in low amplitude cases. Am J Optom
Arch Am Acad Optom 1942;19:279-89.
9. Ciuffreda KJ. Accommodation and its anomalies. In: Charman WN
ed. Vision and visual dysfunction. Vol 1. Visual optics and instrumentation. Boca Raton, FL: CRC Press, 1991:231-79.
10. Ong J, Schuchert J. Dissociated versus monocular cross-cylinder
method. Am J Optom Arch Am Acad Optom 1972;49:762-4.
11. Manas L. Visual analysis, 3rd ed. Chicago: Professional Press; 1965:
155-6.
12. Portello JK, Hong SE, Rosenfield M. Accommodation to astigmatic
stimuli. Optom Vis Sci (suppl) 2001;78:91.
13. Fry GA. Significance of fused cross cylinder test. Optometric Weekly
1940;31:16-9.
14. Goodson RA, Afanador AJ. The accommodative response to the near
point crossed cylinder test. Optometric Weekly 1974;65:1138-40.
15. Rosenfield M, Carrel M. Effect of near-vision addition lenses on the
accuracy of the accommodative response. Optometry 2001;72:19-24.
16. Haynes HM. Clinical observations with dynamic retinoscopy. Optometric Weekly 1960;51:2243-6, 2306-9.
17. Grosvenor T. Primary care optometry, 4th ed. Boston: ButterworthHeinemann, 2002:284-5
18. Chat SWS, Edwards MH. Clinical evaluation of the Shin-Nippon
SRW-5000 autorefractor in children. Ophthal Physiol Opt 2001;21:
87-100.
19. Mallen EAH, Wolffsohn JS, Gilmartin B, et al. Clinical evaluation of
the Shin-Nippon SRW-5000 autorefractor in adults. Ophthal Physiol
Opt 2001;21:101-7.
20. Wolffsohn JS, Gilmartin B, Mallen EAH, et al. Continuous recording
of accommodation and pupil size using the Shin-Nippon SRW-5000
autorefractor. Ophthal Physiol Opt 2001;21:108-13.
21. Bland JM, Altman DG. Statistical methods for assessing agreement
between two methods of clinical assessment. Lancet 1986;1-8476:
307-10.
22. Bannon RE, Walsh R. On astigmatism. Part IIIdSubjective tests. Am J
Optom Arch Am Acad Optom 1945;22:210-8.
23. Cross AJ. Dynamic skiametry in theory and practice. New York: AJ
Cross Optical Co, 1911.
24. Nott IS. Dynamic skiametry, accommodation and convergence. Am J
Phys Opt 1925;6:490-503.
25. Nott IS. Dynamic skiametry. Accommodative convergence and fusion
convergence. Am J Phys Opt 1926;7:366-74.