Spatial frequency dependence of accommodative responses in amblyopic eyes

VisionRes. Vol. 23. No. 12, pp. 1585-1594.1983

0042-698918353.00+ 0.00 Copyright ‘6 1983 Per8amonPressLtd

Printed in Great Britain. All rights reserved

SPATIAL

FREQUENCY DEPENDENCE OF ACCOMMODATIVE RESPONSES IN AM~LYOPlC EYES KENNETH

J.

CIUWREDA

and STEVEN C. HOKODA

Amblyopia Laboratory, Department of Vision Sciences. State College of Optometry. State University of New York. 100 East 24th Street. NY 10010. U.S.A.

Abstract-Monocular. steady-state accommodative responses were measured as a function of spatial frequency of simple sinusoidal gratings presented at high contrast and target vergence levels in amblyopes. as well as in strabismics without amblyopia and in visually-normal control subjects, In general. spatial frequency dependence of the accommodative response was the rule. However, the amblyopic eyes exhibited markedly reduced accommodative responses over most of the spatial frequency range tested. and this was attributed to reduced accommodative controller gain in the sensory pathways involved. in the control of accommodation in the amblyopic eye. Due to the diversity of accommodative response spatial frequency profiles found across all groups. the results suggest that reflex. voluntary. and higherlevel perceptual aspects of accommodation may interplay in a complex manner in the act of

accommodation on a simple sinusoidal grating. Accommodation

Amblyopia

Contrast

Spatial frequency

INTRODUffiON It is

well-known that static aspects of accommodation are abnormal in amblyopic eyes for stimuli, such as square-wave gratings or fetters on a visual acuity chart, containing broadband spatial frequency components. Such accommodative abnormalities include increased steady-state error (Otto and Graemiger. 1967: Wood and Tomhnson, 1974; Otto and Safra, 1976: Kirschen PI ul.. 1981: Ciuffreda et al., 1983). increased depth of focus (Ciuffreda et al., 1983) reduced maximum amplitude (Hokoda and Ciuffreda. 1982; Ciuffreda et al., 1983), and defective vergence drive (Costenbader ef crl.. 1948: Urist, 1959; Kenyon pt a(., 1980, 1981). These accommodative deficits have been attributed to the effects of early abnormal visual experience on the sensory pathways involved in the control of accommodation in the amblyopic eye (Ciuffteda and Kenyon. 1983 : Ciuffreda et al., 1983). The spatial frequency dependence of accommodation has been demonstrated in normal subjects (Heath, 1956: Phillips, 1974; Charman and Tucker, 1977 : Charman and Heron, 1979 : Owens, 1980 : Bour. 1981). but this has not been explored in amblyopic individuals. The results of such an investigation would help to define quantitatively those portions of the spatial frequency spectrum adversely affected by the abnormal visual experience and contributing to the accommodative deficits found in amblyopic eyes when focusing on broadband stimuh

sinusoidal gratings were generated on an oscilloscope (Gould 08255. P7 phosphor) using conventional techniques (Green and Campbell. 1965). Grating spatial frequency ranged from 0.5 to approximately 16.0 cdeg (or maximum focusable spatial frequency) in octave divisions presented in a random sequence. When grating spatial frequency was changed, blank fields were not interleaved. and thus subjects used the previous grating to assist in the initial accon~l~~od~ltive guidance and adjustment to the new grating. Grating vergence was generally 5 or 6 D and was typically at least I D less than the accommodative amplitude of the tested eye. We purposely remained about 1 D or

A

-EE

-

A schematic diagram of the experimental apparatus is presented in Fig. 1. High contrast (SOY,,modulation)

HO

OWG /

u

SL Fig.

METHODS

Abnormal visual experience

1. Schematic diagram of experimental apparatus. top

view. Symbols: HO Hartinger optometer, EE experimenter’s eye. SE subject’s eye. OWG oscilloscope with gratings. FLA . held limiting aperture, SL stimulating lens. CL correcting lens for ametropia, and PRM partially rebecting mirror. Not drawn to scale. 1585

more below this amplitude level to prevent contamination of responses b! accommodative saturation effects and visual fati_pue. However. we also purposely used high accommodative stimulus levels in order to maximise possible changes in accommodative response with target spatial frequency, as the range of accommodative responses is bordered by the accommodative amplitude on the upper end and the tonic accommodative level on the lower end. Preliminary testing in some of the amblyopic subjects at lower accommodative stimulus levels yielded similar results but with somewhat less dramatic differences (dioptricall~). as one’s accommodation typically does not exceed the stimulus level (for moderate to high stimulus values) nor fall below the tonic level. Grating vergence level was accomplished through a change in the value of the stimulating lens and/or distance of the oscilloscope from the subject. Magnification of this lens/target combination was then computed and used in the precise calculation of the grating spatial frequency in the stimulus field. The 3.7 deg field of view was controlled by a circular aperture which only allowed the subject to see the glass oscilloscope face and not adjacent portions of the field. The green homogeneous aperture surround (20 deg) was adjusted by the subject to have the same space-averaged luminance (IO cd
rotated the calibrated ring of the optometcr. which optically displaced its two sets of vertical bars. until the subject reported precise vertical altgnment : this value represented the accommodative state of the eye. Preliminary testing using foveaI]!-presented flashed optometer targets yielded comparable results. Further. followin_p 15 min of practice. all subjects were able to make reliable settings. snnilar to what has been found by others using experienced subjects (Hung and Semmlow. 1980: Ciuffreda ct (I/.. 1983). Eye movements and eye closure were encouraged between measurements to avoid grating adaptanon efiitctb (Blakemoreand Campbell. 1969). St?; to ei_rht measurements were taken at each spattal frequency. wth randomization oftheinitial direction and olfset ma_cnitudc of the optometer bar targets. Dioptric stimuli and average accommodative responses were referenced IO the comeal plane. The Hartinger optometer was also used to measure the tonic or resting level of accommodation. Under our test conditions, it was represented b\- the static accommodatjve response level to a uniform green tield having the same space-averaged ~uIi~~nance as the grating. Following at least I min of fixation into the green field (in an otherwise totally darkened room) to allow for completion of transient accommodative changes (Phillips. 1974). at least four measures of the accommodative response were taken and averaged. and this represented the tonic level of accommodation. The contrast sensitivity function was measured in all subjects in order to determine its relationship IO the accommodative response spatial frequency profile. as has been demonstrated by Owens (1980) in visuallynormal subjects. An Optronix Series 200 Vision Tester was used. Test distance was IO feet. Space-avcraped luminance of the circular (3.8 degrees) test field and the rectangular (8.0 H and 10.0 V dq) surround tield were 75 and 24 cd.!m”. respectively. The full spectacle lens prescription was worn, and the non-tested eye WIS fully patched. Three measurements. using the ascending method of limits, were taken at each spatial frequency and averaged. Subjects from several diagnostic groups were tested. This included 7 amblyopes. 1 former amblyope. and 9 vtsually-normal control subjects. In addition. since most of our amblyopic subjects also had strabismus. we tested two subjects having strabismus without an~blyopia, in order to assess the possible contributi(~r~ of strabismus alone on the ~tccoll~tl~od~ltive rcsponscs in amblyopic eyes. No subjects had a history, or presented with signs and/or symptoms. of ocular or neurological disease. Ail had :I thorough vision examination prior toexperimental testing. The relevant clinical findings are presented in Table I for all subjects. Hk:sl~I.'rs The results in normals are presented in Fig. 2. Rcaponses fell into four groups, with spatial frequency specificity being the rule. Some subjects (1 N-3N) es-

1.587

Spatial frequenry dependence of accommodative responses Table

I. Clinical

data of subjects

Distance SubJcct

Age

Refract&

lyr)

status(D)

IA

strabtsmlc deviation

NOW?

LE

2.75

RE

2.25

1.50

ZA

LE

3.00

-1.00.8

RE

2.25

0.75

3A

LE

- 3.50

RE

3.75

LE

- 1.00

RE

-0.25

LE

- 5.25

4A

(A,

Visual

Fixation

acuity*

status (A)’

amplitude

ht,tor>

(D14

9.1

Early

20 32

0.8 IT

8.0

corrected

10 LET

20,47

1.0 N

7.4

No previous

20 13

Central

1.8

None

20 13

Central

9.4

Earl?

20 :49

Central

8.2

corrected

7.3

Surgeq

7.6

and 6 yr: ,otcro,,ttcot

6.4

Orthoptlcs

20.20

165 176

Central

20 49

4 LE(T)

I.0

20,13

unsteady

unsteadq N

Central

hlhtor)

1.75

3 LET

70

20 51

0.5 ST

20’13

RE PL

spectacles

treatment

historT

of esotropia

wtth

spectacles

for esotropi:t

mitul

6.7

Central

of rwtropia

with

patching 5A

ocular

Significant

Accommodative

past 6 months:

visual

worn 26

7A

LE

- 0.25

- 0.25

80

RE

4.75

3.75

30

0.25

90

0.50

180

LE

29

-0.75

RE 13

FA

0.25

LE

.0.25

180

‘RE

-0.25

0.50

20 13

3RXT

Central

20 52 6 RET

0.5 N

20 14

Central

20 118 8 LET

180

unsteady

6.0 NS

20’17

and

1.2 NS

20 13

3 LHYPER

unsteady

Central

unsteady

itwit)

of correction

past 6 months

6.9

No

6.0

refractive

5.6

No previous

5.0

refractke

6.8

Orthoptics

prewous

imtial

10.3

13>r

age I2 and

20 82: refrnctire 6A

age 4

treatment

or

correction treatment

or

correctmn past 18 months:

visual

acuit)

20 109 LE IS

34 24

2s IN

2.25

0.50

180

2.25

0.25

180

LE

0.50

0.25

> 180

‘RE

0.75

LE

22

2N

LE ‘RE

23

3.00,~-1.00

‘RE

1.00

‘LE

0.25

25 LX(T) 20 AL7

* 150

1.50

ET

None

5

20 13

Central

unsteady

5.6

20 13

Central

unsteady

5.7

20 19

0.5 N unsteady

7.6

20 16

I .O IT unsteady

7.6

20 13

Central

8.4

20 I3

Cetltrkll

8.2

No previous Surgery

a$$ I ?‘r

None

20 I3

Central

7.1

Brief period

20 13

Centr;tl

6.9

to improve

None

20 I3

CcolrZil

6.1

None None

None

RE PL

treatment

of orthopttcs fusion

I yr ago 3N

24

‘LE

4.00

4N

26

ILE

0.25

5N

26

6N

28

7N

28

8N

34

9N

A

2.5

amhl:opc: RE N

nasal:

‘Dominant ‘Assessed probit

None

20 I3

Central

6.8

None

20 I3

Central

6.9

None

20 I3

Central

7.2

20 I3

Central

7.2

LE

0.50

‘RE

. 0.50

‘LE

6.00

0.25

I 15

RE

6.00

0.25

15

LE

2.75

2.50

7

‘RE

2.25

2.00

10

1.50

0.25

180

I.75

0.25

. 180

LE

1.50

m-O.25

90

‘RE

1.50

0.25

90

strabismic L

left:

temporal:

eye determined using a sues

determmed

without

None None

amblyopia:

R

right:

S

superior:

by a sighting of charts

unal~s~s (Finney.

JAssessed b) low-light 40bJectlvely

None

LE

eye: T

None

‘RE

S

right

180

RE PL

1971):

visuoscopy

that lower

ET

FA

esotropta:

I

20’15

Central

7.2

20 I5

Celltrill

7.4

20 15

Ccntr:tl

7.1

20 15

Cer.trZll

7.3

20,13

Central

6.4

20 13

Celllr;tl

6.5

20 I3

Ccotral

7.6

20 I3

CeIlirZd

7.5

former XT

amblyopc; exotropia:

N

normal:

( 1

D

intermittent:

diopter: HYPER

None None None NOW None

&

prism

diopter;

hypertropia:

ALT

I.E

left eye: :titern;tting:

inferior.

test. controlled test limit

(Lawwill.

using heterodynamtc

1966):

for contour

interaction

(From

(‘I cl/.. 1963):

the W,,

threshold

WBS determined

using

is 20 13. the ume-averaged

retinorcopy

(Haynes.

hibited maximum accommodative responses over the middle spatial frequency range, with decrease of accommodative responses toward the tonic accommodative level for the lower and higher spatial frequencies as the stimulus effectiveness (i.e. optimal contrast gradient) decreased. The accommodative response spatial frequency profile was similar in each eye. Other subjects (4N-6N) showed similar response profiles except at the highest spatial frequency: here the accommodative response once again increased, frequently exceeding the accommodative level found over the middle and low spatial frequencies. A third normal variation was found in one subject (7N). The accommodative response was minimum for the lowest spatial frequency, maximum for the highest spatial frequency. and high but relatively constant over the middle spatial

position

of the fovea was estimated

over a 30 set period,

19601.

frequency range. The accommodative response profile was similar in each eye but reduced by 0.5-l .O D in the non-domjnant eye. Responses in two other subjects (8N. 9N) showed little evidence of spatial frequency specificity. especially in the non-dominant eye where the accommodative responses simply appeared to exhibit random variation about the tonic accommodative level; thus, the sinusoidal gratings proved to be a poor accommodative stimulus in each eye of these two subjects. Further, these two subjects’ subjective evaluation of the stimuius was consistent with the low accommodative measures: they saw either a uniform green field or a very low contrast sinusoidal grating. which is to be expected with several diopters of retinal defocus present. The results in amblyopes are presented in Fig. 3.

1588

KENNETH J. CIIJFFREUA and STEQU C. Ho~oD.~

4N 20/13

-AS -AS

-

(0)

-_(A1

6t-

2N

-AS

5N

EN

-AS

-AS

20113

20113

(A) -_(A) --I,)

c

-_(o) -(A)

t05

I 2 4

8

16

Spatial

05

I 2

frequency

4

6

(c

16

/deg

05

12

4

6

16

)

Fig. 2. Accommodation as a function of spatial frequency in 9 normal (N) subjects. Symbols: nondominant eye, A dominant eye. and AS accommodative stimulus. For Figs 2- 5. data values represent X + 1 SEM, and visual acuity of the dominant eye of normals and the non-dominant eye of amblyopes and strabismics is denoted by the Snellen fraction : for Figs 2-4. the arrows indicate the overage accommodative response to a uniform green field. i.e. the tonic accommodation level.

The dominant and amblyopic eyes of all subjects demonstrated spatial frequency specificity, with variation in response patterns similar to those found in our normal subjects, and accommodative responses falling off toward the tonic accommodative level as stimulus effectiveness decreased. However, with only one exception (2A). accommodative responses in the amblyopic eyes were generally reduced, frequently bq several diopters, as compared to the fellow dominant eye. over the tested spatial frequency spectrum. These reduced accommodative responses were found in both anisometropic (5A, 6A) and strabismic (lA4A. 7A, FA) amblyopes. There was no consistent difference in accommodative response variability between the amblyopic and fellow dominant eye (except for subjects 3A and 4A), although all subjects reported the task to be more difficult with the amblyopic eye. There was a difference in the general shape of the accommodative response profile between the amblyopic and fellow dominant eye in many of the subjects (2A. 4A--6A. FA). The results in strabismics without amblyopia are presented in Fig. 4. Subject 1s exhibited spatial frequency specificity in the dominant eye. with the accommodation response maximum at the middle spatial frequencies and reduced for lower and higher spatial frequencies, similar to that found in many of ournormal subjects. However. there was little evidence of spatial frequency specificity in the non-dominant

eye. with accommodation responses showing random variability about the tonic accommodation level. suggesting that the sinusoidal gratings provided a poor stimulus to accommodation. Results in a second subject (2s) were similar. although accommodative responses were markedly reduced in each eye. Contrast sensitivity functions in the amblyopic and formerly amblyopic subjects, as well as in a representative normal and strabismic subject, are presented in Fig. 5. Visually-normal subjects exhibited normal contrast sensitivity functions. peaking over the middle range and showing reduced sensitivity at lower and higher spatial frequencies; further, overall response profiles were similar in each eye of a subject. All amblyopes had normal contrast sensitivity functions in the dominant eye: however, in the amblyopic eye, the contrast sensitivity function was reduced either over theentire spatial frequency spectrum (IA. 4A. FA) or just over the middle and high spatial frequencies (.5AP7A), as has been previously reported (Hess and Howell. 1977). In subject 2A. the contrast sensitivity function in each eye was similar except at the higher spatial frequencies where the amblyopic eye exhibited slightly increased sensitivity. Ofinterest was our finding that this subject’s static accommodative responses, for sinusoidal gratings (Fig. 3) as well as for broadband stimuli (i.e. a Snellen chart), were more accurate and greater with the amblyopic than with the fellow domi-

Spatial frequency dependence of accommodative

6-AS

1589

responses

Is 2-o/13 -AS

54-

-AS

(0)

- AS (not

3

2-

p

I

L

;‘uz ’

(0) --(A)

$ Y a

-_(b) I

I

I I

I

I

2s

6

2 L

-(o)

20119 -AS

5 4 3 I

1

-AS 0.5 Spottal

-AS

-AS

-(Ill

I 2

4

frequency

8

16

f c ldeg I

Fig. 4. Accommodation as a function of spatial frequency in 2 subjects having strabismus without amblyopia (Sk Symbols: t?= non-dominant eye, a = dominant eye. and

AS

accommodative

stimulus.

-CO)

Spatial

frequency ic/deg

1

Fig. 3. Accommodation as a function of spatial frequency in 7 amblyopic (A) subjects and 1 formerly amblyopia (FA) subject. Symbols: C; = amblyopic eye, a = dominant eye. and AS =accommodative stimulus. For subject 6A, open squares represent accommodative responses ofthe dominant eye when AS -6 D, which equalled his acconlmodative amplitude.

nant eye. Subject 3A exhibited similar contrast sensitivity functions (Fig. 5) in each eye: however, accommodative responses to the sinusoidal gratings wereconsistently reduced in the amblyopic eye (Fig. 3). The contrast sensitivity function in each eye of subject 2s was depressed. especially in the non-dominant eye: however, relatively flat accommodative response spatial frequency profiles were measured in each eye. with random variation about the tonic accommodation level.

The results of our investigation clearly demonstrate the spatial frequency dependence of accommodative responses in amblyopic eyes. Similar response dependence was found in the dominant eye of our strabismics without amblyopia. as well as in our. and others (Heath. 1956: Phillips, 1974: Charman and Tucker.

1977; Charman and Heron, 1979; Owens, 1980; Bour, i 98 I ), visually-normal individuals. Absence of an appreciable change in mean accommodative response level in the non-dominant eye of the two strabismics without amblyopia and in each eye of the two normal subjects (8N, 9N) does not necessarily imply lack of spatial frequency dependence, but rather failure of the gratin8 to provide an adequate accommodative stimulus. resulting in random variability about the tonic accommodative level; if the grating provided an adequate accommodative stimulus in the absence of spatial frequency dependence, one would expect the relatively stable mean ac~omm~ative response level to be somewhat higher than the tonic or resting level of accommodation. Thus, when the sinusoidal gratings provided an adequate accommodative stimulus, dependence of the accommodative response on the spatial frequency com~sition of the target appeared to be a general phenomenon, occurring even in amblyopic individuals who have had prolonged periods of abnormal visual experience due to the presence of strabismus. resulting in abnormal binocular interactions (Burian and von Noorden. 1974). and/or anisometropia, resulting in monocular contrast deprivation (6radley and Freeman, 1980). Presence of such spatial frequency dependence of the accommodative responsesuggests that the amblyopic eye visual sensory system contains the normal complement of spatial frequency channels, at least over the spatial frequency spectrum tested. However, the response output and/or the number of neurons per channel may be reduced, and this is reflected in the reduced accommodative

1590

KENNETH

20/13

4N

J.

CIUFFREDAand STEVENC.

20151

5A

6A

20/52

17A

_2O/llS

100

IO

20/l?

FA

44, I00

IO

20/49

I,

05

I

20119

2s

05 I 3 3 6 II 23 Sparta1 frequency Cc/deg

6 II 23

)

Fig. 5. Contrast sensitivity function in amblyopic md formerly amblyopic subjects and in representative normal and strabismic

subjects. Symbols: 2 non-domin~lnt eye and A dominant eye.

responses typically found in the amblyopic eyes. This reduction ofaccommodative response magnitude in the amblyopic eye thus suggests a gain loss in the accommodation system (Hung cl ul., 1981, 1983; Ciulfreda and Kenyon, 1983; Ciuffreda CI ul.. 1983). Furthermore. available evidence strongly suggests a sensory rather than motor or peripheral defect to account for the reduced accommodative responses. First. presence of normal accommodative function in the dominant eye of amblyopes suggests that the motor controller is unaffected, and this is consistent with other investigations involving sensory-motor systems in amblyopes (Mackensen, 1958; Ciuffreda et N/.. 1978a.b). Second, presence of normal consensual but not direct-driven accommodative amplitude in amblyopic eyes (Hokoda and Ciuffreda. 1982) suggests that the site of the accommodative deficit resides neither in the motor controller nor in the peripheral accommodative apparatus. Such a sensory loss might be attributed to decreased responsiveness of the individual neural elements in the amblyopic eye’s afferent visual pathways involved in processing spatial contrast information.

Horct~a

Electrophysiological studies in visually-deprived animals are needed to determmr the site(s) of such abnormal function in the visual patl1aay.s and to characterize the response properties of the mvolved neurons. Our experimental findings using simple sinusoidal. high contrast gratings are consistent with recent expenmental (Ciuffreda and Kenyon. 1Y8i : Ciuffreda CI r/l., 1983) and modeling (Hung cl cl/.. 1981. 1983) studies which also showed reduced static accommodative responses in amblyopic eyes to high contrast. hro~~d~lnd stimuli as found using Snellen letters. In these previous studies (CiulTreda and Kenyon. I083 : Ciuflreda c’l I//. 1983: Hung cl NI.. 1981. 1983) experrmental determination of three model parameter values was sufficient to describe quantitatively the static accommodation system of the amblyopic eye. The three model parameters included accommodative controller gain. depth of focus. and tonic accommodation. It was demons-

trated that accommodative controller gain. which determines most of the accommodative response magnitude at high stimulus levels (Hung and Semmlow, 1980) as used in the present experiment was primarily responsible for the decreased accommodative responses found, as it was typically reduced in the amblyopic eye (Ciuffreda and Kenyon. 1983 : Ciuffreda c’f ol., 1983 : Hung ef N/.. lY81, lY83). Tonic accommodation was similar in the amblyopic and fcllo\+ dominant eye, and the increased depth of focus generally found in the amblyopic eye was insuthcicnt to account for most of the accommodative loss (Ciuffreda et r/I.. 1983; Hung cl/ r/l.. lY81. 1983). In the present experiment. reduced accommodative controller gain in the amblyopic eye would he reflected by reduced accommodative responses over the entire spatial frequency range. In general, this appeared to be true. For example. in subject 3A. the ~tccomtllodati~e response profile was similar in the two eyes hut depressed in the amblyopic eye. However. there were some exceptions. For example. in subject 2A. the accommodative responses in the amblyopic eye were equal to or greater than that found in the fcllob dominant eye; however. this was consistent with Ills atypical accommodative performance using broadband stimuli. where the accommodative responses wcrc again slightly greater over a full range of target vergences in the amblyopic eye. Two other factors may act to reduce, to a small extent. stimulus effectiveness of the sinusoidal gratings and in turn reduce the accommodative response in amblyopic eyes (Ciuffiedn V/ trl., 1983). It is well documented that amblyopiceyesfrequentlyexhibit eccentric (Brock and Givner. 1952) and unsteady (Ciuffreda cv al.. 1979b; Stark PI ~1.. 1982) fixation. Since the majority of amblyopes tested had 0.75 dep of eccentric fixation, and a spatially redundant target configuration was used, it appears unlikely that the eccentricity factor contributed signifcantly to the reduced accommodative responses. with any reduction due solely to asymmetry (with respect to the fovea) in

Spa&l frequency dependence of accommodative response5 stimulation of the near retinal periphery by the grating. Further. recent experimental findings by K&hen et al. (1981) and Ciuffreda e/ al. (1983) demonstrate only a minor effect (-0.25 D) of small eccentric fixation on static accommodative responses to broadband stimuli in amblyopic eyes. and this trend is consistent with findings in normal subjects showing a decrease in closed-loop gain in the accommodation system with increasing stimulus eccentricity (Phillips. 1974). The second factor is unsteady fixation. This includes the presence of saccadic intrusions (CiulTreda er nl., 1979a). increased drift amplitude (Ciuffreda et al., 1980). increased drift velocity (Ciuffreda PI al., 1980). and nystagmus (Ciuffreda cr al., 1979b). Since saccadic intrusions are generally less than I.0 deg in amplitude and only briefly (-200 msec) displace the eccentric fixation locus from the central target region, they would not adversely affect the accommodative response. In fact, saccadic intrusions may be beneficial by providing temporal modulation of the grating retinal image and thus counteract the abnormally-rapid retinal adaptation (Lawwill, 1968) and resultant target fading (Lawwill. 1968; Hess er ul., 1978: Ciuffreda ef al.. 1979a: Sireteanu and Fronius, 1981) frequently found in amblyopic eyes. Such intermittent retinal-image fading, if pronounced and lasting a few seconds, would act to reduce the mean accommodative response level. due to reduced perceived target contrast. and increase response variability; further, for the same reasons. fading would act to produce slightly elevated and more variable estimates of the contrast sensitivity function. especially at the higher spatial frequencies. The increased drift amplitude and nystagmus slow-phase amplitude would result in reduced and variable accommodative responses as the grating moves into the near retinal periphery with its decreasing cone population. Lastly, the increased drift velocity and nystagmus slow-phase velocity may act to “smear” the retinal image and reduce its stimulus effectiveness by reducing retinal image contrast, especially at the high spatial frequencies (Ciuffreda et ul., 1983). Such a notion is consistent with a recent theory of accommodation control in which the high spatial frequency target components are believed necessary to establish accurate accommodation (Charman and Tucker, 1977). Our results demonstrate the diversity of possible accommodative responses to simple sinusoidal gratings. As detailed by Owens (1980). and Ciuffreda and Kenyon (1983), there are presently two primary hypotheses of accommodation control. In the first, called the contrast-control hypothesis, the accommodation system attempts to maximize spatial contrast at the fovea; one would predict accommodative responses to be most accurate over the midrange ( 5 4 c/deg) of spatial frequencies in which visual sensitivity to detection of contrast is greatest. According to the fine-focus-control hypothesis. defocus effects are most evident for the high spatial frequencies; one would predict accommodative responses to be most accurate for the high rather than midrange or low spatial

1591

frequencies. With the exception of Charman and Tucker’s (1977) investigation. all other studies in normal subjects support the contrast-control hypothesis. Our findings were mixed. In the normal subjects. the results of some (IN-3N) support the contrast-control hypothesis. some (7N) support the fine-focus-control hypothesis. and some (4N-6N) support, in part. both hypotheses. In the dominant eye of the amblyopic subjects. some responses (IA. 2A. 4A. 6A) support thecontrast-control hypothesis. some (3A. FA) support the fine-focus-control hypothesis. and some (5A. 7A) support. in part. both hypotheses. In the amblyopic eye of our subjects. the results of some (IA. SA. FA) support the contrast-control hypothesis. some (2A. 3A) support the fine-focus-control hypothesis, and some (4A. 6A, 7A) support. in part. both hypotheses. However, the irregularities in high spatial frequency accommodative responses found in some subjects. which provide support for both hypotheses. may be the result of low contrast. spuriously-resolved images being accommodated on when significant retinal defocus was present (Charman and Tucker. 1977: Smith, 1982). Of particular interest was the finding in some amblyopes (?A, 4A-6A. FA) of accommodative response profiles in each eye which support a different theory. In addition, in one amblyepic subject (7A) following seven months of intensive orthoptic therapy which included accommodation training (Liu et crl.. 1979) and resulted in improved visual acuity in the amblyopic eye (20/120 to 20 75). the accommodative response profile in the amblyopic eye now clearly supported the fine-focus-control rather than the contrast-control hypothesis (as it did pretherapy). Maximum accommodative response was now at 16.0 cideg. and there was a significant increase over the pre-therapy results with respect to both response magnitude and high spatial frequency range which elicited an accommodative response. It has been speculated by Owens (1980) that differences in accommodative response profiles to simple sinusoidal gratings reflected differences in instruction set. Owens (1980), whose results support the contrastcontrol hypothesis, instructed subjects to “focus on all targets naturally, without straining”, and thus it was believed that only “reflex” accommodation was activated. In contrast. Charman and Tucker (1977), whose results support the fine-focus-control hypothesis, instructed subjects to obtain the “best possible” focus, and it was suggested (Owens. 1980) that both reflex and voluntary accommodation were now activated. Such a combined effect might represent the best possible performance of the accommodation system. and accurate accommodative responses to high spatial frequency gratings would now be possible. In our experiment. all subjects were instructed to look at the gratings with the same effort as expended while reading a book. Seeing a grating, but without undue “straining”, was requested. However, with this instruction, being more similar to that used by Owens (1980) than by Charman and Tucker (1977), our results in a

1592

KLXSETH J. CIUFFREVAand STAVESc‘.

relatively large group of naive normal and amblyopic subjects resulted in spatial frequency accommodative response profiles that could support either or a combined hypothesis. This suggests that reflex. voluntary,. and higher-levjel perceptual aspects (Hokoda and Cmlfreda. 1983) of accommodation may interplay in a complex. and perhaps somewhat uncontrollable (for the experimenter) manner. However. this may not be too unexpected considering the stimuli used. Singlesinusoidal gratings provide a rather “unnatural” stimulus. as we typically encounter objects having fairly broadband spatial frequency spectra in everyday viewing. Retinal defocus of a.simple smusoidal grating does not produce a blurred retinal image. as would be the case for broadband stimuli. but rather a lower contrast image. A subject’s attempt to accommodate “appropriately” on such a target would almost. of necessity. demand some degree of voluntary accommodation. until some acceptable level of perceived contrast was attained. and such dynamic changes of accommodation to “search” for the target have been demonstrated (Phillips. 1974: Bour. 1981). When the grating spatial frequency is increased, perhaps now becoming more difficult to see initially and to maintain accurate focus, either of two strategies may be adopted. One could decrease the accommodative effort, resulting in reduced accommodative response and perception of a low contrast grating. or one can increase the accommodative effort (probably by voluntary accommodation). resulting in increased accommodative response and perception of a high contrast grating. Such increased accommodative effort without the sensation of “straining” to see the grating might not be unexpected in the relatively young population tested, who typically had several more diopters of accommodative response remaining (i.e. accommodative amplitude) than either required by the stimulus magnitude or actually expended to see a high-contrast grating. More detailed investigations of the specific control mechanisms of accommodation, especially with respect to the perceptual and voluntary components, are required to understand the neurologic control of the accommodation system and the weightmg assigned to its multi-input stimuli. Lastly. the relationship between acconlmodation and contrast perception must be considered. Implicit in the contrast-control hypothesis is the necessity of close agreement between the spatial frequency region producing the maximum accommodative response and maximum contrast sensitivity ( -4csdeg). In 3 of the 4 subjects tested by Owens (1980). there wpas significant correlation between these tvvo parameters. with peak responses occurring as predicted in the midrange of spatial frequencies. His findings provided some support for the contrast-control hypotheses. at least in normal subjects. Although threshold contrast perception was not assessed in most of our normal subjects. our accommodative response spatial frequency prolilcs. as discussed earlier, do not provide strong support for the contrast-control hypothesisat the exclusion ofall other

HOKOIU

possibilities. However. exploration of the spanal frequency relationship between accommodation and threshold contrast perception IS possible in the amblyoplc subjects. Although correlation analysis cannot be performed due to ditferences m spatial frequencies used in each test. signiticant departures from that predicted by the contrast-control hypothesis were evident in the majority of amblyoptc subjects t 1A. 3A. 5A. 7A. and FA). For example. in subject 3A. the contrast sensitivity function in each eye was approximately equal. yet accommodative responses in the amblyopic eye were reduced over the entire spattal frequency range tested: further. accommodative responses increased as contrast sensitivity decreased with increasing spatial frequency. In subjects .5A and 7A. the contrast sensitivity function in each eye was approximately equal at the lower spatial frequencies. yet accommodative responses were reduced in the amblyopic eye over the entire spatial frequency range tested. In subject IA. the contrast sensitivity function in each eye was approximately equal at the higher spatial frequencies. yet accommodative responses were markedly depressed over most of the range in the amblyopic eye. In the dominant eye of subject FA. contrast sensitivity decreased but accommodative res-onse increased with increasing spatial frequency. IUS. the contrast sensitivity function data in the nblyopic subjects supports our earlier statement that ,ne accommodative response spatial frequency profiles showed significant diversity and did not provide strong support for the contrast-control hypothesis. Hovvever. perhaps under the rather passive viewing conditions used by Owens (1980). the relationship between threshold contrast perception and accommodation may be stronger. With respect to suprathreshold contrast perception in the amblyopic eye. one might expect that the reduced accommodative responses were due to reduced perceived contrast. as accommodation declines as target contrast is decreased in normal observers (Heath, 1956: Phillips. 1974: Bour. 1981). In 4 of the amblyopic subjects (3Am~6A). II contrast matching experiment (Hess and Bradley. 1980) was performed. Interocular comparison of percctved contrast was performed using a lower contrast ( -3o”,,). low or moderate ( - l-6 cdeg) sinusoidal grating with either: (I ) low target vergence ( -2 D) so that the accommodative stimulus and response were similar (Ciuffreda (‘I trl., 1983). thus minimizing the effects of retinal defocus with its consequent reduction in grating retinal image contrast, or (2) high target v’ergence ( -5 or 6 D) as used in our experiments. In this limited group of amblyopes (including both strabismic and anisometropic amblyopes). the grating contrast appeared to be equal or only slightly depressed ( IO”,,) in the amblyopic eye ;IS compared to the fellow dominant eye. This finding provides additional support for our belief that reduced accommodative responses in the amblyopic eye are due primarily to reduced accommodative controller gain in the sensory pathways involved

Spatial

frequency

dependence

in the control of accommodation in the amblyopic eye: thus. the input to either eye alone is approximately the same. yet the output is reduced in the amblyopic eye. A~linolr/c,c/pen~rrlrs-We thank James Kellndorfer for technical assistance and Drs J. Sherman and C. Timpone for allowing us to use their contrast sensitivity apparatus. This work was supported by NIH Grant EY03541 to K.J.C.

REFERENCES

Blakemore C. and Campbe!l F. W. (1969) On the existence of neurones in the human visual system selectively sensitive to the orientation and size of visual images. J. Ph~siol.. Land. 203,237-260. Bour L. J. (1981) The influence of the spatial distribution of a target on the dynamic response and fluctuations of the accommodation of the human eye. b’ision Res. 21, 1287-I 296. Bradley A. and freeman R. (1981) Contrast sensitivity in anisometropic amblyopia. Invest. Ophthal. visual Sri. 21, 467476.

Brock f. W. and Givner 1. (1952) Fixation anomalies in amblyopia. Arc/u Ophthal. 47, 775-786. Burian H. M. and von Noorden G. K. (1974) Binocular Vision and Ocular Motility. Mosby. St Louis. MO. Charman W. N. and Tucker J. (1977) Dependence of accommodation response on the spatial frequency spectrum of the observed object. Vision Res. 17, 129-139. Charman W. N. and Heron G. (1979) Spatial frequency and the dynamics of the accommodation response. Optica Acta 26. 217-228.

Ciuffreda K. J. and Kenyon R. V. (1983) Accommodative vergence and accommodation in normals. amblyopes. and strabismics. In I ‘ergetrce .Ej,e Movements: Basic and Clinical Aspects (Edited by Schor C. M. and Ciuffreda K. J.,), pp. 101-173. Butterworth. London. Ciuffreda K. J.. Kenyon R. V. and Stark L. (1978a) Processing delays in amblyopic eyes: evidence from saccadic latencies. Am. J. Optom. Ph,vsio/. Opt. 55, 187-196. Ciuffreda K. J.. Kenyon R. V. and Stark L. (1978b) Increased saccadic latencies in amblyopic eyes. Invest. Ophthal.

Visual Sci. 17, 697-702.

Ciuffreda K. J., Keny:on R. V. and Stark L. (1979a) Saccadic intrusions m strabismus. .4rchs Ophthal. 97, 1673-1679.

Ciuffreda K. J.. Kenyon R. V. and Stark L. (1979b) Fixational eye movements in amblyopia and strabismus. J. Am. Optom. Assoc. 50, l251- 1258. Ciuffreda K. J., Kenyon R. V. and Stark L. (1980) Increased drift in amblyopic eyes. Br. J. Ophthal. 64. 7-14. CiuKreda K. J.. Hokoda S. C.. Hung G. K.. Semmlow J. L. and Selenow A. (1983) Static aspects of accommodation in human amblyopia. Am. J. Optom. Physiol. Optics 60, 43-9.

Costenbader F.. Baier D. and McPhail A. (1948) Vision in strabismus. Archs Ophthal. 40. 438-453. Finney D. J. (1971) Prohit .4~tr!,xi.v. 3rd edn. Cambridge Univ. Press. Flom M. C.. Weymouth F. W. and Kahneman D. (1963) Visual resolution and contour interaction. J. opt. SK. Am. 53. 1026-1032. Green D. G. and Campbell F. W. (1965) Etfects of focus on the visual response to a sinusoidally modulated spatial stimulus. J. opt. Sot. Am. 55, I 154. I 157. Hartinger H. (1951) Uber ein neues refraktometer. presented at : 57th Mwt. Dwsch Optisch Gose//schuft . pp. 105-l 08. Haynes H. M. (1960) Clinical observations with dynamic retinoscopy. Optom. W’k!,, 51, 224332246.

of accommodative

1593

responses

Heath G. G. (1956) The influence of visual acuity on accommodative responses of the eye. Am. J. Optom. a&.

Am. Acad.

Optom.

33, 513-524.

Hess R. F. and Howell E. f. (1977) The threshold contrast sensitivity function in strabismic amblyopia: evidence for a two-type classification. Vision Rex. 17. 1049-10.55. Hess R. f. and Bradley A. (1980) Contrast perception above threshold is only minimally impaired in human amblyopia. Nature 287.463464. Hess R. F.. Campbell F. W. and Greenhulgh T. (1978) On the nature of the neural abnormality in human amblyopia : neural aberrations and neural sensitivity loss. PHwgcrv Arch. 377.201-207. Hokoda S. C. and Ciuffreda K. J. (1982) Measurement of Opht/~trl. accommodative amplitude in amblyopia. Pltr.riol. Opt. 2, 2055212. Hokoda S. C. and Ciuffreda K. J. (1983) Theoretical and clinical importance of proximal vergence and accommodation. In I’ergettce EFe Movements: Basic trnd Clittkrl Aspects (edited by Schor C. M. and Ciuffreda K. J.). pp. 75-97. Butterworth, London. Hung G. K. and Semmlow J. L. (1980) Static behavior of accommodation and vergence: Computer simulation ol an interactive dual-feedback system. IEEE Trans. Biomed. Engng BME-27,439-447. Hung G. K.. CiuITreda K. J. and Semmlow J. L. (1981) Modeling of human near response disorders. Proc,. 9th .4nn. Narrheast Biomed. Engn~Conf.. Rutgers University. pp. 192-197. Pereamon Press. Oxford. Hung G. K.. CiuffFeda K. J., Semmlow J. L. and Hokoda S. C. (1983) Model of static accommodative behavior in human amblyopia. IEEE Trans. Biomed. EII~II~. In press. Kenyon R. V.. Ciuffreda K. J. and Stark L. (1980) Unexpected role for normal accommodative vergence in strabismus and amblyopia. Am. J. Optom. P/t,&,/. Opt. 57,566-577.

Kenyon R. V.. Ciuffreda K. J. and Stark L. (1981) Dynamic vergence eye movements in strabismus and amblyopia: asymmetric vergence. Br. J. Ophthal. 65, 167-l 76. Kirschen D. G.. Kendall J. H. and Riesen K. S. (1981) An evaluation of the accommodative response in amblyopic eyes. Ant. J. Optom. Ph~siol. Opt. 58. 597-602. Lawwill T. (1966) The fixation pattern of the light-adapted and dark-adapted amblyopic eye. Ant. J. Ophtkrl. 61. 1416-1419. Lawwill T. (I 968) Local adaptation in functional amblyopia. Am. J. Ophthal.

65,903-906.

Liu J. S.. Lee M.. Jane J., Ciuffreda K. J.. Wong J. H.. Grisham D. and Stark L. (1979) Obiective assessment of accommodation orthoptics. 1. Dynamic insufficiency. Am. J. Optom.

Phvsiol.

Ont 56.285-294.

Mackensen. G. (1958) Reaktions~eitmessungen hei amblyopie. Albrec~ht v. Gra& arch. OphthaI. 159, 636 647. Otto J. and Graemiger A. (1967) Uber Unzweckmabige Akkommodation Amplyoper Auaenmit Exrentrischer Fixation. Alhrecht v. Grae/i,.s arc/~ h’lin. e~p. Ophthtrl. 173,125-140. Otto J. and Safra D. (1976) Methods and Results of Quantitative Determination of Accommodation in Amblyopia and Strabismus. In Orthoptics: Past. Preserrt. Future (Edited by Moore S., Mein J. and Stockbridge L.), pp. 45-58. Stratton, New York. Owens D. A. (1980) A comparison of accommodative responsiveness and contrast sensitivity for sinusoidal gratings. Vision Res. 20, 159-167. Phillips S. R. (1974) Ocular neurological control systems: accommodation and the near response triad. Ph.D. Dissertation, Univ. of California, Berkeley. Sireteanu R. and Fronius M. (1981) Nasal-temporal asymmetries in human amblyopia: consequences of long-term interocular suppression. Vision Res. 21, 105551063. Smith G. (1982) Ocular defocus, spurious resolution and contrast reversal. Ophthal. Physiol. Opt. 2, 5-23.

IS94

KENNETHJ. CIUFFREDAand STEVENC. H~UODA

Stark L., Ciuffreda K. J. and Renyon R. V. (1982) Abnormal eye movements in strabismus and amblyopia. In Furwtiotlal Basis qf Ocular Motility Disorders (Edited by Lennerstrand G.. Zee D. and Keller E.), pp. 71-82. Pergamon Press. Oxford.

Urist M. J. (1959) Primary and secondary deviation m comitant squint. Am. J. Ophrho/. 48.647-656. Wood 1. C. J. and Tomlinson A. (1974) The accommodative response in amblyopia. Am. .I. Oprom. f%,t~io/. Opr. S&243-247.