Contrast Detection and Orientation Discrimination Thresholds Associated with Meridional Amblyopia

Pergamon

VisionRes., Vol.37,No. 11,pp. 1451–1457, 1997 @1997ElsevierScienceLtd.All rights resemed

PII: S0042-6989(96)00290-8

Printedin GreatBritain 0042-6989W7 $17.00+ 0,00

Contrast Detection and Orientation Discrimination Thresholds Associated with Meridional Amblyopia R. ST. JOHN* Received

18 April 1996; in revised form 29 July 1996

Humans who have astigmatism resulting in meridional amblyopia exhibit deficits in performing visual tasks at or near detection thresholds. However, there is mounting evidence supporting the idea that performance at threshold may not reliably predict visual capabilities at supra-threshold levels of stimulation. In this study the threshold and supra-threshold performance of six meridional amblyopes were compared. A difference in the pattern of oblique effects was observed between contrast detection thresholds and supra-threshold orientation discriminations. This suggests there exists an independence between populations of neurons subserving contrast detection and the discrimination of visual stimuli. Meridional amblyopia may primarily involve a degradation in those mechanisms subserving visual contrast detection. Populations of cells subserving suprathreshold abilities such as orientation discrimination remain relatively unaffected in humans exhibiting meridional amblyopia. 01997 Elsevier Science Ltd.

Meridionalamblyopia Astigmatism Contrastdetection Orientationanisotropy Obliqueeffect

INTRODUCTION

provide evidence for independentcontrast detection and supra-threshold discrimination. Georgeson & Sullivan (1975)found that in some astigmaticsubjectsa very poor sensitivityat thresholdwas not mirrored by a similar loss in sensitivity for supra-threshold discriminations. In astigmatic eyes contours of different orientationscannot be simultaneouslyfocused on the retina. In cases where the degree of astigmatism is fairly high (e.g. 2 D), the difference in acuity between the two optical meridians may remain even followingcareful refraction and optical correction. This “Meridional Arnblyopia” persists even when acuity is measured using targets generated by diffraction gratings, a method whichbypasses the optics of the eye (Freeman et al., 1972; Mitchell et al., 1967, 1973; Mitchell & Wilkinson, 1974). This suggests that the visual impairments seen in meridional amblyopia as the result of astigmatismare neural rather than optical in origin. The present study examinesthe possibilitythat deficits in contrast detection thresholds in meridional amblyopic subjects are not maintained for supra-thresholdorientation discriminations.

There is mounting evidence supporting the view that performance of the visual system at threshold may not reliably predict performance on a variety of tasks at supra-threshold levels of stimulation. For example, the overall shape of the contrast sensitivity function for gratings of various spatial frequencies is not maintained at supra-thresholdcontrast levels. Georgeson & Sullivan (1975) found that sine wave gratings of equal suprathreshold physical contrast have equal perceived contrast even though their contrast detection thresholds are markedly different. In a similar study, Kulikowski (1976) found that gratingshaving similar supra-threshold contrasts but different spatial frequencies appear to be of equal perceived contrast. Bowne (1990) examined increment thresholds for grating contrast, spatial frequency, orientation and temporal frequency. Contrast increment discrimination was seen to improve at higher levels of target contrast, while different patterns of change were found for spatial frequency, orientation, and temporal frequency discriminations.The results suggest that there is an independence between contrast discrimination and other supra-thresholddiscriminationtasks. METHOD Studies on astigmatic subjects, whose contrast detection thresholds vary greatly with orientation, may also Subjects Data were collected from six astigmatic subjects with meridional amblyopia, and three control subjects who *Departmentof Military Psychologyand Leadership, Royal Military had normal visual acuity with no history of ocular College, Kingston, Ontario, Canada K7K 5L0 IErnail: Stjohnr @rmc.ca]. abnormalities.All subjects were refracted 3 days or less 1451

1452

R. ST. JOHN TAJ3LE1. Clinical data on astigmatic subjects

250 msec, and then decreased sinusoidally with a total rise and fall time of 700 msec. On each trial the phase of Axes of Age when first greatest blur the grating was set at random from a range of O–27C prescribed in defocused radians with respect to the centre of the target. An Refraction results (corrections) corrections eyes observer was required to indicate by pressing a button whether the target had been detected during the interval. S/1 R –0.25 sph./–505OCyl. X 170 7 yr vertical The staircase procedure required two successive “yes” L –0.50 sph./–00OOC)d. X 178 responses on one staircase for the contrast of the next S/2 R –2.00 sph./–2525 cyl. X 2 10 yr vertical L –1.75 sph./–00OOCyl. X 2 grating to be reduced by 3 db. A single “no” response S/3 R +0.50 sph./–7575 cyl. X 90 9 yr horizontal resulted in an increase in contrast of 3 db. The two L +1.00 sph./–00OOcyl. X 90 staircases were randomly interleaved during a testing S/4 R +1 .50 sph./+3.5O cyl. X 180 8.5 yr vertical session and ran independently. Each session was L +1.00 sph./+4.00 cyl. X 180 terminated following at least five reversals of the 11 yr S/5 R –0.75 sph./–00OOcyl. X 20 vertical L –0.50 sph./–7575 Cjt. X 26 directionon both staircases.Contrastdetectionthresholds 10 yr S/6 R +1.25 sph./+3.5O cyl. X 165 vertical were computed as the mean of all the reversals on both L +0.75 sph./+3.75 CJd. X 178 staircases. The decision rule applied in this procedure Note: All subjects reported that usual accommodationcorrespondedto provides a threshold estimate at the 70.7% correct point the least blurred orientation. (Levitt, 1965; Wetherill & Levitt, 1964). Each observer participatedin two sessionsprovidingtwo determinations of thresholdfor each orientation,with the final threshold prior to testing, with retinoscopy used to determine the value computed as the mean of the two sessions. best refractive state. The data from the refractionsand the available clinical history of the astigmatic subjects are SUPRA-THRESHOLDORIENTATION outlined in Table 1. The refractions were performed in DISCRIMINATION order to exclude possible errors due to incorrect optical The difference Iimens for perceived orientation were prescriptionsbeing worn duringtesting.All subjectswere naive psychophysicalobservers at the start of this series determined using a modified version of the method of constantsin combinationwith a two alternativetemporal of experiments. forced choice procedure. Gratings were presented in one Apparatus of two 700 msec intervals, one containing a “standard” Sine-wave gratings were produced conventionally, orientation, the other containing a grating at the “test” using a display generator (Innisfree Picasso) operating orientation.The two stimuliwere presented for the same under computer control. Gratings were displayed on a durationsand had the same contrastmodulationsas those Tektronix 608 CRT monitor with a P31 phosphor. Their used in determining contrast detection thresholds. The mean luminance was 20.0 cd/m2 and the stimuli were intervalswere separatedby an additional700 msec, with visible through a 3.5-deg diameter aperture. The ob- the interval in which the test grating was presented servers viewed the patterns from a distance of 118 cm, chosen randomly for each trial. The test orientation on with light head restraint provided by a chinrest. This each trial was selected randomly from a set of seven system allows for control over orientation of the targets orientations ranging symmetrically about the standard. with a resolution of 0.35 deg and a maximum Michelson The range of the test stimuli was determined in a pilot/ contrast of 0.60 (60%). Contrast calibration was practice set of trials and encompassed the 5–95% performed using a photometer (Model S2 HAGNER frequency of seeing points for each eye for each subject. The task of the observer was to choose which of the UniversalPhotometer)having a circular apertureof 1 deg two intervals contained the grating tilted most counterin diameter. clockwise by depressing one of two buttons. Depression of a third button initiated the next trial resulting in a selfCONTRASTDETECTION THRESHOLDS paced experimental session. Each session contained 20 Contrast detection thresholds were obtained for each presentations of each target orientation paired with the eye in order to evaluate the effects of the meridional standard orientation,resulting in a total of 140 trials per amblyopia on detection of gratings of 10.0 cycles per session. Experimental sessions were run with standard degree at orientations of O,22,45,67,90, 112, 135, 157 orientations of O, 22, 45, 67, 90, 112, 135, 15’7,and and 180 deg. (Note: orientations are specified according 180 deg. The orientation discrimination threshold was deterto ophthalmic convention, where Odeg = horizontal and values increase with rotation in a counterclockwise mined by fittinga cumulativenormal probabilityfunction to the frequency of seeing data using an iterative direction.) Thresholds for contrast detection were determined maximum likelihood procedure via a method of Probits using a double randomly interleaved staircase procedure (Finney, 1971). The cycle of iterations was halted when (Cornsweet, 1962). Each trial consisted of a single the estimateof the slopeon a current cycle failed to differ interval containing a target grating whose contrast was by more than 1.0% of that of the value determined on the increased sinusoidally, maintained at peak contrast for previous iteration. The orientation discrimination

MERIDIONALAMBLYOPIA Subject # 1. R -0.25/-3.50 L -0.50/-4.00

X X

1453

Subject # 2. R -2.00/-4.25 L -1.75/-4.00

170 178

X X

Contrast detection threshold

Contrast detection threshold 8 7

2 2

O Left eye

8

O Left eye

.

7

. Right eye

Right eye

[

[

AQ

o

Q4 ; :3 5 kz

;;/J~

1

o —

I 22

45

/

67

90

112

135

\

I

157

180

1

0

1

22

1

45 /

Orientation of target Orientation discrimination

90

112

1

1

1

135

157

180

\ I Orientation of target

—

Orientation discrimination

8r

O Left eye

7

● Right eye

96

8 7 96 ~s

Z5 34.[

67

0 Left eye ●

Righteye

I

1

[

,.@

o’

1

I

I

I

I

I

I

I

I

o

22

45 /

67

90

112

135

157

180

\

I Orientation of standard

—

FIGURE 1. Figures 1–6: Contrast detection threshold and suprathreshold discrimination data combined from two sessions for meridional amblyopic subjects l&6.Data from refraction are included in top left of figure. L = left eye: R = right eye. Note that contrast detection thresholds are poorest (highest contrast values) for gratings centred around the orientations most blurred in the defocusedleft and right eyes. Supra-threshold orientation discriminations, however, appear to be relatively unaffected by the astigmatisms, and show the “M-shaped” pattern indicative of the normal oblique effect. Data are combinedwith the error bars representing the range of data across the two experimental sessious. Note: in all figureswhere no error bars are shown the range was less than the symbol size.

thresholdwas definedas the reciprocalof the slope of the regression of normalized probability against orientation and represented the difference between 5070 and 84Y0 frequency of seeing points.* (For similar procedures see Heeley, 1991; Heeley & Timney, 1988; Heeley & Buchanan-Smith, 1994.) A Chi-square test was used to ensure that the theoretical error distributiondid not differ significantlyfrom the data. The grating contrast was set at 60Y0for all observers for the orientation discrimination portion of this study. Contrast was defined as: ((L~ax– ~min)l(~rnax + L~i.))*lOO,where L~,Xand L~i. representthe luminance values of the peaks and troughsof the sine-wavegratings. Each observer participated in two sessionsprovidingtwo *I would like to thank Dr. D. Heeley for providingthe algorithmsfor these procedures.

o’

I

1

1

0

22

45

I

1

1

I

90 112 135 / \ I Orientation of standard 67

157 180

FIGURE 2.

determinationsof orientationdiscriminationthresholdfor each orientation, with the final value computed as the mean of the two sessions. RESULTS

The contrast detection thresholds and the orientation discriminationthresholdsare presented in Figs 1–6. Data from the three normal control subjects are combined and presented in Fig. 7. The data from the six astigmatic subjects show clear abnormalities in their contrast detection thresholds. For example, Subject 1 shows poorest contrast detection for targets oriented close to vertical (90 deg). This orientation is approximately orthogonal to the axis of the cylindrical lenses used to correct the astigmatismin the eyes of this subject(left eye 178 deg; right eye 170 deg), and corresponds to the greatest amount of blurring in this subject’s defocused eyes. The astigmatism, even though optically corrected, clearly results in a distortion of the usual “M-shaped” oblique effect seen for normal contrast detection thresholds, In contrast, the data from the orientation discrimination task in this subject show no large elevations of thresholds for vertical gratings. The typical “M-shaped” pattern indicative of a standard oblique effect was found for both eyes. Performance on the discrimination task,

1454

R. ST. JOHN Subject# 3. R +0.50/–3.75X 90 L +1.00/-3,00

X

90

Subject# 4. R +1.50/+3.50 X 180 L +1.00/+4.00X 180

Contrast detection threshold

Contrast detection threshold

O Left eye

87 z :5 o ~4 ~ 23

0 Left eye

8–

.

7

6

~ :5 0 U4 z 83 5 ck2

-

22

‘~ o

—

22

45

/

67

90

112

135

\

I

157

6

-

‘~ 0

180

Right eye

22

45

/

67

90

112

135

\

I

Orientation of target

180

157

Orientation of target Orientation discrimination

Orientation discrimination O Left eye ●

Righteye

‘r

O Left eye

7

.

Right eye

96 =5 %4 s Z3 . g2

J* 1

1

1

1

0

22

45 /

I

1

67

90

112

1

I

1

135

157

180

\ I Orientation of standard

01

1

1

I

1

1

1

1

0

22

45

67

90

112

135

/

1

157

180

\ I Orientation of standard

FIGURE3.

FIGURE4.

unlike the contrast detection task, was best for gratings oriented at 180 and 90 deg and poorest for gratings oriented along the oblique visual axes. A similar pattern of results was seen for the remaining amblyopes. On the contrast detection task, Subject 2 showed deficits for gratings oriented around the vertical axis (90 deg). Subject 3 showed deficitsfor gratings near the horizontal axis (O and 180 deg). Subject 4 showed deficits for gratings around the vertical axis (90 deg). Subject 5 showed deficits for gratings between 112 deg (tilted slightly counter-clockwisefrom vertical), and the 45 deg axis. Subject6 showed deficitsfor gratingsaround the vertical axis (90 deg) and for gratingsat 67 deg (tilted slightly clockwise from vertical). In all cases, performance on the orientation discrimination produced very different results. An “M-shaped” pattern typical of a normal oblique effect was seen for orientation discriminations.It is possible that the effects of meridional amblyopia on orientation discrimination are quite small, and may be masked by the variation and magnitudeof the usual oblique effect seen for orientation discrimination.However, performance was not severely degraded or biased by the meridional amblyopia which had produced quite large distortions in the patterns of contrast detection. In the non-astigmaticcontrol subjects higher contrasts were required to detect the oblique (45 and 135 deg) than

to detect horizontal or vertical gratings (see Fig. 7). The results for the orientation discriminations exhibit the normal “M-shaped” oblique effect pattern typical for gratings of 10.0 cycles per degree (Campbell et al., 1966). There is an overall trend in the results for the astigmatic subjects to have higher thresholds than those seen for the normal control subjects. This suggests the possibleexistenceof an orientation-independenteffect of meridional amblyopia producing a general loss in contrast sensitivity and orientation discrimination. Although contrast detection thresholds are greatly distorted by the amblyopia, further testing of contrast thresholds and orientation discriminationsin meridional amblyopes may confirm these orientation-independent effects. DISCUSSION

It has been suggested that meridional amblyopia in humans is related to losses in sensitivity, or decreased numbersof orientationtuned neuronsas a consequenceof anomalous early visual experience imposed by the blurring of the retinal image along a particular visual axis (Cobb & MacDonald, 1978; Daugman, 1983; Freeman & Thibos, 1975;Harwerth et al., 1980;Mitchell et al., 1973;Mitchell, 1980).Supportfor this idea derives from data on animals raised wearing goggles which

MERIDIONALAMBLYOPIA Subject # 5. R –0.75/–4.00

20 L -0.50/-3.75 X 26

1455

Subject # 6. R +1.20/+3.50 X 160 L +0.75/+4.00 X 178 Contrast detection threshold

X

Contrast detection threshold U

Left

●

Right

eye

O Left eye

8

.

eye

7 z ;5 o ~4 ~ V3 z &2

/

/ $ 4

g/

v I

1

I

1

1

(

(

I

I

o

22

45 /

67

90

112

135

157

180

—

\ I Orientation of target

6 /

●

/ /all.

1~

—

o

—

22

Orientation discrimination 0 Lefteye

01

I

1

o

22

—

1

1

45 /

67

1

1

1

1

I

90

112

135

157

180

\ I Orientation of standard

—

Right eye

+ /

45 /

67

90

112

135

\ I Orientation of target

157

180

Orientation discrimination 8r

0 Left eye

o

.

o’

I

1

I

1

1

1

o

22

45 /

67

90

112

—

Right eye

I

135

I

\

157

180

—

Orientation of standard

FIGURE5.

FIGURE 6.

present one eye with stimuli of a single orientation and the orthogonal orientation to the other eye. These conditions have produced shifts in populations of orientation-specific cortical cells whose development depends upon visual experience during rearing (Stryker et al., 1978). Similar effects occur in animals raised with artificial astigmatismproduced by rearing while wearing a cylindrical lens placed over one eye (1300the& Teller, 1982; Cynader & MitchelI, 1977; Freeman & Pettigrew, 1973).Studiesusing stripe-rearingtechniques,where one or both eyes are exposed to environments having visual contours of one orientation, have produced more mixed effects (Carlson et al., 1986; Hirsch & Spinelli, 1970; Muir & Mitchell, 1973; Singer, 1976). They range from those first described by BIakemore & Cooper (1970), where visual cortical cells show strongbiases towardsthe experienced orientation, to the lack of any clear effects found by Stryker & Sherk (1975). In man, the effects of deprivation due to naturally occurring astigmatism are clouded by the fact that although astigmatic individualsusually accommodate to the least blurred focal plane having the best acuity, they are sometimes capable of accommodating to the other most blurred focal plane (Freeman, 1975). Studies of visual evoked potentials in astigmatic subjects have shown that some have a marked asymmetry in the amplitude of evoked potentials correspondingto the eye

meridian having the lowest refractive error. Others, however, show no clear asymmetries or other obvious effects (Fiorentini & Maffei, 1973). In the present study, human astigmatic subjects exhibited deficits in the contrast sensitivity functions correspondingwith the most blurred orientations during visual development. Conversely, no obvious abnormalities were found for supra-threshold orientation discriminations.An inference based upon these findingsis that there exist contrast-dependentvisual capabilities,such as those involved with contrast detection thresholds,which are strongly affected by the form of visual deprivation associated with naturally occurring astigmatism. Contrast-independentcapabilities, such as those involved in supra-thresholdorientation discriminations,may escape most of the effects of deprivationdue to astigmatism.The thresholds for the astigmatic observers did tend to be higher than those found for the normal control subjects across all orientations tested. This may reflect a small global deficitin orientationdiscriminationresultingfrom severe astigmatism.However, this must be considered to be a very tentative notion in light of the small number of subjects (six) tested. It is not the result of differences between the experience of the normal and astigmatic observers on performing the psychophysical tasks. All were initiallynaive observers,and had similar amountsof practice on the psychophysicalprocedures.

1456

R. ST. JOHN NORMALS (Mean values). Contrast detection threshold 0 Left eye

8

7

. Righteye I

o

8 7

22

45 /

67

90

112

135

157

\ I Orientation of target Orientation discrimination

180

—

o Left eye ●

[

o’

1 0

1

1

22

45

/

67

90

112

Righteye

1

1

1

135

157

180

\ I Orientation of standard

FIGURE 7. Contrast detection threshold and supra-threshold discrimination combineddata for three normal control subjects havingno history of astigmatism or other ocular disorders. Note that the typical “M-shaped” pattern is observed for both contrast detection thresholds and for supra-thresholdorientationdiscriminations.Data are combined with the error bars representing the range of data across two experimental sessions for the three subjects.

An anisotropy in orientation tuning of visual cortical cells in normals has often been used to explain oblique effects (de Valois et al., 1982; Mansfield, 1974; Mansfield & Ronner, 1978). In particular, orientation anisotropiesin S-cells in area 17 of the visual cortex have been linked to anisotropiesin line orientationdiscrimination (Orban & Kennedy, 1981,Orban et al., 1984;Vogels & Orban, 1985). Some authorshave argued that different examples or classes of the oblique effect cannot all be linked to orientation anisotropies seen for area 17 cells. Vogels & Orban (1986) suggest that there may exist at least two kinds of oblique effects, the firstthe result of Scell orientation anisotropies,the second involvinghigher order visual processing. Data from psychophysical studies suggests that oblique effects involve orientation anisotropies in cells or systems at various levels in the visual system (Heeley & Buchanan-Smith, 1992a; Heeley & Timney, 1988, 1989). In addition, magnitudes of the oblique effects depend upon the amountof practice an observer has had performing the psychophysical procedures (Mayer, 1983; Vogels & Orban, 1985), the body posture of the subjects(Buchanan-Smith& Heeley, 1993), the length and size of the targets (Tootle &

Berkely, 1983), and the psychophysicalprocedures used (Heeley & Buchanan-Smith,1992b). Supra-threshold orientation discrimination has been examined in subjects having strabismic and anisometropic amblyopia. Skottun et al. (1986) found that impairments in orientation discrimination for sine wave gratings in amblyopic eyes are spatial frequency but not contrastdependent.Orientationdiscriminationthresholds can be normal or abnormal depending upon the spatial frequency content of the target. When using single lines as stimuli, amblyopes show impairments in orientation discrimination which are dependent upon both the orientation and the length of the target lines (Vandenbussche et al., 1984; Vogels et al., 1984). In the present study, visual deprivation restricted to a particular orientationas a result of astigmatismproduced a decline in grating contrast sensitivity, which reflected the axis of the astigmatism in each subject. However, orientation discrimination for gratings of high contrast was spared,with the resultsexhibitinga relativelynormal oblique effect. The main conclusion is that meridional amblyopes show an oblique effect in orientation discriminationthat does not follow the pattern of deficits seen in their contrastdetectionthresholds.Whether or not this effect persists for gratings of various contrasts from threshold to higher values, and is spatial frequency dependent is currently being evaluated. Bowne (1990) showed that target spatial and temporal frequency, and target orientation discrimination in nonastigmatic subjects are not dependent upon absolute target contrasts.Contrast increment thresholds,however, were strongly influencedby absolute target contrast. He suggested that separate peripheral and central noise in visual processing contributes independently at different levels to performance of different psychophysicaltasks. The findings of the current study provide some support for the idea that various levels in the visual system are involved in processing orientation information. In subjects having meridional amblyopia contrast detection for gratings of 10 cpd is strongly impaired, while high contrast orientation discrimination is not. This suggests that the losses in acuity and contrast detection associated with meridional amblyopiamay involve more peripheral contrast-dependentvisual processes. REFERENCES Blakemore, C. & Cooper, G. F. (1970). Development of the brain dependson the visual environment.Nature (London), 228,477-478. Boothe, R. G. & Teller, D. Y. (1982). Meridionalvariations in acuity and CFSSin monkeys (Macaca rrernestrina) reared with externally applied astigmatism. Vision Research, 22, 801–810. Bowne, S. F. (1990). Contrast discrimination cannot explain spatial frequency,orientation or temporal frequency discrimination. Vision Research, 30, 449-461.

Buchanan-Smith,H. M. & Heeley, D. W. (1993). Anisotropic axes in orientation perception are not retinotopically mapped. Perception, 22, 1389–1402.

Campbell,F. W., Kulikowski,J. J. & Lcvinson,J. (1966).The effect of orientation on the visual resolution of gratings. Journal of Physiologyj 187, 427-436.

Carlson, M., Hubel, D. H. & Wiesel, T. N. (1986). Effects of

MERIDIONALAMBLYOPIA monocular exposure to oriented lines on monkey striate cortex. Developmental Brain Researchj 25, 71–81.

Cobb, S. R. & MacDonald, C. F. (1978). Resolution acuity in astigmats: evidence for a critical period in the humanvisual system. British Journal of Physiological Optics, 32, 38&49.

Cornsweet, T. M. (1962). The staircase method in psychophysics. American Journal of Psychology, 75, 485S491.

Cynader,M. & Mitchell, D. M. (1977).Monocularastigmatismeffects on kitten visual cortex development. Nature (Lorrdorr),270, 177– 178.

Daugman, J. D. (1983). Visual plasticity as revealed in the twodimensionalmodulationtransfer functionof a meridionalamblyope. Human Neurobiology, 2, 71–76.

de Valois, R. L., Yund,E. W. & Helper, N. (1982).The orientationand direction selectivity of cells in the macaque visual cortex. Vision Research, 22, 531-544.

Finney, D. J. (1971). Probit analysis, 3rd edn. Cambridge:Cambridge University Press. Fiorentini, A. & Maffei, L. (1973). Evoked potentials in astigmatic subjects. Vision Research, 13, 1781–1783. Freeman, R. D. (1975). Contrast sensitivity in meridional amblyopia. Investigative Ophthalmology, 14, 78–81. Freeman, R. D., Mitchell, D. E. & Millodot, M. (1972).A neural effect of partial deprivation in humans. Science (NY), 168, 869–871.

Freeman, R. D.& Pettigrew,J. (1973).Alteration of visual cortexfrom environmental asymmetries. Nature (London), 264, 360395. Freeman, R. D. & Thibos, L. N. (1975).Contrastsensitivity in humans with abnormal visual experience. Journal of Physiology, 247, 687– 710.

Georgeson, M. A. & Sullivan, G. D. (1975). Contrast constancy: debhrrring human vision by spatial frequency channels. Journal of Physiology, 252, 627–656.

Harwerth, R. S., Smith III, E. L. & Boltz, R. L. (1980). Meridional amblyopiain monkeys.Experimental Brain Research, 39, 331–338. Heeley, D. W. (1991). Spatial frequency difference thresholdsdepend on stimulus area. Spatial Vision, 5, 205–217. Heeley, D. W. & Buchanan-Smith,H. M. (1992a) Directional acuity for drifting plaids. Vision Research, 32, 97–104. Heeley, D. W. & Buchanan-Smith,H. M. (1992b) Orientation acuity estimated with simultaneous and successive procedures. Spatial Visionj 6, 1–10. Heeley, D. W. & Buchanan-Smith, H. M. (1994). Evidence for separate, task dependent noise processes in orientation and size perception. Vision Research, 34, 2059–2069. Heeley, D. W. & Timney, B. (1988). Meridional anisotropies of orientation discrimination for sine wave gratings. Vision Research, 28, 337-344.

Heeley, D. W. & Timney,B. (1989).Spatial frequencydiscriminations at different orientations. Vision Research, 29, 1221–1228. Hirsch, H. V. B. & Spinelli, D. N. (1970). Visual experience modifies distributionof horizontallyand vertically oriented receptive fields in cats. Science, 168, 869–871. Kulikowski,J. J. (1976). Effective contrast constancy and linearity of contrast sensation. Vision Research, 16, 1419–1431. Levitt, H. (1965).Transformedup-down methodsin psychoaccoustics.

1457

Mitchell, D. E., Freeman, R. D., Millodot, M. & Haegerstrom, G. (1973). Meridional amblyopia: evidence for modification of the human visual system by early visual experience. Vision Research, 13>535-558.

Mitchell, D. E., Freeman, R. D. & Westheimer, G. (1967). Effect of orientation on the modulationsensitivity for interference fringes on the retina. Journal of the Optical Society of America, 57, 246-249. Mitchell,D. E. & Wilkinson,F. (1974).The effect of early astigmatism on the visual resolution of gratings. Journal of Physiology, 243, 739-756.

Muir, D. W. & Mitchell, D. M. (1973). Visual resolution and experience: acuity deficits following early selective visual deprivation. Science, 180, 420-422. Orban, G. & Kennedy, H. (1981). The influence of eccentricity on receptive field types and orientationselectivity in areas 17 and 18 of the cat. Brain Research, 208, 203–208. Orban, G., Vandenbussche, E. & Vogels, R. (1984). Meridional variations and other propertiessuggestingthat acuity and orientation discrimination rely on different neural mechanisms. Ophthalmological and Physiological Optics, 4, 89–93.

Singer,W. (1976).Modificationof orientationand direction selectivity of cortical cells in kittens with monocular vision. Brain Research, 118, 460J168.

Skottun, B. C., Bradley, A. & Freeman, R. D. (1986). Orientation discrimination in amblyopia. Investigative Ophthalmology and Visual Science, 27, 532–537.

Stryker, M. P. & Sherk, H. (1975). Modificationof cortical orientation selectivity in the cat by restricted visual experience: a reexamination.Sciencej 190, 904–905. Stryker, M. P., Sherk, H., Leventhal,A. G. & Hirsch, H. V. B. (1978). Physiologicalconsequencesfor the cat’s visual cortex of effectively restricting early visual experiencewith oriented contours.Journal of Neurophysiology, 41, 896–909.

Tootle, J. S. & Berkely, M. A. (1983). Contrast sensitivity for vertically and obliquely oriented gratings as a function of grating area. Vision Research, 23, 907–910. Vandenbussche, E., Vogels, R. & Orban, G. A. (1984). Human orientation discrimination:changes with eccentricity in normal and amblyopicvision. Investigative Ophthalmology and Visual Science, 27, 237–245.

Vogels,R. & Orban,G. A. (1985).The effect of practice on the oblique effect in line orientation judgments. Vision Research, 25, 1679– 1687.

Vogels, R. & Orban, G. A. (1986). Decision factors affecting line orientationjudgments in the method of single stimuli. Perception and Psychophysics, 40, 74-84.

Vogels, R., Orban, G. A. & Vandenbussche, E. (1984). Meridional variations in orientation discrimination in normal and amblyopic vision. Investigative Ophthalmology and Visual Science, 25, 720728.

Wetherill, G. B. & Levitt, H. (1964).Sequentialestimationof pointson a psychometricfunction. British Journal of Mathematical Statistics and Psychology, 18, 1–10.

Journal of the Acoustical Sociep of America, 49, 467+76.

Mansfield, R. J. W. (1974). Neural basis of orientation preference in primates. Science, 186, 1133–1135. Mansfield, R. J. W. & Ronner, S. F. (1978). Orientation anisotropyin the monkey visual cortex. Brain Research, 149, 229–234. Mayer, M. J. (1983).Practice improvesadults’ sensitivityto diagonals. Vision Research, 23, 551–559.

Mitchell, D. E. (1980). The influence of early visual experience on visual perception. In:Harris C. S. (Ed.), Visual coding and adaptability (pp. 1–50). Hillsdale, NJ: Lawrence Erlbaum.

would like to acknowledge the subjects who participated in this study. Their patience and perseverance were remarkable and are appreciated. I would also like to thank Dr Brian Timney and Dr David Heeley, both of whom provided much insight and commentary during the early stages of the research discussed in this paper. This research was supported by Grant 3705-878from the DND Academic Research Program, Department of National Defence, Canada.

Acknowledgements-I