The influences of visibility and anomalous integration process on the perception of global spatial form versus motion in human amblyopia
Do amblyopes demonstrate general irregularities in processes of global image integration? Or are these anomalies stimulus specific? To address these questions we employed directly analogous global-orientation and global-motion stimuli using a method that allows us to factor out any influence of the low-level visibility loss [Simmers, A. J., Ledgeway, T., Hess, R. F., & McGraw, P. V. (2003). Deficits to global motion processing in human amblyopia. Vision Research 43, pp. 729-738]. The combination of orientation and motion coherence thresholds reported here provides comparable psychophysical measures of global processing by spatial-sensitive and motion-sensitive mechanisms in the amblyopic visual system. The results show deficits in both global-orientation and global-motion processing in amblyopia, which appear independent of any low-level visibility loss, but with the most severe deficit affecting the extraction of global motion. This provides evidence for the existence of a dominant temporal processing deficit in amblyopia.
The inﬂuences of visibility and anomalous integration processes
on the perception of global spatial form versus motion in human
Anita J. Simmers
, Tim Ledgeway
, Robert F. Hess
Department of Optometry and Visual Science, Applied Vision Research Centre, The Henry Wellcome Laboratories for
Vision Sciences, City University, London EC1V OHB 22, UK
School of Psychology, University of Nottingham, Nottingham NG7 2RD, UK
Department of Ophthalmology, McGill Vision Research, McGill University, Montreal, Canada H3A 1A1
Received 6 May 2003; received in revised form 29 May 2004
Do amblyopes demonstrate general irregularities in processes of global image integration? Or are these anomalies stimulus
speciﬁc? To address these questions we employed directly analogous global-orientation and global-motion stimuli using a method
that allows us to factor out any inﬂuence of the low-level visibility loss [Simmers, A. J., Ledgeway, T., Hess, R. F., & McGraw, P. V.
(2003). Deﬁcits to global motion processing in human amblyopia. Vision Research 43, pp. 729–738]. The combination of orientation
and motion coherence thresholds reported here provides comparable psychophysical measures of global processing by spatial-
sensitive and motion-sensitive mechanisms in the amblyopic visual system. The results show deﬁcits in both global-orientation
and global-motion processing in amblyopia, which appear independent of any low-level visibility loss, but with the most severe
deﬁcit aﬀecting the extraction of global motion. This provides evidence for the existence of a dominant temporal processing deﬁcit
2004 Elsevier Ltd. All rights reserved.
Keywords: Amblyopia; Contrast; Global; Spatial; Extrastriate
Amblyopia is characterized by distorted representa-
tions of spatial form and over the past 40 years, the site
of the perceptual deﬁcit in human amblyopia has been
the subject of considerable speculation. Although it is
known that the site of the processing deﬁcit is cortical,
rather than retinal, in both humans and animals (for a
current review see Moseley & Fielder, 2002), little is
known about its extent within the cortex.
Amblyopia is thought to reﬂect alterations in the
neuronal properties of V1, including reduced spatial res-
olution, reduced contrast sensitivity and a reduced num-
ber of binocular cells (Chino, Shansky, Jankowski, &
Banser, 1983; Crewther & Crewther, 1990; Eggers &
Blakemore, 1978; Kiorpes, Kiper, OÕKeefe, Cavanaugh,
& Movshon, 1998). Animal models, however suggest
that whi le the known single cell deﬁcit to the striate cor-
tex may be suﬃci ent to characterize the loss in resolu-
tion and contras t sensitivity, it is insuﬃcient to explain
many of the perceptual anomalies reported in amblyopia
(Barrett, Pacey, Bradley, Thibos, & Morrill, 2003;
Kiorpes et al., 1998; Kiorpes & McKee, 1999). This
0042-6989/$ - see front matter 2004 Elsevier Ltd. All rights reserved.
Corresponding author. Address: Department of Optometry and
Visual Science, Applied Vision Research Centre, The Henry Wellcome
Laboratories for Vision Sciences, City University, London EC1V
OHB, UK. Tel.: +44 207 040 0191; fax: +44 207 608 040 8494.
E-mail address: email@example.com (A.J. Simmers).
Vision Research 45 (2005) 449–460
has led to the suggestion that additional deﬁcits may ex-
ist beyond the striate cortex. Modest evidence from both
primate (Movshon et al., 1987) and cat studies (Sch-
roder, Fries, Roelfsema, Singer, & Engel, 2002) support
a disruption in the binocular organization of extrastriate
cortical areas in amblyopia. It is also clear from neuroi-
maging studies that additional cortical deﬁcits are asso-
ciated with visual processing stages (areas) beyond
striate cortex in human amblyopia (Barnes, Hess,
Dumoulin, Achtman, & Pike, 2001; Imamura et al.,
1997; Sireteanu, Tonhausen, Muckili, Zanella, & Singer,
It follows then that in recent years it has become
increasingly apparent that the extent of the deﬁcit in
amblyopia can depend on the visual functi on measured.
The key question is whether higher integrative levels of
visual processing ‘‘inherit’’ abnormalities from lower
levels, or whether additional developmental abnormali-
ties arise in direct consequence of impoverished visual
In visual space it is often necessary to integrate (i.e.
compare, combine or pool) information over an ex-
tended area to derive a reliable and accurate global per-
cept. This means that infor mation about image structure
over extended areas of visual space must be based on the
combined responses of a number of independent, local
inputs. Studies of the integration of motion and orient a-
tion are thought to reﬂect the integrative properties of
neurons in primate extrastriate cortex and in normal
observers such global estimates of mean direction (Kior-
pes, 2003; Newsome & Pare, 1988; Verghese, Wat-
amaniuk, McKee, & Grzywacz, 1999) and orientation
(Dakin, 2001; Dakin & Watt, 1997) can be combined
with great accuracy in the absence of spatial structure.
These results are consistent with both psy chophysical
(Burr, Morrone, & Vaina, 1998) and neurophysiological
evidence (Felleman & Van Es sen, 1987; Gattass, Gross,
& Sandell, 1981) showing increases in receptive ﬁe ld size
at higher stages of visual processing.
Current work has revealed abnormalities in such glo-
bal processing in amblyopia. Form integ ration para-
digms have shown deﬁcits in contour integration
(Chandna, Pennefather, Kovacs, & Norcia, 2001; Hess,
McIIhagga, & Field, 1997; Kovacs, Polat, Pennefather,
Chandna, & Norcia, 2000; Kozma & Kiorpes, 2003),
judgments of circularity (Hess, Wa ng, Demanins, Wil-
kinson, & Wilson, 1999; Jeﬀrey, Wang, & Birch, 2004),
global orientation discrimination (Simmers & Bex,
2004) and the detection of structure in Glass patterns
(Kiorpes, 2003; Lewis et al., 2002). Motion integration
deﬁcits have also been revealed after early visual depri-
vation in humans (Ellemberg, Lewis, Maurer, Brar, &
Brent, 2002; Simmers, Ledgeway, Hess, & McGraw,
2003) and monkeys (Kiorpes, 2003; Tang et al., 1998).
These results certainly imply a far more complex percep-
tual change in amblyopia than would be predicted by
the well established losses in resolution and contrast
sensitivity. Furthermore higher-level capabilites such
as the individuation of elements or ‘‘counting’’ (Sharma,
Levi, & Klein, 2000) the perception of visual illusions
(Popple & Levi, 2000) and more recently, mirror symme-
try (Levi & Saarinen, 2004) have also been reported
abnormal in amblyopia.
Recently, we have reported global-motion processing
deﬁcits in human amblyopia that are unrelated to (inde-
pendent of) the contrast sensitivity deﬁcit, and that these
may be more extensive for contrast-deﬁned than for
luminance-deﬁned stimuli (Simmers et al., 2003). In
that study we employed random-dot-kinematograms
(RDKs) of the type originally developed by Newsome
and Pare (1988). Normal and amblyopic observers were
required to identify the global (overall) direction of im-
age motion carried by a small proportion of dots that
were displaced in a con sistent direction (‘‘signal’’ dis-
placements) amongst spatially interspersed dots whose
displacement directions were stochastic (‘‘noise’’ dis-
placements). As this task necessarily requires the ability
to integrate local motion information across space and
time, we speculated that the site of these deﬁcits must in-
clude the extrastriate cortex, in particular the dorsal
stream. Several visual areas within the dorsal stream,
such as MT/V5 and MST (and their human homo-
logues), appear well suited to mediate global-motion
processing as they contain cells whose receptive ﬁelds ex-
tend over very large regions of space. However it is cur-
rently unknown whether amblyopes demonstrate more
general irregularities in the processing of global image
structure (e.g. motion, orientation, etc.) or exhibit
anomalies that are speciﬁc to encoding global motion
Therefore, as there is now ﬁrm evidence that global-
motion processing is anomalous in amblyopia and be-
cause of the importance of extrastriate processing for
such a task, the present investigation employs a directly
analogous global-orientation task to assess the relative
involvement of form-sensitive mechanisms in the extrac-
tion of the overall image spatial structure in amblyopia.
It is likely that sensitivity to global form and global
motion are mediated by distinct functional processing
mechanisms in the visual system although these do not
necessarily correspond to the gross anatomical separa-
tion of dorsal and ventral streams (Braddick, OÕBrien,
Wattam-Bell, Atkinson, & Turner, 2000). Indeed there
is accumulating evidence of not only convergence of
the magno- and parvocellular inputs (Devalois, Cottaris,
Mahon, Elfar, & Wilson, 2000; DeYoe & Van Essen,
1988; Sawatari & Callawat, 1996; Takeuchi, De Valois,
& Hardy, 2003) but functional interactions between the
neural mechanisms involved in the processing of shape
and motion information about objects (Kourtzi, Bulth-
oﬀ, Erb, & Grodd, 2001). Therefore this dichotom y
may not be as marked as originally thought.
450 A.J. Simmers et al. / Vision Research 45 (2005) 449–460
It has also been suggested that the course of develop-
ment for dorsal stream mechanisms (motion coherence
tasks) may be more protracted than those underlying
ventral stream mechanisms (form coherence tasks)
(Braddick, Atkinson, & Wattam-Bell, 2003; Gunn
et al., 2002). It is therefore possible that the relative eﬃ-
cacy of each may be diﬀerentially aﬀected during visual
development. However if the impoverished ability of
amblyopes (compared with normals) to extract global-
motion direction is indicative of a more general and
nonspeciﬁc deﬁcit inﬂuencing the ﬁdelity of processes
which serve to integrate local visual cues, regardless of
how they are de ﬁned, performance on a global-orienta-
tion task might well be expected to be compromised to a
Comparing ﬁndings across previous studies that have
investigated global perceptual processing in amblyopia
is diﬃcult because of (1) the variety of paradigms em-
ployed and the likelihood that diﬀerent mechanisms
underlie sensitivity (i.e. Glass patterns versus contour
integration) (2) the diﬀerent types of ambly opia (i.e.
stimulus deprivation versus strabismus or anisometro-
pia). Therefore, in the present investigation and as a
continuation of our global-motion paradigm we have
undertaken an analysis that distinguishes between
contrast-dependent (i.e. visibility-based) as oppos ed to
signal:noise-dependent deﬁcits in strabismic and aniso-
metropic amblyopia, in an attempt to ascertain wheth er
any global-orientation processing deﬁcits observed are
striate or extrastriate in origin. Import antly this task is
directly analogous to the global-motion task therefore
allowing a co mparison between form and motion sensi-
tive mechanisms in amblyopia as well as assessing
performance for both ﬁrst-order and second-order stim-
ulus types. Observers were required to make judgments
of the global statistics (in this case the overall or net ori-
entation) of a stimulus composed of a large number of
randomly positioned dots (elements) as a function of
the dot modulation or contrast (visibility) of the dots
(see Fig. 1 for illustrations). Coherence thresholds were
quantiﬁed through the gradual introduction of incoher-
ently orientated pairs of neighbouring dots until the
global organization could no longer be detected reliably.
Although a local mechanism can encode the orienta-
tion of any single pair or cluste r of elements in the stim-
ulus, performance is ultimately limited by a global
mechanism that integrates local estimates over the entire
Three strabismic, four anisometropic and three
strabismic/anisometropic amblyopes (29.9 ± 10.5 years)
were recruited for the study (see Table 1 for clinical
details). For the purposes of this study amblyopia
was deﬁned as a visual acuity of 20/30 or worse in
the amblyopic eye and anisometropia was deﬁned as
an interocular diﬀerence of greater than 1.00 dioptre
sphere or 1.0 dioptres of cylinder. A control group of
eight observers (mean age 29.4 ± 5.8 years) with nor-
mal visual acuity and normal bin ocular vision were
selected. Viewing was monocular in all cases with
the appropriate refractive correction. All experimental
procedures followed the institutional guidelines, and
informed consent was obtained after the nature and
possible consequences of the experiment had been
explained. All subject s were experienced in psycho-
2.2. Apparatus and stimuli
Stimuli were computer generated and displayed on an
SONY Multiscan 520 GS monitor (with a frame rate of
75 Hz), which was gamm a-corrected with the aid of
internal look-up-tables. As an added precaution psycho-
physical procedures were used to ensure that any resid-
ual luminance non-linearities were minimized
(Ledgeway & Smith, 1994; Nishida, Ledgeway, & Ed-
wards, 1997). The stimuli were presented within a circu-
lar window at the centre of the display, the diameter of
which subtended an angle of 12 at the viewing distance
of 0.84 m. The mean luminance of the remainder of the
display (which was homogeneous) was approximately
In order to illustrate the analogous relationship be-
tween the global-orientation stimuli used in the present
experiment and the global-motion stimuli we have
used previously (Simmers et al., 2003), it is informative
to consider the construction of the latter stimuli ﬁrst.
In a global-motion stimulus (RDK) a random spatial
array of dots is presented on the ﬁrst frame of a mo-
tion sequence and on subsequent frames the dots are
displaced in order to create the impression of motion.
The global-motion coherence level of the stimulus is
manipulated by requiring only a percentage of the
dot displacements on each image update to be in
the same direction (‘‘signal’’ displacements) and the
remainder to be in random directions (‘‘noise’’ dis-
placements). In our previous work, for example, the
direction of the ‘‘signal’’ displacements was chosen
to be either upwards or downwards on each trial with
equal probability. It is evident that over the total time
course of the display the individual dots will trace out
trajectories in space that, in principle, can be consid-
ered as ‘‘motion streaks’’ in space–time. The overall
orientation of these streaks in space–time indicates
the direction of global motion and their spatial extent
depends on the motion coherence level (percentage of
‘‘signal’’ displacements present on each image update).
A.J. Simmers et al. / Vision Research 45 (2005) 449–460 451
The motion streaks become readily apparent if the indi-
vidual frames of a RDK motion sequence are simply spa-
tially superimposed to create a single static image. This
eﬀectively converts the motion information carried by
the dots into spatial orientation information (i.e. the mo-
tion streaks app ear as ‘‘orientation streaks’’ in the result-
ing space–space representation). Perceptually these tasks
are identical to the observer, integrating frames sequen-
tially produces continuous apparent motion (motion
streaks in space–time) and simultaneously produces spa-
tial structure i.e. orientation (orientation streaks in
space–space), thereby allowing us to directly compare
integration over both time and space.
In the present experiment we used this technique (see
below) to construct global-orientation stimuli (Fig. 1)
that are direct ly comparable to the global -motion stim-
uli we have used previously to study amblyopia.
Each global-orientation stimulus was generated anew
immediately prior to its presentation (on any one trial)
and was composed of a single image containing a spatial
array of circular dots. This was achieved in practice by
computing a nominal RDK global-motion sequence
composed of four success ive frames (each containing
50 non-overlapping dots of diameter 0.47). In the ﬁrst
frame the dot positions were determined randomly and
on subsequent frames were displaced by 0.3. The indi-
vidual frames wer e then spatially superimposed to create
a single static image that was then presented for a total
stimulus duration of 426.7 ms. This enabled us to pro-
duce displays in which the consistency of the local orien-
tation signals between matching pairs of neighbouring
dots (separated spatially by 0.3) could be varied (i.e.
signal:noise ratio of the local orientations could be var-
ied to measure a global-orientation coherence thresh-
old). Performance, which is quantiﬁed in term s of the
minimum number of ‘‘signal’’ dots (coherence) required
to support orientati on discrimination, was measured as
a function of the dot modulation or contrast (visibility)
of the dots. The stimulus conditions were therefore di-
rectly comparable to those used previously to investigate
both ﬁrst-order and second-order global motion percep-
tion (Simmers et al., 2003).
Each dot was composed of two-dimensional (2-d),
static noise produced by assigning individual screen
pixels (1.41 · 1.41 arc min) within the area of each dot
to be ‘‘black’’ or ‘‘white’’ with probability 0.5. The dots
were presented on a 2-d, static noise background which
ﬁlled the entire circular display window (mean lumi-
nance of 50 cd/m
and Michelson contrast of 0.1), either
the mean luminance (in the case of ﬁrst-order dots as
shown in Fig. 1a and c) or the mean contrast (in the
case of second-order dots as shown in Fig. 1bandd)
Fig. 1. Illustration of the (a) ﬁrst-order (luminance-deﬁned) and (b) second-order (contrast-deﬁned) global-orientation dot stimuli used in the present
study. The orientation coherence of the stimuli is 100% and ‘‘orientation streaks’’ (in this case vertical), formed by local clusters of spatially
proximal dots, are readily discernible in each image. A magniﬁed view of a single ﬁrst-order dot (c) and a single second-order dot (d) are shown for
clarity (see text for details).
452 A.J. Simmers et al. / Vision Research 45 (2005) 449–460
of which could be less than that of the noise within the
The luminance modulation (visibility) of the ﬁrst-
order dots was deﬁned as
Dot luminance modulation
are the mean luminances of
the noise within the dots and background, respectively,
averaged over pairs of noise elements with opposite
luminance polarity. The luminance modulation of the
ﬁrst-order dots could be varied in the range 0–0.3 (sep-
arated by 1/6 octave steps). This gave us a possible
36 modulat ion amplitudes for the luminance-deﬁned
The contrast modulation (visibility) of the second-
order dots could be varied in an analogous manner
according to the equation:
Dot contrast modulation
are the mean contrasts of the
noise within the dots and background, respectively,
computed over pairs of noise elements with opposite
luminance polarity. The contrast modulation of the sec-
ond-order dots could be varied in the range 0–0.8 (sep-
arated by 1/48 octave steps). This gave us a possible
66 modulation amplitudes for the contrast-deﬁned
In the present task, neighbouring dots within the dis-
play were positioned relative to each other such that
they were oriented either consistently (co-linearly) to
form an elongated streak (‘‘signal’’ orientations) or in
a random manner (‘‘noise’’ orientations). The dots
deﬁning the ‘‘noise’’ orientations had a random ‘‘posi-
tional’’ displacement, but the same separation as the
dots comprising the ‘‘signal’’ orientations, and were oﬀ-
set in a random direction (spanning the 360 range) from
the previous dot in the orientation streak.
The observerÕs task was to indicate whether the ‘‘sig-
nal’’ orientations were either horizontal or vertical
(randomised on each trial). Global-orientation thresh-
olds were measured using an adaptive staircase proce-
dure that varied the proportion of local ‘‘signal’’
orientations (orientation coherence level) present on
each trial, according to the observerÕs recent response
history. At the beginning of each run of trials the stair-
case began with the maximum number of ‘‘signal’’ orien-
tations possible (i.e. with 100% orientation coherence all
dots formed extended streaks oriented along the same
axis). This was decreased following three correct consec-
utive responses and increased following each incorrect
global-orientation judgement. The staircase tracked the
global-orientation coherence level producing 79% cor-
rect responding. Eight staircase reversals were collected
before the staircase terminated and the threshold was
taken as the mean of the last six reversal points. Each
threshold reported is based on the mean of at least ﬁve
such staircases. In those observers with amblyopia,
measurements were repeated with both the amblyopic
eye and non-amblyopic eye in random order. In normal
observers the right or left eye was randomly assigned.
Fig. 2a shows the mean normal result (black symbols)
in which global-orientation thresholds for ﬁrst-order
dots are plotted against the dot modulation (contrast).
Similar to global-motion thresholds (Edwards,
Badcock, & Nishida, 1996; Simmers et al., 2003)
Clinical characteristics of the amblyopic subjects
Subject Visual acuity Spectacle prescription Ocular alignment
RE 20/15 RE + 3.25DS L XOT
LE 20/40 LE + 3.75DS 10D
RE 20/20 Nil LSOT
LE 20/200 14D
RE 20/80 Nil R SOT
LE 20/20 18D
RE 20/80 RE + 5.00DS R SOT
LE 20/20 LE + 1.75 16D
RE 20/20 RE 2.25/1.25 · 180 L XOT
LE 20/50 LE 3.00/1.75 · 170 10D
RE 20/20 RE + 4.00/ 1.00 · 170 L XOT
LE 20/127 LE + 6.00/ 1.75 · 177 10D
RE 20/15 RE piano Straight
LE 20/80 LE + 2.50DS
RE 20/15 RE 0.25DS Straight
LE 20/200 LE + 3.50/0.50 · 90
RE 20/25 RE + 1.50DS Straight
LE 20/50 LE + 3.00/0.5 · 150
RE 20/20 RE + 0.25/1.75 · 135 Straight
LE 20/30 LE + 0.75/3.50 · 55
Red symbols correspond to individual strabismic, green symbols to
strabismic anisometropes and blue symbols to anisometropic
amblyopes—superscript ‘‘a’’ indicates amblyopic subjects who partic-
ipated in both the global-orientation and global-motion experiments.
(For interpretation of colour in this table, the reader is referred to the
web version of this article.)
All observers began each trial at the highest modulation ampli-
tude for both stimuli types this was then reduced to a level at which
performance reached chance. The range of amplitude modulations
chosen in between these two values were tailored for each individual
observer. To ﬁt the data reliably, performance was assessed at no fewer
than 10 modulation depths for both luminance and contrast-deﬁned
A.J. Simmers et al. / Vision Research 45 (2005) 449–460 453
global-orientation thresholds exhibit asymptotic behav-
ior at high levels of dot modulation but increase mark-
edly as the magnitude of the dot modulation
decreases. In the case of normal observers, the relation-
ship between the global-orientation threshold and the
magnitude of the dot modulation is well described by
a power function plus a constant (solid black line).
y = ax
+ c, where a, b and c are constants. For the nor-
mal population the mean corresponding values for the
ﬁrst-order stimuli are a = 1.75e04 (±8.72e05 s.e.m);
b = 2.65(±0.1 s.e.m); c = 15.69 (±0.31 s.e.m) and for
the second-order stimuli are a = 1.37 (±0.25 s.e.m);
b = 3.46 (±0.17 s.e.m); c = 23.88 (±0.79 s.e.m).
Similar to our global-motion task, if performance in
amblyopia is limited by reduced visibility due to the con-
trast sensitivity deﬁcit thought to reside in V1, then we
would expect the response function (mean global-orien-
tation threshold versus dot modulation (contrast)) for
the amblyopic visual system to be well described by a
laterally translated (to the right on this co-ordinate sys-
tem) version of the normal response curve, as modeled
by the dashed curves also depicted in Fig. 2a.
However, if performance is solely limit ed by a deﬁ-
cient global-orientation extraction process, the response
function for the amblyopic visual system should shift
vertically on these axes, as shown by the model predic-
tions (dashed curves) in Fig. 2b.
The raw data for individual amblyopic subjects is dis-
played in Fig. 3 for both ﬁrst-order (Fig. 3a) and
second-order (Fig. 3b) stimuli. Whilst most amblyopes
displayed reduced performance on this task, the
underlying deﬁcit was composed of both a contrast/vis-
ibility component (evident by a lateral shift in the
response function) and a global orientation-based com-
ponent (evident by a vertical shift in the response
Fig. 3 illustrates well the variability of deﬁcit in the
raw data for our group of amblyopic observers; it is dif-
ﬁcult to assess by inspection the degree to which each re-
sponse function is shifted either vertically or laterally.
Therefore a summary of the relative contributions of
visibility and global-orientation processing to the overall
deﬁcit is shown in Fig. 4, where the derived component
anomalies for contrast and global-orientation deﬁcits
are plotted for individual amblyopes. These component
anomalies were best described by independently ﬁtting a
two-parameter model, and taking the ratio of the best
ﬁtting parameters describing the lateral (contrast or vis-
ibility) and vertical (global-orientation sensitivity) shifts
to the raw data, to bring them into correspondence with
the mean performance exhibited by normal subjects. In
all cases the numerator is represented by the mean val-
ues for the normal observers (see above) and the denom-
inator is that of each individual amblyope. The dashed
line(s) represent a ratio of one indicating no diﬀerence
in threshold between the normal and amblyopic observ-
ers with respect to either the contrast or global-orienta-
tion extrac tion of the stimuli.
0.00 0.01 0.10
modeled contrast loss
Orientation-coherence threshold (%)
0.00 0.01 0.10
modeled orientation loss
Orientation-coherence threshold (%)
Fig. 2. The mean ﬁrst-order (luminance-deﬁned) global-orientation
coherence thresholds for the eight normal observers are plotted as a
function of the dot modulation. The relationship between the global-
orientation threshold and the magnitude of the dot modulation is ﬁtted
by a power function plus a constant. (a) The dashed curves
demonstrate hypothetically how a systematic diﬀerence in absolute
sensitivity can be predicted by a simple translation of the threshold
versus dot modulation function along the dot contrast axis (a contrast-
speciﬁc deﬁcit). (b) The dashed curves again demonstrate in theory
how a systematic diﬀerence in the ability to extract global orientation,
will manifest as a simple translation of the threshold versus dot
modulation function along the threshold axis (a global orientation-
454 A.J. Simmers et al. / Vision Research 45 (2005) 449–460
Due to the limited sample size per amblyopic subject
group (strabismic versus anisometropic versus anisome-
tropic strabismics) and in order not to violate the
assumptions of ANOVA, analyses were carried out
for a singl e generic amblyopic subject group. Therefore
to fully explore the pattern of deﬁcits in this subject
group, an analysis of variance was carried out for
amblyopic subject group (all amblyopes) and the factors
of stimulus type (ﬁrst-order versus second-order) and
component anomaly (contrast versus global orientation).
ANOVA revealed no signi ﬁcant eﬀect of stimulus type
(ﬁrst-order versus second-order ) (F = 0.001; p = 0.98)
and a signiﬁcant eﬀect of the component anomaly (spa-
tial versus visibility) (F = 11.369; p = 0.002) with no sig-
niﬁcant interaction (F = 0.64; p = 0.43). Therefore,
0.00 0.01 0.10
Orientation-coherence threshold (%)
Orientation-coherence threshold (%)
Fig. 3. Individual global-orientation thresholds. The solid black line
represents the mean results for eight normal observers: (a) shows the
global-orientation threshold of each individual amblyope for the
luminance-deﬁned, ﬁrst-order dots and (b) shows the global-orienta-
tion threshold of each individual amblyope for the contrast-deﬁned,
second-order dots. Each datum represents the mean of a minimum of
ﬁve blocks of trials and error bars represent ±1 s.e.m. Curves represent
a power function ﬁt to the data.
Fig. 4. Ratio of normal to amblyopic eye performance for both ﬁrst-
order (a) and second-order (b) stimuli. The dashed line(s) represent a
ratio of one indicating no diﬀerence in threshold between the normal
and amblyopic observers with respect to either encoding the contrast
or extracting the global orientation of the stimuli. Values falling along
the horizontal dashed line are consistent with a deﬁcit speciﬁc to
extracting global orientation; values that fall along the vertical dashed
line are consistent with a contrast-speciﬁc deﬁcit.
A.J. Simmers et al. / Vision Research 45 (2005) 449–460 455
unlike our previous global-motion task amblyopes did
not show a greater deﬁcit or indeed a selective loss,
as has previous ly been reported for isolated static stim-
uli (Wong, Levi, & McGraw, 2001), in the processing of
second-order patterns. There was howeve r a signiﬁcant
main eﬀect of component anomaly (F
p = 0.01) demonstrating that global-orientation deﬁcits,
when collapsed across subject group an d stimulus type,
were signiﬁcantly greater than contrast (visibility) deﬁ-
cits. None of the other possible interactions reached
signiﬁcance at the 0.05 probability level. Although the
sample size may have been too small for statistical anal-
ysis of any group diﬀerences, the ﬁgures appear to show
no diﬀerence across patient types (strabismic versus
anisometropic versus anisometropic strabismics). These
results do however provide clear evidence for a global-
orientation processing deﬁcit in amblyopia that cannot
be simply explained by reduced visibility resulting from
the known contrast sensitivity loss.
Conventional visual acuity measures in the ambly-
opic eye were also not found to be a reliable indicator
of overall performance with no signiﬁ cant correlation
between either ﬁrst-order contrast (r
= 0.22;NS) or
global orientation (r
= 0.06; NS) and second-order
= 0.26; NS) or global orientation
= 0.23;NS). This is evident in Fig. 4aandb
where the poorest (
20/200) and best ( 20/30)
individual visual acuities in the amblyopic subject
group do not then consequently dictate the upper and
lower limits of either the contrast or orientation based
In a directly analogous task we have previously re-
ported a global -motion deﬁcit in amblyopia that was
signiﬁcantly larger for second-order stimuli than ﬁrst-
order stimuli. However in the present experiment the
deﬁcits for global-orientation processing for both types
of stimuli were found to be of comparable magnitude
in the amblyopic subject group. To determine whether
amblyopes are anomalous at integrating visual informa-
tion in general, we assessed performance with the com-
parable global-motion task (for details see Simmers
et al., 2003) in eight of our amblyopic subjects (see Table
1 for subject selection).
Fig. 5 compares the ratios obtained with the same
individual amblyopic subjects for global-motion and
global-orientation encoding deﬁcits with both ﬁrst-order
(a) and second-order (b) stimuli. Interestingly there were
signiﬁcant diﬀerences between the results obtained in
each of the four conditions (F
= 8.580; p = 0.03).
Pairwise comparisons (t-test) revealed that the deﬁcits
in global-motion processing were signiﬁcantly more
extensive for the second-order stimuli than the ﬁrst-
order motion stimuli. In addition the second-order,
global-motion processing deﬁcit was signiﬁcantly great-
er than the global -orientation encoding deﬁcits for both
types of stimuli.
Fig. 6 compares the modulation (contrast) processing
deﬁcits in the same individual amblyopic subjects meas-
ured using the global-motion and global-orientation
tasks with both ﬁrst-order (a) and second-order (b) stim-
uli. This ﬁgure clearly illustrates the compara ble nature
Fig. 5. Comparison of the ratios for global-motion and global-
orientation performance for both ﬁrst-order (a) and second-order (b)
stimuli in the same amblyopic observers. The dashed line(s) represent a
ratio of one indicating no diﬀerence in threshold between global-
motion and global-orientation deﬁcits measured using comparable
visual stimuli. Values falling along the horizontal dashed line are
consistent with a deﬁcit aﬀecting only global-motion processing; values
that fall along the vertical dashed line are consistent with a orientation-
speciﬁc deﬁcit. The red dashed line represents the 1:1 line and the
preponderance of values lying beneath this line indicate a greater
deﬁcit in global motion extraction, especially for the second-order
stimuli. (For interpretation of colour in this ﬁgure, the reader is
referred to the web version of this article.)
456 A.J. Simmers et al. / Vision Research 45 (2005) 449–460
of any visibility loss with respect to either experimental
paradigm, with no signiﬁcant diﬀerences being evident
These results are important in that not only have we
shown deﬁcits in global-orientation and global-motion
processing in amblyopia, which appear to be independ-
ent of any low-level visibility loss, the ability to integrate
local second-order motion signals across space and time
(to derive a global percept of image motion) appears to
show the greatest impairment overall.
This is readily evident in Fig. 7, which summarises the
magnitudes of the component anomalies found using the
two tasks. This clearly suggests that there is a temporal
processing deﬁcit in amblyopia, whi ch is particularly
marked for the encoding of second-order, global-motion
Our study shows convincingly that there is also a glo-
bal-orientation processing deﬁcit in human amblyopia
that consists of both contrast (visibility) and signal:noise
dependent components. These deﬁcits are present for
both luminance-deﬁned and contrast-deﬁned sti muli,
but interestingly the spatial integration based loss is
not as extensive as that found for the processing of ana-
logous global-motion stimuli, particularly in the case of
contrast-deﬁned patterns (see Fig. 7). This is a strong
conclusion because the motion and spatial tasks were
identical in every respect except for the fact that the mo-
tion stimuli were presented sequentially whereas the spa-
tial stimuli were presented simultaneously. For motion,
second-order stimuli were more aﬀected than ﬁrst-order
ones, a result not found for spatial processing. These re-
sults suggest independent deﬁcits for spatial and motion
coding in amblyopia with global-motion coding being
aﬀected more severely, particularly for second-order
In a comparison of monocular versus binocular dep-
rivation (congenital cataracts) two separate studies have
looked at sensitivity to form (Lewis, Ellemberg, Maurer,
Wilkinson, & Wilson, 2002) and motion (Ellemberg
et al., 2002) integration. In both studies binocular depri-
vation consistently resulted in a more profound loss of
visual function and as in the present study a greater loss
in sensitivity was evident for the perception of global
motion. Although obvious comparisons can be made be-
tween these and our own study, deprivation amblyopia
is in itself a rare and uncommon form of amblyopia typ-
ically associated with very poor residual visual function
and widely believed to be both qualitatively and quanti-
ﬁably diﬀerent from the more common forms of ambly-
opia (strabismus and anisometropia), therefore this
paradigm may not be especially useful in understanding
Useful comparisons may however be drawn from a
recent behavioral study, which looked at contour inte-
gration in amblyopia as a form of global percept ual
organization in non-human primates (Kozma & Kior-
pes, 2003). In this study contour integration was im-
paired in both strabismic and anisometropic animals
and similar to our study the integration deﬁcits were
Fig. 6. Comparison of the ratios for the contrast-speciﬁc deﬁcits
present for both ﬁrst-order (a) and second-order (b) stimuli. The
dashed line(s) represent a ratio of one indicating no diﬀerence in the
contrast-speciﬁc deﬁcit between the global-motion and global-orien-
tation experiment. The red dashed line represents the 1:1 line and the
preponderance of values lying on this line indicates that any contrast-
speciﬁc loss present in individual amblyopic subjects was comparable
in terms of magnitude between the global-motion and global-orienta-
tion tasks. (For interpretation of colour in this ﬁgure, the reader is
referred to the web version of this article.)
A.J. Simmers et al. / Vision Research 45 (2005) 449–460 457
also clearly unrelated to losses in contrast sensitivity,
supporting our ﬁndings in the present study that deﬁcits
in global processing appear independent of any low-level
visibility loss. Coupled with deﬁcits to the fellow eye
(Kovacs et al., 2000; Simmers et al., 2003) this would
indicate a disruption to processing mechanisms beyond
V1. The authors suggest that elevated centra l noise in
the amblyopic visual system could be responsible for
losses in sensitivity.
On the basis of the available neurophysiological evi-
dence and the likelihood that more complex visual
processing occurs in a roughly hierarchical fashion fur-
ther along the visual pathway, it seems not unreasonable
to locate the contrast-dependent anomaly in striate cor-
tex where it has been shown by numerous studies that
cells in the amblyopic visual system have altered spatial
responses and lowered contrast sensitivity (Chino et al.,
1983; Crewther & Crewther, 1990; Eggers & Blakemore,
1978; Kiorpes et al., 1998; Movshon et al., 1987). The
signal:noise de ﬁcit (ability to integrate local information
across the visual ﬁeld) is likely to involve extrastriate vis-
ual areas where single cell neurophysiology has docu-
mented cells with much larger receptive ﬁelds
composed of multiple subunits capable of such an inte-
grative role (Movshon, Adelson, Gizzi, & Newsome,
1985). This is not meant to imply, however, that a com-
mon mechanism would necessarily mediate bot h the
integration of oriented static spatial cues and the inte-
gration of moving signals. Based on the littl e direct
physiology that is available, the former could take place
within the ventral stream; predominantly concerned
with the processing of spatial form and may be associ-
ated with the perceptual discrimination of overall image
shape and contour orientation and the latter almost cer-
tainly within the dorsal processing stream; which is con-
cerned with motion processing with the ultimate role of
motor actions (Ungerleider & Mishkin, 1982). Although
functional interactions between the two now seem highly
likely (Kourtzi et al., 2001). Within the context of this
framework, our results for the present spatial task and
the previous motion task (Simmers et al., 2003) suggest
that there are deﬁcits at the level of both the ventral and
dorsal extrastriate cortex, it is interesting to speculate
that the dorsal stream may be more disadvantaged in
amblyopia, as evidenced by the relatively impai red per-
formance of our observers on the global-motion detec-
tion task. The pattern of results reported here appears
consistent with prev ious suggestions of dorsal stream
vulnerability in speciﬁc developmental disorders and
neurological impairment (Atkinson et al., 2001; Brad-
dick et al., 2003; Gunn et al., 2002).
Interestingly, evidence is emerging that higher corti-
cal areas are relatively delayed in their development
compared to V1 (Bachevalier, Hagger, & Mishkin,
1991; Distler, Bachevalier, Kennedy, Mishkin , &
Ungerleider, 1996; Rodman, 1994), and moreover that
the extraction of local features appears to develop rela-
tively early, whereas the integration of these local fea-
tures may emerge substantially later and at diﬀerent
times for diﬀerent types of image structure independent
to the development of spatial resolution (Ellemberg
et al., 2002; Kiorpes & Bassin, 2003; Kovacs et al.,
1999; Lewis et al., 2002). These observations raise the
intriguing possibility that speciﬁc visual deﬁcits may
co-vary with the age of onset of amblyopia.
Most perceptual anomalies that have been reported
in amblyopia have been explicable in terms of the striate
deﬁcit that we know is present in animals made artiﬁ-
cially strab ismic, anisometropic or form-deprived
(Chino et al., 1983; Crewther & Crewther, 1990; Eggers
Stimulus and Component Anomalies
Mean Ratio of Performance
Fig. 7. Summary of the mean deﬁcits (±1 standard deviation) in amblyopic performance consistent with losses speciﬁc to encoding stimulus contrast
(visibility), global orientation and global motion for both ﬁrst-order and second-order stimuli.
458 A.J. Simmers et al. / Vision Research 45 (2005) 449–460
& Blakemore, 1978; Kiorpes et al., 1998; Movshon
et al., 1987). Few tasks requiring global integration of
the kind reported in this study have been used on the
same human amblyopes with comparable form (orienta-
tion) and motion stimuli and when they have the conclu-
sions have not been deﬁnitive when it comes to
extrastriate function. For example, contour integration
tasks of the kind reported by Field, Hayes, and Hess
(1993) and Kovacs and Julesz (1993) involve a global
integrative process at which amblyopes have demon-
strated reduced performance (Hess et al., 1997). How-
ever the cortical site of this type of integration is
currently unknown, suggestions have been made impli-
cating both striate and extrastriate sites (Altmann,
Bulthoﬀ, & Kourtzi, 2003; Kourtzi, Tolias, Altmann,
Augath, & Logothetis, 2003; Kozma & Kiorpes, 2003).
With respect to the reduced performance found in
amblyopia one proposal that has been advanced, is in
terms of the known positional deﬁcit amblyopes demon-
strate in judging the relative position of a target for well-
separated stimuli (Hess & Holliday, 1992). Whether this
positional deﬁcit is downstream in V1 or situated in the
extrastriate cortex where the integration proper may oc-
cur is presently unknown.
In another form of global perceptual processing the
ability with which amblyopes can integrate local samples
of orientation to judge the mean orientation of a popu-
lation of diﬀerently oriented elements has been ques-
tioned. These studies would suggest that amblyopes
are able to integrate orientation information across vis-
ual space with either no performance diﬀerence evident
between the amblyopic and fellow eye (Mansouri, Allen,
Hess, Dakin, & Ehrt, 2004) or conversely that the neural
representation of local image structure appears to show
a greater variability compared to normal performance in
both the amblyopic and fellow eye (Simmers & Bex,
2004). The latter result being more consistent with the
present study and a deﬁcit at global stages of visual
processing in amblyopia. Thus, it may well be that not
all forms of global integration are equally aﬀected in
Taken together the deﬁcits in global-orientation and
global-motion processing reported in the present study,
suggest that global perceptual organization is impaired
in amblyopia consistent with deﬁcits to extrastriate
processing mechanisms. Our ﬁnding that the integration
of local motion signals is aﬀected to a g reater extent
than comparable spatial integration would argue that
the deﬁcit in amblyopia diﬀerentially aﬀects the dorsal
extrastriate processing stream.
A Medical Research Council Research Fellowship to
AJS, a University of Nottingham, Research Committee
grant to TL and a CIHR Operating grant to RFH
(MT108-18) supported this work.
Altmann, C. F., Bulthoﬀ, H. H., & Kourtzi, Z. (2003). Perceptual
organization of local elements into global shapes in the human
visual cortex. Current Biology, 13, 342–349.
Atkinson, J., Anker, S., Braddick, O., Nokes, L., Mason, A., &
Braddick, F. (2001). Visual and visuospatial development in young
children with Williams syndrome. Developmental Medicine and
Child Neurology, 43, 330–337.
Bachevalier, J., Hagger, C., & Mishkin, M. (1991). Functional
maturation of the occipitotemporal pathway in infant rhesus
monkeys. In N. A. Lassen, D. H. Ingvar, & M. E. Raichle (Eds.),
Alfred Benzon symposium, brain work and mental activity
(pp. 231–240). Copenhagen: Munksgaard.
Barnes, G. R., Hess, R. F., Dumoulin, S. O., Achtman, R. L., & Pike,
G. B. (2001). The cortical deﬁcit in humans with strabismic
amblyopia. Journal of Physiology—London, 533, 281–
Barrett, B. T., Pacey, I. E., Bradley, A., Thibos, L. N., & Morrill, P.
(2003). Nonveridical visual perception in human amblyopia.
Investigative Ophthalmology and Visual Science, 44, 1555–1567.
Braddick, O., Atkinson, J., & Wattam-Bell, J. (2003). Normal and
anomalous development of visual motion processing: Motion
coherence and Ôdorsal-stream vulnerabilityÕ. Neuropsychologia, 41,
Braddick, O. J., OÕBrien, J. M. D., Wattam-Bell, J., Atkinson, J., &
Turner, R. (2000). Form and motion coherence activate independ-
ent, but not dorsal/ventral segregated, networks in the human
brain. Current Biology, 10, 731–734.
Burr, D. C., Morrone, M. C., & Vaina, L. M. (1998). Large receptive
ﬁelds for optic ﬂow detectors in humans. Vision Research, 38,
Chandna, A., Pennefather, P. M., Kovacs, I., & Norcia, A. M. (2001).
Contour integration deﬁcits in anisometropic amblyopia. Investi-
gative Ophthalmology and Visual Science, 42, 875–878.
Chino, Y. M., Shansky, M. S., Jankowski, W. L., & Banser, F. A.
(1983). Eﬀects of rearing kittens with convergent strabismus on
development of receptive-ﬁeld properties in striate cortex neurons.
Journal of Neurophysiology, 50, 265–286.
Crewther, D. P., & Crewther, S. G. (1990). Neural site of strabismic
amblyopia in cats: Spatial frequency deﬁcit in primary cortical
neurons. Experimental Brain Research, 79, 615–622.
Dakin, S. C. (2001). Information limit on the spatial integration of
local orientation signals. Journal of the Optical Society of America
a-Optics Image Science and Vision, 18, 1016–1026.
Dakin, S. C., & Watt, R. J. (1997). The computation of orientation
statistics from visual texture. Vision Research, 37, 3181–
Devalois, R. L., Cottaris, N. P., Mahon, L. E., Elfar, S. D., & Wilson,
J. A. (2000). Spatial and temporal receptive ﬁelds of geniculate and
cortical cells and directional selectivity. Vision Research, 30,
DeYoe, A., & Van Essen, D. C. (1988). Concurrent processing streams
in monkey visual cortex. Trends in Neurosciences, 11, 219–226.
Distler, C., Bachevalier, J., Kennedy, C., Mishkin, M., & Ungerleider,
L. G. (1996). Functional development of the corticocortical
pathway for motion analysis in the macaque monkey: A C-14-2-
deoxyglucose study. Cerebral Cortex, 6, 184–195.
Edwards, M., Badcock, D. R., & Nishida, S. (1996). Contrast
sensitivity of the motion system. Vision Research, 36, 2411–2421.
Eggers, H. M., & Blakemore, C. (1978). Physiological basis of
anisometropic amblyopia. Science, 201, 264–267.
A.J. Simmers et al. / Vision Research 45 (2005) 449–460 459
Ellemberg, D., Lewis, T. L., Maurer, D., Brar, S., & Brent, H. P.
(2002). Better perception of global motion after monocular than
after binocular deprivation. Vision Research, 42, 169–179.
Felleman, D. J., & Essen, D. C. (1987). Van Receptive ﬁeld properties
of neurons in area V3 of macaque monkey extrastriate cortex.
Journal of Neurophysiology, 57, 889–920.
Field, D. J., Hayes, A., & Hess, R. F. (1993). Contour integration by
the human visual-system––evidence for a local association ﬁeld.
Vision Research, 33, 173–193.
Gattass, R., Gross, C. G., & Sandell, J. H. (1981). Visual topography
of V2 in the macaque. The Journal of Comparative Neurology, 201,
Gunn, A., Cory, E., Atkinson, J., Braddick, O., Wattam-Bell, J.,
Guzzetta, A., et al. (2002). Dorsal and ventral stream sensitivity in
normal development and hemiplegia. Neuroreport, 13, 843–847.
Hess, R. F., & Holliday, I. E. (1992). The spatial localization deﬁcit in
amblyopia. Vision Research, 32, 1319–1339.
Hess, R. F., McIIhagga, W., & Field, D. J. (1997). Contour integration
in strabismic amblyopia: The suﬃciency of an explanation based
on positional uncertainty. Vision Research, 37, 3145–3161.
Hess, R. F., Wang, Y. Z., Demanins, R., Wilkinson, F., & Wilson, H.
R. (1999). A deﬁcit in strabismic amblyopia for global shape
detection. Vision Research, 39, 901–914.
Imamura, K., Richter, H., Fischer, H., Lennerstrand, G., Franze
Rydberg, A., et al. (1997). Reduced activity in the extrastriate
visual cortex of individuals with strabismic amblyopia. Neuro-
science Letters, 225, 173–176.
Jeﬀrey, B. G., Wang, Y. Z., & Birch, E. E. (2004). Altered global shape
discrimination in deprivation amblyopia. Vision Research, 44, 167–177.
Kiorpes, L. (2003). Amblyopic deﬁcits in contrast sensitivity do not
predict deﬁcits in global perception. Investigative Ophthalmology
and Visual Science, 44, E-abstract 3185.
Kiorpes, L., & Bassin, S. A. (2003). Development of contour
integration. Visual Neuroscience, 20, 567–575.
Kiorpes, L., Kiper, D. C., OÕKeefe, L. P., Cavanaugh, J. R., &
Movshon, J. A. (1998). Neuronal correlates of amblyopia in the
visual cortex of macaque monkeys with experimental strabismus
and anisometropia. Journal of Neuroscience, 18, 6411–6424.
Kiorpes, L., & McKee, S. P. (1999). Neural mechanisms underlying
amblyopia. Current Opinion in Neurobiology, 9, 480–486.
Kourtzi, Z., Bulthoﬀ, H. H., Erb, M., & Grodd, W. (2001). Object
selective responses in the human motion area MT/MST. Nature and
Neuroscience, 5, 17–18.
Kourtzi, Z., Tolias, A. S., Altmann, C. F., Augath, M., & Logothetis,
N. K. (2003). Integration of local features into global shapes:
Monkey and human fMRI studies. Neuron, 37, 333–346.
Kovacs, I., & Julesz, B. (1993). A closed curve is much more than an
incomplete one: Eﬀect of closure in ﬁgure-ground segmentation.
Proceedings of the National Academy of Science, 90, 7495–7497.
Kovacs, I., Kozma, P., Feher, A., & Benedek, G. (1999). Late
maturation of visual spatial integration in humans. Proceedings of
the National Academy of Sciences of the United States of America,
Kovacs, I., Polat, U., Pennefather, P. M., Chandna, A., & Norcia, A.
M. (2000). A new test of contour integration deﬁcits in patients
with a history of disrupted binocular experience during visual
development. Vision Research, 40, 1775–1783.
Kozma, P., & Kiorpes, L. (2003). Contour integration in amblyopic
monkeys. Visual Neuroscience, 20, 577–588.
Ledgeway, T., & Smith, A. T. (1994). Evidence for separate motion-
detecting mechanisms for ﬁrst- and second-order motion in human
vision. Vision Research, 34, 2727–2740.
Lewis, T. L., Ellemberg, D., Maurer, D., Wilkinson, F., Wilson, H. R.,
et al. (2002). Sensitivity to global form in glass patterns after early
visual deprivation in humans. Vision Research, 42, 939–948.
Levi, D. M., & Saarinen, J. (2004). Perception of mirror symmetry in
amblyopic vision. Vision Research, 44, 2475–2482.
Mansouri, B., Allen, H. A., Hess, R. F., Dakin, S. C., & Ehrt, O.
(2004). Integration of orientation information in amblyopia. Vision
Research, 44, 2955–2969.
Moseley, M., & Fielder, A. (Eds.). (2002). Amblypoia: A multidiscipli-
nary approach. Oxford: Butterworth-Heineman.
Movshon, J. A., Adelson, E. H., Gizzi, M. S., & Newsome, W. T.
(1985). The analysis of moving visual patterns. In C. Chagas, R.
Gattass, & C. G. Gross (Eds.), Study group on pattern recognition
mechanisms. Vatican City: Pontiﬁca Academia Scientiarum.
Movshon, A. J., Eggers, H. M., Gizzi, M. S., Hendrickson, A. E.,
Kiorpes, L., & Boothe, R. G. (1987). Eﬀects of unilateral blur on
the macaqueÕs visual system. III. Physiological observations.
Journal of Neuroscience, 7, 1340–1351.
Newsome, W. T., & Pare, E. B. (1988). A selective impairment of
motion perception following lesions of the middle temporal visual
area (MT). Journal of Neuroscience, 8, 2201–2211.
Nishida, S., Ledgeway, T., & Edwards, M. (1997). Dual multiple-scale
processing for motion in the human visual system. Vision Research,
Popple, A. V., & Levi, D. M. (2000). Amblyopes see true alignment
where normal observers see illusory tilt. Proceedings of the National
Academy of Sciences of the United States of America, 97,
Rodman, H. R. (1994). Development of inferior temporal cortex in the
monkey. Cerebral Cortex, 4, 484–498.
Sawatari, A., & Callawat, E. M. (1996). Convergence of magno- and
parvocellular pathways in layer 4B of macaque primary visual
cortex. Nature, 380, 442–446.
Schroder, J. H., Fries, P., Roelfsema, P. R., Singer, W., & Engel, A. K.
(2002). Ocular dominance in extrastriate cortex of strabismic
amblyopic cats. Vision Research, 42, 29–39.
Sharma, V., Levi, D. M., & Klein, S. A. (2000). Undercounting
features and missing features: Evidence for a high-level deﬁcit in
strabismic amblyopia. Nature and Neuroscience, 3, 496–501.
Simmers, A. J., & Bex, P. J. (2004). The representation of global spatial
structure in amblyopia. Vision Research, 44, 523–533.
Simmers, A. J., Ledgeway, T., Hess, R. F., & McGraw, P. V. (2003).
Deﬁcits to global motion processing in human amblyopia. Vision
Research, 43, 729–738.
Sireteanu, R., Tonhausen, N., Muckili, L., Zanella, F. F., & Singer, W.
(1998). Cortical site of the amblyopic deﬁcit in strabismic and
anisometropic subjects investigated with fMRI. Investigative Oph-
thalmology and Visual Science, 39, S909.
Takeuchi, T., De Valois, K. K., & Hardy, J. L. (2003). The inﬂuence of
color on the perception of luminance motion. Vision Research, 43,
Tang, C., Kiorpes, L., & Movhson, A. J. (1998). Motion detection in
amblyopic monkeys. Investigative Ophthalmology and Visual Sci-
ence, 39, S330.
Ungerleider, L. G., & Mishkin, M. (1982). Two cortical visual systems.
Analysis of visual behaviour. London: MIT press, pp. 549–587.
Verghese, P., Watamaniuk, S. N. J., McKee, S. P., & Grzywacz, N. M.
(1999). Local motion detectors cannot account for the detectability
of an extended trajectory in noise. Vision Research, 39, 19–30.
Wong, E. H., Levi, D. M., & McGraw, P. V. (2001). Is second-order
spatial loss in amblyopia explained by the loss of ﬁrst-order spatial
input? Vision Research, 41, 2951–2960.
460 A.J. Simmers et al. / Vision Research 45 (2005) 449–460