See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/11104897

Circuits for Local and Global Signal Integration in Primary Visual Cortex.

Article in The Journal of Neuroscience : The Official Journal of the Society for Neuroscience · November 2002
Source: PubMed

CITATIONS READS

664 273

6 authors, including:

Alessandra Angelucci Jonathan B Levitt

University of Utah City College of New York
60 PUBLICATIONS 4,240 CITATIONS 68 PUBLICATIONS 4,634 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Connectomics View project

Development of hybrid array for optogenetic stimulation and electrical recording View project

All content following this page was uploaded by Jonathan B Levitt on 03 January 2014.

The user has requested enhancement of the downloaded file.

 The Journal of Neuroscience, October 1, 2002, 22(19):8633–8646

Circuits for Local and Global Signal Integration in Primary

Visual Cortex

Alessandra Angelucci,1 Jonathan B. Levitt,2 Emma J. S. Walton,3 Jean-Michel Hupé,4 Jean Bullier,4 and
Jennifer S. Lund1
1Department of Ophthalmology and Visual Science, Moran Eye Center, University of Utah, Salt Lake City, Utah 84132,
2Department of Biology, City College of the City University of New York, New York, New York 10031, 3Department of
Visual Sciences, Institute of Ophthalmology, University College London, London EC1V 9EL, United Kingdom, and 4Centre
de Recherche Cerveau et Cognition, Centre National de la Recherche Scientifique–Unité Propre de Recherche Unité
Mixte de Recherche 5549, Université Paul Sabatier, Toulouse 31062, France

Contrast-dependent changes in spatial summation and contex- dimensions of the surround field. The spatial scale of feedback
tual modulation of primary visual cortex (V1) neuron responses circuits from extrastriate cortex to V1 is, instead, commensu-
to stimulation of their receptive field reveal long-distance inte- rate with the full spatial range of center–surround interactions.
gration of visual signals within V1, well beyond the classical Thus these connections could represent an anatomical sub-
receptive field (cRF) of single neurons. To identify the cortical strate for contextual modulation and global-to-local integration
circuits mediating these long-distance computations, we have of visual signals. Feedback projections connect corresponding
used a combination of anatomical and physiological recording and equal-sized regions of the visual field in striate and extra-
methods to determine the spatial scale and retinotopic logic of striate cortices and cover anisotropic parts of visual space,
intra-areal V1 horizontal connections and inter-areal feedback unlike V1 horizontal connections that are isotropic in the ma-
connections to V1. We have then compared the spatial scales caque. V1 isotropic connectivity demonstrates that anisotropic
of these connectional systems to the spatial dimensions of the horizontal connections are not necessary to generate orienta-
cRF, spatial summation field (SF), and modulatory surround tion selectivity. Anisotropic feedback connections may play a
field of macaque V1 neurons. We find that monosynaptic hori- role in contour completion.
zontal connections within area V1 are of an appropriate spatial
scale to mediate interactions within the SF of V1 neurons and to Key words: primary visual cortex; extrastriate cortex; feed-
underlie contrast-dependent changes in SF size. Contrary to back connections; lateral connections; surround modulation;
common beliefs, these connections cannot fully account for the macaque

A central question in visual cortical processing is how local and Wiesel, 1990; Levitt and Lund, 1997; Walker et al., 1999).
signals are integrated across space to generate global percepts. Furthermore, V1 RFs show dynamic spatial properties, changing
Traditionally, visual information has been seen as ascending in size depending on stimulus contrast (Kapadia et al., 1999;
through a hierarchy of cortical areas, with cells at each successive Sceniak et al., 1999). These neurophysiological responses require
stage processing inputs from increasingly larger regions of space. integration of visual signals beyond the RF of single V1 neurons
However, long-distance integration of visual signals can occur at and thus cannot be easily explained by classical RF concepts.
very early stages of processing. The response of cells in the Identifying the neural circuitry underlying these long-distance
primary visual cortex (V1) to stimulation of their receptive field computations is crucial, because they may represent the neural
(RF) can be modulated in a selective way by contextual stimuli substrates for feature grouping (Kapadia et al., 1995; Mizobe et
lying far outside the RF in the RF surround (Blakemore and al., 2001) and figure–ground segregation (Knierim and Van Es-
Tobin 1972; Nelson and Frost, 1978; Allman et al., 1985; Gilbert sen, 1992; Nothdurft et al., 1999).
Currently, most models of center–surround interaction in V1
Received March 8, 2002; revised June 19, 2002; accepted June 27, 2002. are based on intrinsic horizontal (or lateral) connections (Gilbert
This work was supported by Medical Research Council Grants G9203679 and et al., 1996; Somers et al., 2002). These are long-range, reciprocal,
G9408137, by European Community Grant Viprom Biomed 2, by Wellcome Trust intralaminar projections made by excitatory neurons in layers 2/3,
Grant 061113, in part by National Institutes of Health Grants EY12781 and Re-
search Centers in Minority Institutions G12RR-03060, and by a grant from Research 4B/upper 4C␣, and 5/6 of macaque area V1 (Rockland and Lund,
to Prevent Blindness, Inc. (New York, NY) to the Department of Ophthalmology, 1983). These connections show a periodic, patchy pattern of
University of Utah. We thank Drs. Paul Bressloff and Robert Shapley for useful
discussions, Dr. Paul Bressloff and Gina Cantone for help with data analysis, and
termination and preferentially link cortical domains of similar
Kesi Sainsbury for expert technical assistance. Dr. Niall McLoughlin participated in functional properties (Malach et al., 1993; Yoshioka et al., 1996).
some of the experiments. On the basis of laminar origin and termination, connections
Correspondence should be addressed to Alessandra Angelucci, Department of
Ophthalmology and Visual Science, Moran Eye Center, University of Utah, 50 between visual cortical areas have been classified as feedforward
North Medical Drive, Salt Lake City, UT 84132. E-mail: alessandra.angelucci@ (FF) or feedback (FB), and a hierarchical organization of cortical
hsc.utah.edu.
J.-M. Hupé’s present address: Center for Neural Science, New York University, 4
areas has been proposed previously (Rockland and Pandya, 1979;
Washington Place, New York, NY 10003. Felleman and Van Essen, 1991). V1, at the bottom of the hierar-
Copyright © 2002 Society for Neuroscience 0270-6474/02/228633-14$15.00/0 chy, receives its main FF inputs from the thalamus and sends
8634 J. Neurosci., October 1, 2002, 22(19):8633–8646 Angelucci et al. • Anatomical Circuits for Spatial Integration in V1

partially segregated FF projections to several extrastriate cortical (mrf; Barlow et al., 1967) of the neuron were initially hand-mapped
areas, which, in turn, send FB projections to V1. It has been through the dominant eye; this was then confirmed by computer map-
ping. All subsequent quantitative experiments proceeded under com-
suggested that FB connections have a less precise retinotopic puter control. The preferred orientation, direction of motion, spatial and
organization than FF projections (Perkel et al., 1986; Salin and temporal frequency, and size of stimuli of the neuron were determined.
Bullier, 1995), and that only FF inputs can drive V1 neurons, The optimal parameters for the recorded cell were then used to measure
whereas FB connections would have a modulatory influence RF and surround sizes. Stimuli were presented for 2– 4 sec within each
(Crick and Koch, 1998). FB connections may therefore represent block of trials in a randomized order, and results of four to eight repeated
blocks were averaged. To measure the spontaneous firing rate of the cell,
an additional or alternative substrate for contextual modulation interleaved blanks of the same mean luminance as the stimuli were
in V1. presented. To spatial summation data we fit f unctions representing a
The aim of this study was to provide a basis for disentangling difference of the integrals of excitatory and inhibitory Gaussian mecha-
the relative roles of inter-areal FB and intra-areal horizontal nisms (Sceniak et al., 2001) as described in detail by Levitt and L und
connections in integrating signals within and beyond the RF of (2002). From these f unctions we determined the stimulus diameter at
which responses peaked and asymptoted. Population values are ex-
V1 neurons. We reasoned that to mediate interactions within or pressed as mean ⫾ SEM. Statistical significance of laminar variation was
beyond the RF, a given connectional system must be commensu- tested with the Kruskal –Wallis test.
rate with the spatial extent of the RF or modulatory surround Combined tracer injections and physiolog ical recording. In a separate set
field of the neuron. Thus, we have compared the visuotopic scale of animals, combined anatomical and electrophysiological recording sur-
vival experiments were performed on nine adult macaque monkeys (M.
of each connectional system with the spatial extent of the classical
fascicularis or M. mulatta). Tracer injections (n ⫽ 17, all clearly confined
RF, spatial summation field (SF), and modulatory surround field to the cortical gray matter) were made in electrophysiologically charac-
of macaque V1 neurons. Our results demonstrate that monosyn- terized cortical loci between 2.2 and 7.5° eccentricity in the lower visual
aptic V1 horizontal connections are of an appropriate scale to field representation of areas V1, V2, or V3. The animals were prepared
mediate interactions within the SF and could represent an ana- as described above, intubated, and anesthetized with 0.5–2% isoflurane
in a 70:30 mixture of N2O and O2 (three animals were anesthetized with
tomical substrate for dynamic changes in SF size such as induced Sufentanil as described above). Fluids (lactated Ringer’s solution at 1–2
by stimulus contrast or scotomata (Das and Gilbert, 1995). FB ml 䡠 kg ⫺1 䡠 hr ⫺1) were continuously inf used intravenously to support
circuits from extrastriate cortex to V1, on the other hand, are of cardiovascular f unction. Monitoring of vital signs, surgery, recordings,
an appropriate scale to play an important role in global integra- and hand plotting of RFs were performed as described above. A short-
tion of visual signals and modulation of responses far beyond the duration topical mydriatic agent (cyclopentolate) was applied to the
corneas, and the eyes were fitted with contact lenses.
SF of V1 cells. Areas V2 and V3 were identified electrophysiologically using the
Parts of this work have been published previously in abstract known sequence of gray –white matter transitions as described previously
form (Angelucci et al., 1998, 2000). (Gegenf urtner et al., 1997). Corresponding retinotopic loci in V1 and V2
were identified in three animals as described by Hupé et al. (2001b).
MATERIALS AND METHODS Once the appropriate cortical sites were found, the position of the
Electrophysiolog ical recording. In a first set of animals, quantitative elec- microelectrode was recorded, the electrode was withdrawn, and the
trophysiological recording terminal experiments were performed on animal was paralyzed by intravenous inf usion of vecuronium bromide
seven adult macaque monkeys (Macaca fascicularis or M. mulatta). All (0.1 mg 䡠 kg ⫺1 䡠 hr ⫺1, with a loading dose of 1 hr). The foveas were
procedures conformed to British Home Office and United States Na- plotted, and a foveal mrf was mapped in V1 (and periodically throughout
tional Institute of Health guidelines. Animals were premedicated with the experiment) to monitor eye movements. A recording electrode glued
atropine sulfate (0.02– 0.04 mg / kg) and acepromazine maleate (0.05 to a glass micropipette (intertip distance, ⬍50 ␮m; Hupé et al., 1999)
mg / kg) and preanesthetized with ketamine (10 –30 mg / kg, i.m.). The filled with a tracer solution was lowered into the same striate or extra-
trachea and saphenous veins were cannulated; the animal was artificially striate locus where the initial recordings were performed. RF size and
ventilated with room air or with a 50:50 mixture of O2 and N2O; and eccentricity of cells through the depth of the penetration were first
anesthesia was maintained by continuous intravenous inf usion of sufen- remapped, and then the tracer was injected through the attached pipette
tanil citrate (4 – 8 ␮g 䡠 kg ⫺1 䡠 hr ⫺1). The animal’s head was fixed to a (inner tip diameter, 13–17 ␮m). The tracers used were cholera toxin
stereotaxic apparatus; a small craniotomy and durotomy were made over subunit B [C TB, low salt; List Biologic, C ampbell, CA; 1% in 0.1 M
the occipital cortex; and a tungsten-in-glass microelectrode (Merril and phosphate buffer (PB), pH 6.0] or biotinylated dextran amine (BDA,
Ainsworth, 1972) was positioned over the exposed cortex, which was then MW 3000; Molecular Probes, Eugene, OR; 10% in 0.01 M PB, pH 7.25).
covered with warm agar. To minimize eye movements, on completion of The tracers were delivered iontophoretically using 2 ␮A for C TB and 6
surgery, the animal was paralyzed by continuous intravenous inf usion of ␮A for BDA of positive current in 7 sec on – off cycles for 10 –30 min.
vecuronium bromide (0.1 mg 䡠 kg ⫺1 䡠 hr ⫺1) in lactated Ringer’s solution C TB was preferred for these studies, because its sensitive anterograde
with glucose (5.4 ml / hr). Electroencephalogram and electrocardiogram and retrograde transport reveals reciprocal connections to an injected
were monitored continuously. Peak expired C O2 was maintained near point (Angelucci et al., 1996). However, in some cases BDA was used to
4.0%, rectal temperature near 37°C, and blood oxygenation near 100%. compare the extent of the label with C TB. At the end of the injection, to
The pupils were dilated and accommodation paralyzed with topical avoid leakage of tracer along the pipette track, the pipette was left in
atropine; the corneas were protected with zero power rigid gas- place for at least 30 min and then withdrawn while reversing the current
permeable contact lenses; and the eyes were refracted. The location of to negative. In five V1 and three V2 injection cases (n ⫽ 4 animals), after
the foveas was plotted (and checked periodically throughout the exper- the tracer injection in physiologically characterized loci, additional re-
iment) on a tangent screen using a reversible ophthalmoscope. E xtracel- cordings were made of RF size and location along a few electrode
lular recordings were made in the opercular region of area V1 between penetrations made at 1 mm intervals around the injection site; recording
2 and 8° retinal eccentricity in the lower visual field. Spikes were sites were marked by electrolytic lesions. In all remaining cases (n ⫽ 5 V1
conventionally amplified and displayed and stored on a personal com- and 4 V3 injections) recordings were made only at the injection site;
puter (resolution, 250 ␮sec). Small electrolytic lesions (1–2 ␮A for 2–5 retinotopic maps in these cases could not be recorded because of lack of
sec) were made along the electrode track and later reconstructed on time and need to recover the animal. The animal was recovered from
Nissl- and cytochrome oxidase (C O)-stained tissue sections. For quan- paralysis and anesthesia, allowed to survive for 10 –20 d, and finally killed
titative studies, visual stimuli were displayed on a Barco ICD 451B color with an overdose of sodium pentobarbital (100 mg / kg, i.v.) and perf used
TV monitor driven by an AT Truevision Vista Graphics board. At a transcardially with saline followed by 4% paraformaldehyde in 0.1 M PB,
viewing distance of 114 cm, the screen subtended 13 ⫻ 13° of visual pH 7.4, for 30 min.
angle. Stimuli consisted of square patches of drifting achromatic sinusoi- Areas V1 and V2 were dissected free, flattened between glass slides,
dal gratings of average luminance of 37.5 cd /m 2. Contrast was held fixed postfixed overnight, cryoprotected by sinking in 30% phosphate-buffered
at 75% (i.e., below response saturation for most cells). sucrose, and finally sectioned on a freezing microtome at 40 ␮m tangen-
The location and size of the classical RF or minimum response field tially to the pial surface. The block containing areas V3 and MT was
Angelucci et al. • Anatomical Circuits for Spatial Integration in V1 J. Neurosci., October 1, 2002, 22(19):8633–8646 8635

similarly postfixed and cryoprotected, and then cut parasagittally. C TB

was revealed immunohistochemically using the protocol of Angelucci et
al. (1996); BDA was revealed using standard Vector Laboratories (Bur-
lingame, CA) ABC V I P-based reactions. Some (n ⫽ 3) animals received
an injection of C TB and one of BDA in corresponding retinotopic loci in
areas V1 and V2. In these brains, C TB was revealed using a standard
peroxidase – antiperoxidase method (Lanciego et al., 1998). To reveal
areal and layer boundaries, interleaved sections were stained for C O
(one in three sections of tissue containing areas V1 and V2) or for myelin
(Gallyas, 1979) or Nissl substance (one in five sections, for each method,
of tissue containing V3 and MT).
Data anal ysis. C TB and BDA anterograde and retrograde labels were
mapped in each available section using a camera lucida. Layer bound-
aries were identified and drawn by overlaying the maps of the label to
adjacent C O- or Nissl-stained sections. Areal boundaries were identified
using the pattern of C O staining (for V1 and V2) or Gallyas staining (for
V3 and MT) as well as the specific pattern of anterograde and retrograde
label. Surface-view two-dimensional (2D) composite reconstructions of
the label were made separately for each V1 and V2 layer by overlaying
maps of serial tangential sections using vascular landmarks as alignment
points. Surface-view reconstructions of the label in V3 and MT were
made from serial sagittal sections as described in detail previously (John-
son et al., 1989). Briefly, the mapped label in each section was projected
onto a line running through midlayer 4, using a radial segmentation
scheme, and serial sections were aligned using a combination of sulcal
and vascular landmarks. C ells were counted, and plots of label density
were computer-generated. T ypically, label density showed a Gaussian-
like distribution. Our definition of a labeled field included all bins with label
density within 95% of the peak density (see Fig. 4). For eight injections (five
in V1 and three in V2), the density maps of the V1 and V2 label were
overlaid to physiological maps obtained from the same animal in these two
cortical areas, using the location of electrolytic lesions as alignment points;
the visuotopic extent of the labeled fields was measured directly on the
retinotopic maps as well as estimated as described below.
In all remaining cases (n ⫽ 5 V1 and 4 V3 injection cases) in which
retinotopic maps were not recorded, the visuotopic extent of labeled Figure 1. Estimated extent of corticocortical connectional fields in visual
connectional fields was estimated as follows. All labeled fields in striate field coordinates. a, Visual field measurements. VM, HM, Vertical and
and extrastriate cortex were anisotropic in cortical space their long axis horizontal meridian, respectively. Ec, Retinal eccentricity of the center of
corresponding to the cortical area elevation axis, and to the axis of the injection site or of the labeled field determined experimentally for all
anisotropy of the magnification factor (M F), demonstrated at least for cases by electrophysiological recording. E⫹, E⫺, estimated retinal eccen-
V1 (where it runs parallel to the vertical meridian; Van Essen et al., 1984; tricity of RF center of cells at the end points of the axis of the labeled field
Tootell et al., 1988; Blasdel and C ampbell, 2001) and V2 (where it runs (see Eq. 2). Dashed circles, Mean RF size of neurons at the eccentricity of
orthogonal to the C O stripes; Roe and Ts’o, 1995). We measured the the end points of the labeled field, measured experimentally in the same
density map of each labeled field along (elevation axis) and across or different animals. D°, estimated linear visuotopic extent of the axis of
(azimuth axis) the representation of the isopolar lines of the visual field, the labeled field ( gray circle diameter). ARF, Aggregate receptive field
using as a reference published retinotopic maps of striate cortex (Van size of the labeled field’s axis, calculated as D° ⫹ mean RF size of cells at
Essen et al., 1984; Dow et al., 1985; Tootell et al., 1988) and extrastriate the end points of the labeled field. b, Cortical measurements. Xc, Cortical
cortex [V2 (Gattas et al., 1981; Roe and Ts’o, 1995), V3 (Burkhalter et location of the injection site or of the center of the labeled field. X⫹, X⫺,
al., 1986; Gattas et al., 1988), and MT (Albright and Desimone, 1987; Cortical location of cells at the end points of the labeled field’s axis. ⌬X⫹,
Maunsell and Van Essen, 1987)]. Cortical measurements were corrected ⌬X⫺, Measured cortical distance of the end points of the labeled field
for tissue shrinkage caused by histological processing. This was estimated from the center. D(mm), Measured cortical extent of the axis of the labeled
on a case-by-case basis for 12 of 17 injections using the measured in vivo field ( gray oval diameter).
distance between electrolytic lesions, between injection sites, or both. In
our hands, shrinkage caused by C TB and BDA processing ranged in
different cases between 30 –33 and 8 –13%, respectively. We also esti-
mated shrinkage caused by perf usion and cryoprotection (⬃12%) and where D(mm) (Fig. 1b) is the measured length (in millimeters) of the
applied this correction factor to the intersection distance, i.e., to the labeled field along the isoeccentricity axis, and S was estimated at the
anteroposterior dimension of the 2D maps obtained from serial sagittal injection eccentricity using the equation from Dow et al. (1981). Because
section reconstructions (see Fig. 4). For five injections in which shrinkage retinal eccentricity ( E), and thus MF and S, vary along the isopolar axis,
could not be estimated directly, we applied a 30% shrinkage correction we determined the retinotopic location of the end points of this axis of
for C TB and 10% for BDA (a possible 2–3% error in shrinkage correc- label, denoted as E⫹ and E⫺, respectively (Fig. 1a; with E⫹ ⬎ E⫺) by
tion in these cases would not have affected our results). Knowing the integrating the equations (Van Essen et al., 1984; Tootell et al., 1988)
cortical extent of labeled fields and the eccentricity of injection sites relating MF to E along the isopolar axis of V1. Thus, for example,
(recorded in all cases), to estimate the visuotopic extent of the two axes integrating equation MF( E) ⫽ aE ⫺b mm /°, from Van Essen et al. (1984),
of label, we used published equations relating MF and scatter ( S) in RF it follows that:
center position to retinal eccentricity in areas V1–V5. For labeled fields

冋 册 冋 册
1 1
within V1, we used equations from Van Essen et al. (1984) and Tootell a b⫺1 a b⫺1

et al. (1988), because these authors measured MF separately along and E⫹ ⫽ , E⫺ ⫽ ,

aE 1⫺b
c ⫺ 共 b ⫺ 1 兲 ⌬X ⫹ aE 1⫺b
c ⫹ 共 b ⫺ 1 兲 ⌬X ⫺
across the isopolar axis of V1. Because MF and S are constant along (2)
isoeccentricity contours, the linear visuotopic extent (designated D°)
(Fig. 1a) of the axis of the labeled field along these contours was where a and b are constants, Ec is the physiologically recorded eccen-
estimated as: tricity (in degrees) of the injection site or of the center of the long axis
of the label (Fig. 1a), and ⌬X⫹ and ⌬X⫺ are the measured cortical
⬚
separation (in millimeters) of the two end points (X⫹ and X⫺) from the
D (deg) ⫽ D(mm)/MF(mm /deg) ⫹ S(deg) , (1) center, Xc (Fig. 1b).
8636 J. Neurosci., October 1, 2002, 22(19):8633–8646 Angelucci et al. • Anatomical Circuits for Spatial Integration in V1

The linear visuotopic extent (D°) for the long axis of label is then given estimated D° only for the long axis of the label field, substituting in
by: Equation 1 the largest published values of M F (Gattas et al., 1981;
Albright and Desimone, 1987; Gattas et al., 1988) at the retinal eccen-
⬚
D (deg) ⫽ 共E⫹ ⫺ E⫺兲(deg) ⫹ S(deg) . (3) tricity of the injection site. The rationale for this choice was that the
largest values of M F at a given retinal eccentricity most likely reflect M F
The aggregate receptive field size (ARF) (Fig. 1a) was calculated as: values along the anisotropy axis. Using the largest values of M F might
have caused us to underestimate the extent of retrogradely labeled FB
⬚
ARF (deg) ⫽ D(deg) ⫹ mRF(deg) , (4) fields in visual field coordinates. This error would not have altered our
conclusions that the visuotopic extent of FB fields is larger than that of
where mR F is the mean RF size of cells at the end points of the axis of V1 intrinsic horizontal connections. To avoid introducing additional
label, and can reflect the mrf or summation field of the neurons (see errors, we did not attempt to estimate the visuotopic extent of FB fields
Results). We used our own physiological measures of RF sizes appro- along their shorter axis or their visual field anisotropy; thus, the retro-
priate for the end points eccentricity and cortical layer location. These gradely labeled fields of cells of origin of FB connections in extrastriate
were measured either in the same animal in which the tracer injection cortex are represented in visual space as circles (see Fig. 7a) rather than
was made or in a separate set of quantitative physiological experiments ovals (as are the fields of V1 horizontal connections, or of terminal FB
performed in different animals (see above). connections inV1) (see Figs. 6a, 7a, 8a,b). To estimate ARFs in V2 and
Below we provide a detailed example of our estimates of the visuotopic V3, we used our own measures of mrf and summation field sizes (Levitt
extent of the injection site and resulting labeled horizontal connections et al., 1994; Gegenf urtner et al., 1997; this study). MT RF sizes were
for the case in Figure 6a. The V1 injection was made at 6.5° eccentricity taken from studies by Albright and Desimone (1987) and Maunsell and
in the lower visual field, 4° from the vertical meridian. D° along the Van Essen (1987). Estimated sizes of D° and ARF were consistent with
anteroposterior (AP; i.e., isoeccentricity) axis of V1 was estimated those determined physiologically (see Fig. 8a).
substituting in Equation 1 the following values: D(mm) ⫽ 1.13 mm
(measured cortical extent of injection site AP axis) or 4.34 mm (mea-
sured cortical extent of horizontal connections AP axis); MF (at E ⫽ RESULTS
6.5°) ⫽ 1.3 mm /° (using equation MF ⫽ 13E ⫺1.22 mm /°; Van Essen et al., In a first set of single-unit recording experiments, we determined
1984); and S (at E ⫽ 6.5°) ⫽ 0.16° (using equation S ⫽ 0.314 ⫻ mrf size
quantitatively the spatial dimension of the RF and modulatory
⫺0.86 min; Dow et al., 1981).
D° of the injection site and resulting horizontal connections along the surround field of V1 cells. In a second set of combined anatomical
mediolateral (ML; i.e., isopolar) axis of V1 was estimated substituting in and physiological experiments, we determined the visuotopic
Equations 2 and 3 the following values: a and b ⫽ 11.7 and 1.01 extent of V1 horizontal connections and of feedback connections
(constants from equation MF ⫽ 11.7E ⫺1.01 mm /°; Van Essen et al., 1984); from extrastriate cortex to V1 and compared them with V1 cells’
Ec ⫽ 6.5° (physiologically recorded E of injection site); ⌬X⫹ and ⌬X⫺ (for
injection site) ⫽ 0.55 mm (measured cortical extent of injection site ML
receptive field and surround field sizes measured in the previous
radius); ⌬X⫹ and ⌬X⫺ (for horizontal connections) ⫽ 2.1 and 3.1 mm set of experiments. The two sets of experiments were performed
(measured cortical separation of f urthest medial and lateral labeled in different animals but in the same region of visual space (2– 8°
points, respectively, from the injection center); and S (at E ⫽ 6.5°) ⫽ retinal eccentricity in the lower visual field representation of V1).
0.16°.
From Equation 2 we obtained E⫹ and E⫺ (for injection site) ⫽ 6.8 and Spatial extent of V1 neuron receptive field and
6.2°, and E⫹ and E⫺ (for horizontal connections) ⫽ 7.8 and 4.97°. From modulatory surround field
Equations 1 and 3 we obtained D° (for injection site) ⫽ 1.03° (AP axis) ⫻
0.78° (ML axis), and D° (for horizontal connections) ⫽ 3.5° (AP axis) ⫻ Area summation curves were measured for 59 neurons sampled
3° (ML axis). from all layers of macaque V1 (n ⫽ 18 in layers 2/3; n ⫽ 24 in layer
The ARF size of the V1 injection site in Figure 6a was calculated as 4; and n ⫽ 17 in layers 5/6) between 2 and 8° eccentricity. Of these
follows: ARF of AP axis ⫽ D° of injection (1.03°) ⫹ mean RF size in V1 cells, 69% had complex RFs; the rest were simple cells (for a
layer 2/3 at 6.5° eccentricity [mrf ⫽ 0.55°; high- and low-contrast sum- detailed report of these data, see Levitt and Lund, 2002). A
mation field (SF ) ⫽ 1.15 and 2.65°, respectively]. ARF of ML axis ⫽ D°
of injection (0.78°) ⫹ mean RF size/2 in V1 layers 2/3 at E⫹ (6.8°) high-contrast (75%) drifting grating patch of optimal stimulus
eccentricity (mrf ⫽ 0.56°; high- and low-contrast SF ⫽ 1.18 and 2.7°, parameters for the recorded neuron was centered over the
respectively) ⫹ mean RF size/2 in V1 layers 2/3 at E⫺ (6.2°) eccentricity computer-mapped RF of the cell, and its diameter was systemat-
(mrf ⫽ 0.54°; high and low contrast SF ⫽ 1.13° and 2.6°, respectively). RF ically increased. We measured response amplitude as a function
sizes in this case were obtained from our own equations relating RF size
to eccentricity in the different layers of V1 and derived from a separate
of stimulus diameter. Typically, responses increased with patch
set of quantitative physiological experiments (see above). size to a peak and either asymptoted at the peak or showed
The above estimates were obtained using M F values from Van Essen response suppression as stimulus size was further increased (i.e.,
et al. (1984). Although M F values reported by Tootell et al. (1988) and surround suppression) (Fig. 2a). We took as a measure of RF size
Blasdel and C ampbell (2001) tend to be slightly larger than those of Van the smallest stimulus diameter at peak response (Fig. 2a; we
Essen et al (1984), applying M F values from these other authors to the
above estimates yielded only slightly smaller values of D° and ARF size. included in this analysis only cells that showed surround suppres-
Thus, for example, the aggregate high-contrast SF size of the V1 injec- sion). We refer to this measure of RF size as the high-contrast SF.
tion (along the ML axis) in Figure 6a measured 1.9° using the M F of Van SFs for our sample of V1 cells averaged 1.0 ⫾ 0.1° (Fig. 2b),
Essen et al. (1984) but was 1.8° using the M F of Tootell et al. (1988). The increased with retinal eccentricity (Fig. 2c, middle function), and
ratio of D° of lateral connections to the aggregate high-contrast SF size
of the V1 injection was 1.6 (for the ML axis) using the M F from Van
showed no statistically significant variation across cortical layers
Essen et al. (1984) and 1.3 using the M F from Tootell et al. (1988). Thus, despite a trend for the largest SFs to be found in deeper layers.
using M Fs in the literature from different authors, we observed minimal These results on layer differences in SF sizes are consistent with
differences in our estimates. those from other studies (Sceniak et al., 2001; Cavanaugh et al.,
For labeled fields within the central 5° of V2, we used M F and S values 2002; Levitt and Lund, 2002). RF size has been shown to vary
from Roe and Ts’o (1995), because they reported separate measures of
M F along and across C O stripes. Published measurements of M F in depending on the method and the stimulus contrast used to
more peripheral V2 (Gattas et al., 1981) and in V3 (Gattas et al., 1988) measure it. Figure 2c compares RF size as a function of eccen-
and MT (Albright and Desimone, 1987; Maunsell and Van Essen, 1987) tricity using three different test conditions. Figure 2c, middle
are averaged across isopolar and isoeccentricity axes, thus not taking into function, shows our measure of high-contrast SF size as a function
account possible anisotropies in M F. Similar to V1 and V2, anatomical
anisotropies of labeled connectional fields in areas V3 and MT likely
of retinal eccentricity. Figure 2c, bottom function, shows data
reflect anisotropies in M F within these areas. Thus, for the label in more from Dow et al. (1981), who measured RF size by moving a
peripheral (⬎5°) V2 and for all labeled fields in V3 and MT, we high-contrast bar of light and hand drawing the contours of the
Angelucci et al. • Anatomical Circuits for Spatial Integration in V1 J. Neurosci., October 1, 2002, 22(19):8633–8646 8637

Figure 2. Extent of RF and surround field for a population of macaque V1 cells. a, Response of a representative V1 neuron to an optimal high-contrast
grating patch of increasing diameter (top right symbol ). Patch diameter at peak response (left arrow) was taken to be the size of the SF of the cell in b
and c (middle function). Patch diameter at asymptotic response (right arrow) was taken to be the size of the surround field of the cell in d and e. b,
Distribution of SF diameters for a population of V1 neurons (n ⫽ 59), measured as in a. Arrowhead, Mean. c, RF size as a function of retinal eccentricity
measured under three different test conditions, each one indicated by symbols to the right of each line. Straight lines are regression lines. Middle function,
SFs measured using expanding high-contrast (75%) gratings; data from this study (n ⫽ 59 cells). Bottom function, Hand-mapped mrf; based on data from
Dow et al. (1981). Top function, SFs measured using expanding low-contrast gratings; based on data from Sceniak et al. (1999) and obtained by
multiplying our high-contrast SF function (middle function) by 2.3. Stars, Means. d–f, Distributions of surround field diameters for a population of V1
cells measured under three different test conditions, each one indicated by top right symbols. d, Expanding high-contrast optimal grating stimulus,
including only cells with suppressive surrounds (n ⫽ 59, same cells as in b; note different scale on x-axis in b and d). e, Optimal center grating stimulus
surrounded by expanding most suppressive grating stimulus (n ⫽ 30, subset of cells in b and d). f, Optimal center grating and most suppressive surround
grating stimuli plus blank annulus expanding in the surround (n ⫽ 30 cells, same cells as in e). Arrowheads in d–f, Means.

area of visual space that elicited spikes from the neuron. This of surround suppression has been shown to be highest in the
measure of RF size is commonly known as mrf or “classical” RF region abutting the RF center (Walker et al., 1999), by blanking
(Barlow et al., 1967). Our mean high-contrast SFs were 2.2-fold out the region of maximal surround strength (as in Fig. 2f ), we
greater than the mean mrf sizes of Dow et al. (1981). Further- aimed at revealing the most remote surround influences. As
more, RF size depends on stimulus contrast (Kapadia et al., 1999; surround outer diameter increased, responses decreased; as an-
Sceniak et al., 1999). Sceniak et al. (1999) found that, for the same nulus outer diameter increased, responses increased. Surround
V1 cells, SFs were on average 2.3-fold greater when measured and annulus outer diameters at which responses asymptoted were
using low-contrast rather than high-contrast gratings (Fig. 2c, top taken as measures of surround sizes (Fig. 2a,f, respectively).
function). Under all test conditions, surround diameters were found to
For the same V1 cells, we also measured the extent of the extend up to and ⬎13°. However, different distributions and mean
modulatory surround field using three different methods (Fig. values [mean, 5.0 ⫾ 0.6° (Fig. 2e) and 7.1 ⫾ 0.2° (Fig. 2f )] were
2d–f ). Figure 2d shows the distribution of surround sizes mea- obtained with the different test methods, reflecting the fact that
sured for 59 neurons using expanding high-contrast gratings. As the surround region is most suppressive close to the RF. Thus,
stimulus size increased beyond the high-contrast SF of the cells, masking out the most suppressive near surround, as in Figure 2f,
responses decreased. Surround size was defined as the smallest revealed more distant influences, whereas the latter were masked
stimulus diameter at which the response of the neuron asymp- by optimally stimulating the most suppressive near-surround re-
toted (Fig. 2a). Surround sizes ranged between 1.2 and ⬎13° (13° gion, as in Figure 2e. Surround sizes did not vary significantly with
was the largest stimulus diameter we could produce on our retinal eccentricity or cortical layer. For each neuron, we calcu-
monitor), averaging 5.1 ⫾ 0.6°. For a subset (n ⫽ 30) of these lated a “relative” surround extent as the ratio of surround field
cells, surround sizes were measured using two additional meth- diameter (measured as in Fig. 2d) to the high-contrast SF diam-
ods. A center optimal grating stimulus was confined to a central eter; the population average was 4.6 ⫾ 0.7.
region the size of the high-contrast SF of the cell and was To summarize, in V1 at 2– 8° eccentricity, surround field sizes
surrounded by the most suppressive grating stimulus configura- were on average 4.6 (up to ⬎13) times larger than high-contrast
tion (usually, but not always, a grating at the same orientation as SF sizes and at least 10 (up to ⬎27) times larger than the mean
in the center). We then systematically varied either the outer mrf size of V1 cells.
diameter of the surround stimulus from 0° (i.e., central stimulus Several other recent studies have examined the spatial summa-
alone) to 13° (diameter of the display screen) (Fig. 2e) or of a tion properties of macaque V1 neurons (Sceniak et al., 2001;
blank annulus introduced between the central stimulus and a Cavanaugh et al., 2002). In these studies, center and surround
full-field (13°) surround stimulus (Fig. 2f ). Because the strength responses were modeled as independent, spatially overlapped
8638 J. Neurosci., October 1, 2002, 22(19):8633–8646 Angelucci et al. • Anatomical Circuits for Spatial Integration in V1

Table 1. Cortical extent and anisotropy of V1 lateral connections and of

feedback connections to V1

Anisotropy
Connections Cortical layer Long axisa ratiob
V1 lateral 2/3 (n ⫽ 10) 6 ⫾ 0.7 (3–9) 1.56 ⫾ 0.12
4B/4C␣ (n ⫽ 8) 6.7 ⫾ 0.7 (4.7–10) 1.5 ⫾ 0.1
5/6 (n ⫽ 3) 7.9 ⫾ 1.6 (6.3–9.5) 1.76 ⫾ 0.2

FB in V2 2/3A (n ⫽ 5) 6.1 ⫾ 0.6 (4.6 – 8.3) 3.4 ⫾ 0.8

5/6 (n ⫽ 6) 6.4 ⫾ 1.2 (4 –9.4) 3.9 ⫾ 1.0
FB in V3 2/3A (n ⫽ 5) 5.2 ⫾ 1.2 (2.7– 8.1) 3.3 ⫾ 0.6
5/6 (n ⫽ 5) 7.9 ⫾ 1.2 (4.5–9.8) 3.2 ⫾ 0.2
FB in MT 2/3A (n ⫽ 1) 4.5 2.8
5/6 (n ⫽ 2) 8.9 ⫾ 2.1 (6.8 –10.9) 2 ⫾ 0.4

FB in V1
from V2 all (n ⫽ 3) 6.8 ⫾ 0.4 (6.4 –7.6) 2 ⫾ 0.1
FB in V1
from V3 all (n ⫽ 3) 13.4 ⫾ 0.5 (12.9 –13.9) 2 ⫾ 0.2

All values are mean ⫾ SEM. Values in parenthesis are minimum and maximum; n,
number of labeled connectional fields in the layer.
a
Extent (D(mm) ; see Fig. 1b) of labeled field long axis (i.e., along the cortical area’s
elevation axis).
b
Extent of long axis/extent of short axis of labeled field.

Figure 3. Patchy lateral (or horizontal) connections in layers 2/3 of

macaque area V1. A surface view 2D composite reconstruction of CTB- Sceniak et al. (2001), although our model parameters are
labeled connections is shown. The labeled field axes measured 9 ⫻ 6 mm. somewhat larger than those reported by C avanaugh et al.
Black oval, CTB uptake zone; blank annulus, region of heavy label. Note
anisotropic distribution of overall label. The foveal representation is (2002).
toward the bottom (lateral V1); the V1–V2 border is to the right (anterior Cortical extent and patterns of horizontal and
V1). Small square, Labeled patch shown at higher power in the inset. Scale
bar, 500 ␮m (corrected for 30% shrinkage). Inset, High-power drawing of
feedback connections
patch in the small square, showing labeled fibers and somata (dots), CTB (n ⫽ 8) or BDA (n ⫽ 2) injections (uptake zone diameters,
indicating reciprocity of connections. Scale bar, 100 ␮m. 0.27–1.2 mm) were made in physiologically characterized V1 loci
at different cortical depths (n ⫽ 2 in layers 1–3, 5 in layers 1– 4C,
excitatory and inhibitory mechanisms, each with a Gaussian spa- 1 in layers 1–5, and 2 in layers 1– 6) between 2.5 and 7.5° eccen-
tial sensitivity profile and with the inhibitory mechanism being tricity in the lower visual field representation. Consistent with
broader than the excitatory one (also see DeAngelis et al., 1994). previous results obtained with different anatomical tracers (Rock-
Cavanaugh et al. (2002) found that a ratio of Gaussian model was land and Lund, 1983; Yoshioka et al., 1996), CTB or BDA
the best fit to their data, whereas Sceniak et al. (2001) favored injections in macaque V1 layers 2/3 produced patches of terminal
a difference of Gaussian (DOG) model. The spatial spread of label surrounding the injected V1 column (Fig. 3). CTB addition-
the center and surround mechanisms in these studies was ally retrogradely labeled cell bodies (but not fibers) within each
estimated directly, and in one study (Sceniak et al., 2001) patch, indicating the reciprocal nature of these connections (Fig.
exclusively, from the fitted curves. In the present study, we fit 3, inset). Reciprocal lateral connections were also labeled in layers
our spatial summation data to a DOG model and used the fits 4B/upper 4C␣ and 5/6, when the tracer injection involved these
mainly to derive robust estimates of SF and surround field V1 laminas. Both tracers revealed different patterns of label in
sizes. Because the parameters derived from the Gaussian sen- these layers: bar-shaped fields in 4B/upper 4C␣ (Asi et al., 1996;
sitivity f unctions depend strongly on assumptions about the Angelucci et al., 2002) and a less patchy, more diffuse label in 5/6
mechanisms underlying center-surround interactions that may (Rockland and Knutson, 2001). The labeled fields of lateral
not be valid, we chose to report empirical measurements of SF connections in all V1 layers were anisotropic in cortical space
and surround sizes. However, because the DOG model is a (Fig. 3). In layers 2/3, the longer axis of label (D(mm)) (Fig. 1b),
good descriptor of our summation data, and to allow for known to extend orthogonal to the ocular dominance domains
comparison with previous studies, we also derived from the (Yoshioka et al., 1996), measured on average 6 ⫾ 0.7 mm (ex-
fitted curves the Gaussian spread (radius) of the excitatory and tending up to 9 mm). The distance from the edge of the tracer
inhibitory components (Sceniak et al., 2001) (for details, see uptake zone to the farthest labeled cell averaged 2.9 ⫾ 0.4 mm.
Levitt and L und, 2002). The population means were 1.2° for The most distant labeled cells were consistently located laterally
the excitatory radius and 2.7° for the inhibitory radius, reveal- to the injection site, i.e., toward the foveal representation of V1
ing a somewhat larger mean RF center mechanism than our (Fig. 3). Furthermore, the mean anisotropy ratio (extent of long/
empirical measurements of high-contrast SF size. The width of short axis) of CTB-labeled layer 2/3 lateral connection fields was
the surround inhibitory mechanism instead agreed well with 1.56 ⫾ 0.1, closely matching the anisotropy ratio (1.6) of the V1
our empirical measurements of surround size as described in magnification factor in these layers due to the ocular dominance
Figure 2, d and e. These results are consistent with data from domains (Blasdel and Campbell, 2001). The latter two observa-
Angelucci et al. • Anatomical Circuits for Spatial Integration in V1 J. Neurosci., October 1, 2002, 22(19):8633–8646 8639

dense (Barone et al., 2000) and significantly ( p ⬍ 0.01) less

extensive than in layers 5/6 (Fig. 4b), and decreased in density and
spatial extent with distance from V1, whereas the label in the
lower layers increased in spatial extent (Table 1, middle). The
density of labeled cells within the FB fields gradually declined
with distance from a denser center core region. The retrograde
label appeared clustered in the upper layers and showed fluctu-
ations in cell density in the lower layers. Labeled FB fields in
extrastriate cortex were anisotropic in cortical space (Fig. 4b;
Table 1, middle), their long axis following the overall anisotropy
of visual field representation of the cortical area. Thus, in V2,
the longer axis of the label extended orthogonal to the C O
bands, and in all areas was approximately parallel to the longer
(i.e., elevation) axis of the cortical area itself. E xtent and
anisotropy ratios for retrogradely labeled FB connections in
the upper and lower layers of extrastriate cortex are reported
in Table 1, middle.
Our V1 injections were either confined to the CO interblob
columns (n ⫽ 2 injections through layers 1–3, and n ⫽ 1 injection
through layers 1– 4B) or involved both CO blob and interblob
compartments (n ⫽ 7). Resulting retrograde FB label involved all
V2 stripe compartments, even in cases in which the V1 injection
was clearly confined to an interblob column. The anterograde
label in V2 (terminals of feedforward axons from V1) arising
from these same V1 injections in the interblob columns (n ⫽ 3)
Figure 4. Cells of origin in area V3 of feedback connections to V1. a, was more focused than the retrograde (FB) label and was either
Micrograph (left) of a sagittal section through dorsal V3 (shaded box on confined to the pale CO stripes (n ⫽ 1; V1 injection in layers 1–3)
the right shows location of the photographed region on the annectant or involved both the thick and pale stripes (n ⫽ 2). There was no
gyrus), showing CTB-labeled cell bodies (arrows) in layers 2/3A and 5/6. obvious difference in extent between FB fields in extrastriate
Cortical layers are indicated at the bottom; WM, White matter; arrowhead,
cortex labeled by V1 injections involving CO blob or interblob
labeled fibers in layers 4 and 3B (terminals of feedforward connections
from V1). The composite surface map for this case is shown in b. The columns.
injection site involved V1 layers 1– 4C and was made at 6.5° eccentricity in We also determined the extent of the divergence region of FB
the lower visual field (same injection case as in Figs. 6a, 7a). D, Dorsal; P, connections to V1. This is the V1 region containing terminals of
posterior. Scale bar, 100 ␮m. b, Surface view plots of cell label density in FB axons, anterogradely labeled by small tracer injections in
the upper (left) and lower (right) layers of dorsal V3, generated using
custom software written in Matlab. Color scale represents cell density extrastriate cortex. CTB or BDA injections (uptake zone diam-
(numbers are cells per 500 ␮m 2). Bins containing ⬍5% of peak cell eter, 300 –1500 ␮m) were made in physiologically characterized
density were removed from the image. Label anterior to the crown of the loci in areas V2 (n ⫽ 2 CTB and 1 BDA injections) or dorsal V3
annectant gyrus ( purple triangles in a, b) is in area V3A. Purple squares (in (n ⫽ 4 CTB injections) (Fig. 5a) between 2.2 and 6.5° eccentricity
a, b), Location of the fundus of the lunate sulcus. The long axis of these
in the lower visual field. To investigate the retinotopic organiza-
feedback fields measured 7.8 mm in the upper layers and 9.8 mm in the
lower layers. The visual field map of the lower-layer feedback field is tion of FB connections to V1, all the V2 injected cases (n ⫽ 3)
shown in Figure 7a. M, Medial (away from the fovea); P, posterior (toward received a second injection of a different tracer (BDA or CTB) in
the V1–V2 border). Scale bar, 1 mm (corrected for 30 and 12% shrinkage V1 at the same retinal eccentricity as the V2 injection. Upper-
in the anteroposterior and mediolateral axis, respectively). layer injections in V2 or V3 produced large fields of patchy
terminal and cell body label in V1 layers 2/3 and 4B (Fig. 5b).
tions suggest that lateral connections follow the overall anisotropy Injections involving all V2 or V3 cortical laminas resulted in even
of visual field representation in V1. CTB and BDA injections larger labeled fields within V1 and additionally produced terminal
produced similar results. We found no statistically significant and sparse cell body label at the layer 5/6 border of V1. The layer
difference in the extent or anisotropy ratio between lateral con- 5/6 label appeared less clearly patchy than in the layers above.
nections in different V1 layers, despite a trend for connections to Labeled patches in different V1 layers arising from the same
be more extensive and more anisotropic in the deeper layers. The extrastriate injection were vertically aligned. Anterograde and
extent and anisotropy ratio for lateral connections in different V1 retrograde labels overlapped in the V1 patches, indicating the
laminas are reported in Table 1, top. reciprocal nature of feedforward and feedback connections to and
The same V1 injections (n ⫽ 10) used to determine the extent from V1. FB fields within V1 arising from tracer injections in
of intra-areal V1 lateral connections labeled retrogradely the cells V2 or V3 greatly exceeded the size of the intra-areal V1 fields
of origin of FB connections to V1 in extrastriate cortex (Fig. 4a). produced by similarly sized V1 injections, the V3 injections
We confined our analysis to FB from areas V2, V3, and MT. labeling larger fields in V1 than the V2 injections (Table 1,
Small (⬃300-␮m-diameter) tracer injections confined to V1 layers bottom). The FB fields in V1 were anisotropic in cortical space
1–3 retrogradely labeled extensive fields of somata in the super- (Fig. 5c), their longer axis extending parallel to the V1–V2
ficial (2/3A) and deeper (at the 5/6 border) layers of area V2; border when near that border; their mean anisotropy ratio was
injections involving layers 1– 4B or 1– 6 additionally labeled cells greater than that of V1 intra-areal lateral connections (Table
in the upper and lower layers of areas V3 and MT. Within each 1,bottom).
extrastriate area, the retrograde label in the upper layers was less Our V2 injections (one confined to a pale CO stripe and two
8640 J. Neurosci., October 1, 2002, 22(19):8633–8646 Angelucci et al. • Anatomical Circuits for Spatial Integration in V1

Figure 6. Visuotopic extent of V1 lateral connections. a, Visual field map

of a representative CTB injection site and resulting labeled lateral con-
nections in V1 layers 2/3. The injection site was in the lower visual field
representation of V1 at 6.5° eccentricity, 4° from the vertical meridian
(VM ). HM, Horizontal meridian. D° of the V1 connectional field (dashed
gray oval, 3 ⫻ 3.5°) and ARF size of the V1 injection site (black ovals)
were estimated as detailed in Materials and Methods. The black ovals
represent ARF sizes computed using three different measures of RF size,
each indicated by symbols as in Figure 2c (aggregate mrf, 1.3 ⫻ 1.6°;
aggregate high-contrast SF, 1.9 ⫻ 2.2°; aggregate low-contrast SF, 3.4 ⫻
3.7°). b, Histogram of the population means (n ⫽ 21) of the relative
visuotopic extent of labeled V1 lateral connections along the isopolar
(black bars) and isoeccentricity (hatched bars) axes of the labeled fields.
Data from all layers are pooled together. The visuotopic extent is ex-
pressed as the ratio of D° of V1 connections to the ARF size of neurons
at the V1 injection site and is shown for each of three different methods
of measuring RF (and thus ARF) size (symbols on x-axis, as in Fig. 2c).
The trend for ratios to be smaller along the isoeccentricity axis of the field
was not statistically significant. Error bars indicate SEM. The dashed
horizontal line marks a ratio of 1.

(Angelucci and Levitt, 2002). We observed no obvious difference

in extent between FB fields terminating in different CO compart-
ments of V1.
Visuotopic extent of horizontal and
Figure 5. Patchy terminal label of feedback connections in layer 4B of V1 feedback connections
arising from a tracer injection in area V3. a, Micrograph of a sagittal Cortical measurements of intrinsic V1 and inter-areal FB fields
section through dorsal area V3, showing a CTB injection site involving all
cortical layers (layer 1 is involved in the injection but not in this specific were converted into visual field coordinates and related to the
section). The injection was made at 6.4° eccentricity in the lower visual spatial extent of the receptive and surround fields of V1 cells.
field. 1, Layer 1; WM, white matter; AG, annectant gyrus. Scale bar, 200 Here we use two visuotopic measures: (1) D° is the extent, in
␮m (corrected for 30% shrinkage). b, Micrograph showing a surface view degrees of visual angle, of the axis of the connectional field,
of CTB-labeled terminals and cell bodies (arising from the injection site
derived from the cortical retinotopic map (see Eqs. 1, 3) (Fig. 1a);
in a) in a single tissue section cut tangentially through V1 layer 4B. c, 2D
composite serial tangential section reconstruction of anterograde (i.e., its extent is independent of RF size; and (2) ARF is the cumu-
feedback) terminal label through the whole thickness of layer 4B. Arrow- lative RF size of all labeled neurons in a given labeled region;
heads in b and c point to the same two patches. Note anisotropic distri- thus its extent is dependent on the method and stimulus contrast
bution of overall label. The axes of the labeled field measured 13.8 ⫻ 8.1 used to measure RF size (see Eq. 4) (Fig. 1a). The visuotopic
mm. The visual field extent of this layer 4B-labeled field is represented as
a gray oval in Figure 8b. Scale bar, 1 mm, for b and c (corrected for 30% extent (D°) of lateral connections is shown for a representative
shrinkage). Medial, Away from the fovea; Anterior: toward the V1–V2 case in Figure 6a. D° of CTB-labeled layer 2/3 lateral connections
border. (dashed gray oval ) is shown centered onto three different esti-
mates of the ARF size of the V1 injection (the three black ovals
involving both thick and pale stripes) produced terminal and cell indicate aggregate mrf and aggregate high- and low-contrast SF,
body labels in the interblob columns of V1 (consistent with results respectively). D° of V1 lateral connections in this case was 2.2-fold
of Sincich and Horton, 2002). Terminal (FB) label in V1 arising greater than the aggregate mrf of the connections cells of origin
from V3 injections was confined either to the CO blob or to the and closely matched the aggregate low-contrast SF size of the
interblob columns; larger injections (⬃1.5 mm in diameter) cells of origin. Across the population, the monosynaptic spread
labeled both V1 compartments. These observations suggest (D°) of V1 horizontal connections averaged 2.47 ⫾ 0.3° in extent.
that, similarly to corticocortical projections between V1 and V2 V1 injection sites ranged in size between 0.2° and 0.8° (D°) or 0.5°
(Sincich and Horton, 2002), the connections between V1 and V3 and 1.3° (aggregate mrf). Because our interest lay in determining
form two parallel, segregated pathways, one related to the CO whether lateral connections extend beyond the limits of V1 cells
blob columns of V1 and the other one related to the interblobs RFs, we calculated a “relative” visuotopic extent for these con-
Angelucci et al. • Anatomical Circuits for Spatial Integration in V1 J. Neurosci., October 1, 2002, 22(19):8633–8646 8641

Table 2. Visual field extent and anisotropy of V1 lateral connections the elevation axis of the connections. In addition, the mean
anisotropy ratio of their visual field extent (ARF size along the
long axis/ARF size along the short axis of the labeled field)
approached 1. There was no statistically significant difference in
anisotropy ratio across cortical layers, despite a tendency for
connections to be more anisotropic in layers 5/6 (Table 2).
We then asked whether the dimensions of feedback connec-
tions from extrastriate cortex to V1 are commensurate with the

Ratios are shown for ARFs computed using three different measures of RF size
(symbols on the left; as in Fig. 2c). All values are mean ⫾ SEM. Values in
parenthesis are minimum and maximum; n, same as in Table 1, V1 lateral.
1
D° along the labeled field’s long axis/ARF size of the injection site.
2
ARF size along the labeled field’s long axis/ARF size of the injection site.
†
ARF size along long axis/ARF size along short axis of labeled field.

nections as the ratio of the visuotopic extent of the connectional

field (either D° or ARF size) to the ARF size of its cells of origin.
Figure 6b shows population means of the relative visuotopic
extent of V1 connectional fields across all cortical layers (n ⫽ 21
connectional fields). Mean population values for the different V1
layers are shown in Table 2. Ratios in Figure 6b and Table 2 are
additionally shown for each of three different methods of mea-
suring RF size (mrf and high- and low-contrast SF). On average,
across the population, the monosynaptic spread (D°) of V1 lateral
connections was approximately three times larger than the mrf
and approximately two times larger than the high-contrast SF of Figure 7. Visuotopic extent of retrogradely labeled fields of cells of
their cells of origin but was commensurate with the low-contrast origin of FB connections in extrastriate cortex. a, Visual field map of FB
SF of the cells. Therefore, these connections could account mono- fields of neurons in layers 5/6 of areas V2 (top left), V3 (middle right), and
synaptically for the apparent expansion of the SF at low contrast MT (top right) labeled by a CTB injection through V1 layers 1– 4C at 6.5°
eccentricity (same injection case as in Fig. 6a). Visual field maps of V1
(Kapadia et al., 1999; Sceniak et al., 1999). However, because V1 lateral connections in layers 2/3 (bottom left) and 4B (bottom right) labeled
neuron surround fields were on average approximately five (up to by the same V1 injection are also shown. Gray circles, D° of the connec-
⬎13) times larger than the high-contrast SF of the neurons (see tional fields. Black ovals, aggregate mrf size of neurons at the V1 injection
above), V1 horizontal connections are significantly less extensive site (1.3 ⫻ 1.6° in layers 2/3; 1.1 ⫻ 1.2° in layer 4B). Dashed black circles,
mean mrf size of cells at the edge of labeled fields. The aggregate mrf size
than the mean surround size of V1 cells. Comparison of the mean of each connectional field is the sum of the diameter of the gray circle plus
visuotopic extent of these connections (2.47°) with the mean the diameter of one dashed black circle. This was estimated as described
Gaussian spread (diameter) of the excitatory center (2.4°) and in Materials and Methods and measured 3.5 ⫻ 4.1° (V1 layers 2/3
inhibitory surround (5.4°) mechanisms also revealed that the horizontal connections), 4.1 ⫻ 4.8° (V1 layer 4B horizontal connections),
monosynaptic spread of horizontal connections is too small to 6.1° (V2 FB), 8.7° (V3 FB), and 23.6° (MT FB). The aggregate mrf of
retrogradely labeled neuronal FB fields in the upper layers of extrastriate
account for mean surround size and is instead commensurate cortex (data not shown) measured 5.4° (V2), 7.6° (V3), and 15.3° (MT).
with the size of the RF center mechanism. Scale bar, 2°. b, Histogram of the population means of the relative
Despite being anisotropic in cortical space (Fig. 3; Table 1, visuotopic extent of labeled layer 5/6 FB fields (black bars) in areas V2
top), V1 lateral connections covered isotropic regions of visual (n ⫽ 6), V3 (n ⫽ 5), and MT (n ⫽ 2), arising from the same V1 tracer
injections. The visuotopic extent is expressed as the ratio of the aggregate
space (Fig. 6a; Table 2). The ratio of D° of the foveal half of the mrf size of the FB field along its long axis to the aggregate mrf size of
connections to D° of their peripheral half averaged 0.94 ⫾ 0.5, neurons at the V1 injection site. White bar, Mean aggregate mrf ratio (3.3 ⫾
indicating that visual space is represented symmetrically along 0.24) for V1 lateral connections (n ⫽ 21). Note cut on the y-axis scale.
8642 J. Neurosci., October 1, 2002, 22(19):8633–8646 Angelucci et al. • Anatomical Circuits for Spatial Integration in V1

scale of V1 neuron modulatory surround fields. Because the cells Table 3. Visual field extent of feedback neuron fields in extrastriate
of origin of FB connections in extrastriate cortex and of lateral cortex
connections in V1 were labeled by the same V1 tracer injections
(n ⫽ 10), we were able to directly compare the extent of the visual
field region that these two different connectional systems convey
to the same V1 column. Figure 7a shows an example of the
visuotopic extent of retrogradely labeled fields of cells of origin of
FB connections in layers 5/6 of extrastriate cortex. The visuotopic
extent of V1 horizontal connections to the same injection site is
also shown for comparison. The aggregate mrf sizes of the FB
fields in V2, V3, and MT were 4.6-, 7.7-, and 21-fold larger,
respectively, than the aggregate mrf size of neurons at the V1 Ratios are shown for ARFs computed using two different measures of RF size (mrf
injection sites. In comparison, the aggregate mrf sizes of V1 and high contrast SF; symbols on the left). Other conventions as in Table 2; n, same
intra-areal lateral connections in layers 2/3 and 4B were only 2.7 as in Table 1, FB in V2, V3, and MT.
1
ARF size along the labeled field’s long axis/ARF size of the injection site.
and 3.7 times larger than the aggregate mrf size of the same V1
injection. Across the population, aggregate mrf sizes of retro-
gradely labeled FB fields in the lower layers of extrastriate cortex those of the terminal FB fields in V1 layers 4B and 5/6 labeled by
averaged 3.8 ⫾ 0.6° (in V2), 6.7 ⫾ 0.7° (in V3), and 26.6 ⫾ 3° (in the V3 injection were 1.1 and 1, respectively. Across the popula-
MT); those of V1 horizontal connections across all layers instead tion, the visuotopic extent (D°) of FB terminal fields in V1 labeled
averaged 2.9 ⫾ 0.4. Figure 7b shows population means of the by V2 or V3 injections averaged 3.42 ⫾ 1.2° (D° of the injection
relative visuotopic extent of retrogradely labeled FB fields in the sites ranging between 0.3 and 2.7° and the aggregate mrf between
lower layers of areas V2, V3, and MT and, for comparison, of V1 1.5 and 7.2°). Figure 8c and Table 4 show population means of the
lateral connections. Relative visuotopic extent values are shown relative visuotopic extent (D° or aggregate mrf of FB field in
in Table 3 for each of two different methods of measuring RF V1/aggregate mrf of extrastriate injection site) of anterogradely
size. These results indicate that the region of visual space con- labeled FB fields across all layers of V1. These results indicate
veyed by FB connections from extrastriate cortex to V1 is larger that the aggregate mrf of FB terminal fields within V1 arising
than that conveyed by horizontal connections to the same V1 from V2 or V3 injections is commensurate with the aggregate mrf
column. Furthermore, such visual space region increases with of FB neurons at the injected site; i.e., FB connections link equal
cortical distance from V1, relating to the magnification factor and regions of visual field in striate and extrastriate cortices. Injec-
RF size of neurons in the extrastriate region giving rise to the FB tions in overlapping retinotopic locations in V1 and V2 further
projections. This was the case for both upper and lower layer FB emphasized the orderly topographic organization of these con-
fields, but within each extrastriate area, the lower layer fields were nections, demonstrating that FB neurons project symmetrically to
always more extensive than the upper layer fields (Table 3). V1 around a central point at the same retinotopic location as the
We then compared the scale of FB fields with the scale of injected V2 column (Fig. 8a, bottom). Unlike V1 intra-areal
physiologically measured surround field sizes of V1 neurons. lateral connections, FB connectional fields in V1 were anisotropic
Specifically, we compared the relative visuotopic extent of FB in visual space (Fig. 8a, top, b; Table 4).
fields (Table 3) with the relative extent of surround fields of V1
neurons (see above). On average, depending on the cortical area DISCUSSION
of origin, FB connections from the lower layers of extrastriate Our results show that the monosynaptic spread of horizontal
cortex conveyed information to a V1 column from regions of connections in macaque V1 is commensurate with the low-
visual space 5–25 (up to 29) times the aggregate mrf of the V1 contrast summation field of V1 neurons. These connections are
column, and 6 –27 (up to 32) times the aggregate high-contrast SF not sufficiently extensive to account for the mean size of V1 cells’
of the V1 injection (Table 3). Surround field sizes of V1 cells were surround fields or longer-range center–surround interactions.
on average 10 (up to ⬎27) times larger than the mrf of the cells Feedback connections from extrastriate cortex to V1, instead, are
and 5 (up to ⬎13) times larger than their high-contrast SF (see commensurate with the full range of V1 cells’ center and sur-
above). Thus, the spatial scale of FB connections from extrastri- round field sizes; they show an orderly topographic organization
ate cortex to V1 is commensurate with the full spatial range of and terminate in a patch-like manner within V1 (Fig. 9). These
empirically measured modulatory surround fields of single V1 results strongly suggest that V1 horizontal connections integrate
cells. Similarly, comparison of the mean visuotopic extent of FB signals within the SF, whereas feedback connections underlie
fields with the mean Gaussian spread (diameter) of the inhibitory interactions within and beyond the SF of V1 neurons.
surround mechanism (5.4°) revealed the mean visuotopic extent The size of the RF depends on the method used to measure it.
of FB from V3 (5.6° in layers 2/3 and 6.7° in layers 5/6) to be The mrf is the low-threshold, spiking region, which can be
commensurate with the mean size of the RF surround mecha- mapped using moving small stimuli. This region is surrounded by
nism, and FB from V2 (3.4° in layers 2/3 and 3.8° in layers 5/6) and a higher-threshold, depolarizing field, incapable of driving the
MT (15.3° and 26.6° in the upper and lower layers) with shorter- cell when stimulated in isolation but capable of increasing the
and longer-range surround sizes, respectively. response of the cell to stimulation of its mrf (Bringuier et al.,
Figure 8, a and b, shows the visuotopic extent of anterogradely 1999). The size of this subthreshold field surrounding the mrf has
labeled fields of terminals of FB connections within V1 arising been measured intracellularly in cat V1 neurons and was found to
from a BDA injection in V2 (Fig. 8a) or a CTB injection in V3 be coextensive with the monosynaptic spread of V1 horizontal
(Fig. 8b). The relative visuotopic extent (D° of FB field/aggregate connections in this species (Bringuier et al., 1999). This larger
mrf of neurons at the injection site) of the FB terminal field in subthreshold region can be revealed extracellularly in areal sum-
layers 2/3 of V1 labeled by the V2 injection was 0.85 (Fig. 8a); mation experiments, using expanding gratings, and has been
Angelucci et al. • Anatomical Circuits for Spatial Integration in V1 J. Neurosci., October 1, 2002, 22(19):8633–8646 8643

Table 4. Visual field extent and anisotropy of feedback terminal fields

in V1

Ratios are shown for only one method of measuring RF size (mrf; symbol on the
left). n, Same as in Table 1, FB in V1.
1,2,†
As in Table 2.

shown to be larger when measured at low stimulus contrast

(Kapadia et al., 1999; Sceniak et al., 1999). In the present study,
we found the monosynaptic spread of horizontal connections in
monkey V1 to be coextensive with the empirically measured
low-contrast SF of V1 cells and with the mean Gaussian spread of
the excitatory center mechanism. Thus, the subthreshold depo-
larizing synaptic integration field of V1 neurons, measured intra-
cellularly by Bringuier et al. (1999), most likely represents the
low-contrast summation field of the neurons measured extracel-
lularly. In macaque, feedforward thalamic afferents to single layer
4C V1 neurons are of appropriate scale to underlie the mrf and,
possibly, the high contrast SF of these cells (Angelucci et al.,
2002) but are not sufficiently extensive to mediate longer-range
interactions. Thus, we suggest that intra-areal V1 horizontal
connections play an important role in shaping the spatial summa-
tion properties of V1 neurons at low contrast. Somehow, the
inputs from laterally offset neurons are more effective in driving
the center neuron in a low-contrast regimen, when feedforward
(lateral geniculate nucleus) inputs are only moderately driving
the response of the center neuron (for a model, see Angelucci et
al., 2002) (Fig. 9).
We were surprised to find that horizontal connections in ma-
caque V1 are isotropic in visual space. This contrasts with results
in other species, in which these connections are anisotropic along
an axis colinear to the optimal orientation in the visual field map
[tree shrew (Bosking et al., 1997), cat (Schmidt et al., 1997), and
owl monkey (Sincich and Blasdel, 2001)]. Anisotropic lateral
connections in visual field have been suggested to serve as an

4
aggregate mrf of neurons at the V2 injection site (1.6 ⫻ 1.15°); gray oval,
estimated D° of resulting labeled FB terminal field in V1 (1.35°x1°).
Dashed rectangles, Three RFs (mrf) at V2 injection site recorded in the
same vertical penetration at different cortical depths (L3, L5, L6, V2
cortical layers 3, 5, 6, respectively). Gray rectangles, Four RFs (mrf) at V1
injection site (star at the bottom) recorded in the same vertical penetration
in different layers (most superficial in layer 2, deepest in layer 6). Note
good overlap of RFs at V1 and V2 injected points, and their location at
the center of the FB terminal field. Rectangles 1–9, mrf sizes of neurons at
Figure 8. Visuotopic extent of feedback terminal fields anterogradely V1 recording sites 1–9 shown at the bottom. Filled black rectangle, Foveal
labeled in V1 by tracer injections in V2 or V3. a, FB field in V1 arising RF mapped to monitor eye movements. Note good agreement between
from a V2 injection. Bottom, Surface-view 2D serial tangential section estimated and empirically measured visuotopic extents of connections
reconstruction of the FB terminal field in layers 2/3 of V1 labeled by a and injection sites. b, Estimated visuotopic extent of labeled FB terminal
BDA injection through V2 layers 1– 6 at ⬃2° eccentricity in the lower fields in V1 layers 4B ( gray oval, 7.7 ⫻ 5.6°) and 5/6 (dashed gray oval,
visual field. The arrow shows the approximate location of the V1–V2 7.1 ⫻ 5.3°) arising from a CTB injection (black oval, 7.2 ⫻ 6.6°) through
border [vertical meridian (VM )] and points toward the fovea. Dots 1–9, V3 layers 1– 6 at 6.4° eccentricity in the lower visual field. c, Histogram of
V1 recording sites; numbers correspond to RFs mapped at the top. Star, the population means of the relative visuotopic extent of labeled FB
Center of CTB V1 injection made at the same retinal eccentricity as the terminal fields in V1 (isopolar axis) arising from V2 or V3 injections. The
V2 injection. The labeled field axes measured 7.6 ⫻ 4 mm. Scale bar, 1 visuotopic extent is expressed as the ratio of D° of the FB fields to the
mm (corrected for 8% shrinkage). Top, Visual field map of the BDA- aggregate mrf size of neurons at the extrastriate injection site. Data from
labeled FB terminal field shown at the bottom. Black oval, Estimated all V1 layers are pooled together.
8644 J. Neurosci., October 1, 2002, 22(19):8633–8646 Angelucci et al. • Anatomical Circuits for Spatial Integration in V1

anatomical substrate for intracortical generation of orientation

selectivity or for contour completion in V1 (Bosking et al., 1997;
Li, 1999; Sincich and Blasdel, 2001). The visual space anisotropy
of V1 lateral connections seen in other species might, instead,
reflect the longer-length summation receptive fields of V1 cells,
which have been demonstrated at least in the tree shrew (Bosking
and Fitzpatrick, 1995). Consistent with this hypothesis are the
observations that summation fields in macaque V1 are generally
isotropic (Sceniak et al., 2001; Levitt and Lund, 2002), and that
both horizontal connections (this study) and summation fields
(Sceniak et al., 2001) are more anisotropic in layer 6 than in more
superficial layers of macaque V1. Our results suggest that visual
field anisotropy is not required to generate orientation-selective
RFs. Our finding that feedback terminal fields in V1 show visual
field anisotropy suggests that contour completion in macaque V1
might instead be mediated by feedback connections.
The monosynaptic spread of horizontal connections is not
sufficiently extensive to account for the scale of the modulatory
surround region beyond the low-contrast SF of V1 neurons.
However, the region between the high- and low-contrast SF (Fig.
9, hatched annulus) can suppress or facilitate the center response
depending on the contrast of the center stimulus (Levitt and
Lund, 1997; Polat et al., 1998; Sceniak et al., 1999, 2001; Mizobe
et al., 2001) (for discussion, see Angelucci and Bullier 2002).
Thus, monosynaptic horizontal connections could mediate sur-
round modulation within this region of space. One example of
such “short-range” surround modulation is colinear facilitation,
i.e., enhancement of the mrf center response to an optimally
oriented low-contrast stimulus by flanking co-oriented and coax-
ial high-contrast stimuli; a phenomenon thought to underlie per-
ceptual grouping of contour elements (Hess and Field, 2000).
Figure 9. Summary diagram showing the spatial scales of V1 lateral and One interpretation of colinear facilitation is that it simply reflects
feedback connections relative to the spatial scales of empirically measured placement of flank stimuli within the low-contrast SF of the same
summation receptive field and modulatory surround field of V1 neurons. cell, and could thus be explained by the same mechanism under-
Gray area, Region over which presentation of stimuli at the same orien-
tation as the center stimulus can suppress the center response to an
lying the expansion of the RF at low contrast (for “long-range”
optimally oriented high contrast stimulus. White area, Region over which colinear facilitation, see Mizobe et al., 2001). Recent evidence
presentation of optimally oriented high-contrast stimuli evokes or facili- that GABA inactivation of laterally displaced V1 sites reduces
tates a response from the neuron (high-contrast SF). Hatched gray annu- colinear facilitation suggests that horizontal connections may be
lus, Region over which presentation of stimuli at the same orientation as
the underlying anatomical substrate (Crook et al., 2002). Short-
the center stimulus can suppress or facilitate the center response to an
optimally oriented stimulus depending on the center stimulus contrast. range surround suppression can also be observed for high-
Lateral connections within V1 (red) extend beyond the high-contrast contrast center stimuli. The same connectional system, and thus a
summation field (black circle) and are commensurate with the low- similar inhibitory mechanism, could account for “shrinkage” of
contrast SF (dashed black circle) of the V1 neurons from which they arise. the RF and short-range surround suppression at high contrast
Feedback connections (FB; blue) from extrastriate cortex to V1 are
commensurate with the full spatial scale of the SF and surround field. FB (Angelucci et al., 2002) (Fig. 9) [see Sceniak et al. (1999) for an
from “higher” cortical areas is more extensive than FB from “lower” alternative model]. Whether short-range surround suppression is
areas. Both connectional systems (lateral and FB) are patchy in V1. Scale generated via lateral (Hupé et al., 2001b) or feedback connec-
bar, 1°. We have previously proposed a model of how FB and lateral tions, it is clear that lateral connections are less extensive than
connections might mediate modulation of RF responses (Angelucci et al.,
2002). In this model, the output of each excitatory pyramidal neuron (e.g., even the empirically measured mean SF size of V1 neurons or the
the center recorded neuron) in V1 is controlled by a local inhibitory mean Gaussian spread of the inhibitory surround mechanism.
neuron having higher response gain and contrast threshold than the Several lines of evidence suggest that polysynaptic circuits of
pyramid (Lund et al., 1995; Somers et al., 2002). FB and lateral inputs horizontal connections within V1 are unlikely to underlie long-
contact directly both neuron types, whereas feedforward inputs are only
to the pyramid. The divergent–convergent organization of FB and lateral range center–surround interactions. First, the strong inhibitory
axons is such that these two systems overlap in space and are active for nature of most surround effects would preclude propagation of
any stimulus diameter, even for stimuli confined to the RF center. At low signals through a cascade of lateral connections. Because lateral
contrast, RF excitation predominates; lateral and FB input to the pyramid axons are known to target the same population of neurons at
can be summed from more distant cortical (and visual space) locations
before inhibition begins to rise. Suppression of the center neuron re- every synaptic location (⬃80% excitatory and 20% inhibitory
sponse would result from increasing the weight of excitation onto the neurons; McGuire et al., 1991), inhibition would occur at each
pyramid and its local inhibitory neuron, either via high-contrast feedfor- relay step. Second, the slow conduction velocity of horizontal
ward drive or via lateral and FB inputs, such as by increasing stimulus axons (Bringuier et al., 1999; Girard et al., 2001) would preclude
diameter.
them from processing fast information across long distances, at
least in V1 and V2 where, because of the large magnification
Angelucci et al. • Anatomical Circuits for Spatial Integration in V1 J. Neurosci., October 1, 2002, 22(19):8633–8646 8645

factor and small RF sizes, these connections connect relatively to occur in the responses of V1 and V2 cells to illusory contours
small regions of space. (Lee and Nguyen, 2001) or to occluded contours defined by
Inter-areal feedback projections provide information to a small contextual depth cues (Sugita, 1999; Bakin et al., 2000).
column of V1 neurons from larger regions of space than V1
lateral connections, the size of the visual field region conveyed to REFERENCES
V1 increasing with cortical distance from V1. The visuotopic size Albright TD, Desimone R (1987) Local precision of visuotopic organi-
of the feedback projection field within V1 arising from a small zation in the middle temporal area (MT) of the macaque. Exp Brain
Res 65:582–592.
column of extrastriate cortical neurons matches the aggregate RF Allman J, Miezin F, Mc Guinness E (1985) Stimulus specific responses
size of the extrastriate neurons of origin, demonstrating an or- from beyond the classical receptive field: neurophysiological mecha-
derly topographic arrangement of these connections. Our results nisms for local-global comparisons in visual neurons. Annu Rev Neu-
rosci 8:407– 430.
show that feedback connections are of appropriate scale to un- Angelucci A, Bullier J (2002) Reaching beyond the classical receptive
derlie long-range modulatory effects within and beyond the low- field of V1 neurons: horizontal or feedback axons? J Physiol (Paris), in
press.
contrast SF of V1 neurons (Fig. 9). The influence of feedback Angelucci A, Levitt JB (2002) Convergence of color, motion and form
connections on the RF center response of recipient neurons has pathways in macaque V3. Soc Neurosci Abstr, in press.
long been known. Specifically, inactivation of areas V2 and MT Angelucci A, Clasca F, Sur M (1996) Anterograde axonal tracing with
the subunit B of cholera toxin: a highly sensitive immunohistochemical
reduces the response of neurons in lower-order areas to visual protocol for revealing fine axonal morphology in adult and neonatal
stimulation of their RF center (Sandell and Schiller, 1982; brains. J Neurosci Methods 65:101–112.
Mignard and Malpeli, 1991; Hupé et al., 1998, 2001a), suggesting Angelucci A, Lund JS, Walton E, Levitt JB (1998) Retinotopy of con-
nections within and between areas V1 to V5 of macaque visual cortex.
that at least part of the FB input normally sums with FF inputs. A Soc Neurosci Abstr 24:897.
role for feedback connections in mediating center-surround in- Angelucci A, Levitt JB, Hupé J-M, Walton EJS, Bullier J, Lund JS
(2000) Anatomical circuits for local and global integration of visual
teractions is supported by recent evidence that inactivation of information: intrinsic and feedback connections in macaque visual
area MT reduces the suppressive effect of surround motion stim- cortical area V1. Eur J Neurosci 12 [Suppl 11]:285.
ulation in V3, V2, and V1 neurons (Hupé et al., 1998). Further- Angelucci A, Levitt JB, Lund JS (2002) Anatomical origins of the clas-
sical receptive field and modulatory surround field of single neurons in
more, feedback from MT has been shown to act on the early part macaque visual cortical area V1. Prog Brain Res 136:373–388.
of the response, suggesting that it may act on V1 neurons at the Asi H, Levitt JB, Lund JS (1996) In macaque V1 lateral connections in
same time as feedforward signals from the thalamus (Hupé et al., layer 4B have a different topography than in layers 2/3. Soc Neurosci
Abstr 22:1608.
2001a). Feedback connections’ conduction velocities are as rapid Bakin JS, Nakayama K, Gilbert CD (2000) Visual responses of monkey
as those of feedforward connections and 10 times faster than areas V1 and V2 to three-dimensional surface configurations. J Neu-
rosci 20:8188 – 8198.
those of horizontal connections (Girard et al., 2001). Because of Barlow HB, Blakemore C, Pettigrew JD (1967) The neural mechanisms
these spatial and temporal properties, feedback connections are of binocular depth discrimination. J Physiol (Lond) 193:327–342.
well suited to convey fast visual signals across distant parts of Barone P, Batardiere A, Knoblauch K, Kennedy H (2000) Laminar
distribution of neurons in extrastriate areas projecting to visual areas
space. Contrary to previous reports that feedback connections are V1 and V4 correlates with the hierarchical rank and indicates the
more diffuse and nonspecific than feedforward connections operation of a distance rule. J Neurosci 20:3263–3281.
(Rockland and Virga, 1989; Shipp and Zeki, 1989), we show here Blakemore C, Tobin EA (1972) Lateral inhibition between orientation
detectors in the cat’s visual cortex. Exp Brain Res 15:439 – 440.
that feedback projections are parcellated into discrete patches in Blasdel GG, Campbell D (2001) Functional retinotopy of monkey visual
V1 and overlap with patches of V1 feedforward projecting neu- cortex. J Neurosci 21:8286 – 8301.
Bosking WH, Fitzpatrick D (1995) Physiological correlates of anisot-
rons. The precise alignment of labeled feedback terminal clusters ropy in horizontal connections: length summation properties of neu-
in V1 with clusters of feedforward efferent cells (Salin et al., 1995; rons in layers 2 and 3 of tree shrew striate cortex. Soc Neurosci Abstr
Johnson and Burkhalter, 1997), is well suited to fast transfer of 21:1751.
Bosking WH, Zhang Y, Schofield B, Fitzpatrick D (1997) Orientation
specific functional information to and from V1, and it may also selectivity and the arrangement of horizontal connections in tree shrew
explain the orientation specificity of interactions between sur- striate cortex. J Neurosci 17:2112–2127.
round and RF stimulus configuration. The periodicity of feedback Bringuier V, Chavane F, Glaeser L, Frégnac Y (1999) Horizontal prop-
agation of visual activity in the synaptic integration field of area 17
connections, much like that of intra-areal connections, matches neurons. Science 283:695– 699.
the periodic distribution of specific functional properties in V1, Burkhalter A, Felleman DJ, Newsome WT, Van Essen DC (1986) An-
atomical and physiological asymmetries related to visual areas V3 and
such as orientation preference. Similarly to V1 lateral connections VP in macaque extrastriate cortex. Vision Res 26:63– 80.
(Malach et al., 1993), feedback connections appear to link points Cavanaugh JR, Bair W, Movshon JA (2002) Nature and interaction of
of like-orientation preference (Gilbert and Wiesel, 1989; Shmuel signals from the receptive field center and surround in macaque V1
neurons. J Neurophysiol, in press.
et al., 1998). Previously we have proposed a detailed model of Crick F, Koch C (1998) Constraints on cortical and thalamic projections:
how feedback connections might mediate orientation specific the no-strong-loops hypothesis. Nature 391:245–250.
modulation of RF responses (Angelucci et al., 2002) (Fig. 9). Crook JM, Engelmann R, Löwell S (2002) GABA-inactivation attenu-
ates colinear facilitation in cat primary visual cortex. Exp Brain Res
Although patchy feedback connections in V1 appear to reflect 143:295–302.
linking of similar stimulus attributes in striate and extrastriate Das A, Gilbert CD (1995) Receptive field expansion in adult visual
cortex is linked to dynamic changes in strength of cortical connections.
cortex, the relationship of the feedback terminal patches to the J Neurophysiol 74:779 –792.
CO columns in V1 suggests also the existence of specific com- DeAngelis GC, Freeman RD, Ohzawa I (1994) Length and width tuning
partments within V2 and V3, each compartment having a segre- of neurons in the cat’s primary visual cortex. J Neurophysiol
71:347–374.
gated terminal territory in V1 (Angelucci and Levitt, 2002). Dow BM, Snyder AZ, Vautin RG, Bauer R (1981) Magnification factor
Although we are proposing that feedback connections may and receptive field size in foveal striate cortex of the monkey. Exp
mediate center–surround interactions in V1, a more general role Brain Res 44:213–228.
Dow BM, Vautin RG, Bauer R (1985) The mapping of visual space onto
for this connectional system, consistent with their spatial scale, foveal striate cortex in the macaque monkey. J Neurosci 5:890 –902.
would be to mediate global-to-local, top-down integration of Felleman DJ, Van Essen DC (1991) Distributed hierarchical processing
in the primate cerebral cortex. Cereb Cortex 1:1– 47.
visual information. Global-to-local signal integration might rep- Gallyas F (1979) Silver staining of myelin by means of physical develop-
resent an essential step in visual processing and has been shown ment. Neurol Res 1:203–209.
8646 J. Neurosci., October 1, 2002, 22(19):8633–8646 Angelucci et al. • Anatomical Circuits for Spatial Integration in V1

Gattas R, Gross CG, Sandell JH (1981) Visual topography of V2 in the Mignard M, Malpeli JG (1991) Paths of information flow through visual
macaque. J Comp Neurol 201:519 –539. cortex. Science 251:1249 –1251.
Gattas R, Sousa APB, Gross CG (1988) Visuotopic organization and Mizobe K, Polat U, Pettet MW, Kasamatsu T (2001) Facilitation and
extent of V3 and V4 of the macaque. J Neurosci 8:1831–1845. suppression of single striate-cell activity by spatially discrete pattern
Gegenfurtner KR, Kiper DC, Levitt JB (1997) Functional properties of stimuli presented beyond the receptive field. Vis Neurosci 18:377–391.
neurons in macaque area V3. J Neurophysiol 77:1906 –1923. Nelson JI, Frost B (1978) Orientation selective inhibition from beyond
Gilbert CD, Wiesel TN (1989) columnar specificity of intrinsic horizon- the classical receptive field. Brain Res 139:359 –365.
tal and corticocortical connections in cat visual cortex. J Neurosci Nothdurft HC, Gallant JL, Van Essen DC (1999) Response modulation
9:2432–2442. by texture surround in primate area V1: correlates of “popout” under
Gilbert CD, Wiesel TN (1990) The influence of contextual stimuli on the anesthesia. Vis Neurosci 16:15–34.
orientation selectivity of cells in primary visual cortex of the cat. Vision Perkel DJ, Bullier J, Kennedy H (1986) Topography of the afferent
Res 30:1689 –1701. connectivity of area 17 in the macaque monkey: a double-labeling
Gilbert C, Das A, Ito M, Kapadia M, Westheimer G (1996) Spatial study. J Comp Neurol 253:374 – 402.
integration and cortical dynamics. Proc Natl Acad Sci USA 93:615– 622. Polat U, Mizobe K, Pettet MW, Kasamatsu T, Norcia AM (1998) Col-
Girard P, Hupé J-M, Bullier J (2001) Feedforward and feedback con- linear stimuli regulate visual responses depending on cell’s contrast
nections between areas V1 and V2 of the monkey have similar rapid threshold. Nature 391:580 –584.
conduction velocities. J Neurophysiol 85:1328 –1331. Rockland KS, Knutson T (2001) Axon collaterals of Meynert cells di-
Hess R, Field D (2000) Integration of contours: new insights. Trends verge over large portions of area V1 in the macaque monkey. J Comp
Cognit Sci 3:480 – 486. Neurol 441:134 –147.
Hupé J-M, James AC, Payne BR, Lomber SG, Girard P, Bullier J (1998) Rockland KS, Lund JS (1983) Intrinsic laminar lattice connections in
Cortical feedback improves discrimination between figure and back- primate visual cortex. J Comp Neurol 216:303–318.
ground by V1, V2 and V3 neurons. Nature 394:784 –787. Rockland KS, Pandya DN (1979) Laminar origins and terminations of
Hupé J-M, Chouvet G, Bullier J (1999) Spatial and temporal parameters cortical connections to the occipital lobe in the rhesus monkey. Brain
of cortical inactivation by GABA. J Neurosci Methods 86:129 –143. Res 179:3–20.
Hupé J-M, James AC, Girard P, Lomber SG, Payne BR, Bullier J Rockland KS, Virga A (1989) Terminal arbors of individual “feedback”
(2001a) Feedback connections act on the early part of the responses in axons projecting from area V2 to V1 in the macaque monkey: a study
monkey visual cortex. J Neurophysiol 85:134 –145. using immunohistochemistry of anterogradely transported phaseolus
Hupé J-M, James AC, Girard P, Bullier J (2001b) Response modulations vulgaris-leucoagglutinin. J Comp Neurol 285:54 –72.
by static texture surround in area V1 of the macaque monkey do not Roe AW, Ts’o DY (1995) Visual topography in primate V2: multiple
depend on feedback connections from V2. J Neurophysiol 85:146 –163. representation across functional stripes. J Neurosci 15:3689 –3715.
Johnson PB, Angelucci A, Ziparo RM, Minciacchi D, Bentivoglio M, Salin PA, Bullier J (1995) Corticocortical connections in the visual sys-
Caminiti R (1989) Segregation and overlap of callosal and association tem: structure and function. Physiol Rev 75:107–154.
neurons in frontal and parietal cortices of primates: a spectral and Salin PA, Kennedy H, Bullier J (1995) Spatial reciprocity of connections
coherence analysis. J Neurosci 9:2313–2326. between areas 17 and 18 in the cat. Can J Physiol Pharmacol
Johnson RR, Burkhalter A (1997) A polysynaptic feedback circuit in rat 73:1339 –1347.
visual cortex. J Neurosci 17:7129 –7140. Sandell JH, Schiller PH (1982) Effect of cooling area 18 on striate cortex
Kapadia MK, Ito M, Gilbert CD, Westheimer G (1995) Improvement in
visual sensitivity by changes in local context: parallel studies in human cells in the squirrel monkey. J Neurophysiol 48:38 – 48.
observers and in V1 of alert monkeys. Neuron 15:843– 856. Sceniak MP, Ringach DL, Hawken MJ, Shapley RM (1999) Contrast’s
Kapadia MK, Westheimer G, Gilbert CD (1999) Dynamics of spatial effect on spatial summation by macaque V1 neurons. Nat Neurosci
summation in primary visual cortex of alert monkeys. Proc Natl Acad 2:733–739.
Sci USA 96:12073–12078. Sceniak MP, Hawken MJ, Shapley RM (2001) Visual spatial character-
Knierim JJ, Van Essen DC (1992) Neuronal responses to static texture ization of macaque V1 neurons. J Neurophysiol 85:1873–1887.
patterns in area V1 of the alert macaque monkey. J Neurophysiol Schmidt KE, Goebel R, Löwell S, Singer W (1997) The perceptual
67:961–980. grouping criterion of colinearity is reflected by anisotropies of connec-
Lanciego JL, Luquin MR, Guillén J, Giménez-Amaya JM (1998) Mul- tions in the primary visual cortex. Eur J Neurosci 9:1083–1089.
tiple neuroanatomical tracing in primates. Brain Res Protocols Shipp S, Zeki S (1989) The organization of connections between areas
2:323–332. V5 and V1 in macaque monkey visual cortex. Eur J Neurosci
Lee TS, Nguyen M (2001) Dynamics of subjective contour formation in 1:308 –331.
the early visual cortex. Proc Natl Acad Sci USA 98:1907–1911. Shmuel A, Korman M, Harel M, Grinvald A, Malach R (1998) Rela-
Levitt JB, Lund JS (1997) Contrast dependence of contextual effects in tionship of feedback connections from area V2 to orientation domains
primate visual cortex. Nature 387:73–76. in area V1 of the primate. Soc Neurosci Abstr 24:767.
Levitt JB, Lund JS (2002) The spatial extent over which neurons in Sincich LC, Blasdel GG (2001) Oriented axon projections in primary
macaque striate cortex pool visual signals. Vis Neurosci, in press. visual cortex of the monkey. J Neurosci 21:4416 – 4426.
Levitt JB, Kiper DC, Movshon JA (1994) Receptive fields and func- Sincich LC, Horton JC (2002) Divided by cytochrome oxidase: a map of
tional architecture of macaque V2. J Neurophysiol 71:2517–2542. the projections from V1 to V2 in macaques. Science 295:1734 –1737.
Li Z (1999) Pre-attentive segmentation in the primary visual cortex. Somers D, Dragoi V, Sur M (2002) Orientation selectivity and its mod-
Spatial Vision 13:25–50. ulation by local and long-range connections in visual cortex. In: The cat
Lund JS, Wu Q, Hadingham PT, Levitt JB (1995) Cells and circuits primary visual cortex (Payne BR, Peters A, eds), pp 471–520. San
contributing to functional properties in area V1 of macaque monkey Diego: Academic.
cerebral cortex: bases for neuroanatomically realistic models. J Anat Sugita Y (1999) Grouping of image fragments in primary visual cortex.
187:563–581. Nature 401:269 –272.
Malach R, Amir Y, Harel M, Grinvald A (1993) Relationship between Tootell RBH, Switkes E, Silverman MS, Hamilton SL (1988) Functional
intrinsic connections and functional architecture revealed by optical anatomy of macaque striate cortex II: retinotopic organization. J Neu-
imaging and in vivo targeted biocytin injections in primate striate rosci 8:1531–1568.
cortex. Proc Natl Acad Sci USA 90:10469 –10473. Van Essen DC, Newsome WT, Maunsell JH (1984) The visual field
Maunsell JHR, Van Essen DC (1987) Topographic organization of the representation in striate cortex of the macaque monkey: asymmetries,
middle temporal visual area in the macaque monkey: representational anisotropies, and individual variability. Vision Res 24:429 – 448.
biases and the relationship to callosal connections and myeloarchitec- Walker GA, Ohzawa I, Freeman RD (1999) Asymmetric suppression
tonic boundaries. J Comp Neurol 266:535–555. outside the classical receptive field of the visual cortex. J Neurosci
McGuire BA, Gilbert CD, Rivlin PK, Wiesel TN (1991) Targets of 19:10536 –10553.
horizontal connections in macaque primary visual cortex. J Comp Yoshioka T, Blasdel GG, Levitt JB, Lund JS (1996) Relation between
Neurol 305:370 –392. patterns of intrinsic lateral connectivity, ocular dominance and cyto-
Merril EG, Ainsworth A (1972) Glass-coated platinum-plated tungsten chrome oxidase reactive regions in macaque monkey striate cortex.
microelectrodes. Med Biol Eng 10:662– 672. Cereb Cortex 6:297–310.

View publication stats