Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 1996 Nov 1;16(21):6945-64.
doi: 10.1523/JNEUROSCI.16-21-06945.1996.

Functional organization for direction of motion and its relationship to orientation maps in cat area 18

A Shmuel  1 , A Grinvald
Affiliations

Functional organization for direction of motion and its relationship to orientation maps in cat area 18

A Shmuel et al. J Neurosci. .

Abstract

The goal of this study was to explore the functional organization of direction of motion in cat area 18. Optical imaging was used to record the activity of populations of neurons. We found a patchy distribution of cortical regions exhibiting preference for one direction over the opposite direction of motion. The degree of clustering according to preference of direction was two to four times smaller than that observed for orientation. In general, direction preference changed smoothly along the cortical surface; however, discontinuities in the direction maps were observed. These discontinuities formed lines that separated pairs of patches with preference for opposite directions. The functional maps for direction and for orientation preference were closely related; typically, an iso-orientation patch was divided into regions that exhibited preference for opposite directions, orthogonal to the orientation. In addition, the lines of discontinuity within the direction map often connected points of singularity in the orientation map. Although the organization of both domains was related, the direction and the orientation selective responses were separable; whereas the selective response according to direction of motion was nearly independent of the length of bars used for visual stimulation, the selective response to orientation decreased significantly with decreasing length of the bars. Extensive single and multiunit electrical recordings, targeted to selected domains of the functional maps, confirmed the features revealed by optical imaging. We conclude that significant processing of direction of motion is performed early in the cat visual pathway.

PubMed Disclaimer

Figures

Fig. 6.
Fig. 6.
Confirmation of the optical maps by targeted unit recordings. A and B present the differential direction maps for upward versus downward motion and leftward versus rightward motion, respectively. These maps were used to guide vertical penetrations of an electrode for multiunit recording.Crosses are superimposed on both direction maps marking the locations of penetrations guided by each map. C, The map of preferred angle of direction computed by pixel-wise vectorial addition.D, Results from the guided penetrations. Thenumber at the top of each separate column indicates the number of penetration, as marked in A–C. The vertical axis of each column is scaled to the range 0–2000 μm, representing estimated cortical depth. The preference of multiunit for orientation (red lines) and direction (green arrows) is presented as a function of cortical depth. The lengths of lines and arrows are proportional to the normalized magnitudes of selective responses for orientation and direction, respectively. Thecolored arrows above the columns correspond to the preferred direction detected by optical imaging. The colors correspond to the colors of the preferred direction angle map, at the location of each penetration. The mean absolute deviation of the preferred angle detected by optical imaging relative to the angle measured by electrical recording was 15.7 ± 15°. E, The differential orientation map obtained by stimulating with the grating pattern. Dark and bright areas represent cortical locations that preferred vertical and horizontal orientation, respectively. The preferred orientation detected by multiunit recording along each column (as presented in D) was relatively constant. As in the case of direction, most of the preferred orientations detected by electrical recording were highly correlated with those detected by optical imaging. Note that the panels whose multiunits presented oblique preferred orientation—such as columns4, 5, 15, and 17—were located as expected ingray zones between the black/white patches of the orientation map. Thus, their preference for oblique orientation is consistent with the preference detected by optical imaging.F, The locations of all penetrations are marked ontop of the imaged cortical area. The pattern of superficial blood vessels was used to guide the electrode to the location selected for recording. G, The red curve represents the average columnar multiunit response of units whose depth was <1000 μm. The specific columnar responses were cyclically shifted before averaging to be aligned by their maximal response. The blue curve represents the mean columnar response obtained by optical imaging, at the locations of electrode penetrations (the two curves were normalized).
Fig. 1.
Fig. 1.
Activity patterns evoked by gratings moving in various directions. Eight different gratings, each moving in two opposite directions, were presented to the cat while activity maps were collected. A subset of eight orientation and direction single-condition optical maps is presented. Dark patches in each image represent cortical areas that were active while the cat was stimulated with gratings whose orientation and direction are marked on the image. The four sets of crosses are marked on identical cortical locations to enable easy comparison of the patterns of activity. Theimage in the center is the image of the cortical surface and superficial blood vessels, taken under green illumination (540 nm). Each of the different stimuli selectively activated a small number of patches in the imaged area. Maps produced by gratings of similar orientation, but opposite direction of motion (180° apart), were similar to one other. Gratings at orthogonal orientations activated complementary patches. The maps are scaled such that the whole range of gray levels corresponds to a fractional change of 3.2 × 10−4 for presenting the activity maps. The wavelength of illumination used for imaging (hereafter wavelength) was 650 ± 10 nm. A, Anterior; P, posterior; M, medial; L, lateral. Scale bar, 1 mm.
Fig. 2.
Fig. 2.
Correlations between single-condition maps.A, The features of similarity indicated for orientation and direction single-condition maps are demonstrated by the matrix of correlation coefficients among these activity maps. The comparison of each map to another was performed by calculating the correlation coefficient between corresponding pixels. The entry (i,j) presents the correlation coefficient of the single-condition map corresponding to direction (i−1)π/8 with that of the single-condition map corresponding to direction (j−1)π/8 (i,j = 1…16).B, The observations of similarity are summarized in a correlation graph. The average correlation coefficient between pairs of single-condition maps is presented as a function of the angular difference in the direction corresponding to the compared maps. The average was computed by first cyclically shifting each row of the matrix in A such that the entry of the diagonal was shifted to the first column, and then by averaging over the rows. Gratings of similar orientation that moved in opposite directions of motion produced similar maps (r = 0.60). Gratings of orthogonal orientation produced complementary maps (r = −0.60). The average correlation between activity patterns produced by gratings whose relative orientation was oblique was ∼0.
Fig. 3.
Fig. 3.
Clustering according to direction of motion in area 18. A and B illustrate orientation and direction single-condition maps obtained from the same cortical area by stimulating with horizontal gratings moving upward and downward, respectively. The dark patches represent cortical areas activated by the stimulus marked next to the image. The entire range of gray levels represents a fractional change of 5 × 10−4 (wavelength 650 ± 10 nm). The two sets of crosses are located in identical cortical locations. C, A differential direction map between activity produced by a horizontal grating moving upward and a horizontal grating moving downward. Thedark regions correspond to an upward direction preference and the bright regions to a downward direction preference. The map was computed by subtracting the single-condition map presented in B from that presented in A. The differential direction map, of amplitude 2.9 × 10−4, is scaled to the full range of gray levels. D, The same direction map as seen in C, presented in a scaled gray level whose entire range represents a fractional change of 5 × 10−4. This scale was used to facilitate comparison of amplitudes withA and B. E and F present the same direction map obtained from interlaced complementary blocks of data. The sets of crosses superimposed on both images correspond to identical cortical locations. Black andwhite crosses correspond to regions activated by downward and upward motion, respectively. The average S/N (signal-to-noise ratio; see for formal definition) of the eight direction maps in this experiment was 7.6 ± 1.1. The range of gray levels here represents a fractional change of 3.2 × 10−4. Scale bar, 1 mm.
Fig. 4.
Fig. 4.
Clustering according to direction of motion in adult cat area 18. The results presented here were obtained from an adult cat. A–F illustrate the results in a format identical to that of Figure 3. The presented maps were obtained from vertical gratings moving left and right. The whole range of gray levels represents a fractional change of 2 × 10−3 inA, B, D, 0.9 × 10−3 in C, and 1 × 10−3 in E and F(wavelength 630 ± 10 nm). The ratio of the amplitudes here was 3.5, and the average S/N of the direction maps was 4.8 ± 0.9. Scale bar, 1 mm.
Fig. 5.
Fig. 5.
Direction and orientation maps evoked by various visual patterns. The set of visual stimuli included five subsets. The pattern of stimulus used to obtain the maps of each row of images is marked at the left of the row. The pattern used for thefirst row of images was composed of randomly located squares whose size was 1.85 × 1.85°. In the next three rows, the length of the randomly located bars was doubled from one row to the next. The width of the elements was not changed. In the fifth row, full-screen rectangular gratings were used. The spatial frequency of the gratings was 0.18 cycles per degree; one cycle was divided into 33% (1.85°) white and 67% (3.7°) black. Each pattern was moved in one of four directions orthogonal to the orientation of bars during different presentations. The left column of images presents the differential direction maps obtained by upward versus downward motion. The middle column presents data obtained by leftward versus rightward motion. The right column presents the “orientation” maps obtained by dividing the sum of responses to leftward and rightward motion by the sum of responses to upward and downward motion. Within each column, marks are placed at identical locations. The full scale of gray levels corresponds to 5.4 × 10−4 for the direction maps and to 7.2 × 10−4 for the orientation maps (wavelength 630 ± 10 nm). The selective response to direction was virtually invariant to the different stimuli used, in terms of both the activity pattern and its magnitude. The pattern of the selective response of orientation was invariant to the different stimuli as well; however, the magnitude of the activity selective to orientation increased for increasing the length of the bars. Roughly, the amplitude of the maps changed in a logarithmic manner for increasing the length of the bars.A, Anterior; P, posterior; M, medial;L, lateral. Scale bar, 1 mm.
Fig. 7.
Fig. 7.
Relationship between the amplitude of the mapping signal and the underlying spike activity. The columnar normalized response measured by electrical recording at each site was integrated across the upper 1000 μm of cortical depth. At each cortical location, the response measured by optical imaging was sampled from the corresponding single-condition map. The 68 responses measured optically were plotted as a function of the respective responses measured by electrical recording (68 = 17 columns × 4 stimuli). The line is the best linear fit to the data obtained by linear regression. A high degree of linearity is demonstrated (r = 0.78).
Fig. 8.
Fig. 8.
Direction selectivity of single neurons at the center of direction-selective domains. One hundred thirty-nine isolated units were recorded from two kittens. All recordings were performed along tracks perpendicular to the cortex, targeted to the center of direction-selective domains. The sum of spikes of each cell during the presentation of each condition was computed. The direction that evoked the maximal response was referred to as the preferred direction (PD). The direction index (DI) was defined as DI = (response to (PD) − response to (PD + 180°))/response to (PD). The distribution of direction indices is illustrated. It is skewed toward high values of direction selectivity. Eighty-one percent of the cells exhibited a direction index ≥0.5.
Fig. 9.
Fig. 9.
A full set of differential direction maps. Eight different gratings, each moving in one of two opposite directions, were presented to the cat while activity maps were collected. The set of eight differential direction maps is presented. Dark andbright patches in each image represent cortical areas that preferred motion in the direction of the black andwhite arrows marked on the image, respectively. The entire range of gray levels represents a fractional change of 3.6 × 10−4 (wavelength 650 ± 10 nm). The image of the corresponding cortical surface and superficial blood vessels is presented in the center of Figure 1. The maps contain significant portions of gray areas, implying no preference for any of the two directions of motion. Pairs of differential direction maps produced by gratings of similar orientations (adjacent images in the figure) were similar (the middle image in the left columnshould be compared to the image above it only after reversing the white and black patches in one of the images, because the color look-up table used here is not cyclical). The similarity between direction maps decreased as the difference in the corresponding axes of motion increased. These trends of similarity between differential direction maps are quantified by the plot in the center. The comparison of maps to one another was performed by calculating the correlation coefficient between corresponding pixels. The average correlation coefficient between pairs of direction maps is presented as a function of the angular difference in axis of motion. Direction maps that corresponded to similar axes of motion were similar (r = 0.43). Direction maps that corresponded to orthogonal axes of motion were uncorrelated (r = 0.00). Δaom, Difference in the axis of motion. A, Anterior; P, posterior; M, medial; L, lateral. Scale bar, 1 mm.
Fig. 10.
Fig. 10.
The overall organization for direction. A complete set of single-condition maps was combined by pixel-wise vectorial addition. A, The angle of the preferred direction at any location is color-coded according to the color codepresented on the left. Superimposed are vectors that represent the local preferred direction (direction of the arrow) and the magnitude of preference (length of the arrow). The preferred direction was spatially continuous along the surface of the cortex in most of the imaged cortical area. B, The rate of change of the preferred direction. The map was computed using the gradient transform (bright denotes high rate of change). The bright areas in the form of lines here correspond to lines across which preferred direction reversed or nearly reversed. The green arrows represent the angle and magnitude of the preference of the local population. C, The magnitude of the direction selectivity is coded with a gray level scale. This map represents the length of the pixel-wise vectorial sum of the single-condition maps.Bright regions correspond to high direction selectivity anddark regions correspond to low direction selectivity. Apart from the lines of low magnitude of preference, most of the cortical surface here exhibited clustering according to direction. Scale bar, 1 mm.
Fig. 11.
Fig. 11.
Top. Spatial frequencies of maps for orientation and direction preference. A and Bpresent the angle maps of preferred orientation and preferred direction, respectively. Although both maps use color look-up tables that have a resolution of 45°, the size of patches within the map for direction preference is smaller than the size of patches for orientation preference. The ratio of the number of patches of iso-orientation and the number of patches of iso-direction presented here is ∼1:2.
Fig. 13.
Fig. 13.
The overall relationship between the organization of orientation and direction. The red lines represent the preferred orientation and the magnitude of preference by their angle and length, respectively. Superimposed are green arrows, representing the local preferred direction. At the background, in gray-level presentation, is the map of magnitude for direction preference (dark represents low magnitude). Locations coded in yellow exhibited a high rate of change of preferred orientation angle (orientation singularities, computed using the gradient transform). A and B are examples from a kitten and an adult cat, respectively. A patch that exhibited orientation selectivity often included two or more patches of direction selectivity. The preferred directions of motion exhibited within these patches were mostly orthogonal to the orientation. Thus, the direction patches within a single orientation patch often represented preferences for opposite directions of motion. The singularities of orientation preference tended to be point-like. In contrast, the areas of low magnitude of preference for the direction domain were in the form of long curved lines. These tended to run across the center of orientation domains, in which the magnitude of preference for orientation exhibited a ridge-like maximum. Many of the orientation singularities were connected to the lines depicting regions of low directionality (examples are marked by ellipses); however, endpoints of direction discontinuity lines in locations other than orientation singularities existed (examples are marked by arrows). Scale bar, 1 mm.
Fig. 14.
Fig. 14.
Top. An instance of a 360° layout for direction preference. The format used here is identical to that of Figure 13. Because the amplitude of direction selectivity is smaller than the amplitude of the orientation selectivity, the lengths of the orientation lines were scaled down by 1.5 for clarity. The orientation singularity at the right gives rise to one direction discontinuity line, whereas three such lines extend from the leftsingularity. Scale bar, 1 mm.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Albright TD, Desimone R, Gross CG. Columnar organization of directionally selective cells in visual area MT of the Macaque. J Neurophysiol. 1984;51:16–31. - PubMed
    1. Albus K. The detection of movement direction and the effects of contrast reversal in the cat’s striate cortex. Vision Res. 1980;20:289–293. - PubMed
    1. Baker JF, Petersen SE, Newsome WT, Allman JM. Visual response properties of neurons in four extrastriate visual areas of the owl monkey (Aotus trivirgatus): a quantitative comparison of medial, dorsomedial, dorsolateral and middle temporal areas. J Neurophysiol. 1981;45:397–416. - PubMed
    1. Berman NEJ, Wilkes ME, Payne BR. Organization of orientation and direction selectivity in areas 17 and 18 of cat cerebral cortex. J Neurophysiol. 1987;58:676–699. - PubMed
    1. Blasdel GG, Salama G. Voltage-sensitive dyes reveal a modular organization in monkey striate cortex. Nature. 1986;321:579–585. - PubMed

Publication types

LinkOut - more resources

-