Directional tuning of cells in area 18 of the feline visual cortex for visual noise, bar and spot stimuli: a comparison with area 17

The mechanism for processing random-dot motion at various speeds in early visual cortices

All moving objects generate sequential retinotopic activations representing a series of discrete locations in space and time (motion trajectory). How direction-selective neurons in mammalian early visual cortices process motion trajectory remains to be clarified. Using single-cell recording and optical imaging of intrinsic signals along with mathematical simulation, we studied response properties of cat visual areas 17 and 18 to random dots moving at various speeds. We found that, the motion trajectory at low speed was encoded primarily as a direction signal by groups of neurons preferring that motion direction. Above certain transition speeds, the motion trajectory is perceived as a spatial orientation representing the motion axis of the moving dots. In both areas studied, above these speeds, other groups of direction-selective neurons with perpendicular direction preferences were activated to encode the motion trajectory as motion-axis information. This applied to both simple and complex neurons. The average transition speed for switching between encoding motion direction and axis was about 31°/s in area 18 and 15°/s in area 17. A spatio-temporal energy model predicted the transition speeds accurately in both areas, but not the direction-selective indexes to random-dot stimuli in area 18. In addition, above transition speeds, the change of direction preferences of population responses recorded by optical imaging can be revealed using vector maximum but not vector summation method. Together, this combined processing of motion direction and axis by neurons with orthogonal direction preferences associated with speed may serve as a common principle of early visual motion processing.

Speed and direction response profiles of neurons in macaque MT and MST show modest constraint line tuning

Several models of heading detection during smooth pursuit rely on the assumption of local constraint line tuning to exist in large scale motion detection templates. A motion detector that exhibits pure constraint line tuning responds maximally to any 2D-velocity in the set of vectors that can be decomposed into the central, or classic, preferred velocity (the shortest vector that still yields the maximum response) and any vector orthogonal to that. To test this assumption, we measured the firing rates of isolated middle temporal (MT) and medial superior temporal (MST) neurons to random dot stimuli moving in a range of directions and speeds. We found that as a function of 2D velocity, the pooled responses were best fit with a 2D Gaussian profile with a factor of elongation, orthogonal to the central preferred velocity, of roughly 1.5 for MST and 1.7 for MT. This means that MT and MST cells are more sharply tuned for speed than they are for direction; and that they indeed show some level of constraint line tuning. However, we argue that the observed elongation is insufficient to achieve behavioral heading discrimination accuracy on the order of 1-2 degrees as reported before.

Temporal interactions in direction-selective complex cells of area 18 and the posteromedial lateral suprasylvian cortex (PMLS) of the cat

Temporal interactions in direction-sensitive complex cells in area 18 and the posteromedial lateral suprasylvian cortex (PMLS) were studied using a reverse correlation method. Reverse correlograms to combinations of two temporally separated motion directions were examined and compared in the two areas. A comparison to the first-order reverse correlograms allowed us to identify nonlinear suppression or facilitation due to pairwise combinations of motion directions. Results for area 18 and PMLS were very different. Area 18 showed a single type of nonlinear behavior: similar directions facilitated and opposite directions suppressed spike probability. This effect was most pronounced for motion steps that followed each other immediately and decreased with increasing delay between steps. In PMLS, the picture was much more diverse. Some cells exhibited nonlinear interactions, that were opposite to those in area 18 (facilitation for opposite directions and suppression for similar ones), while the majority did not show a systematic interaction profile. We conclude that nonlinear second-order reverse correlation characteristics reveal different functional properties, despite similarities in the first-order reverse correlation profiles. Directional interactions in time revealed optimal integration of similar directions in area 18, but motion opponency--at least in some cells--in PMLS.

Cortical cartography revisited: a frequency perspective on the functional architecture of visual cortex

Viewed in the plane of the cortical surface, the visual cortex is composed of overlapping functional maps that represent stimulus features such as edge orientation, direction of motion, and spatial frequency. Spatial relationships between these maps are thought to ensure that all combinations of stimulus features are represented uniformly across the visual field. Implicit in this view is the assumption that feature combinations are represented in the form of a place code such that a given pattern of activity uniquely signifies a specific combination of stimulus features. Here we review results of experiments that challenge the place code model for the representation of feature combinations. Rather than overlapping maps of stimulus features, we suggest that patterns of activity evoked by complex stimuli are best understood in the context of a single map of spatiotemporal energy.

Responses to moving texture patterns of cells in the striate-recipient zone of the cat's lateral posterior-pulvinar complex

We have studied the response properties of cells in the lateral part of the lateral posterior nucleus or striate-recipient zone (LPl) of the lateral posterior nucleus-pulvinar complex to the motion of textured patterns [visual noise]. Our purpose was to determine basic noise response characteristics and to compare these properties to that of cells in area 17 known to project to the LPl. Practically all LPl cells (87%) responded to the motion of visual noise. The evoked discharges were either sustained or characterized by several bursts. On average, as found in cortex, LPl neurons were more broadly tuned for the direction of noise than that of gratings (bandwidths of 49 and 36 degrees, respectively; t-test, P < 0.005). Noise tuning function in LPl was comparable to that found in cortex (mean of 48 degrees). One third of the LPl units did not exhibit any preferences for drift direction of noise. Such cells were virtually not encountered in the striate cortex. This group of LPl cells was generally not tuned for grating direction. For practically all LPl cells, responses to noise varied as a function of drift velocity. The mean optimal velocity was 27.5 degrees/s with mean bandwidth of 2.5 octaves. LPl cells were sensitive to a broader range of velocities than complex cells in area 17. The results of the present study showed that visual noise is an appropriate stimulus for studying motion sensitivity of cells in the LPl. It also revealed that the noise response properties, such as direction and velocity tuning functions, are very similar to those reported in the striate cortex. The exact contribution of area 17 in the visual noise responsiveness of LPl cells remains to be determined. This study provides additional evidence that the lateral posterior nucleus-pulvinar complex may be involved in many aspects of visual processing.

Influence of GABA-induced remote inactivation on the orientation tuning of cells in area 18 of feline visual cortex: a comparison with area 17

We have investigated the effect of iontophoretically applying the inhibitory transmitter gamma-aminobutyric acid (GABA) through four pipettes, each located at a horizontal distance of some 500-600 microns from the recording site, on the orientation tuning of cells in areas 17 and 18 of the cat visual cortex for moving the stationary flash-presented bar stimuli. Forty-five of 74 cells tested in area 18 (61%) showed a significant (greater than 25%) increase in orientation tuning width (at half the maximum response) during GABA application, which reflected an increase in response to non-optimal orientations. The mean orientation tuning width of these cells increased by 79%, and the ratio of responses to the orientation orthogonal to the optimum and to the optimum increased from 0.16 to 0.46. The results were similar to those from area 17, in which 36 of 54 cells (66%) showed significant broadening of orientation tuning during GABA application, with a 90% increase in mean tuning width and an increase in the relative response to the orientation orthogonal to the optimum from 0.17 to 0.42. The distributions of cells in areas 17 and 18 with respect to the magnitude of GABA-induced effects on orientation tuning width were not significantly different (mean increase in tuning width: area 17, 102%; area 18, 87%). Although most cells were tested only with moving bars, comparable effects of remote GABA application on orientation tuning were observed when stationary flash-presented bars were used. Of 11 cells thus tested in area 18, seven showed significantly broader tuning during GABA application, with a 132% increase in mean tuning width. In some 25% of cells in each area which showed a significant effect of GABA application on orientation tuning the response to at least one non-optimal orientation exceeded, during GABA application, the response to the previous optimum. There was essentially no correlation between the changes in orientation tuning and changes in the level of spontaneous activity or in the response to the optimum orientation during GABA application. Thus, an increase in the general excitability of recorded cells or the loss of an unspecific inhibitory input cannot account for the effects of GABA application on orientation tuning. Remote GABA application presumably inactivated cells with different preferred orientations from that of the recorded cell. It is thus argued that the observed broadening of orientation tuning during GABA application reflected the loss of an inhibitory input tuned to non-optimal orientations.(ABSTRACT TRUNCATED AT 400 WORDS)