The representation of Kanizsa illusory contours in the monkey inferior temporal cortex.

Stimulus reduction is an effective way to study visual performance. Cues such as surface characteristics, colour and inner lines can be removed from stimuli, revealing how the change affects recognition and neural processing. An extreme reduction is the removal of the very stimulus, defining it with illusory lines. Perceived boundaries without physical differences between shape and background are called illusory (or subjective) contours. Illusory and real contours activate early stages of the macaque visual pathway in similar ways. However, data relating to the processing of illusory contours in higher visual areas are scarce. We recently reported how illusory contours based on abutting-line gratings affect neurones in the monkey inferotemporal cortex, an area essential for object and shape vision. We now present data on how inferotemporal cortical neurones of monkeys react to another type of shapes, the Kanizsa figures. A set of line drawings, silhouettes, their illusory contour-based counterparts, and control shapes have been presented to awake, fixating rhesus monkeys while single-cell activity was recorded in the anterior part of the inferotemporal cortex. Most of the recorded neurones were responsive and selective to shapes presented as illusory contours. Shape selectivity was proved to be different for line drawings and illusory contours, and also for silhouettes and illusory contours. Neuronal response latencies for Kanizsa figures were significantly longer than those for line drawings and silhouettes. These results reveal differences in processing for Kanizsa figures and shapes having real contours in the monkey inferotemporal cortex.

Illusory shape representation in the monkey inferior temporal cortex.

Perceived boundaries without physical differences between shape and background are called illusory contours (ICs). ICs and real contours (RCs) activate the early processing stages of the macaque visual pathway and the occipitotemporal areas of the human visual system in a similar way. However, it is not known how these contours are processed further in the highest visual areas. We tested how the responses of inferior temporal cortical (IT) neurons of macaque monkeys change in relationship to figures with RCs or ICs. The same set of figures [coloured pictures, ICs and silhouettes (SILs)] was presented to awake, fixating rhesus monkeys while the single-cell activity was recorded in the anterior part of the IT. Most of the neurons responsive to RCs were also responsive to the same shapes presented as ICs. The average net firing rates, however, were significantly lower for the illusory stimuli than for the stimuli in the RC conditions, and the latency of the responses was significantly longer for the ICs than for the RCs. The shape selectivity was found to be different for coloured stimuli and ICs, and similar for SILs and ICs, suggesting the invariance of selectivity to shapes having the same contour but lacking internal surface information. These results suggest different modes of processing of RCs and ICs in the IT, which might explain the differences in their perception.

Relationship between stimulus complexity and neuronal activity in the inferotemporal cortex of the macaque monkey.

The single-unit activity of 217 cells was recorded from the inferotemporal cortex (IT) of two awake macaque monkeys while they performed a fixation task. The stimuli were coloured geometrical shapes or coloured representations of natural or artificial objects. To determine whether the stimuli could be separated into groups on the basis on neuronal population behaviour, the responses to the images were analysed by factor analysis and cluster analysis. It was a common result of each analysis that, on the basis of neuronal responses, the stimulus set could be separated into two groups, despite the lack of difference in mean response rate to them. Similar groups were formed when only the first half of the responses was analysed. The results suggest a differential coding of the images of simple geometrical shapes and of the images of complex, real (photographic) objects. We found significant differences between the two stimulus groups in physical features, other than size or luminance. Our results suggest that the same neurone population might respond differently to simple and complex images in the first 150 ms of their responses. The differences might be attributed to "non-obvious" physical features of the stimuli, such as the amount of internal lines in the images, colourfulness and the length of perimeter of the stimuli.

Effects of surface cues on macaque inferior temporal cortical responses.

Humans are able to recognize objects when surface details, such as colour, texture and luminance gradients, are not available. By systematically eliminating colour, texture, shading, contrast and inner contours from given objects, we tested whether certain shape-selective inferior temporal cortex (IT) neurons of awake rhesus monkeys remain selective for these objects as the surface information is reduced. In psychophysical experiments, we established that the rhesus monkey can identify the shape of a coloured object largely independently of its surface characteristics and, to a lesser degree, of its inner contours. Shape selectivity of the neurons does not change when texture and shading are concealed. The responsiveness of the neurons is also affected by the removal of these surface attributes. The IT neurons were found to respond highly similarly to objects brighter or darker than their background. Selectivity for shape is preserved when the contrast is reversed. Deletion of the inner contours, outlining the main parts of the objects, did not affect the responses and selectivity of the IT neurons. These findings indicate that the IT can contribute to the invariant perception of objects having different surface details.