Breaking Down the Image

The process of breaking the visual field down into constituent features begins before visual information has even left the eye. Whereas receptor cells simply fire action potentials more rapidly the brighter the light that hits them, ganglion cells, the neurons whose axons leave the eye through the optic nerve, respond to particular

Figure 2 Unlike the situation shown in Fig. 1, real-world object perception almost always requires that the to-be-perceived object be separated from other elements in the visual field, as illustrated here. For example, the coffee mug is partially occluded by the plate and salad bowl and itself occludesparts of the tabletop, fruit bowl, and an apple. To deal with this complexity, the visual system breaks the image into constituent features such as edges and color patches, decides which features belong together and which belong in separate groups, and then recombines feature subsets to form representations of single objects.

Figure 2 Unlike the situation shown in Fig. 1, real-world object perception almost always requires that the to-be-perceived object be separated from other elements in the visual field, as illustrated here. For example, the coffee mug is partially occluded by the plate and salad bowl and itself occludesparts of the tabletop, fruit bowl, and an apple. To deal with this complexity, the visual system breaks the image into constituent features such as edges and color patches, decides which features belong together and which belong in separate groups, and then recombines feature subsets to form representations of single objects.

patterns of light over a small patch of the retina. More specifically, most ganglion cells exhibit center-surround receptive fields, firing most rapidly either to a small spot of light surrounded by darkness (an on-center receptive field) or to a small spot of darkness surrounded by light (an off-center receptive field).

By the time visual information has reached the primary visual cortex (also known as striate cortex and area V1), it has been recoded again. Here, neurons called simple cells respond most strongly to stationary bars (rather than spots) of light surrounded by darkness (or to dark bars surrounded by light). Furthermore, each of these neurons responds best to a bar in a particular orientation. A second type of neuron in V1, the complex cell, also responds to oriented bars but is tolerant of small shifts in location of the bars. In addition to bars, simple and complex cells also respond to the borders between regions of light and darkness. One prominent place where such borders occur is at the edges of objects, so that simple and complex cells can be considered edge detectors (see Figs. 4A and 4B). Oriented edges are thought to be some of the most important features coded by early visual processes.

Other types of elementary features thought to be coded in the primary and secondary visual cortex include color, line size, line curvature, edge corners and intersections, motion direction, and distance in 3D space. Each feature can be thought of as being coded in a topologically organized feature map, which holds information about which parts of the visual field contain the feature. Early visual cortical areas contain a large number of such feature maps, e.g., one map for right-tilting lines, one for horizontal lines, and one each for the colors blue, green, and red. For example, a pencil rolling across a desktop might engender activity on the yellow color, vertical line orientation, and rightward motion feature maps. By splitting up the information in this way, the visual system is able to process each type ofelementary feature separately, but at the same time.

Was this article helpful?

0 0
Breaking Bulimia

Breaking Bulimia

We have all been there: turning to the refrigerator if feeling lonely or bored or indulging in seconds or thirds if strained. But if you suffer from bulimia, the from time to time urge to overeat is more like an obsession.

Get My Free Ebook


Post a comment