Introduction
This activity simulates an experiment in which we map out the receptive fields of neurons in striate cortex, the first area of the cerebral cortex to receive information from the eyes. Neurons in striate cortex respond best to bars of light, rather than the spots of light that maximally stimulate retinal ganglion cells.
Experimental Procedure
After determining the response characteristics of retinal ganglion cells (as demonstrated in the activity on Ganglion Receptive Fields in Chapter 2), some researchers turned their attention to cells farther along in the visual system, in the first area of the cerebral cortex to receive information from the eyes. This area is called striate cortex in humans, and primary visual cortex in the cats that were the subjects of these early single-cell recording studies.
As described in your textbook, Harvard neuroscientists Hubel and Wiesel were attempting to stimulate visual cortex cells with spots of light that they manipulated using glass slides. They failed to stimulate any cells with the spots, but serendipitously discovered that the edges of the slides, which effectively showed bars of light to the cells, did cause many neurons to respond. By switching to light bar stimuli and carefully observing neuronal responses as they moved and rotated the bars, Hubel and Wiesel were able to discover several important characteristics of these cells’ receptive fields.*
Real neurons can fire at rates ranging from zero (if the neuron is dead or extremely inhibited) to hundreds of action potentials per second. To simplify things, our simulation of Hubel and Wiesel’s experiment represents neural firing rates on a scale of 0 to 100. Note that there is a certain amount of randomness associated with neural firing rates, so the firing rate is constantly changing a little bit, even if you don’t move or change the bar.
*More recent research has indicated that Gabor patches (see the activity on these stimuli) actually stimulate striate cortex cells better than bars of light, but we will use the simpler, “classic” version of the experiment in this activity.
Review: What Is a Receptive Field?
As you learned in Chapter 3, every neuron in the visual system has a distinctive receptive field—an area of the retina that the cell responds to, along with a particular pattern of light that must be present in that area for the neuron to respond. As we move farther along in the visual system, receptive fields generally get larger (that is, neurons respond to larger areas of the retina) and more complex. In the retina, ganglion cells come in two types, on-center and off-center (see the activity on Ganglion Receptive Fields to review these receptive field types). In contrast, some 32 types of striate cortex receptive fields are illustrated in this activity, yet several important receptive field characteristics of these cells have still been left out for simplicity’s sake.
As you read about and explore the receptive field characteristics of striate cortex cells, note that receptive fields are not “all-or-nothing” propositions. For example, a cell may respond best to a vertically oriented bar of light. If the bar is tilted slightly to the left or right, however, the cell will probably still respond somewhat. It just won’t quite respond at its maximum firing rate. Such response gradation is characteristic of almost all areas of the nervous system, and this mechanism is quite different than that used in human-made computers such as the one you are using now.
Receptive Field Characteristics of Striate Cortex Neurons
As noted in the previous section, striate cortex neurons vary in many different ways, some of which are listed below. Descriptions of the receptive field characteristics are listed below.
-
Bar width: Neurons may respond best to narrow or wide bars of light (in reality there are many more gradations of receptive field widths than illustrated here). In our simulation, the firing rate is cut in half if the bar of light is the wrong width for the cell’s receptive field.
-
Bar orientation: Neurons may respond best to vertical, left-oblique, right-oblique, or horizontal bars (again, in reality some cells might respond best to intermediate angles like 20 degrees). In our simulation, if the bar orientation is off by 23 degrees, the firing rate is cut in half; if the bar orientation is off by 45 degrees, the firing rate decreases 80%; and orientations off by more than 45 degrees cause no response in the cell at all.
-
Cell type: Simple cells respond maximally only when the bar of light is in the center of the cell’s receptive field, whereas complex cells respond at the same rate across the entire width of the receptive field.
-
End stopping: Some neurons, known as end-stopped cells, respond best when the bar of light terminates within its receptive field. Neurons that are not end-stopped respond best if the bar extends through the entire receptive field.
-
Edge- and bar-detectors: While many neurons respond best to bars of light surrounded on either side by darkness, as illustrated in our demonstration, other neurons respond best to single edges—light on one side and darkness on the other. Moreover, some edge-detection neurons respond best when the light is on the left side of the edge, while others respond best when the light is on the right. (This characteristic is not illustrated in our demonstration.)
-
Receptive field size: As with retinal ganglion cells, striate cells vary in the size of their receptive fields—the extent of the area of the visual field to which each cell responds. (This characteristic is not illustrated in our demonstration. The bar width in our simulation varies, but that is not the same as the receptive field size.)
-
Motion: Some neurons respond best when bars are in motion across their receptive field, while others respond best to static stimuli. Moreover, a motion-sensitive cell might respond strongly to leftwards movement but show no response at all to rightwards movement. (This characteristic is not illustrated in our demonstration.)
-
Color: Some neurons are color-selective, responding best to green, red, blue or some other color. Other neurons respond equally well to any light color. (This characteristic is not illustrated in our demonstration.)
-
Ocular dominance: Some neurons respond better to light presented in the left eye (these neurons are said to be left-eye dominant), while others are right-eye dominant. (This characteristic is not illustrated in our demonstration.)