Chapter 3
A Brain Change That Contributes To Reorganization In Cognition
In chapter 2, I described Piaget’s first four sensorimotor stages, and proposed that Stage 4 was fundamentally different from the earlier stages. The operational mode of the schemes or neural circuits during the first three stages is undiscriminating in their assimilation of aliment or sensorimotor activation —that is, the schemes are almost completely undifferentiated. Although Stage4 schemes are far from being differentiated from each other, they begin to be so.
In this chapter I explain how a type of maturation would assist, if it does not actually induce, the shift from Stage-3 cognition to Stage-4 cognition. I think that the shift from Stage-3 cognition to Stage-4 cognition constitutes a major reorganization of cognition.
Piaget proposed that the change from Stage-3 cognition to Stage-4 cognition was incremental —that it did not constitute a major reorganization. The Stage-3 child’s imitation of the movement of a swinging object, or her imitation of herself striking at the object by making an abbreviated striking movement, was as if she were saying, "I could if I wanted to. " From that point, it is only a half step to acting intentionally on objects, as the child does in Stage 4. Piaget saw making-spectacles-last —striking while watching or listening and the variations of these schemes —as transitional to Stage 4.
It is possible, as was proposed by Piaget, that the shift from Stage 3 to Stage 4 is merely an incremental change. Perhaps the emotions that signal success or failure guide the change from Stage 3 to Stage 4. The child might discover that she is successful at causing an object to move when she strikes a proximal object with her foot, and is unsuccessful when the object is out of reach. The child might discover that she succeeds in causing an object to move by striking her crib with her arm only when that object is attached to the crib, or that she succeeds in causing an object to move only when her arm is connected to the object by a string that is taut. Also, recall Piaget’s description of the steps involved in removing an intervening object to grasp a toy. First, the child reached for the toy (something the child has done since she was 4 months old). In the course of reaching, the child inadvertently depressed an intervening pillow. Next, the child learned to depress the pillow to get the toy, and then to knock the pillow aside to grasp the toy. Incremental transitions —such as the ones just described —could totally account for the change from Stage 3 to Stage 4.
But I think there is another, very different way to account for this change. I think that maturation of part of the visual system, which takes place toward the end of Stage 3, would certainly support the kinds of differentiation that characterize Stage-4 cognition. Complete myelination of the peripheral visual tracts would support the beginning discrimination of visual schemes of one object from those of another —a differentiation that is first manifest in Stage 4 (Malerstein, 1986).
Complete myelination of the peripheral visual tracts —the neural tracts that transmit electrical impulses from the retina to the cerebral cortex —takes place just before Stage 4. As long as myelination of the visual tracts that connect the retina to the cerebral cortex is incomplete, any organization of visual information coming to the cerebral cortex would be unstable. Not only is the organization of information coming from the eye stabilized by complete myelination of the visual tracts, but more significantly, the kind of organization of visual information that is stabilized is particularly suited to differentiate schemes that correspond to a world of separate objects —objects that have edges.
To understand my proposal, it is necessary to understand something about neural tracts and about the myelination of neural tracts. A neural tract is similar to an electric cable. An electric cable is a bundle of many thin wires. Each of the wires that make up the cable may be surrounded by insulation material —for example, plastic. A neural tract is made up of a bundle of axons, sometimes referred to as fibers. In a mature organism, a fatty substance called myelin surrounds the axons, or fibers, of some neural tracts, giving such tracts the appearance of insulated wire cables. Although myelination of the individual axons provides some insulation of one axon from another, the primary function of myelination of an axon is to increase the speed of transmission of electrical impulses along that axon. See Figure 1. The axons of neuron A and B are myelinated, and the axon of neuron C is not.
Three findings support the proposal that complete myelination of the visual tracts influences the shift from Stage 3 to Stage 4. First, complete myelination of the visual tracts occurs at the right time. Yakovlev and Lecours (1967) found that the visual tracts that transmit electrical impulses from the retina to neurons of the primary visual areas (V1) of the cerebral cortex are not completely myelinated until the end of month 4 or 5. 24, 25 This takes place just before the shift from Stage-3 cognition to Stage-4 cognition. The neurons of V1 are the points of entry into the cerebral cortex for visual information. Second, although myelination merely speeds transmission of electrical impulses along the axon of one nerve cell to the next cell, rapid transmission of electrical impulses is critical if a neural tract —a bundle of these axons —is to have stable downstream effects. Toward the end of Stage 3, once the visual tracts are fully myelinated, the schemes or neuronal circuits downstream from the V1 neurons have stable activations to work with as they interact with the world. Third, V1 neurons respond to particular visual patterns that are cast on the retina. Such selective responses to visual patterns organize neuronal circuits that are downstream from V1 in a manner that would promote differentiated circuits for different objects. As we saw in chapter 2, schemes or neuronal circuits of different objects begin to be differentiated in Stage 4.
I will describe how V1 neurons respond selectively to patterns of light, and how that selectivity promotes differentiated circuits for different objects. But first, I must explain how the complete myelination of a neural tract stabilizes activation of downstream neuronal circuits.
Complete Myelination of the Sensory Tracts Stabilizes Activation
As I noted above, myelin —the fatty sheath that surrounds the axons of certain neurons —speeds the transmission of electrical impulses along the axon of a neuron to downstream neurons. Speed of transmission of these impulses is critical for a neuron and the neuron, which is downstream from it, to retain their relationship. This is because the first impulses to reach downstream neurons will activate them and inhibit activation by impulses from other neurons that arrive a bit later. As I explained in chapter 1, the more often the relationship between two neurons is repeated, the stronger the relationship becomes. Also, there is evidence that neurons that are not activated lose their downstream connections (Lichtman & Coleman, 2000).
Until a tract —a bundle of neuronal axons —is completely myelinated , any transmission of impulses conducted by the axons of that tract to downstream neurons will be unstable over time. In order to understand this, consider two neurons that I will call P and Q, whose axons are part of a tract. At the downstream end of the tract are two other neurons that I will call P1 and Q1, which are neighbors. Assume that neuron P’s axon has the shortest route to neuron P1, and that neuron Q’s axon has the shortest route to neuron Q1. The first impulses to reach a downstream neuron activate that neuron. Hence, all things being equal, impulses from neuron P will activate neuron P1, and impulses from neuron Q will activate neuron Q1. However, because myelinated axons transmit impulses more rapidly than unmyelinated axons, if P’s axon is not myelinated and Q’s is, impulses from neuron Q may activate not only neuron Q1, but also neuron P1. Once neuron P’s axon is fully myelinated, however, impulses conducted by axon P will reach neuron P1 earlier than impulses conducted by axon Q, because axon P’s route to P1 is shorter. Thereafter, neuron P will activate neuron P1, and neuron Q will activate neuron Q1. The relationship of P to P1 and Q to Q1 and activation of neuronal circuits that are downstream from P1 and Q1 will then be stable. However, prior to the myelination of P’s axon, as a consequence of this fluctuating activation of P1 and Q1, activation of neuronal circuits that are downstream from P1 and Q1 will have been unstable.
Thus, if speed of transmission of impulses conducted by a neural tract to downstream neuronal circuits becomes stable only when myelination of a neural tract is complete, only then, does activation of the downstream neuronal circuits also become stable. 26, 27 Furthermore, the cells that make up the myelin sheath secrete proteins that inhibit axonal growth cones (Woolf & Blocklinger, 2002). These proteins are thought to inhibit the repair of damaged nerves. Such proteins would contribute stability to existing myelinated tracts, and thus to any downstream neuronal circuits.
The Primary Visual Area (V1) Cells of the Cerebral Cortex —Selective Gates into the Brain
Fast —hence stable —neural transmission of impulses from the cells of the retina to the visual cortex 28 provides a new tool to be exploited in the differentiation of schemes of different objects, differentiation that is manifest in Stage 4. To understand how this works, it is necessary to understand how the V1 cells operate. This understanding is based on studies in monkeys done by Hubel & Weisel (1979). Some V1 cells are most active when lines or boundaries of light of a specific orientation fall on the retina. 29 For example, certain V1 cells are most active when a vertical line or vertical boundary of light falls on the retina, and basically inactive when that line of light is off-the-vertical by as little as 10 to 20 degrees (On a clock, 15 degrees is the difference between 12 and 12:30). These V1 cells are seldom spontaneously active. When stimulated, they go from 0 to 40 firings per second (De Valois & De Valois, 1988). Other V1 cells are most active when the line of light that falls on the retina is at an angle that is 10 to 20 degrees off the vertical. Other V1 cells are most active when the angle is still greater, and so on. Still other V1 cells are most active when the light that falls on the retina is moving in a particular direction. It follows that until the neurological connections between the retina and the cerebral cortex are stable —that is, until the visual tracts are completely myelinated —the responsiveness of these V1 cells to orientation of lines of light, and to the movement of light, would be unstable. Accordingly, until these visual tracts are completely myelinated, activation of neuronal circuits that are downstream from the V1 cells would be unstable.
Movement of edges of light as a set is the single most reliable visual measure for distinguishing one object from another. A set of edges of light —lines or boundaries of light —that belong to one object, which differ from a set of edges of light that belong to another object, generally discriminates the two objects. This is especially true, if one of the sets moves relative to the other set.
Colors, patterns, and textures also help to distinguish one object from another. But they are less reliable in this respect than are sets of edges that move relative to each other. Different objects may share the same colors; a green plate may rest on a green table. Discrete objects may share patterns or textures, making it difficult to distinguish one object from the other. A single object may be multicolored or have more than one pattern or texture. In such instances, differences in color, pattern, or texture could mislead one into thinking that an object was two or more discrete objects.
Toward the end of Stage 3, the neural tracts from the retina to V1 are completely myelinated. V1 cells may then act as reliable gates to downstream neuronal circuits in the cerebral cortex. Consider a V1 cell that is most active when an edge of light cast on the retina is vertical. When this cell is activated, it will transmit electrical impulses to a particular downstream group of cells —that is, to a particular neuronal circuit or scheme. Another cell, which is most active when the edge of light cast on the retina is 20 degrees off vertical, will transmit electrical impulses to different downstream neurons. Similarly, impulses from a V1 cell that is most active when the light cast on the retina is moving in a particular direction, activates its own set of neurons.
Once input to the primary visual area cells is stable, they act as gates to downstream neuronal circuitry —or scheme —activation. They reliably segregate downstream schemes, based on the orientation of an edge of light and on the movement of light that is cast on the retina. Downstream schemes, distinguished from each other in terms of edges and movement of light, are then available for developing constructs —scheme relationships —that work better in a world composed of separate objects that are impervious to light.
Taking into account these facts, let us look again at an experiment that Piaget conducted on his Stage-4 child —the experiment that involved an object that was placed on a platform. When Piaget placed a matchbook on the platform, his child did not reach for the matchbook unless Piaget tilted the platform and the matchbook slid. When Piaget placed a goblet on the platform, his child reached for the goblet.
Two points are to be made. First, the Stage-4 child’s failure to reach for the matchbook on a platform indicates incomplete division between the schemes of two different objects. The scheme of the matchbook and the scheme of the platform are not distinct from each other if the matchbook does not move relative to the platform. Second, the child apparently uses edges and their movement to differentiate the scheme of one object from the scheme of another. The edges of the sliding matchbook and the edges of the goblet are distinct from the edges of the platform. When the edges of objects are distinct from the edges of a platform, the Stage-4 child reaches for the objects. His schemes of the objects are distinct from his scheme of the platform.
It is opportune that, just before Stage 4, the gate cells of V1 are brought on-line by full myelination of the visual tracts to the cerebral cortex. These gate cells, which are differentially sensitive to orientation of edges and movement of light, would automatically segregate downstream schemes —or neuronal circuits —based on the orientation of edges and on whether they move in a particular direction. In late Stage 3, based on edges and movement of light cast on the retina, visual schemes are newly organized into reliably segregated schemes. The schemes or neuronal circuits downstream from V1 that correspond to two different objects would be segregated from each other depending on whether the sets of edges of the two objects were oriented differently and/or on whether the edges moved relative to each other. Late in Stage 3, these schemes that are now segregated in terms of edges and movement work more successfully in their interaction with a world of different objects. The infant is better able to grasp the things he wants.
What I have said here about the visual system applies to the touch system as well. The touch system operates analogously to the visual system in terms of edges and movement (Gardener & Kandel, 2000). The neural tracts of the touch system —from the skin to the primary somatosensory area of the cerebral cortex (S1) 30 —is completely myelinated later than the neural tracts of the visual system, at about 12 months (Yakovlev & Lecours, 1967). However, the neural tracts that serve the upper limbs and the head —the relevant parts for early scheme organization —are probably completely myelinated earlier. 31 Maturation of the touch system could supplement object differentiation for sighted infants. In blind infants, it could be a primary aid in the differentiation of object schemes. 32
During Stage 4, children are students of objects —presumably of their edges and of their displacement. As they watch a new object, they will pick it up, rotate it, bring it closer to their eyes, and then extend their arms. They may do this over and over with the same object. Sometimes they pick up a familiar object and examine it in the same detail , as if it were new to them.
Children in the next stage —Stage 5 (12-16 months) — are advanced students of edges. They pull on a string, a refinement or extension of edges, to obtain an object. They pick up a matchbook that rests on a platform even when the matchbook is not moving, and they may pull on the platform to bring the matchbook within reach. The Stage-5 child apparently distinguishes the edges of the matchbook scheme from those of the platform scheme even when the matchbook and the platform are not moving relative to each other. The child has learned more about certain edges.
The Stage-5 child’s schemes that are segregated by edges works increasingly better in a world composed of separate objects that are impenetrable to light or to touch. Nonetheless, the Stage-5 child may at first attempt to put a ring on a stick by touching the ring to the side of the stick. Apparently, the child’s construct —the edge quality of his schemes that segregates schemes of different objects —remains permeable. Our adult construct is that edges are often good indicators of separate, solid objects.
The Stage-5 child exploits the edge segregation of his schemes as they interact with the world. He discovers that he cannot pass a ring through a stick. As noted, he pulls on a platform to retrieve an object that rests on the platform, and pulls on a string —an extension of the edges of an object —in order to get the object. The child discovers how his schemes of different objects, segregated by different sets of edges and movement of light and touch stimuli, work or do not work —that is, bring satisfaction or dissatisfaction.
I argue that, ushering in Stage 4, complete myelination of the visual and somatosensory tracts to their respective primary sensory areas of the cortex is an important mechanism for distinguishing the schemes of different objects. Complete myelination of these tracts provides rapid, hence stable , transmission from peripheral visual and touch cells to the primary visual and somatosensory areas, respectively, of the cortex. When activated, the schemes, separated by visual and by somatosensory edge orientation and movement, are strengthened as they interact successfully with a world of solid objects. These schemes find what works, what is successful, what brings satisfaction or dissatisfaction.
Complete myelination of the visual and somatosensory tracts to the cerebral cortex need not be the only mechanism used to shift from Stage 3 to Stage 4 cognition. 33 Other maturational factors that are yet to be discovered could be in play. Typically, biological systems have built-in redundancy. That is, they often have several ways of achieving the same end.
Additionally, as valuable as the orientation of edges and their movement as a set are in learning to differentiate discrete objects, they are insufficient in themselves to differentiate certain objects. Some objects cannot be moved —for example, a mountain. The edges of some objects are not visible —for example, the earth; or are not touchable —for example, the moon. Yet adults talk about such objects as if they were distinct. Children, however, may continue to use edge orientation and movement to define objects as late as Piaget’s Preoperational Period (2-7 years). I will give examples of this in chapter 6.
Some theoreticians infer that synaptic pruning results in major cognitive reorganization. Synaptic change —whether strengthening or pruning —probably is responsible for incremental change —that is, assimilation and accommodation. I doubt that synaptic change is ordinarily responsible for major reorganizations, such as the transition from Stage 3 to Stage 4 appears to be.
More About Stage 4
There are advantages to the infant’s waiting until Stage 4 before he differentiates schemes of different objects. If we think back to the cumulative mode of the first three stages, we see how rich and interconnected the schemes have become during the first 7 or 8 months. A particular toy scheme is perhaps pleasurable, graspable, bangable, suckable, watchable, and so on. In Stage 4, when the schemes begin to separate into different object schemes, the aliment that has been assimilated to form a toy’s scheme is rich and interconnected (and, though interconnected, perhaps multiple). Being interconnected, the toy’s scheme is accessible for activation —that is, it may be remembered or reconstructed —through many and varied routes (and perhaps in essentially distinct locations). The toy’s scheme may be activated through pleasurable, graspable, suckable, and so on. Once differentiation begins, it is a rich, highly accessible scheme. Since differentiation is by its nature constraining —for example, the toy is or is not reachable —both the richness of the schemes and their accessibility would be reduced had the differentiation been made earlier.
In chapter 2, I proposed that scheme organization shifts significantly between Stage 3 and Stage 4. In this chapter, I theorize that maturation of two sensory tracts to the cerebral cortex —their complete myelination —late in Stage 3 (for the sighted), just before Stage 4, fosters differentiation of schemes in terms of object edges and their movement. It is reasonable to assume that this maturational factor assists the shift in scheme organization that characterizes Stage 4. Other findings suggest that this maturational factor continues to play a role in the incremental progress from Stage s 4 to Stage 6 of the Sensorimotor Period. This incremental progress in scheme organization depends on monitoring by emotions, which signal success or failure.
Before going on to the next chapter, I wish to make one additional point. I have described how an undifferentiated, global scheme —a widespread neuronal circuit —would be affected by a maturational factor —the full myelination of neural tracts to edge-movement gate cells, entry points to the cerebral cortex. This factor assists the differentiation of schemes —more modular neuronal circuits —for objects, including the self.
My thesis of how a maturational factor interacts with undifferentiated schemes obviates what investigators who study neurophysiological function in mature organisms refer to as the binding problem. The binding problem arises from the fact that sensory input is broken down into small components that may be located in distant areas of the brain. For example, a scene is deconstructed into colors, edges, and movement, and "what" and "where" take different pathways in the cerebral cortex. The problem, as posed by these investigators, is how does the brain integrate these components and different pathways into a coherent perception of an object or the sense of a self?
If we accept that the functional unit of early brain is an undifferentiated and global scheme —a whole —there is no binding problem. We have no problem of how the components get together. They are together to begin with.
Components are what in the course of development are sculpted from an early whole —that is, a scheme. Complete myelination of a neural tract (as I described above), a surge in the number of synapses, a surge in hormones, or environmental influences, including a change in culture —such as the development of a written language, or of a system of mathematics —may carve out components.
In the next chapter, I will reexamine the sensorimotor schemes to suggest how the brain can give rise to conscious schemes. I begin by describing the reticular activating system of the midbrain and its role in control of the sleep-wake cycle. I will argue that, beginning in Stage 3, all stage-typical schemes are conscious schemes —that a portion of each stage-typical scheme is waking-state activation.
24 Yakovlev and Lecours (1967) acknowledged that their study was crude. They worked with postmortem specimens. Their standard of complete myelination was the depth of color of haematoxylin-stained sections of brain tissue in a healthy 28-year-old male. In order to assess stage of myelination at different ages, they compared the depth of color of a neural tract of their standard to the same neural tract in the brains of fetuses, children, and young adults. Now it is possible to visualize brain function and anatomy in live humans, using different types of brain-scanning devices. Each device has its strengths and its limitations. Thus far, for tracing the development of myelination, the work of Yakovlev and Lecours, especially for small or scattered tracts, constitutes the best data we have.
25V1 is located on the inner surface of the occipital lobe of each cerebral hemisphere. See Figures 2 and 4.
26 See chapter 1, footnote 9. Merzenich (1998) provides evidence of the potential plasticity of neuronal relationships in the cerebral cortex, even in mature brains.
27 In the discussion that follows, I will continue to discuss complete myelination of a tract as if stabilization of downstream circuits required absolutely complete myelination. I assume, however, that when most of a tract is fully myelinated, transmission by the tract to downstream circuitry is reasonably stable.
28 Figure 5 is a schematic drawing of the visual system. The axons of ganglion cells constitute the optic nerve, which connects the retina to the lateral geniculate ganglion of the thalamus —a relay station of the visual system. The geniculocalcarine tract connects the lateral geniculate ganglion to V1 cells located in the calcarine fissure of the occipital lobe of the cerebral cortex.
29De Valois and De Valois (1988) found that X and Y ganglion cells also respond differentially to edges and movement of light. The axons of the ganglion cells are fully myelinated earlier than those that make up the geniculocalcarine tract, which is the last leg of the visual tracts to be fully myelinated. The connection of X and Y cells to the downstream cells of the geniculocalcarine tract, however, is not coordinated. Thus, the X and Y cells could not be expected to play a role in differentiating object schemes in Stage 4 in humans. It is possible, however, that these cells might play such a role in animals that are less cortically dependent, or even in newborn human infants. This might help to explain any findings of very early, but temporary, object differentiations in newborns.
30 S1 is a strip of cortex on the outside surface of both cerebral hemispheres just behind the central sulcus (a crevice in the surface of the brain). See Figures 2, 3 and 4. Cells of area 1 of S1 receive stimuli from nerve endings that are activated by touch. Electrical stimulation of area 1 produces a sensation of tingling at a particular point on the opposite side of the body. For the most part, not only is body representation in the brain transposed right to left —so that the right side of the brain connects to the left side of the body —but representation is also inverted. For example, sensation will be felt at a point of the upper part of the body —the tongue and thumb —if the bottom of area 1 is stimulated. See Figures 3 and 4.
It should be noted that area 1 cells respond differentially to edge orientation and direction of movement of touch stimuli.
31 Maturation generally begins at the head end of an organism and proceeds toward its tail.
32 Proprioception —the sensory system that responds to position and movement of muscles, tendons, and joints —plays a role in differentiation of objects. For example, cells of area 2 —a part of S1 that receives both touch and proprioceptive stimuli —respond selectively to round and rectangular objects (Gardner & Kandel (2000). What might be analogous to edge-orientation and motion selectivity of visual and touch system cells of the cerebral cortex is not known for the proprioception system. Spencer’s finding of a constant relationship of shoulder torque to elbow torque at about 4 months when the infant reaches for an object is a lead regarding the timing of maturation and the function of the proprioception system that serves the arm. See Chapter 1.
33 Like I, Diamond (2001) proposed a role for myelination in cognitive development. Drawing on studies of children and monkeys, she proposed that increased myelination of the dorsolateral prefrontal cortex enables children and monkeys to overcome the A-not-B error. Adults with prefrontal cortical damage tend to be uninhibited —have difficulty adhering to a plan of action —and tend to perseverate —repeat a behavior that no longer makes sense. Apparently, her theory is that a more mature prefrontal cortex should help the child to inhibit searching for an object where he last found it —the A-not-B error —rather than searching for it where he last saw it. If Diamond is correct, then prefrontal myelination would assist the transition from Stage 4 to Stage 5. It should be noted, however, that while there are spurts of myelination of the prefrontal cortex in early childhood, myelination of the prefrontal cortex is not complete until adolescence or early adulthood (Yakovlev & Lecours, 1967).
< Chapter 2 | Contents | Chapter 4 >
Do you have a comment or would like to open a discussion? The author would be pleased to hear from you. Contact AJ Malerstein here.
