Neural Basis Of Sensitive Periods Research Paper

1. Examples Of Developmental Phenomena With Sensitive Periods

1.1 Imprinting And Song Learning

Probably the best known example of early external influences of the environment on the organization of behavior is the so-called ‘imprinting’ process (Lorenz 1935), by which a young bird restricts its social preference to a particular animal or object. In the course of ‘filial imprinting,’ for example, a young chick or duck learns about the object that it has followed when leaving the nest (Hess 1973). Young zebra finches, in the course of ‘sexual imprinting’ (Immelmann 1972), learn the features of an object that subsequently releases courtship behavior in fullygrown birds. In addition to these two phenomena, many other paradigms of imprinting have been described, as for example homing in salmons, habitat imprinting, acoustic imprinting, or celestial orientation in birds.

All imprinting phenomena are characterized by at least two criteria (Immelmann and Suomi 1982): First, learning about the object which the bird is following later on, or to which courtship behavior is directed, is restricted to a sensitive period early in development. In the case of filial imprinting, this phase is quite short (several hours) and it starts directly after hatching. In sexual imprinting, which has been investigated mainly in birds, which are hatching underdeveloped and with closed eyes, the sensitive period starts at the day of eye opening, and may last for several days. The second feature of all imprinting paradigms examined so far is that the information storage is rather stable. The preference for an object to follow or to court, which has been established in the course of the sensitive period, cannot be altered later on. Whenever the bird has a chance to choose between the imprinted object and another one, it will choose the familiar imprinted object.

Song learning is, at the first glance, a little bit more complicated than imprinting. It has been shown to comprise two parts (Konishi 1965, Marler 1970). Early in life a young male bird (only males, in most avian species, sing) learns about the song he himself is singing later when adult, and he is learning mainly from his father. At the time of learning, the young male is not able to sing by himself. This time span is called the acquisition period, and it is thought that the male is storing some template of the song he has heard during this phase of learning. When the bird grows older, he starts singing by himself, and it has been shown that he tries to match his own song with the previously acquired template. During this ‘crystallization period,’ the young male selects its final song from a bigger set of other songs that he was singing at the beginning. Thereafter, this selected song remains stable and shows only minor variation. The other songs are, in most cases, not uttered any longer.

Song learning thus shows the same characteristics as imprinting. It occurs during a sensitive period, and after the crystallization period, the song that is selected cannot easily be altered. Recent research indicated that a second event like crystallization could also be demanded in imprinting. At least formally, one can separate an acquisition period (which is the ‘classical’ sensitive period) and a second event that may be called stabilization also in imprinting (Bischof 1994).

As already mentioned, development is an interplay between genetic instruction and acquired information. This is also the case in imprinting and song learning (Bischof 1997). In imprinting, not only the behavior for which the object is learned is genetically determined, there are also some genetic constraints which at least lead to advantages for certain objects to be learned easier than others. In filial imprinting, it has been shown for example that there is a natural preference for red over other colors, and this preference can be enhanced or diminished by selective breeding. In song learning, only a small variety of songs can be learned by most species, and there is some indication that certain features of song are innate and not alterable by early learning.

1.2 Plasticity Of The Visual Cortex In Cats

The best-known example of sensitive periods in neural development is the plasticity of neurons of the visual cortex in cats (Hubel and Wiesel 1970). In the adult cat, most neurons in area 17 are driven by visual stimulation of the left as well as the right eye, and are thus defined as binocular. If one eye of a kitten is briefly sutured closed in its early postnatal life, the access of the eyes to cortical neurons is dramatically altered. There is an obvious lack of binocular neurons in the visual cortex of such kittens, and most of the neurons are driven exclusively or are at least strongly dominated by, the non-deprived eye. These changes in ‘ocular dominance distribution’ are only observed if monocular deprivation occurs during postnatal development; the same deprivation in an adult cat does not cause any change. Thus, there is a sensitive period during which the alteration of the visual input affects the wiring of neurons within the visual cortex. The wiring, which has been established after the end of this sensitive period, remains stable for the rest of life. Most of the results obtained in cats were later on confirmed and extended by research on other animals, including monkey, ferret, rat, and mouse (Berardi et al. 2000).

As in imprinting and song learning, there is evidence that in addition to early learning, genetic instruction plays an important role for the organization of the visual cortex. The basic pattern of wiring is already there at birth, and this basic pattern is either stabilized during the sensitive period if the sensory information is adequate, or can be altered if the information coming from the eyes deviates from normal. For example, if one eye is closed as in the experiments above, or if the eyes are not aligned appropriately as it is the case in squinting.

2. Similarity Of Events With Sensitive Periods On The Behavioral And On The Neuronal Level

The short overviews in Sect. 1 already show that in addition to the fact that the environment affects the organization of brain and behavior only during a sensitive period, there are three features that are similar in all examples. First, there is some genetic preorganization that is maintained or altered by sensory information during the sensitive period. Second, the shape of the sensitive period over time is very similar in all cases: the effect of environmental information increases quickly to a maximum, and becomes smaller much slower, going asymptotically to zero. Third, if the sensitive period is over, the brain structure or the behavior is not easily altered again. However, so far our comparison was only on phenomena, the effects may be due to very different mechanisms. It is therefore necessary to go into more detail and to discuss the mechanisms by which storage of information occurs in the different paradigms, and how the start, the duration and the end of sensitive periods, respectively, is determined. Sects. 2.1–2.3 will show that there is indeed a lot of similarity on the level of mechanisms.

2.1 Control Of Sensitivity

The control of the time span over which external information is able to affect brain and behavior, was initially thought to be due to a genetically determined window which was opened for some time during development allowing external information to access the CNS. However, this turned out to be too simple an idea because the environmental sensitivity could be shifted in time due to experimental conditions (Bateson 1979). Dark rearing, for example, delays the sensitive period during which ocular dominance can be shifted by monocular deprivation, and the sensitive period for imprinting lasts longer if the young bird is isolated and thus not able to see the appropriate stimulus which leads to imprinting. In imprinting, one can also show that the sensitive period is prolonged if the stimulus is not optimal. For example, exposing a zebra finch male on another species like a Bengalese finch leads to a prolongation of the sensitive period.

The ideas, which were raised to explain these phenomena, were as follows (Bischof 1997): the natural onset of the sensitive periods coincides with the functioning of the sensory systems involved. Thus, the sensitive period for sexual imprinting in zebra finches starts with about 10 days when the eyes are fully open, while in the precocial chicks, filial imprinting starts directly after birth because these birds are born with open eyes. The sensitive period for monocular deprivation should also start at eye opening. This is roughly correct, but the effect is quite low at the beginning, and some recent results indicate that eye opening may trigger some intermediate events that then lead to enhanced sensitivity of the affected area of the brain.

Why does the sensitive period have a time course with a quite sharp increase to a maximum of sensitivity, but a slow, asymptotic decrease? One idea is that the sensitive period is some self-terminating process. If we suppose that an object is described by a limited amount of information bits, or, on the storage side, there is a limited store for the information which has to be acquired, it is easy to imagine that the probability for storage of a given bit of information is high at the beginning and goes asymptotically to zero, dependent on the amount of information already stored (Bischof 1985).

2.2 Sites And Modes Of Storage

For all the examples of phase specific developmental phenomena in Sect. 1, the locations within the brain are known where the plastic events can be observed. Therefore, it is possible to compare the changes of wiring between different examples. It was the visual cortex plasticity where it was detected first that the anatomical basis for the development of neuronal specificity was a segregation of previously overlapping neuronal elements within area 17 (LeVay et al. 1978). Thus, the specification of neurons was caused by a reduction of preexisting neuronal elements. This principle was also found in imprinting and song learning. In both paradigms, the spine density of neurons within the areas that are involved in the learning process was substantially reduced in the course of the sensitive period (Bischof 1997). This indicates that pruning of preexisting elements is an essential part of the physiological mechanisms underlying phase dependent developmental plasticity and learning. The reduction of spine density is stable thereafter; it can not be enhanced by any treatment when it has occurred once. This is also an indication that the reduction of spines may be the anatomical basis of imprinting like learning and developmental plasticity.

2.3 Ultrastructural Events

It has long been speculated that the machinery causing learning induced changes during imprinting, song learning, and cortical plasticity may be different from that causing changes in adult learning, because there is a decrease in spine density instead of an increase, and the changes are stable and cannot be reversed. However, concerning the basic machinery, no significant differences were found. Developmental learning can obviously also be explained by ‘Hebbian’ synapses which strengthen their connections if pre-and postsynaptic neurons fire together, and disconnect if the activity is asynchronous. To cause changes in the postsynaptic neuron, NMDA receptors are involved as in other learning paradigms, causing Long Term Potentiation (LTP) or Long Term Depression (LTD) as well as the cascades of second messengers which finally lead to the activation of the genome which then causes long term changes in synaptic efficiency.

The difference to adult learning obviously lies in the fact that plasticity is limited to a certain time span. Many ideas have been developed which systems may gate plasticity (Katz 1999). One of the earliest ideas was that myelination delimits plasticity. Unspecified projections, adrenergic, serotoninergic or cholinergic have been shown to play a role where it was investigated. Recent links from experiments with knockout animals point towards neurotrophic agents which may only be available to a limited amount; when the resource is exhausted, plasticity is no longer possible (Berardi and Maffei 1999). Another very interesting finding is that inhibition plays a role (Fagiolini and Hensch 2000); neurons may have to reach a genetically determined balance of inhibition and excitation to become plastic. Whether these latter ideas can also be applied to the other early learning paradigms has to be examined.

3. Is Generalization To Humans Possible?

One has to be very careful if one generalizes examples from one species to another, and this is even more true for generalization from animals to humans. However, there are some hints that at least part of the results described in Sect. 2 can be applied to humans (Braun 1996). On the neuronal level, it is globally accepted that amblyopia, a visual deficit, is based on the mechanisms described for the development of the visual cortex. It has been shown that if one corrects in humans the misalignment of the eyes which causes amblyopia during early development, the connection between eyes and cortical neurons, and this is no longer possible in adults (Hohmann and Creutzfeldt 1975).

On the behavioral level, it has been shown that language learning has so much in common with song learning (Doupe and Kuhl 1999) that it is intriguing to speculate that the neuronal machinery may be similar in both paradigms. However, one has to be aware that the similarity is as yet only on the phenomenological level.

Since the early days of imprinting research, it has also been discussed whether aggressiveness, social competence, and similar things are imprinted (Leiderman 1981), and whether this can also be applied to humans. The frightening idea was that in this case parents could easily make big mistakes if they did not confront their children with the appropriate surrounding. However, evidence is sparse even in animals that social competence is severely influenced by early experience. If there is an influence, ways are available, even in the case of sexual imprinting in the zebra finch, to at least overcome temporarily the effects of imprinting. However, that imprinting effects are in most cases only covered but not eliminated, may be reason enough to pay some attention to the conditions under which children grow up.

Bibliography:

1. Bateson P 1979 How do sensitive periods arise and what are they for? Animal Behaviour 27: 470–86
2. Berardi N, Maffei L 1999 From visual experience to visual function: Roles of neurotrophins. Journal of Neurobiology 41: 119–26
3. Berardi N, Pizzorusso T, Maffei L 2000 Critical periods during sensory development. Current Opinion Neurobiology 10: 138– 45
4. Bischof H-J 1985 Environmental influences on early development: A comparison of imprinting and cortical plasticity. In: Bateson P P G, Klopfer P H (eds.) Perspectives in Ethnology, Vol. 6. Mechanisms. Plenum Press, New York, pp. 169–217
5. Bischof H-J 1994 Sexual imprinting as a two-stage process. In: Hogan J A, Bolhuis J J (eds.) Causal Mechanisms of Behavioural Development. Cambridge University Press, Cambridge, UK, pp. 82–7
6. Bischof H-J 1997 Song learning, filial imprinting, and sexual imprinting: Three variations of a common theme? Biomedical Research–Tokyo 18(Suppl. 1): 133–46
7. Braun K 1996 Synaptic reorganization in early childhood experience and learning processes: Relevance for the development of mental diseases. Zeitschrift fur Klinische Psychologie Psychiatrie und Psychotherapie 44: 253–66
8. Doupe A J, Kuhl P K 1999 Birdsong and human speech: Common themes and mechanisms. Annual Review of Neuroscience 22: 567–631
9. Fagiolini M, Hensch T K 2000 Inhibitory threshold for criticalperiod activation in primary visual cortex. Nature 404: 183–6
10. Hess E H 1973 Imprinting: Early Experience and the Developmental Psychobiology of Attachment. Van Nostrand Rein-hold, New York
11. Hohmann A, Creutzfeldt O D 1975 Squint and the development of binocularity in humans. Nature 254: 613–4
12. Hubel D H, Wiesel T N 1970 The period of susceptibility to the physiological effects of unilateral eye closure in kittens. Journal Physiology (London) 206: 419–36
13. Immelmann K 1972 The influence of early experience upon the development of social behaviour in estrildine finches. Proceedings of the 15th International Ornithological Congress. The Hague 316–38
14. Immelmann K, Suomi S J 1982 Sensitive phases in development. In: Immelmann K, Barlow G W, Petrinovich L, Main M (eds.) Behavioural Development. Cambridge University Press, Cambridge, UK, pp. 395–431
15. Katz L C 1999 What’s critical for the critical period in visual cortex? Cell 99: 673–6
16. Konishi M 1965 The role of auditory feedback in the control of vocalization in the white-crowned sparrow. Zeitschrift fur Tierpsychologie 22: 770–83
17. Leiderman P H 1981 Human mother-infant social bonding: Is there a sensitive phase? In: Immelmann K (ed.) Behavioral Development. The Bielefeld Interdisciplinary Project. Cambridge University Press, Cambridge, UK
18. Le Vay S, Stryker M P, Shatz C J 1978 Ocular dominance columns and their development in layer IV of the cat’s visual cortex: A quantitative study. Journal of Comparative Neurology 179: 223–44
19. Lorenz K 1935 Der Kumpan in der Umwelt des Vogels. Journal fur Ornithologie. 83: 137–213, 289–413
20. Marler P 1970 A comparative approach to vocal learning: Song development in white-crowned sparrows. Journal of Comparative and Physiological Psychology. 71(Suppl.): 1–25