Persistent neuronal density changes related to the establishment of a motor memory

Behavioural Brain Research 99 (1999) 115 – 121

Research report

Persistent neuronal density changes related to the establishment of a motor memory Paola Morales a,*, Teresa Pinto-Hamuy b, Vı´ctor Ferna´ndez b, Eugenia Dı´az c a

Morphology Program, Biomedical Sciences Institute, School of Medicine, Uni6ersity of Chile, P.O. Box 70005, Correo 7, Santiago, Chile b Physiology and Biophysics Program, Biomedical Sciences Institute, School of Medicine, Uni6ersity of Chile, Correo 7, Santiago, Chile c Department of Anatomy, Charing Cross and Westminster Medical School, Uni6ersity of London, London, UK Received 7 August 1997; received in revised form 23 January 1998; accepted 15 May 1998

Abstract Rats were trained in a lateralized reaching motor task during either an ‘early’ (22 – 31 days old) or a ‘late’ (62 – 71 days old) postnatal period. The ‘late’ group showed significant neuronal density reduction in cortical layers II – III of the contralateral motor forelimb representation. The ‘early’ group evidenced a similar localized contralateral effect that persisted after a subsequent period without training. Furthermore, in this group, a bilateral overall decrease in neuronal density was found throughout the motor cortex. This bilateral experience and age-dependent effect is conceivably related to a critical period of motor cortical development. The localized reduction of neuronal density strongly indicates a morphological expression of the motor engram. Our present study supports the concept that the acquisition and retention of motor learning involves the persistence of structural changes in the brain. © 1999 Elsevier Science B.V. All rights reserved. Keywords: Motor cortex; Training; Persistence; Plasticity; Neuronal density; Critical period; Rat

1. Introduction Although available data indicate the permanence of some cerebral and behavioral effects in rats reared in complex environments [6,10,22], to our knowledge, no reports have been issued on the persistence after training of a structural change in the motor cortex. In every day life, rats feeding, grooming, playing and performing other basic behaviors, display bilateral forelimb performance according to their species strategies. However, under special conditions, they may show manual preference. Using Peterson’s model of lateralized motor training [37], localized physiological

* Corresponding author. Tel.: +56 2 6786307, +56 2 6786289; Fax: +56 2 7776916; E-mail: [email protected]

[1,15,18,28] and morphological changes [16,44] in the motor cortex of the rat have been demonstrated. Further, previous results using this paradigm showed that rats trained immediately after weaning display a localized decrease in cell density and increase in cortical thickness in the forelimb motor representation [12]. In the present study, we investigated the persistence of an experimentally induced structural change related to a lateralized reaching task. In order to address this issue, two objectives were pursued: first, to examine interhemispheric neuronal density changes induced at two different stages of development; and, second, to determine whether early-induced structural modification persists after a relatively long period without training. Our findings suggest that learned motor behavior involves a permanent modification of brain structure.

0166-4328/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved. PII S0166-4328(98)00099-0

116

P. Morales et al. / Beha6ioural Brain Research 99 (1999) 115–121

2. Materials and methods

2.1. Subjects Sixteen male pigmented rats (AxC) from a local colony were divided into the following groups: (i) ‘early’ trained, 22–31 days old (n =6); (ii) ‘late’ trained, 62 – 71 days old (n =6); and (iii) control subjects matched by age to the first two groups (n = 4).

2.2. Training A brief and intensive period of training (four daily sessions of 25 trials each, during 10 days, making a total of 1000 trials) was imposed on the experimental rats as described in Dı´az et al. [12]. Experimental subjects were selected only if they demonstrated at least a 92% preferred paw reaching response (right or left). Two testing boxes were used: one for early post-weaned rats (17 ×20 ×20 cm3) and another for young-adult subjects (27× 27× 30 cm3). The front wall of the chamber consisted of an upper glass panel separated from a lower wooden part by a 1.3 or 1.7 cm wide slot for each box, respectively. The trial consisted of a number of attempts required by the subject to reach and grasp, with its preferred paw, a grain of cereal held at the tip of a hatpin. The grain was presented at the level of the slot far enough from the box in order for the animal to perform a full extension of its forelimb. Control rats were subjected to an identical testing box, number of exposures and amount of cereal, except that the food was available inside the box. At 72 days of age, before the sacrifice, controls and early experimental rats were placed in the training box and exposed to 25 trials (similar to the original training condition) in order to verify the manual preference or the persistence of behavioral response). All the rats were anesthetized with diethyl ether and perfused intracardially with a 0.9% saline solution followed by a 4% formaldehyde/saline solution. Brains were removed immediately from the cranium, post-fixed in the above fixative and cryoprotected for 48 h in a 30% sucrose solution. Frozen serial coronal sections (30 mm thick) were obtained using a sliding microtome. These sections were stained with cresyl violet. The region to be investigated was located between stereotaxic coordinates 2.4 – 0.1 mm rostral to Bregma. It was selected on the basis of previous, well known microstimulation maps [13,33,39]. Neuronal density was studied in layers II – III of the lateral agranular motor cortex (AGl) since previous findings indicate that neurons in these layers are involved in motor skill learning [20,36]. AGl is cytoarchitectonically characterized by the homogeneous appearance of its superficial layers and a broad layer V [13].

Neuronal density was quantified using the optical disector method [17,42,43]. We sampled one out of three sections, counting the neuronal nuclei with an ocular micrometer grid under 100× oil immersion objective. Criteria for recognition of neurons were based on Ling et al. [29] and Braendgaard et al. [9]. Optical disector of a known area (0.068× 0.068 mm2) was applied focusing down the microscope through 8 mm in depth, as read from the fine focusing knob. We discarded all the nuclei that intersected the left and the bottom sides of the frame, as well as the first nuclei that came into focus in the upper surface of the brain sections. For each section, six unbiased counting frames were sampled in a systematically random fashion inside of an area of 0.11 mm2. This area was situated 2.5 mm from the midline in order to avoid sampling the medial agranular cortex and lateral somatosensory neurons. The numerical density (Nv) of neurons was estimated by using the equation: Nv = Q − /h×a(fra), where Q − is the total average of neurons counted, h is the height of the optical disector (8 mm) and a(fra) corresponds to the disector area (4624 mm2).

2.3. Statistical analysis In a within-subject hemisphere comparison, the ‘trained hemisphere’ is defined as the one contralateral to the forelimb used, while the ‘nontrained hemisphere’ corresponds to the ipsilateral one. Statistical analysis was performed using the correlated one-tailed Student’s t-test. For intergroup comparisons, the one-tailed Student’s t-test for independent data was applied. Neuronal counting was performed independently by two persons, who were not aware of experimental designations of brains.

3. Results Interhemispheric comparisons of brain serial sections within AGl cortex (2.4–0.1 mm rostral to Bregma and 2.5 mm lateral to midline) were performed along the antero-posterior axis (A–P). This area of analysis showed throughout its extension a homogeneous appearance of pyramidal cells in layers II–III which are densely packed, and a layer V that is broader and contains larger and more densely staining cells. The results of the neuronal density counts evidenced for each subject a significant neuronal density reduction (PB0.005 and PB0.0005) in a middle sector of the trained hemisphere with respect to the corresponding region in the nontrained hemisphere. We define this area of change as sector B and the two adjacent sectors anterior (A) and posterior (C) without detectable modifications. Sector B varied in length and location from subject to subject (Table 1) and was present in the

P. Morales et al. / Beha6ioural Brain Research 99 (1999) 115–121

117

Table 1 Characterization of sector B for each subject found in the ‘early’ and ‘late’ trained groups Subjects

‘Early’ R1 R2 R3 R4 R5 R6 Means ‘Late’ R7 R8 R9 R10 R11 R12 Means

Length of sector B (mm)

1.17 1.10 1.20 1.23 0.96 1.11

Location of sector B (mm from Bregma)

Interhemispheric neuronal reduction (in %)

1.74–0.57 1.74–0.64 1.71–0.51 1.89–0.66 1.86–0.90 1.80–0.69

14.83 21.04 14.26 10.07 13.02 13.82

1.79–0.66*

14.51 91.47

1.29 1.26 1.41 1.26 1.32 1.29

1.49–0.20 1.56–0.30 1.74–0.33 1.68–0.42 1.56–0.24 1.65–0.36

12.63 16.37 15.19 9.52 9.03 12.96

1.309 0.02

1.61–0.31

12.62 91.20

1.13 90.04*

Column II: antero-posterior axis length ( 9 SEM) of sector B. Column III: location with respect to Bregma. Column IV: percentage of neuronal reduction ( 9 SEM) in an interhemispheric comparison for each subject. For statistical analysis between scores of these groups, an independent t-test was used: * PB0.05.

‘early trained’ and ‘late trained’ groups but not in control groups (Fig. 1). Note that the entire area of analysis is cytoarchitectonically homogeneous, thus sector B is defined solely on statistical analysis of neuronal density counts. The region cannot be distinguished based on morphological landmarks. In the ‘early’ group, the structural modification and behavioral response persisted even after 50 days without training. The performance was similar in speed and accuracy to the original response. In the control group, no significant differences in neuronal density were found along the A–P axis between the right and left hemispheres, regardless of handedness or age of manipulation. Quantitative morphological determinations of sector B for both groups at day 72 are presented in Table 1. Differences can be found between the ‘early’ and ‘late’ trained groups. In the ‘early’ group the zone of decrement in neuronal density was smaller and more rostrally located than in the ‘late’ group (Table 1, P B 0.05). The difference in the percentage of neuronal density reduction of sector B (trained versus nontrained hemispheres) did not reach significance when scores of ‘early’ and ‘late’ groups were compared (Table 1). A striking result was the difference in the overall neuronal density between the two experimental groups of rats (Fig. 1). In the ‘late’ group, the average scores of each subject were above the reference dashed line (22 × 104 neurons/mm3). In contrast, the ‘early’ group scores were either at reference line level or below (with the

exception of R3). For a more quantitative expression of this global effect, neuronal density ranges of ‘early’, ‘late’ and control groups were analyzed (Table 2). For this purpose, average scores of the most rostral and caudal end (210 mm each) of AGl were studied in order to avoid sampling in sector B where local asymmetries were detected. Significant differences were found only between ‘early’ scores (trained or nontrained hemispheres) (PB 0.0005) with respect to both ‘late’ (trained or nontrained hemispheres) and ‘early’ and ‘late’ control (right or left hemispheres) scores (Table 2). Scores of both control groups (‘early’ and ‘late’) were comparable to those of the ‘late’ experimental hemispheres.

4. Discussion The present study shows that the acquisition and retention of motor learning involves a persistent decrease in neuronal density within a focal area of the contralateral motor cortex (layers II–III). This locus of plasticity (Table 1) should be located within the forelimb representation according to Sanes’ electrophysiological motor cortical map [39]. It is important to point out that nontrained control groups did not show cortical asymmetries in spite of their handedness. In line with our findings, Karni et al. [21] found that in the human primary motor cortex, an enlarged local blood oxygenation area induced by motor learning persists after several months without training. Similarly, an increase in the

118

P. Morales et al. / Beha6ioural Brain Research 99 (1999) 115–121

Fig. 1. Average of neuronal density (neurons/mm3 9 SEM) for each subject trained during postnatal days 22 – 31 (‘early’) and 62 – 71 days (‘late’), respectively. “, scores for each trained hemisphere, plotted at anterior (A), intermediate (B) and posterior (C) sectors of the AGl motor cortex; , scores found in the homotopic sectors of nontrained hemispheres; - - -, arbitrary reference line at the level of 22 × 104 neurons/mm3. A correlated Student t-test was used; a total of 186 individual scores per hemisphere for each subject were analyzed; In sector B, significant interhemispheric differences were found: * P B0.005; ** P B 0.0005 (standard error bars are too short to be visible).

number of synapses per Purkinje cell in rats trained in a complex motor task persisted after 28 days when compared with active and inactive controls [26]. These results imply that motor skill learning rather than mere

motor activity is required to induce permanent structural changes. A reduced length of the focus (Table 1, sector B) in the ‘early’ group as compared to the ‘late’ group was

P. Morales et al. / Beha6ioural Brain Research 99 (1999) 115–121

detected. This could be a consequence of a progressive selectivity of connections during the time elapsed between training and sacrifice. This selectivity process may be involved in fine motor skills acquisition [14]. The more frontal location of sector B in the ‘early’ versus the ‘late’ group, is probably due to a rostral displacement and/or enlargement of forelimb representation. In fact, learning of motor skills produces changes in the movement representation maps in rats [24], monkeys [32] and humans [34,35]. In addition to the focal change in the contralateral forelimb representation, early training induced a global, age-dependent modification in the motor cortex of both hemispheres (see Table 2 and Fig. 1; compare neuronal density ranges between ‘early’ and ‘late’). The experience-dependent global effect demonstrated here occurs in parallel with the time-course of the postnatal differentiation process. Black and Greenough [7] refer to this period as one of experience expectation. In fact, as the animal changes from a liquid to a solid diet, organism– environmental relationships are drastically modified. For the subject, it implies awareness of external resources and significant postural adjustments which allow reaching and grasping for food. In this context, the new feeding habit is imposed during a time of exceptional cortical susceptibility [2]. The results of Dı´az et al. [12] demonstrate the relevance of the critical period since in early trained subjects, they obtained a reversal of the rostro-caudal gradient of cell differentiation; the face– forelimb sequence switched to forelimb – face. Further, the global effect described may have an explanation at the cellular level. Recent studies indicate that a certain type of LTP recorded in rat slices is induced only during Table 2 Average neuronal density scores (×104 neurons/mm3 9 SEM) found in the rostral and caudal 210 mm of the motor cortex analyzed (2.4–0.1 mm from Bregma) Hemispheres

Rostral (×104 neurons/ mm3 9 SEM)

Caudal (×104 neurons/ mm3 9 SEM)

Early trained Early nontrained Late trained Late nontrained Early control right Early control left Late control right Late control left

20.3 9 0.46** 19.8 9 0.45** 25.9 9 0.56 24.8 9 0.54 26.8 9 0.80 25.5 90.83 26.1 9 0.86 27.0 9 0.78

23.7 9 0.45** 24.29 0.43** 27.29 0.63 28.99 0.71 27.8 9 0.86 31.1 90.79 29.3 90.87 29.4 9 0.97

The scores correspond to the trained hemispheres and nontrained hemispheres of experimental groups and right or left (naive) hemispheres of ‘early’ and ‘late’ controls. An independent Student t test was used. Significant differences were detected only between ‘early’ scores (trained or nontrained hemispheres) with respect to both ‘late’ (trained or nontrained hemispheres) and ‘early’ and ‘late’ controls (right or left hemispheres) scores: ** PB0.0005.

119

the critical period [11,25]. This LTP, induced from infragranular layers, is age dependent and modulated by afferent stimulation. The age dependence may be due to inhibitory circuitry that develops after the critical period in the cerebral cortex [3,30]. In fact, our ‘early’ group, trained at the end of the critical period, presents a bilateral global effect that can be induced only in this period. It should be pointed out that a second form of LTP, experience dependent but not age related, can be recorded in layers II–III [25]. This kind of LTP could underlie the focal effect found in the ‘early’ and ‘late’ experimental groups. It is reasonable to assume that the reduction in neuronal density in layers II–III, reported here, may reflect neuropile growth (increase of dendritic branching and synaptogenesis) which provides more connectivity. This interpretation is supported by Golgi studies that demonstrated an enlargement of apical dendritic branches in layers II–III [44] and in layer V [16] of the somatosensory-motor cortex after a practically identical motor task. Furthermore, synaptogenesis in layers II– III of motor cortex has been found after training on a complex motor learning task [27] and after repetitive stimulation of its afferents [23]. It is important to point out that an inverse relation between cortical thickness and cellular density after environmental conditions has been reported [5]. Recent results of our group show that training induce a significant increase in cortical thickness (PB 0.05 and PB 0.005, unpublished data) which is restricted to the specific area of neuronal density changes. This result is in agreement with Dı´az et al. [12], in rats trained in an identical motor task during the early postweaned period. Taking these findings together, it would be reasonable to expect that the reduction in neuronal density reflects an increment in dendritic arborization associated with changes in cortical volume. There is no reason to assume that the lowered neuronal density is due to loss of neurons in the focus. To our knowledge, no evidence of apoptotic mechanisms induced by training has been reported so far. However, on the basis of our findings, we cannot rule out other factors such as angiogenesis [8] and/or gliosis and glial surface area increase [40,41], which might contribute to reduction in neuronal density in layers II–III of the motor cortex as observed in the present study. Pertinent findings of Asanuma and Pavlides [4] suggest that during repetitive motor practice, sensory inputs associated with movements could produce LTP in layers II–III of motor cortex [38]. Motor learning may be related to the formation of loop circuits between motor cortex, somatosensory cortex and thalamic nuclei (ventrolateral (VL) [19] and ventroposterolateral nucleus [23]). At the initial stage of learning, the VL projections to motor cortex are diffuse and become specific through practice [4]. These circuits could be at the bases of the focal, persistent and age-independent morphological

120

P. Morales et al. / Beha6ioural Brain Research 99 (1999) 115–121

modification found in our study. On the other hand, we also demonstrate that there is a global age-dependent effect, found across the cerebral hemispheres after early training, that may be attributed to the stage of development at which the motor task was imposed. This may be due to an increased excitability level and/or reduced inhibitory influences at this period of development. In summary, this work demonstrates the persistence of structural changes after motor learning. Recently, Morales and Pinto-Hamuy [31] reported that the motor engram can persist for at least a year. These results strongly support the idea of long-lasting morphological cortical plasticity underlying procedural memory.

Acknowledgements We thank Dr Joaquı´n Fuster, Dr Francisco Aboitiz, Dr Vicente Montero and Dr Marc Zeise for their comments on the manuscript, and M. Canitrot and A. Martinez for their technical assistance. Supported by Grants FONDECYT (1930-962) and DID-S/9608 to TPH and FONDECYT (1950649) to VFH.

References [1] Alca´ntara AA, Saks ND, Greenough WT. Fos is expressed in the rat during a forelimb reaching task. Soc Neurosci Abstr 1991;17:141. [2] Altman J. Postnatal growth and differentiation of the mammalian brain, with implications for a morphological theory of memory. In: Quarton GC, Melnechuk T, Schmitt FO, editors. The neuroscience, a study program. New York: The Rockefeller University Press, 1967:723–743. [3] Aroniadou VA, Keller A. The patterns and synaptic properties of horizontal intracortical connections in the rat motor cortex. J Neurophysiol 1993;70:1553–69. [4] Asanuma H, Pavlides C. Neurobiological basis of motor learning in mammals. Neuroreport 1997;8:1–6. [5] Beaulieu C, Colonnier M. Effects of the richness of the environment on six different cortical areas of the cat cerebral cortex. Brain Res 1989;495:382–6. [6] Bennett EL, Rosenzweig MR, Diamond MC, Morimoto H, Herbert M. Effects of successive environments on brain measures. Physiol Behav 1974;12:621–31. [7] Black JE, Greenough WT. Developmental approaches to the memory process. In: Martinez JL, Kesner RP, editors. Learning and memory, a biological view. San Diego, CA: Academic Press, 1991:61 – 91. [8] Black JE, Isaacs KR, Anderson BJ, Alca´ntara AA, Greenough WT. Learning causes synaptogenesis, whereas motor activity causes angiogenesis, in cerebellar cortex of adult rats. Proc Natl Acad Sci USA 1990;87:5568–72. [9] Braendgaard H, Evans SM, Howard CV, Gundersen HJG. The total number of neurons in the human neocortex unbiasedly estimated using optical disectors. J Microsc 1990;157:285 – 304. [10] Camel JE, Withers GS, Greenough WT. Persistence of visual cortex dendritic alterations induced by postweaning exposure to a ‘superenriched’ environment in rats. Behav Neurosci 1986;100:810 – 3.

[11] Crair MC, Malenka RC. A critical period for long-term potentiation at thalamocortical synapses. Nature 1995;375:325–8. [12] Dı´az E, Pinto-Hamuy T, Ferna´ndez V. Interhemispheric structural asymmetry induced by a lateralized reaching task in the rat motor cortex. Eur J Neurosci 1994;6:1235 – 8. [13] Donoghue JP, Wise SP. The motor cortex of the rat: cytoarchitecture and microstimulation mapping. J Comp Neurol 1982;212:76 – 88. [14] Ebbesson SOE. Evolution and ontogeny of neural circuits. Behav Brain Sci 1984;7:321 – 32. [15] Fuchs JL, Bajjalieh SM, Hoffman CA, Greenough WT. Regional brain 2-deoxyglucose uptake during performance of a learned reaching task. Soc Neurosci Abstr 1983;9:54. [16] Greenough WT, Larson JR, Whithers GS. Effects of unilateral and bilateral training in a reaching task on dendritic branching of neurons in the rat motor-sensory forelimb cortex. Behav Neural Biol 1985;44:301 – 14. [17] Gundersen HJG. Stereology of arbitrary particles. A review of unbiased number and size estimators and the presentation of some new ones, in memory of William R Thompson. J Microsc 1986;143:3 – 45. [18] Hyde´n H, Egyha´zi E. Changes in RNA content and base composition in cortical neurons of rats in a learning experiment involving transfer of handedness. Proc Natl Acad Sci USA 1964;52:1030 – 5. [19] Iriki A, Pavlides C, Keller A, Asanuma H. Long-term potentiation of thalamic input to the motor cortex induced by coactivation of thalamocortical and corticocortical afferents. J Neurophysiol 1991;65:1435 – 41. [20] Kaneko T, Caria MA, Asanuma H. Information processing within the motor cortex. II. Intracortical connections between neurons receiving somatosensory cortical input and motor output neurons of the cortex. J Comp Neurol 1994;345:172–84. [21] Karni A, Meyer G, Jezzard P, Adams MM, Turner R, Ungerleider LG. Functional MRI evidence for adult cortex plasticity during motor skill learning. Nature 1995;377:155 – 8. [22] Katz HB, Davies CA. Effects of differential environments on the cerebral anatomy of rats as a function of previous and subsequent housing conditions. Exp Neurol 1984;83:274 – 87. [23] Keller A, Arissian K, Asanuma H. Synaptic proliferation in the motor cortex of adult cats after long-term thalamic stimulation. J Neurophysiol 1992;68:295 – 308. [24] Keller A, Weintraub ND, Miyashita E. Tactile experience determines the organization of movement representations in rat motor cortex. NeuroReport 1996;7:2373 – 8. [25] Kirkwood A, Hey-Kyoung L, Bear MF. Co-regulation of longterm potentiation and experience-dependent synaptic plasticity in visual cortex by age and experience. Nature 1995;375:328–30. [26] Kleim JA, Ballard D, Vij K, Greenough WT. Learning-dependent synaptic modification in the cerebellar cortex of adult rat persist for at least four weeks. J Neurosci 1997;17:717 –21. [27] Kleim JA, Lussnig E, Schwarz ER, Comery TA, Greenough WT. Synaptogenesis and FOS expression in the motor cortex of the adult rat after motor skill learning. J Neurosci 1996;16:4529– 35. [28] Kriva´nek J, Buresova´ O. Cortical acetylcholinesterase and ‘handedness’ in rats. Experientia 1972;28:291 – 2. [29] Ling EA, Paterson JA, Privat A, Mori S, Leblond CP. Investigation of glial cells in semithin sections I. Identification of glial cells in the brain of young rats. J Comp Neurol 1973;149:43–72. [30] Luhmann HJ, Prince DA. Postnatal maturation of the GABAergic system in rat neocortex. J Neurophysiol 1991;65:247–63. [31] Morales P, Pinto-Hamuy T. Persistence after a year of a cortical structural modification early induced by training: structural correlate of a procedural memory? Not Biol 1996;4:213.

P. Morales et al. / Beha6ioural Brain Research 99 (1999) 115–121 [32] Nudo RJ, Milliken GW, Jenkins WM, Merzenich MM. Usedependent alterations of movement representation in primary motor cortex of adult squirrel monkeys. J Neurosci 1996;15:785 – 807. [33] Neafsey EJ, Bold EL, Haas G, Hurley-Gius KM, Quirk G, Sievert CF, Terreberry RR. The organization of the rat motor cortex: A microstimulation mapping study. Brain Res Rev 1986;11:77 – 96. [34] Pascual-Leone A, Dang N, Cohen LG, Brasil-Neto JP, Cammarota A, Hallett M. Modulation of muscle responses evoked by transcranial magnetic stimulation during the acquisition of new fine motor skills. J Neurophysiol 1995;74:1037–45. [35] Pascual-Leone A, Grafman J, Hallett M. Modulation of cortical motor output maps during development of implicit and explicit knowledge. Science 1994;263:1287–9. [36] Pavlides C, Miyashita E, Asanuma H. Projection from the sensory to the motor cortex is important in learning motor skills in the monkey. J Neurophysiol 1993;70:733–41. [37] Peterson GM. Mechanisms of handedness in the rat. Comp Psychol Monogr 1934;9:1–67. [38] Sakamoto T, Porter LL, Asanuma H. Long-lasting potentiation of synaptic potentials in the motor cortex produced by stimula-

.

[39]

[40]

[41]

[42]

[43]

[44]

121

tion of the sensory cortex in the cat: a basis of motor learning. Brain Res 1987;413:360 – 4. Sanes JN, Suner S, Donoghue JP. Dynamic organization of primary motor cortex output to target muscles in adult rats. I. Long-term patterns of reorganization following motor or mixed peripheral nerve lesions. Exp Brain Res 1990;79:479 – 91. Sirevaag AM, Greenough WT. Differential rearing effects on rat visual cortex synapses. III. Neuronal and glial nuclei, boutons, dendrites and capillaries. Brain Res 1987;424:320 – 32. Sirevaag AM, Greenough WT. Plasticity of GFAP-immunoreactive astrocyte size and number in visual cortex of rats reared in complex environments. Brain Res 1991;540:273 – 8. Sterio DC. The unbiased estimation of number and sizes of arbitrary particles using the disector. J Microsc 1984;134:127– 36. West MJ, Gundersen HJG. Unbiased stereological estimation of the number of neurons in the human hippocampus. J Comp Neurol 1990;296:1 – 22. Withers GS, Greenough WT. Reach training selectively alters dendritic branching in subpopulations of layers II – III pyramids in rat motor-somatosensory forelimb cortex. Neuropsychology 1989;27:61 – 9.