Книги по МРТ КТ на английском языке / Functional Neuroimaging in Child Psychiatry Ernst 1 ed 2000

104L. E. Arnold, A. J. Zametkin, L. Caravella and N. Korbly

Particularly for some developmentally disabled children or individuals with thought disorders, visits to the scanner ahead of time may only serve to increase anxiety or even allow time for paranoid delusions to develop. Each individual situation requires careful assessment informed by advice from parents, who may have previous experience with the child in similar situations.

Assessment of beneWt±risk ratio

Assessment of the beneWt±risk ratio should consider the entire study. It should evaluate risks comprehensively and consider the known in vitro eVects (for example, the eVects of a drug on enzymatic processes identiWed in cell preparation) and in vivo eVects (in animals, adults, and children), as well as potential eVects of the technician's or clinician's skills on the risks. The database utilized for judgment about risk should also include noncontrolled clinical Wndings and other observations. It should also evaluate beneWts comprehensively and consider the health and welfare of the subject in general, as well as the advancement of medical knowledge that would beneWt the subject in the long run. Finally an assessment should consider possibilities for improving the beneWt±risk ratio by increasing beneWts, decreasing risk, or both (see above).

A problem in extrapolating reports of adverse eVects from clinical samples is that the clinical experience is often confounded by complicating factors, including the illness for which the clinical procedure was carried out. Since a research procedure is undertaken on a voluntary basis rather than under clinical necessity, it is less likely to involve those individuals who anticipate the most distress from it. This probably reduces the psychologic risk for the sample as a whole, although there still may be individuals who assent/ consent despite negative anticipation. An interesting related question concerns the subsequent eVect on the child's later psychologic reaction to the same test if it is clinically indicated. Will the child weather it better because of success in the research administration or experience greater stress because of being sensitized to the risks? Systematic follow-up of child and adolescent research subjects regarding both physical and emotional consequences would be of great value for future formal risk estimation. For example, Kruesi et al. (1988) reported that lumbar punctures for research purposes were rated by child subjects as equivalent in both acceptability and side eVects/discomfort to procedures such as blood sampling, electroencephalography, and attending school in the research setting, thus providing a comparison with which to gauge psychologic risk.

The beneWt±risk ratio usually diVers for patients who can directly beneWt from the research through an increase

in knowledge about their disorder and for healthy controls, who do not directly beneWt from the study of a disorder that they do not have. In this case, the US Code of Federal Regulation (CFR), Subtitle A, Part 46, clearly indicates that studies carrying higher than minimal risk with no direct beneWt for the subjects are generally disallowed for children. However, when the risk is ªa minor increase above minimal riskº, this regulation allows children to participate if the beneWts to society are sizable. Just as ªminimal riskº is subject to the interpretations of an IRB, so is ªminor increaseº. Therefore, healthy children can participate in brain imaging studies if the risk can be judged to be ªminimal' or ªjust above minimalº. Whether this describes the risk from radiation exposure from PET/SPECT is diYcult to determine. Although excess health hazards from low-level radiation exposure have not been detected in any large studies so far (Mossman et al., 1996; Ernst et al., 1998), the concept of radiation exposure carries overriding emotional freight that can distort the objective assessment of radiation risks.

A beneWt usually not addressed and highly dependent on the way research is conducted by the investigators is the potentially positive experience a child may experience by participating as an autonomous individual in an adult enterprise with the purpose of helping society at large. Examination of this aspect of research with children might identify strategies that could promote the development of altruism.

To insure that the risk experienced by healthy normal controls is not ªwastedº, it is important to insure they are indeed normal and, therefore, able to yield the needed normative data. This can be done through a multistage screening analogous to that used to insure that patients who are subjects have the relevant disorder. Parents return standard behavior ratings and questionnaires. If no pathology is noted on the paper-and-pencil screen, the parent is interviewed by phone. If no pathology is evident from the phone interview, the parent and child/adolescent are invited to the clinic and observed and interviewed with structured instruments. If no relevant pathology is detected here, the child/adolescent is declared ªnormalº and oVered an opportunity to participate in the study as a healthy control.

Informed assent and consent principle of respect for persons

Although a favorable beneWt±risk ratio is likely in the study of severe mental disorders where potential beneWts usually outweigh the experimental risks, an issue remains, that of

informed consent/assent. In fact, some of the concerns about validity of assent in a mentally impaired population similarly apply to very young children.

By deWnition and US federal statute, minors cannot actually consent but are permittedªand, above certain ages requiredº, to assent to participate in research, whereas their parents are required to give permission in order for the child to participate. As a reminder, valid informed consent (assent) requires (i) disclosure of relevant information, (ii) comprehension of the information, and (iii) voluntary agreement to participation, free of coercion and undue inXuences.

Developmentally sensitive consent

As children mature, they concurrently develop a better capability of understanding the risks of research and acquire a better ability to consent (Laor, 1987; Leikin, 1993), that is, as cognitive ability increases gradually, so may consent ability. A considerable amount is now known about minors' understanding of the consent process. The basic work of Piaget and Inhelder (1958) suggests increments of cognitive ability around 7 years of age and when adult-like formal operations are attained during puberty. Lewis et al. (1978) found signiWcant increases by grade level (elementary school) in awareness of risk, understanding of future consequences, wariness of individuals with vested interests, and need for independent professional advice. KeithSpiegel and Maas (1981) (quoted by Leikin, 1993) found that children, particularly those under age 9 years, had diYculty understanding the scientiWc purposes of research, failed to recall their role in research, and were unaware of their freedom to withdraw. Those 9 years and under tended to focus on only one or two salient pieces of information, while those above age 9 were more similar to adults in their reasoning about research. Weithorn and Campbell (1982) found that 14-year-olds show the same beneWt±risk reasoning as 21-year-olds, and that 9-year-olds reach the same conclusions albeit through a diVerent reasoning strategy. Lewis et al. (1978) found that children aged 7±9 years (but not those aged 6 years) asked all the relevant questions needed for an informed consent. Leikin (1993), after reviewing the literature, concluded that by age 9 children have enough cognitive capacity to participate in the decision. Weithorn and Scherer (1994) concluded that schoolage children have the capacity to assent and that children aged 14 years, on average, have the capacity to consent in the same manner as adults. Therefore, enough seems to be known about children's understanding and judgment to justify more empowerment in the consent process, including the right to make some altruistic decisions.

Laor (1987), in fact, oVers this resolution: ª. . . every individual can achieve, and exercise, diVerent levels of autonomy at diVerent times. . . . The question as to who is or is not autonomous should be replaced by the questions of how much autonomy can/should be ascribed to/required of the individual in diVerent circumstances, be the individual a child or an adult. . . .º He proposes that children age 12 years and above be able to consent to research on their own (like adults); that between ages 7 and 12 years they can consent but parents' assent is also needed (the reverse of the current rules); and below age 7 parental consent is needed, whenever possible coupled with the child's assent. The participation of children under 7 years of age without their assent is a thorny issue, but, given the severity of certain early childhood disease (leukemia, autism), parents could be empowered to make decisions for young children when the research is critical to the understanding of severe illness. Laor's recommendation was not adopted in 1977 by the American National Committee for the Protection of Human Subjects of Biomedical and Behavioral Research because of a lack of public consensus at the time (Grisso and Vierling, 1978; Melton, 1980; Laor, 1987). Is it time to reconsider the recommendation, perhaps with some modiWcation (for example, age 14 rather than 12 years for adult-like consent)?

Motivation

Motivation varies widely among children as a function of their cognitive maturity, individual circumstances, and personality. Motivation to participate in research may be altruistic or may be driven by the compensation oVered for participation (see below). A child may Wnd that missing a day of school is rewarding enough to induce participation in a research study. The opportunity to impress peers or to feel special may inXuence another child. Other motivations may include unconscious seeking of punishment to assuage guilt feelings. For example, healthy siblings of a patient might feel guilty about not sharing the patient's problems. Therefore, the investigator needs to be alert to the motivation of children during the process of informed assent. Particularly, the investigators have the responsibility to insure the absence of coercion or undue inXuences. Therefore, the assent with children should be conducted separate from the parents, though the child could be present at the parental consent.

Inducements (incentives, coercion, undue inXuences)

The possibility of coercion (i.e., real or perceived negative consequences of refusing to participate in the research) or

106L. E. Arnold, A. J. Zametkin, L. Caravella and N. Korbly

coercive inducement (e.g., excessive reward to obtain compliance) needs to be carefully monitored. A generally acceptable incentive may become an undue inXuence if the individual is particularly vulnerable. Deciding how much is too large a compensation for the risk, discomfort, inconvenience, and lost time of subjects and their parents requires some thought. Paradoxically, the greater the risk, the harder it may be to justify a large compensation. The inducement of compensation should not be so large as to override judgment about the potential risks.

A complication in determining appropriate compensation is the varying economic status of the families. The same payment may be much more of an inducement for a poor subject than a wealthy one. However, it is unfeasible (and unfair) to equalize inducement by varying the compensation with economic status. The child's view of dollar amounts also needs to be considered. For example, KeithSpiegel and Maas (1981) found that minors ages 9 to 15 years were interested in compensation of any amount, whereas adults were mainly interested in higher amounts.

Parents need to be compensated for lost time from work and transportation costs. How generously parents are compensated may be an issue. Would a poor parent be tempted to coerce the child into research for the compensation? Such exploitation would run counter to the usual expectation that the parent's judgment about the best interests of the child would protect the child from undue inducement. Working middle-class parents should be reimbursed fairly for lost time from work, but this rate may verge on coerciveness for some poor families and yet it is inconceivable to pay poor or unemployed parents less.

Some IRBs and even some ethics experts are tempted to avoid such thorny decisions by banning compensation in child research, at least for the children themselves. However, it is diYcult to justify this to the subjects and their parents. It is the practice in some places to reimburse only transportation for the child and one parent, particularly where treatment is part of the study. However, a reasonable approach could be to oVer fair compensation at middle-class levels when the study budget can aVord it; in high-risk studies where there is a concern about the compensation becoming an undue inXuence for the poor subjects, some provision could be made for a neutral clinician to monitor the consent/assent process.

Other suggested solutions have not elicited enthusiasm or general acceptance. For example, some ethicists suggest that there be no discussion about compensation until after consent in order to preclude any inducement. However, others feel that this might unfairly exclude some families who could not aVord to take time oV from work and pay transportation without some reimbursement. In practice,

most IRBs take the position that knowledge about compensation is part of a fully informed consent and many even require it to be in the consent form, at least for parents. Some IRBs prohibit the information about parents' compensation to be given to the child. Another controversial recommendation not commonly followed is that none of the funding comes from pharmaceutical companies or other interested parties (Small et al., 1994).

When the IRB's policy is to reimburse the child, then arises the question of who receives the compensation and how. Is the cash given to the parents or the child? Is the check made out to the adolescent's name or parent's name? Is compensation given as a gift certiWcate for children's stores? The alternatives are many and should be carefully chosen and, unless against the IRB's policy, clearly outlined in protocol and the consent and assent forms.

Paradoxically, insurance for injury during research, to which no one would object ethically, is an inducement not universally used despite general agreement about its desirability. This particular inducement, in fact, would make any research more ethical by reducing economic risk to the subject.

A nonmonetary inducement may occur when the sibling of a patient is used as normal control. Is the sibling psychologically coerced by guilt and by the expectation of taking responsibility to help the ill family member? A partial answer can come from observations such as those reported by Amiel (1985), who found that siblings of patients were enthusiastic about volunteering for a procedure involving two indwelling venous lines for a day because it allowed them to share the attention the sick sibling had been getting and to satisfy their curiosity about what the patient had been going through.

Another nonmonetary inducement is the wish to please the investigator, especially if the investigator is also the child's therapist. Where is the line between cultivating an alliance and seductive trading on friendship? In a longitudinal study, what is too much pressure for retention?Where is the line between healthy, respectful expectation of fulWlling the agreement and manipulating guilt? As already mentioned above, a therapist, case manager, or neutral clinician not involved in the research can be one safeguard against undue inducement of this type.

Neutral clinician to insure understanding

Within the consent/assent empowerment remains an obligation to make sure the child or adolescent understands the information provided. Because of wide individual variation, one cannot depend on the subject`s age to insure adequate understanding; even some adults have diYculty

understanding consent information. When the investigator is not the child's regular clinician, it would be desirable for the investigator and the child's clinician to join in making sure the child understands the information provided.When the investigator is the child's regular clinician, Weithorn and Scherer (1994) suggest that consent be obtained by another clinician. The participation of a neutral clinician whose only interest and responsibility are the child's welfare could help to justify consent (assent) empowerment for children younger than those currently allowed to consent. The neutral clinician would act as a counterbalance to the investigator's bias, as well as to the potential, perceived or real, parental coercion or undue inXuence.

It is tempting for researchers to believe that children should be free to participate in research to a greater degree. The possibility of self-serving blind spots within the investigator's assessment makes a dialogue with noninvestigator clinicians and other neutral parties essential, at least for policy-making. These neutral clinicians may need to be recruited from outside the research institution to insure neutrality. They should be paid for their time, perhaps through the IRB from the study budget. This expense would be justiWed if it makes a valuable but more than minimal risk study ethically possible. Of course, cost and eVort considerations suggest that involvement of an additional clinician in day-to-day consent with individual child subjects is necessary and advisable only for research deemed more than minimal risk by the IRB. More-than- minimal risk research is permissible under certain carefully speciWed situations (see Code of Federal Regulations as cited above).

An example of using an independent clinician to insure that children were not being coerced can be found in a recent National Institute of Mental Health PET study that recruited normal siblings of patients with autism, Tourette's disorder, and Lesch±Nyhan disease. A member of the Bioethics Program of the NIH Clinical Center interviewed each sibling control teenager (minimum age 12 years), alone without investigators or parents present, to screen for understanding of the study, coercion, and undue inXuence.

It is also critical to make sure that children understand that they have a choice. Weithorn and Scherer (1994) reviewed the development of locus of control (internal in 40% of prepubertal children, $80% of late adolescents) and other relevant psychologic dimensions. They list several ways to help the child to feel more autonomous (e.g., ask them to choose which chair to sit in; ask assent neutrally). They note that this not only promotes valid consent/assent but also improves compliance with the research procedures.

Selection of subjects (principle of justice)

Patients versus healthy controls

A highly controversial question remains as to whether the principle of justice is served or violated by excluding healthy children from research because they may not beneWt directly from the research that studies a disorder they do not have and because the research procedure is estimated to be more than minimal risk, because it is one they are not commonly exposed to (for example, an intravenous line). Should the individuals already suVering from a disorder also be the ones to bear the burden of research? True, they are the ones who can mostly beneWt from the research directly. However, society at large may also beneWt from advanced knowledge and potential prevention of costly diseases. Each research protocol needs to be assessed individually in an eVort to respect all three principles of ethics, including justice.

Socioeconomic status and ethnic issues

An important goal in recruiting subjects is to achieve a sample that represents the general population. Control subjects should be representative of the general population, and a clinical sample should reXect the clinical population. This condition requires some attention to the socioeconomic status and ethnic composition of the sample. Cultural and socioeconomic preferences, biases, opportunities, and confounds with exclusion criteria may skew a sample. If major ethnic groups are not represented in the sample, important Wndings may be missed that might be useful to the absent or under-represented ethnic groups and to the advancement of clinical science. There has been oYcial recognition of the need to insure equal access to research and its beneWts (Ellis, 1994; Varmus, 1994) for both minorities and minors (deWned in the USA as below 18 years of age). Recent initiatives mandate the participation of minorities, women, and minors in all US NIH-funded research unless their exclusion is explicitly and well justiWed.

Underrepresentation of ethnic or socioeconomic groups poses a safety and statistical dilemma. Stratifying a sample by ethnicity or socioeconomic group can reduce statistical power and require a larger sample, thus exposing more children to research-related risks. The investigator must make an informed decision as to whether ethnicity is likely to aVect the question under study. If not, then all subjects can be pooled with no special recruitment or exclusion criteria other than equal access. If ethnicity is likely to make a diVerence, then a decision has to be made whether (i) to

initially concentrate eVorts on one ethnic group to minimize the number who need to be exposed to the risk of the procedure, and then study other groups later if the results warrant it; or (ii) to increase the sample size suYciently to provide the statistical power for ethnic stratiWcation.

The opinions expressed herein are the views of the authors and do not necessarily reXect the oYcial position of the National Institute of Mental Health or the US Department of Health and Human Services.

Conclusion

Literature review and recent experience in neuroimaging research suggest that the degree of risk (including aversiveness of neuroimaging procedures to children) is often assumed to be higher than indicated by the data. The available data suggest that, in normal circumstances, the degree of discomfort associated with most neuroimaging studies is comparable to that associated with children's everyday experiences. Likewise, the degree of physical risk appears to be less than commonly assumed (e.g., Huda and Scrimger, 1989). Furthermore, many of these risks can be minimized by appropriate sensitivity and planning.

Despite popular notions to the contrary, available evidence suggests that adolescents can meaningfully evaluate risks and beneWts in a fashion similar to adults. Younger children, while cognitively less able, nonetheless reach sensible conclusions concerning risks and beneWts, albeit through rationales somewhat diVerent from those used by adolescents and adults. Current regulations may excessively limit children's opportunities both to contribute altruistic service and to beneWt as a group from research. Developmentally enlightened revisions of current regulations and research practices could ensure that children and adolescents reap the full beneWts of neuroimaging research on their neuropsychiatric disorders.When a study is considered to pose more than minimal risk and does not oVer equivalent beneWt for the subject, a neutral clinician could insure that the child is not psychologically coerced or unduly inXuenced by parent or investigator. The risks of neuroimaging research with children and adolescents should be carefully balanced against the risks of depriving them of the beneWts of state-of-the-art knowledge concerning the severe neuropsychiatric illnesses that aZict many of them. This delicate balance requires the collaboration of investigators with physicists, ethicists, and patients and their parents.

Acknowledgement

This chapter is adapted and expanded from an article in the Journal of the American Academy of Child and Adolescent Psychiatry, (1995) 34, 929±39.

iReferencesi

Amiel, S. A. (1985). Pediatric research on diabetes: the problem of hospitalizing youthful subjects. IRB Rev. Hum. Subjects Res., 7, 4±5.

Arnold, L. E., StoV, D., Cooke, E. et al. (1995). Ethics of biological psychiatric research with children and adolescents. J. Am. Acad. Child Adolesc. Psychiatry, 347, 929±39.

Arnold, L. E., StoV, D., Cooke, E. et al. (1996). Ethical issues in biological mental health research with children and adolescents. In

Ethical Issues in Mental Health Research with Children and Adolescents, eds. K. Hoagwood and P. S. Jensen, pp. 89±111. Mahwah, NJ: Lawrence Erlbaum.

Belmont Report (1979). Ethical Guidelines for the Protection of Human Subjects of Research. (OPRR Reports). Washington, DC: NIH OYce for Protection from Research Risks and The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research.

Billen, D. (1990). Spontaneous DNA damage and its signiWcance for the ªnegligible doseº controversy in radiation protection.

Radiat. Res., 124, 242±5.

Castellanos, F. X., Elia, J., Kruesi, M. J. P. et al. (1994). Cerebrospinal

Xuid monoamine metabolites in ADHD boys. Psychiatr. Res., 52, 305±16.

Douglas, J. W. B. (1975). Early hospital admissions and later disturbances of behavior and learning. Dev. Med. Child Neurol., 17, 456±80.

Ellis, G. B. (1994). Inclusion of Women and Minorities in Research

(OPRR Reports 94±01). Bethseda, MD: NIH OYce of Protection from Research Risks.

Ernst, M., Zametkin, A. J., Matochik, J. A. et al. (1996). Presynaptic dopaminergic deWcits in Lesch±Nyhan Disease. N. Engl. J. Med.,

334, 1568±1604.

Ernst, M., Zametkin, A. J., Matochik, J. A., Pascualvaca, D. and Cohen, R. M. (1997a). Low medial prefrontal dopaminergic activity in autistic children. Lancet, 350, 638.

Ernst, M., Cohen, R. M. Liebenauer, L. L., Jons, P. H. and Zametkin, A. J. (1997b). Cerebral glucose metabolism in adolescent girls with attention-deWcit/hyperactivity disorder. J. Am. Acad. Child Adolesc. Psychiatry, 36, 1399±1406.

Ernst, M., Freed, M. E. and Zametkin, A. J. (1998). Health hazards of radiation exposure in the context of brain imaging research: special consideration for children. J. Nucl. Med., 39, 689±98.

Grisso, T. and Vierling, L. (1978). Minors' consent to treatment: a developmental perspective. Professional Psychol., 9, 412±27.

Huda, W. and Scrimger, J. W. (1989). Irradiation of normal volunteers in nuclear medicine. J. Nucl. Med., 30, 260±4.

Jay, S. M., Elliot, C. H., Ozolins, M., Olsen, R. A. and Pruitt, S. D.

(1982). Behavioral management of children's distress during painful medical procedures. Behav. Res. Ther., 23, 513±20.

Jons, P. H., Ernst, M., Hankerson, J., Hardy, K. and Zametkin, A. J. (1997). Follow-up of radial catheterization for positron emission tomography scans. Hum. Brain Map., 5, 119±23.

Keith-Spiegel, P. and Maas, T. (1981). Consent to research: are there developmental diVerences? In Proceedings of American Psychological Association, 1981. Washington, DC: American Psychological Association.

Kopelman, L. M. (1989). When is the risk minimal enough for children to be research subjects? In Children and Health Care: Moral and Social Issues, eds. L. M. Kopelman and J. C. Moskop, pp. 89±99. Boston, MA: Kluwer.

Kruesi, M. J. P., Swedo, S. E., CoVey, M. L., Hamburger, S. D., Leonard, H. and Rapoport, J. L. (1988). Objective and subjective side eVects of research lumbar punctures in children and adolescents. Psychiatr. Res., 25, 59±63.

Laor, N. (1987). Toward liberal guidelines for clinical research with children. Med. Law, 6, 127±37.

Leikin, S. (1993). Minors' assent, consent, or dissent to medical research. IRB Rev. Hum. Subjects Res., 15, 1±7.

Lewis, C., Lewis, M. and Ifekwunigue, M. (1978). Informed consent by children and participation in an inXuenza vaccine trial. Am. J. Public Health, 68, 1079±82.

Melton, G. (1980). Children's concepts of their rights. J. Clin. Child Psychol., 9, 186±190.

Mossman, K., Goldman, M., Masse, F. et al. (1996). Radiation risk in perspective. Health Physics Society Position Statement. Soc. Nucl. Med. Newsletter.

Nicholson, R. (1986). Medical Research with Children: Ethics, Law, and Practice, pp. 82±6. Oxford: Oxford University Press.

Piaget, J. and Inhelder, B. (1958). The Growth of Logical Thinking from Childhood to Adolescence, New York: Basic Books.

Quinton, D. and Rutter, M. (1976). Early hospital admissions and later disturbances of behavior: An attempted replication of Douglas' Wndings. Dev. Med. Child Neurol., 18, 447±59.

Rosenberg, D. R., Sweeney, J. A. Gillen, J. S. et al. (1997). Simulation for desensitization of children requiring MRI. J. Am. Acad. Child Adolesc. Psychiatry, 36, 853±9.

Small, A. M., Campbell, M., Shay, J. and Goodman, I. S. (1994). Ethical guidelines for psychopharmacological research in children. In Ethics and Child Mental Health, ed. J. Hattab, Jerusalem, Israel: Gefen Publishing.

Varmus, H. (1994). NIH guidelines on the inclusion of women and minorities as subjects in clinical research. Fed. Reg., 59, 14508±13.

Weithorn, L. and Campbell, S. (1982). The competency of children and adolescents to make informed decisions. Child Dev., 53, 1589±98.

Weithorn, L. A. and Scherer, D. G. (1994). Children's involvement in research participation decisions: psychological considerations. In Children as Research Subjects: Science, Ethics, and Law, ed. M. A. Grodin and L. H. Glantz, Cary, NC: Oxford University Press.

Wertz, D. C., Fanos, J. H. and Reilly, P. R. (1994). Genetic testing for children and adolescents. J. Am. Med. Assoc., 272, 875±81.

Williams, K. S., Hankerson, J. G., Ernst, M. and Zametkin, A. (1997). Use of propofol anesthesia during outpatient radiographic imaging studies in patients with Lesch±Nyhan syndrome. J. Clin. Anesth., 9, 61.

Introduction

The neocortex is a thin layer about 2 to 4mm thick that covers the surface of the cerebrum. Neuronal elements occupy 30% of the neocortex, the remainder being accounted for by glial, vascular, and pericytal elements. Only 10% of the neural somata belong to the Golgi type I cells, the source of cortical eVerents. Interneurons (Golgi type II cells) account for the majority of cells in the cortex as well as for a majority (80%) of total cortical synapses. In humans, enlargement of the neuronal somata and its dendritic arbor is no longer detectable after the 6th year of life (Scammon, 1932; Caviness et al., 1997).

The brain achieves its maximum volume by the middle of the second decade of life (for both males and females), reaching a plateau by the 12th year (Dekaban and Sadowsky, 1978). According to Wilmer (1940), brain weight comprises 21% of total body weight at 6 months (fetal period), 15% at term, and only 3% in the adult. The brain is dominated by the cerebrum, which occupies 90% of its volume, and 60% of the cerebrum is gray matter. Most of the gray matter is neocortex. It is 60 times the volume of the diencephalon, the second largest cerebral gray matter structure. A weight diVerence between the brains of males and females is usually the result of a greater volume of central white matter and a larger cerebellum in the male (Caviness et al., 1997).

The neurons and glial cells of the cortex are anatomically grouped according to a vertical and horizontal organization. The horizontal conWguration is known as lamina, and the neocortex generally contains six such layers. When utilizing cell-stained slides, each layer has a diVerent appearance that results from the cell types and their arrangement. Each layer has a speciWc type of input and output as well. Layer IV is always the sensory input layer, receiving incoming connections from the thalamus. The other form of

organization, the one which is actually hierarchical to the lamina, is the vertical organization of the cortex. The vertical array of cells and Wbers uniWes the neurons and laminae into units of function that are known as cell columns.

The Wrst recognition of the anatomic cell column was described by Lorente de No in 1938. Mountcastle (1957) was the Wrst to oVer a general hypothesis of columnar organization. Since then, it has become increasingly clear that the mammalian cortex is organized both anatomically and functionally into discrete units called columns. The word ªcolumnarº has acquired multiple meanings. Columns can be recognized by histologic examination, physiologic studies, and immunocytochemistry. A general deWnition that applies to all forms is that they are an arrangement of operations, neurites, and cell bodies perpendicular to and extending as units through the six laminae of the cerebral cortex.

The smallest unit of columnar organization, the anatomic minicolumn, is based on the ontogenetic cell column (Szentagothai, 1968, 1978; Hubel and Wiesel, 1972, 1974; Mountcastle, 1978; GoYnet, 1984; Leise, 1990). This is a line of cells that arises in the fetal brain and is the Wrst anatomic organization of the cortex, well before the horizontal lamination, which begins around 24±28 weeks. A minicolumn contains approximately 80±120 neurons which extend from the lowest layer VI all the way through layer I and are readily identiWed under the microscope, especially in areas that are cytoarchitecturally deWned as columnar. Larger units of function are known as macrocolumns. The ocular dominance and orientation of columns in the primary visual cortex are examples of macrocolumns. These large units of function are made up of hundreds of co-activating minicolumns and can range from about 300mm to 1500mm in width.

The beauty of columnar organization is the plasticity and dynamic interaction it oVers. Individual minicolumns can