National Academies Press: OpenBook

Placing Children in Special Education: A Strategy for Equity (1982)

Chapter: Biological and Social Factors Contributing to Mild Mental Retardation

« Previous: References
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 133
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 134
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 135
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 136
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 137
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 138
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 139
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 140
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 141
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 142
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 143
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 144
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 145
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 146
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 147
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 148
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 149
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 150
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 151
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 152
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 153
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 154
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 155
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 156
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 157
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 158
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 159
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 160
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 161
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 162
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 163
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 164
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 165
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 166
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 167
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 168
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 169
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 170
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 171
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 172
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 173
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 174
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 175
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 176
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 177
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 178
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 179
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 180
Suggested Citation:"Biological and Social Factors Contributing to Mild Mental Retardation." National Research Council. 1982. Placing Children in Special Education: A Strategy for Equity. Washington, DC: The National Academies Press. doi: 10.17226/9440.
×
Page 181

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Biological and Social Factors Contnbut~ng to Mild Mental Retardation JACK P. SHONKOFF The concept of mental retardation has eluded clear definition for centu- r~es. From the simplistic moralisms of preindustrial times to the complex "scientific" determinations of contemporary societies, the mentally re- tarded population has been to a great extent a cultural creation. As social and economic demands have changed, so have the names and the charac- teristics of the categories of intellectual deficit. The debate over the relative etiological contributions of biological attri- butes in the individual, both inborn and acquired, and sociocultural factors in the environment has raged fiercely. It assumes particular significance in American society today with regard to the phenomenon of mild mental retardation. This paper provides an overview of recent research in areas directly relevant to these issues, formulates the current state of the art, and provides a framework for conceptualizing the available data in their imperfect form. In so doing, it attempts to specifically examine the contri- bution of biological and social factors to the disproportionate representa- tion of minority students and males in education programs for the mildly mentally retarded. HISTORICAL OVERVIEW Shifting criteria for mental defectiveness have clearly mirrored changes in society. In early Western civilizations, handicapped children were fre I am grateful to Ian Canino, C. Keith Conners, Allen Crocker, Leon Eisenberg, Robert Hag- gerty, Jane Mercer, Julius Richmond, and Arnold Sameroff for constructive reactions to an earlier draft of this paper. 133

134 SHONKOFF quently put to death, and those left to survive were often ostracized and cared for by the clergy (Menolascino, 1977~. Before the development of the industrial revolution and universal public education, almost all of those now categorized as mildly retarded were undoubtedly indistinguish- able from the general population. In medieval England, for example, a person merely had to be able "to count twenty pence, to tell one's age, and to name one's parents" in order to avoid designation as an idiot and thereby retain the right to the profits of his own property (Kirman and Bicknell, 1975:51. In the aftermath of the political consciousness of individual rights stirred up by the American and Prench revolutions in the 18th century, attention began to be directed toward the human needs of the mentally handicapped. During much of the 19th century, medicine greatly influenced the societal response to the problem of mental deficiency. While detailed classifica- tions of brain pathology were being compiled by such eminent neurolo- gists as Jean Martin Charcot, the possibilities of education for the "feeble- minded" were being championed by such physicians as Edward Seguin (Blanton, 19751. In an era when universal public education was viewed in the United States as a solution to the growing social problems associated with the industrialization, urbanization, and ethnic diversity resulting from increased immigration from Europe, institutions for the feeble-minded were established in a spirit of educational optimism, not simply as custodial enterprises. As the belief in the reversibility of significant mental retarda- tion weakened, however, the climate of hope and idealism diminished. With the growth of the intelligence testing movement at the turn of the 20th century came fierce battles over the need to protect society from the threat of its defective members who could now be more readily identified. Inspired by the tenets of social Darwinism, some of the most influential American psychologists of the early 20th century, including such luminar- ies as Lewis Terman, Henry Goddard, and Robert Yerkes, joined well-or- ganized efforts to advance the eugenic philosophy, popularized by Sir Francis Galton, by advocating compulsory sterilization and severe restric- tions on immigration. Terman singled out the mildly mentally impaired as a serious threat to the health of the society. In the first edition of the man- ual for the Stanford-Binet scales, he wrote (Terman, 1916:6-71: Intelligence tests will bring tens of thousands of these high grade detectives under the surveillance and protection of society. It is hardly necessary to emphasize that the high grade cases of the type now so frequently overlooked, are precisely the ones whose guardianship it is most important for the state to assume. Mildly retarded people were feared for their assumed tendencies toward immorality, delinquency, criminality, and the propagation of "defective"

Biological and Social Factors 135 children who would further dilute the competence and vitality of American society. The residential institutions that originated in a spirit of salvation evolved into bastions of isolation and educational vacuum. In the years following World War II, encouraged by the work of such re- searchers as Heinz Werner and Alfred Strauss, interest in special educa- tion had a rebirth. In the decades that followed, with the increasing mili- tancy of many parents of handicapped children, the dramatic focus in the 1960s on civil rights for victims of institutionalized discrimination, and the critical support given by President Kennedy to the needs of mentally re- tarded persons, a revolution began in the status of the developmentally dis- abled population in American society. The widely held belief in the bene- fits of segregated special education gave way to arguments for normalized "mainstreaming" in the public school system (Dunn, 1968), which culmi- nated in the passage of the Education for All Handicapped Children Act of 1975 (P.L. 94-142). Historically, the problem of the classification of children for educational purposes has been problematic. In England the passage of the Defective and Epileptic Children (Education) Act in 1899 authorized special classes for children who were deemed incapable of performing adequately in ordi- nary classes but who were not seriously enough impaired to be assigned to an institutional setting. The Education Act of 1921 specifically addressed the needs of the mildly retarded by creating a category of mental defect re- stricted to children ages 7-16 and based on educational but not social de- ficiencies (Blanton, 1975~. At the turn of the century, when the French minister of education com- missioned Alfred Binet to develop a test to facilitate the early identification of children who could not meet the demands of regular schooling, the die was cast and the classification of school children was irrevocably altered. Although Binet himself believed in the value of compensatory instruction, his instrument has sometimes been used as a tool for limiting the educa- tional options for intellectually impaired youngsters. The Binet-Simon scales were adapted for use in school systems throughout Europe and the United States. Data obtained in Belgium and Italy revealed significant dif- ferences in scores related to social class, and eminent cultural anthropolo- gists argued that this "scientific" concept of measured intelligence was very much culturally determined (Blanton, 1975~. In the United States, revisions of the Binet scales were developed by Goddard, Kuhlman, and Terman, and the history of the use of these and other intelligence tests for the educational classification of children has been rich and controversial. At the heart of much ongoing debate has been the conflict between the "scientific," quantitative data obtained from stan- dardized tests and the practical matter of educational classification and

136 SHONKOFF class placement, which is always affected by social values, attitudes, and beliefs. The changing nature of these values has been reflected in the changing definitions of mental retardation. In a presentation to the Na- tional Education Association in 1910, Goddard defined a "subnormal child [as] one who is unable to do school work at the usual rate, or any child who is behind his grade" (Goddard, 1910:242~. He suggested the fol- lowing classification (p. 242~: The temporarily subnormal ... whose backwardness is due to sickness, physical impairment, or unfavorable environment, [and the] permanently subnormal or "feeble-minded" which consists of three subgroups "idiots" twho] are totally ar- rested before the age of three, [the "imbeciles" who] become permanently arrested between the ages of three and seven, [and the "morons" who] become arrested be- tween the ages of seven through twelve. Little attention was paid to individual differences in the mentally retarded population. Generally speaking, a simple qllantitat*e concept of back- wardness was accepted in educational circles, and similar curriculum ma- terials were applied for a variety of children with diverse learning handi- caps. It was not until Werner and Strauss (1939) began to talk about the importance of functional analyses of individual strengths and weaknesses rather than standardized test scores that the concept of mental retarda- tion as a homogeneous condition was seriously challenged. Their popular- ization of the notions of endogenous (familial) and exogenous (secondary to prenatal, perinatal, or postnatal brain insult) mental retardation ushered in a new era of special education and laid the foundation for many of the modern concepts of specific learning disabilities. In 1953 a committee of the World Health Organization defined mental deficiency as incomplete or insufficient general development of the mental capacities secondary to biological factors and defined mental retardation as the same condition secondary to social factors. The upper boundary of deficit was conventionally defined as two standard deviations below the mean on a standardized intelligence test. In 1959 the American Association on Mental Deficiency (AAMD) pro- posed a system of classification that included a requirement for assessing adaptive behavior and created the category "borderline retardation" for those individuals with "subaverage intellectual functioning" as defined by a test score of between one and two standard deviations below the mean. Among the novel features of this model were its emphasis on current level of functioning and its focus on individuals whose deficits are manifested during the developmental period (Heber, l9S9~. In 1973 the AAMD announced that "since 1959 numerous changes have

Biological and Social Factors 137 taken place in the field and in the society which necessitate a new manual to reflect the knowledge and philosophy of the seventies" (Grossman, 1973:4-5~. This new definition of mental retardation, which is still cur- rent, required "significantly subaverage general intellectual functioning," which was defined as two standard deviations below the mean, thereby eliminating the category of "borderline retardation." In their acknowl- edgment of "changing concepts regarding the social capability of persons with low intelligence" (p. 5), the AAMD arbitrarily transferred a segment of the mentally retarded population back into the "normal" fold with a simple stroke of the pen. As observed by MacMillan et al. (1980:112), "many of the children in a mildly retarded sample study conducted in 1965 would be 'nonretarded control' subjects today if they achieved an IQ of 75 to 85." Diagnostic systems for retardation have changed in their conceptual as well as their quantitative dimensions. They have alternately stressed the functional interests of psychometricians and educators and the etiological curiosities of the medical profession. Perhaps the best analysis of the dif- ferential impact of diverse models of diagnosis is that of Mercer (1971~. She defines the clinical perspective as one that considers retardation to be an intrinsic handicapping condition. The current AAMD definition re- flects this perspective. It is a statistical and pathological model designed to serve the needs of the helping professions, e.g., medicine, psychology, and education. The clinical perspective implies that a person who fits the criteria is in fact mentally retarded, even if no one is aware of that fact and a definitive diagnosis has not been made. The social system perspective, by contrast, implies that the status is assigned to an individual within a specific social milieu. The implication of this sociological model is that a person is in fact mentally retarded only when he or she is designated as such by a social system and therefore is perceived that way by its other members. Generally speaking, the school has traditionally been the sys- tem that most frequently assigns the social status of mental retardation. It is therefore critical that we gain greater insight into the factors that con- tribute to those administrative decisions that can so dramatically affect children's lives. The need to recognize that we are dealing with values and not objective truths is an important beginning. In summary, the concept of mental retardation is fluid and defies perma- nent definition. In its mild manifestations, it is less a vehicle for understand- ing those people whom it labels than a mirror of the society that determines its boundaries (Sarason and Doris, 1979~. In this context of uncertainty this paper explores the data regarding the biological and social roots of mild retardation.

138 EPIDEMIOLOGY OF MILD MENTAL RETARDATION SHONKOFF In view of the continually changing definition of the mildly retarded popu- lation, it is not at all surprising to discover that this group is very difficult to count. Indeed, the search for valid epidemiological data has been fraught with frustration and inevitable limitations. Some of the confounding factors are related to methodological difficulties, while others are inherent in the chameleonlike nature of the condition itself. TYPE S OF DATA Two types of data have been the focus of study: incidence and prevalence rates. Incidence refers to the number of new cases of a condition that occur in a given time interval. These kinds of data have been particularly prob- lematic for the study of mental retardation because of the difficulty in de- termining the point at which the condition begins to exist. For children whose diagnosis is specific and unequivocal (e.g., Down's syndrome), this question has been relatively easy to answer. For the mildly retarded popu- lation, however, the point at which the diagnosis may appropriately be made is often difficult to ascertain. The empirical observation that an indi- vidual may move in and out of the mildly retarded category further clouds the usefulness of incidence data. Prevalence refers to the number of individuals who have a given condi- tion at a specific point in time. Although they are related to incidence data, prevalence rates are affected by the duration of a condition and are there- fore lowered by the removal of persons from the target population through death, "cure," or diagnostic revision. This paper focuses primarily on prevalence data, as these numbers are the most relevant for defining and planning intervention services. LIMITATIONS OF THE DATA The most fundamental dilemma is clearly related to the absence of a con- sistent definition of mild retardation. Whereas moderate and severe men- tal retardation have been relatively easy to identify, regardless of changing nosologies, the boundary between "mildly defective" and "low normal" re- mains ambiguous and tentative. As discussed above, diagnostic criteria have been altered as the values of the society have changed, and it is likely that further modifications will be developed in the future. Moreover, the present emphasis on concurrent adaptive behavior requires consideration of abilities that have traditionally eluded reliable and valid quantification. In the absence of a permanent, universally acceptable definition, it is

Biological and Social Factors 139 not at all surprising that much of the available epidemiological data on all levels of mental retardation have been significantly influenced by the era during which they were collected, the target groups studied, and the disci- plinary orientations of the investigators. Clinical and school populations, for example, are not at all comparable. On one hand, medically based studies are generally skewed by populations with a disproportionate num- ber of "patients" with medically diagnosable conditions characterized by abnormal neurological signs and well-described clusters of findings (syn- dromes). Educationally based studies, on the other hand, understandably rely heavily on classifications related to school placement and pedagogical strategies. Thus, in some instances, a reported low prevalence of mental retardation may simply reflect limited resources for special education or a strong commitment to "mainstreaming" and individualized instruction; alternatively, a high prevalence rate may reflect artificially inflated figures designed to secure increased funding for service programs. MacMillan et al. (1980) examined the implications of these variations related to the sources of data for the planning and interpretation of relevant research. They differentiated between the mission of the school (which is to deliver education services) and the mandate of the psychological researcher (which is to build a model of retardation based on scientific rigor). The former is heavily influenced by variations in teacher behavior regarding re- ferrals, differences in the way those referrals are screened, and the range of alternative placements and education options available within each school system. The latter should be characterized by strict adherence to objective and highly reproducible data. Consequently, meaningful comparisons among studies clearly require explicit information on the criteria for selec- tion of each target group. Sociological and anthropological investigations have employed yet an- other framework whereby retardation is defined in terms of a broad ecolo- gical analysis of social status within a specific cultural milieu. Robinson (1978) noted that the reported prevalence of mild mental retardation in the People's Republic of China is essentially zero; their technologically unso- phisticated society places minimal value on individual achievement and maximal emphasis on social cohesion and mutual support. In Sweden, where industrial modernization and emphasis on achievement are more ev- ident, the reported prevalence of mild retardation is also relatively low, in part because of social acceptance of educational mainstreaming of intellec- tually limited children (Grunewald, 1979~. In both countries the preva- lence of mental retardation at all levels is significantly lower than reported in most studies because they primarily consider the severely impaired. From the clinical perspective, the mildly retarded have been overlooked; from a sociological perspective, they do not exist as a discrete group.

140 SHONKOFF In addition to the problems of disciplinary variation and changes over time in the definitions employed, methodological rigor within disciplines and contemporary studies has been wanting. The bulk of the epidemiologi- cal literature does not conform to the AAMD requirement that a diagnosis of mental retardation be based on well-standardized measurement of both adaptive and intellectual deficits. Smith and Polloway (1979), for example, found the inclusion of adaptive behavior measures in less than 10 percent of the recent research efforts that they reviewed. Cleland (1979) reported that many studies mismatched individuals' test scores with the appropriate level of retardation. In an analysis of 566 articles in the American Journal of Mental Deficiency and Mental Retardation from 1973 through 1979, Taylor (1980) found that only 28 percent included terminology consistent with the AAMD classifications, confirming Cleland's assertions by demon- strating that almost 20 percent of the studies he reviewed included subjects who had been inappropriately classified based on data presented in the article itself. Interpretation of such information clearly presents major problems. The variety of data-collection methods employed has contributed addi- tional confusion to the literature. Lemkau et al. (1942) studied the preva- lence of mental disorders in Baltimore, Maryland, through an examination of the records of community and state agencies. Bremer (1951) surveyed the entire population ( 1,300 people) of a small Norwegian fishing village through interviews and personal observations. Wishik (1964) studied two Georgia counties through a combination of a communitywide campaign to solicit voluntary referrals and a canvass of 10 percent of the households in the area. Lapouse and Weitzner (1970) reviewed these and nine other epi- demiological studies, whose case-finding mechanisms ranged from reviews of school and other agency records to sample surveys, interviews with key community informants, and individual testing by the investigators them- selves. The prevalence rates for all levels of mental retardation generated by this wide variety of methods ranged from a low of 3.4/1,000 to a high of 77.0/1,000. When broken down by severity, the percentages of mild retar- dation within each group ranged from 63 to 92 percent, with a median of 80 percent. Clearly, the limitations of the available epidemiological data are formidable. With these caveats in mind, we now examine the numbers. PREVALENCE OF MILD MENTAL RETARDATION If intelligence were, in reality, normally distributed on a Gaussian curve, the prevalence of all degrees of mental retardation would be 2.28 percent. In fact, however, this is not the case. Several explanations have been of

Biological and Social Factors 141 fered to identify the reasons for the empirically observed variations from the statistically predicted rates. Tarjan et al. (1973) have asserted that the true prevalence of mental re- tardation is closer to 1 percent. They explain the lower figure largely on the basis of the fact that not all people with IQ scores below two standard devi- ations from the mean have deficits in adaptive behavior (and therefore would not be appropriately classified as retarded). This position is sup- ported by Mercer (1973), who found a prevalence estimate for IQ scores below 70 of 21.4/1,000 in Riverside, California, but a rate of mental retar- dation of 9.7/1,000 when an evaluation of adaptive behavior was added to the diagnostic criteria. Further arguments advanced by Tarjan et al. (1973) to support the lower prevalence figure include the assumption that severely retarded individuals have a shortened life span and the observa- tion that "about two-thirds of the individuals diagnosed as "mildly] re- tarded lose this label during late adolescence or early adulthood" (p. 3721. Rutter et al. (1970) have added another consideration. They report an overall prevalence rate of 2.53 percent (based on IQ scores alone) among the 2,334 children ages 9-11 on the Isle of Wight and note that this con- firmed a slightly higher prevalence than theoretically expected (2.28~o) be- cause of the increased number with severe mental retardation. Given the small absolute number of retarded children in their population (59), the authors did not subdivide their group by levels of severity. The classic studies of Birch et al. (1970) in Aberdeen, Scotland, provide additional data, collected in a somewhat different fashion. Initial preval- ence rates were obtained by ascertaining the number of children (ages 8-10) who were identified as subnormal by the local school authorities and placed in special programs based on evaluation of their social competence, school performance, medical status, and psychometric test scores. These children, whose diagnoses were confirmed after reexamination by the in- vestigators, represented 1.26 percent of the population. Subsequent review of the scores of a psychometric test universally administered at school entry revealed an additional group of children who scored below the cutoff point at age 7 but who were not administratively designated as subnormal in the schools. This group represented 1.49 percent of the population of 8,274 children ages 8-10, giving a best estimate of overall prevalence of mental retardation of 2.75 percent. In the study, 50 percent of the children admin- istratively diagnosed as subnormal had IQ scores of 60 or more, compared with 77 percent of the total group. The authors noted that their prevalence data for Aberdeen reflect the "demands of a modern industrial society with free, universal, and compulsory education and the psychometric screening of virtually all children at 7 years of age" (Birch et al., 1970:9~.

142 SHONKOFF In summary, valid prevalence rates for mild retardation are hard to come by. The overall prevalence of all levels of mental retardation is likely to be between 1 and 3 percent, with at least three quarters of that group probably falling within the range of mild impairment. Of all the method- ological weaknesses throughout this literature, however, the major factor that sabotages efforts to get better numbers is the problem of definition. If it is true that mild retardation will always be a reflection of contemporary cultural values, and if it is true that the boundary between normality and subnormality is inevitably blurred, the hope for more precise prevalence data is fantasy. VARIATIONS RELATED TO POPULATION SUBGROUPS Despite the problems and disagreements described above, a number of strong relationships have consistently been reported regarding the relative prevalence rates of mild retardation among specific demographic subgroups. SOCIOECONOMIC DIFFERENCES In 1962, The Report to the President of the President's Panel on Mental Retardation noted (p. 9~: Epidemiological data from many reliable studies show a remarkably heavy correla- tion between the incidence of mental retardation, particularly in its milder manifes- tations, and the adverse social, economic and cultural status of families in these groups in our population. These are for the most part the low income groups- who often live in slums and are frequently minority groups where the mother and the children receive inadequate medical care, where family breakdown is common, where individuals are without motivation and opportunity and without adequate ed- ucation. In short, the conditions which spawn many other health and social problems are to a large extent the same ones which generate the problem of mental retardation. The documentation of this phenomenon has been extensive and almost uniformly reproducible, although most reports have not included mea- sures of adaptive behavior. In a 1937 study of educational backwardness in children in the regular public schools of London, Burt reported a fre- quency of greater than 20 percent in the poor districts as compared with 1 percent in the well-to-do areas (cited in Rutter et al., 1970~. The New York State Department of Mental Hygiene (1955) in the early 1950s found a fourfold increase in the prevalence of mental retardation (loosely defined to include a variety of problems) from the highest to the lowest socioeco- nomic areas in Syracuse for children and youth under age 18. Stein and Susser (1969) collected data in the industrial city of Salford in northwest

Biological and Social Factors 143 England and found very few children with IQ scores between 50 and 79 in school districts with "high social standing," in contrast to large numbers in districts of "low social standing." The Isle of Wight investigations con- firmed the reproducibility of these findings for small-town as well as inner- city populations (Rutter et al., 1970~. In their elegant studies in Aberdeen, Birch et al. (1970) reported a prevalence of mild retardation approaching zero in the upper socioeconomic classes, with an increase in prevalence rates by a factor of two for each step down the class ladder, resulting in a summary conclusion that the prevalence of mild retardation (based on IQ greater than or equal to 60) was nine times higher in the lowest class than in the highest class. When within-class differences were examined, it was found that approximately 91 percent of the lower-class population of re- tarded children were mildly impaired (IQ greater than or equal to 50), while 89 percent of the retarded children in the highest class were moder- ately to severely subnormal (IQ less than 50~. Detailed analysis of the data confirmed the fact that these marked discrepancies were accurate reflec- tions of the prevalence rates based on the diagnostic criteria accepted for the study and were not an artifact related to class differences in adminis- trative identification by the school system. Lapouse and Weitzner (1970) reviewed 12 epidemiological studies that further confirmed this inverse re- lationship between socioeconomic status and prevalence rates for mild retardation. A recent analysis of data on more than 35,000 children from the Collab- orative Perinatal Project of the National Institute of Neurological and Communicative Disorders and Stroke specifically looked at the relation- ship of race and socioeconomic status to the prevalence of mild retardation based on test scores of 50 to 69 on the (WISC-R at age 71. Rates for the white population were 3.34 percent for the lower socioeconomic group (bottom 25 percent), 1.31 percent for the middle group (middle 50 percent), and 0.30 percent for the upper group (top 25 percent), with an overall prevalence of mild retardation for the white children of 1.17 per- cent. Data for the black youngsters revealed a rate of 7.75 percent for the lower socioeconomic group, 3.59 percent for the middle group, and 1.19 percent for the upper group, with an overall rate of 4.83 percent (Broman, unpublished data, 1981~. Many investigators have tended to subsume the demographic character- istics of the lower socioeconomic classes under conceptualization desig- nated as the culture of poverty, which implies a pervasive psychological sense of hopelessness and the inevitability of competitive disadvantage. Others have observed that such a view merely serves as an excuse for policy makers and educators to expect minimal benefits from intervention efforts (Ryan, 1971, cited in Eisenberg and Earls, 1975~. Attempts to analyze var

144 SHONKOFF tables within the lower socioeconomic groups have yielded inconsistent findings. In examining the relationship between mild retardation and class status, Birch et al. (1970) found an ever greater prevalence in the portion of the lowest socioeconomic classes living in large families in areas with particularly poor and overcrowded housing. Zajonc (1976) suggests that regional and ethnic differences in intellectual test performance are signifi- cantly related to family configuration, including factors such as the order and number of children and the time interval between their births. Firkow- ska et al. (1978) found that, although family size was an influential factor, parental education and occupation were the major variables affecting scores on the Ravens Progressive Matrices among 11-year-old children in Warsaw, Poland, where housing and community resources were of equal quality in the socially and economically heterogeneous neighborhoods that were created by the government following World War II. RACIAL AND ETHNIC DIFFERENCE S Those studies that have systematically examined epidemiological data for racial differences in the reported prevalence of mental retardation have demonstrated consistent findings of disadvantage for minority groups. Four of the projects reviewed by Lapouse and Weitzner (1970) provide in- teresting insight into some fundamental issues. In a survey of the total pop- ulation under age 18 in Onondaga County, in New York, in 1953 (342,000, 98 percent white), based on requested referrals from all possible commu- nity agencies, an overall mental retardation prevalence of 35.2/1,000 was found, based on an IQ cutoff score of 90. When analyzed for racial differ- ences, the rate in the city of Syracuse for nonwhite children was 125/1,000 compared with 30.9/1,000 for white children. This fourfold discrepancy was reduced to a twofold difference (130.7/1,000 versus 63.9/1,000) when children from the same socioeconomic area in the city were compared. Rates for the remainder of the county were reported to be 88.9/1,000 for nonwhites and 30.0/1,000 for whites (New York State Department of Mental Hygiene, 1955~. A major question obviously raised by this study re- lates to the validity of data obtained through soliciting records from com- munity agencies whose individual identification and selection criteria are not clearly defined, especially with regard to race and ethnicity. Moreover, the establishment of an IQ score of 90 as the criterion for subnormality is highly problematic. Studies by Lemkau et al. (1941, 1942) in the urban Eastern Health District in Baltimore, Maryland, provide fascinating data related to the in- teraction between race and age. Case finding was accomplished through record reviews of a wide range of community agencies, including schools,

Biological and Social Factors 145 prisons, and courts for all age groups. Using an IQ score of 69 as the cut- off, a prevalence rate of 12.2/1,000 was calculated for the entire popula- tion of 54,600. Analysis of children ages 10-14 revealed prevalences of 98.2/1,000 for nonwhites and 26.1/1,000 for whites. Further examination of the data for people ages 20-60, however, revealed essentially no racial differences (7.2/1,000 for nonwhites versus 6.5/1,000 for whites) in IQ scores below 69. It appears that the racial differences as well as the overall changes in prevalence rates are related to issues that are peculiar to the school years. Wishik (1964) reported an overall prevalence rate of 36.6/1,000 in a study population of 55,000 under age 21 in two counties in Georgia se- lected as being representative of the state regarding racial (27 percent black) and urban-rural characteristics. Individuals were located through a solicitation of referrals and a random household survey and were identified as retarded based on an IQ score less than 80 and the clinical judgment of pediatricians. Analysis of the target group revealed no significant racial differences in prevalence rates. Reschly and Jip son (1976) administered individual IQ tests (WISC-R) to 950 of a stratified sample of 1,040 children in Pima County, Arizona. Scores revealed markedly increased prevalence rates of mild mental retar- dation for black, Mexican-American, and Pap ago Indian children com- pared to Anglo children when full-scale IQ scores were examined and a cutoff at 75 was used. When the cutoff score was reduced to 69 and the per- formance IQ was used as the criterion, however, the disproportionate clas- sification was eliminated for the Mexican-Americans and greatly reduced for the black and Pap ago Indian children. Data from the Collaborative Perinatal Project of the National Institute of Neurological and Communi- cative Disorders and Stroke, based on IQ scores (WISC-R) between 50 and 69, revealed a prevalence rate for mild retardation of 4.83 percent among blacks and 1.17 percent among whites, with a persistence of at least a two- fold difference across all socioeconomic groups (Broman, unpublished data, 1981). Perhaps the best-known and most influential work on ethnic dispropor- tion in the classification of school children has been the studies conducted in Riverside, California. By critically examining the validity of standard- ized intelligence tests, Mercer (1973) demonstrated the overwhelming im- portance of culturally appropriate evaluations of adaptive behavior in order to justify a diagnosis of mental retardation. The addition of an assess- ment of adaptive behavior to the criterion of an IQ score less than 70 reduced the prevalence rate from 21.4/1,000 to 9.7/1,000. Of greater importance, however, was the observation that the decrease in diagnosed retardation was even more dramatic for black and Mexican-American

146 SHONKOFF children, with reductions of 44.9 to 4.1 and 149.0 to 60.0 per 1,000, re- spectively. The lack of change in the prevalence rate for Anglo children (4.4/1,000) clearly demonstrated the cultural discrimination of the IQ test and highlighted its contribution to the disproportionate classification of children from ethnic and racial minority groups. Controversies over racial and ethnic differences in IQ scores have been passionately raging since the introduction of intelligence testing in the early part of this century. These issues will not be addressed further in this paper. SEX DIFFERENCE S There can be little argument against the claim that much of the difference between the behavior of males and females in a given society is culturally determined. Nevertheless, biological differences between the sexes that are independent of social milieu have been well documented and must also be considered whenever specific characteristics are found to be distributed in a disproportionate manner. Two issues that bear some consideration in this regard are the greater susceptibility of males to a range of adverse con- ditions and their relatively slower rate of maturation for a variety of biolog- ical functions. A substantial amount of data has accumulated demonstrating the greater biological vulnerability of males (Childs, 1965; Hutt, 1978; Winter, 19721. A review of mortality indices reveals a higher proportion of males reported in spontaneous abortions (Stevenson and McClarin, 1957) as well as in neonatal deaths (Naeye et al., 1971~. In developing countries, male infants succumb to the intestinal complications of poor sanitation in greater numbers than females (Potts, 1970~. Males are more susceptible to infec- tious diseases, including neonatal septicemia (Smith et al., 1956) and those that affect the central nervous system, such as meningitis and en- cephalitis (Carpenter and Petersdorf, 19623. The ratio of febrile seizures in boys compared with girls has been reported as 1.4:1.0 (Flor-Henry, 1974~. Males have repeatedly been shown to have a much greater rate of involvement in accidents, especially after the first two years of life (Huts, 1978; Winter, 19721. The relatively slower rate of maturation of boys has also been well stud- ied. Boys have lower growth velocity and later bone ossification and begin puberty on average about 2.5 years after girls (Nicolson and Hanley, 1953~. Although some inconsistencies have been reported, a fair amount of data has been generated that indicates that girls mature cognitively and linguis- tically at a faster rate than boys in the early years (Waber, 1976~. Hutt (1978) suggests that the relatively protracted period of development in boys may increase the length of any theoretically sensitive periods during which

Biological and Social Factors 147 negative influences, such as malnutrition, could have an effect on brain de- velopment. Moreover, preliminary evidence suggesting more complete lat- eralization of language and spatial abilities in male brains might mean that the lesser degree of cerebral lateralization in female brains may reflect greater plasticity and therefore less susceptibility to the effects of unilateral insults (Lake and Bryden, 1976; Witelson, 19761. Although it is generally said that mental retardation is more common in males than in females (Farber, 1968; Goodman et al., 1956; Kirk and Weiner, 1959), the literature on sex differences in prevalence rates is ac- tually somewhat equivocal. Rutter et al. (1970) reports that although there is widespread agreement that severe mental retardation is somewhat more common in boys than girls, the sex distribution for mild retardation is fairly equal. They explain this discrepancy by distinguishing between mental re- tardation per se and educational backwardness. Data collected on the Isle of Wight, for example, revealed a prevalence rate for "intellectual retarda- tion" of about 2.5 percent; a prevalence rate for specific reading retarda- tion of about 4 percent; a prevalence rate for general "reading backward- ness'' of 6.5 percent; and, with some overlap among the groups, an overall prevalence rate for "severe intellectual or educational difficulties" of 8 per- cent. Although the male-female ratio for intellectual retardation was found to be essentially equal (0.9:1), the ratio for specific reading retarda- tion was 3.3:1. It was suggested that the greater prevalence of school fail- ure in boys is related to specific reading problems rather than global intel- lectual deficits. Other investigators report different conclusions. Birch et al. ~ 1970) found a slightly higher ratio of boys to girls who were rated abnormal (56 percent versus 44 percent), due largely to significant sex differences in those with IQ scores greater than 70, compared with little or no differences in the more severely impaired children. Lapouse and Weitzner (1970), in their review of 12 studies, reported a range of male-female ratios of mental retardation from a low of 1.1:1 to a high of 1.9:1, with only one exception reflecting a greater proportion of females. When levels of retardation were examined separately, however, the sex differences were inconsistent. In a study designed specifically to look at the prevalence of mild retardation based on IQ scores, Reschly and Jipson (1976) actually found a higher rate among females, although the differences were not statistically significant. Data from the Collaborative Perinatal Project corroborated that, for whites, girls have a higher rate of mild retardation (using scores of 50 to 69 on the WISC-R for children age 7) than boys (1.29 percent versus 1.03 per- cent) and, for blacks, boys have a higher rate than girls (4.99 percent ver- sus 4.24 percent) (Broman, unpublished data, 1981~. Despite a substantial amount of evidence to suggest the greater biologi

148 SHONKOFF Cal vulnerability of males than females and in the face of well-documented greater numbers of boys than girls in special education placements, the ep- idemiological literature does not confirm a consistently higher prevalence of mild mental retardation in males. GE OGRAPHIC DIFFERENCE S The differential impact of rural versus urban life Oh the prevalence of men- tal retardation has been difficult to ascertain. Some investigators have sug- gested that urban residence is correlated with higher levels of intelligence (Lehman, 1959; McNemar, 1942), while others have found no consistent differences (Jastak et al., 1963; Lapouse and Weitzner, 19701. Careful analysis of the confounding influences of socioeconomic and ethnic factors has not been done, and the data in this regard are therefore inconclusive. AGE DIFFERENCE S One of the most consistent findings among epidemiological studies of men- tal retardation is the dramatic change in prevalence rates with age. Gener- ally speaking, most retarded persons are mildly impaired, and the bulk of this group is not identified until the school years, with subsequent loss of official diagnostic classification in adult life (Farber, 1968; Goodman and Tizard, 19621. The 12 studies reviewed by Lapouse and Weitzner (1970) showed an increased prevalence in retardation (regardless of the definition used) between the first two 5-year periods (from birth to age 5 and ages 5-9), a larger increment during the next 5-year period (ages 10 14), a de- crease in the prevalence rate by half during the next 5 years (ages 15-19), and a further decrease beyond age 20 to a prevalence rate that remains es- sentially stable throughout adult life. Gruenberg (1964) reported different prevalence rates in many countries (England, Formosa, Scandinavia, and the United States) but similarly shaped curves for age-specific rates with a steady rise to peak levels during the school years and a steady decline thereafter. MacMillan et al. (1980) noted that the school has been the ma- jor identifier of mildly retarded people, those who are "not easily differen- tiated from non-retarded children in playground, marketplace, and employment situations that do not make school-like cognitive demands" (p. 109~. These trends in age-related rates, perhaps more than any other data, un- derline the role of the school in the pathogenesis of mild mental retarda- tion. Although some might attribute the rising number during the school years to more effective diagnostic systems, the subsequent declining preva- lence in late adolescence and adulthood provides a strong argument for the

Biological and Social Factors 149 significance of extrinsic social factors in the assignment of this label Fur- ther studies of the complex relationships among the demands of formal ed- ucation and the requirements for competence in adult life are clearly need- ed to inform the development of policy guidelines in this critical area. SUMMARY In summary, the overall prevalence of mental retardation ranges between 1 and 3 percent, at least three quarters of whom are probably mildly re- tarded. Although precise data are most likely unachievable, there can be little question that ethnic minority groups and those in the lowest socioeco- nomic strata in society comprise a significantly disproportionate segment of those labeled as impaired. The data on sex distributions are more com- plex in that the numbers of boys assigned to categories of special educa- tional need far outnumber those of girls, yet the epidemiological data on prevalence of mental retardation are less consistent and somewhat equivo- cal with regard to sex differences. Perhaps the most striking finding in the epidemiological literature is the critical influence of age on diagnosis. In a variety of social contexts and re- gardless of the definition employed, the numbers of children identified as mentally retarded have been demonstrated to peak consistently in the ele- mentary and junior high school years. This relationship between preva- lence data and age confirms without question the fact that mild retarda- tion is largely a creation of universal compulsory education. Despite all the definitional confusion and methodological variation, the data show a consistent tension between the demands of the school and the performance of poor, nonwhite children, especially boys. The sociocul- tural explanations for this phenomenon are most compelling and unargu- able, yet, as Birch and Gussow (1970) so eloquently warned (pp. 6-7~: There is some danger, however, that our initial focus on the social and cultural vari- ables relevant to educational achievement may lead us to neglect certain big-social factors which can directly-or indirectly influence the developing child and alter his primary characteristics as a learner.... The fact is that the child who is both the subject and the object of all this concern, the individual who is interacting with these social, cultural and educational settings, is a biological organism.... As an organism the child is not only a mind and a personality capable of being unmoti- vated, unprepared, hostile, frustrated, understimulated, inattentive, distracted or bored; he is also a body which can be tired, hungry, sick, feverish, parasitized, brain-damaged or otherwise organically impaired. The remainder of this paper examines the interplay between biological and social factors that may affect school achievement.

150 THE BIOSOCIAL ROOTS OF MILD RETARDATION SHONKOFF The question of etiology at all levels of mental retardation frequently goes unanswered. In a survey of 800 severely retarded, institutionalized per- sons, Berg (1963) was able to identify a definite cause or known syndrome in only one third of the cases. In a more recent study at the Fernald State School in Massachusetts, 34 percent of the 1,077 residents with IQ scores below 50 were designated as retarded for unknown reasons (Moser and Wolff, 19711. When one seeks to identify etiological mechanisms in the mildly retarded population, the task is even more formidable and unre- vealing. In view of the fact that the frequency of abnormal neurological findings is negligible in most of the mildly impaired population (Rutter et al., 1970; Birch et al., 1970), there is often a tendency to minimize the im- portance of organic factors among the unknown (or at least unproven) causes of a child's diminished abilities. Optimal competence and performance for any child, however, are de- pendent on the interplay between intrinsic biological integrity and an envi- ronment that facilitates the development of skills and positive self-esteem. The relationship between mild retardation and lower socioeconomic and ethnic minority status as well as the greater prevalence of school failure among boys have been extensively analyzed from educational, sociological, and political perspectives. Without minimizing the validity of cultural in- fluences, however, it is important to keep in mind the very real discrepan- cies in the distribution of biological factors that predispose children to poor school performance. In very simple terms, brain function is a critical determinant of intelligence, and factors that may adversely affect brain function are found with greater frequency among males as well as in groups that are victims of institutionalized social disadvantage, such as members of ethnic minorities and the poor. Although the complexity of the data have so far precluded a clarification of the differential contributions of nature and nurture, we cannot justify a summary disregard for the causal role of organic vulnerabilities in mildly retarded school children. pREcoNcEprIoN INFLUENCES Before conception there are already two sets of variables that have poten- tial effect on the developmental competence of the child who is ultimately born. The first involves the genetic contribution of each parent, and the second relates to those demographic factors that correlate with increased risks for the successful completion of gestation.

Biological and Social Factors Genetic Factors 151 Genetic causes of mental retardation can be related to abnormalities of chromosomes, single genes, or multifactorial inheritance. Chromosomal disorders with associated mental retardation are generally characterized by moderate to severe intellectual deficits and/or atypical physical find- ings, including neurological abnormalities. Down's syndrome is the most common example. Others, such as Turner's and Kleinfelter's syndromes, may have associated mild retardation, but the majority of these children are of normal intelligence. The relatively low incidence of these conditions (1/10,000 female births and 1/1,000 male births, respectively), the low rate of intellectual impairment involved, and the absence of data to suggest disproportionate distribution among socioeconomic classes or ethnic mi- norities suggest that major chromosomal disorders do not contribute to the numbers of mildly retarded children in any appreciable way. Specific sex chromosome abnormalities have been associated with suggestions of devel- opmental vulnerability, particularly for language but not with mental re- tardation (Leonard et al., 1974; Tennes et al., 1977~. The recent discovery of the so-called fragile X chromosome in a number of institutionalized re- tarded males, whose causes of impairment were previously unknown, how- ever, has opened up new areas of investigation that may shed light on the disproportionate number of males among the severely retarded (Gerald, 1980~. The association of the fragile X chromosome with mild retardation in females has recently been noted, and further study is clearly needed (Turner et al., 1980~. Single-gene abnormalities may be inherited through autosomal domi- nant, autosomal recessive, X-linked dominant, or X-linked recessive mech- anisms. Although many of these disorders (such as sickle-cell disease and cystic fibrosis) are not associated with intellectual deficits, a large number of inborn errors of metabolism (such as phenylketonuria) that are inher- ited as autosomal recessive disorders are accompanied by moderate to severe retardation. Some of these inborn errors have a high incidence in certain ethnic groups (e.g., Tay-Sachs disease among Ashkenazi Jews). No associations have been demonstrated, however, with social class or with those ethnic groups that have been disproportionately identified in the mildly retarded population. Again, the relatively low incidence of these metabolic disorders (e.g., 1/14,000 births for phenylketonuria) and their usual association with severe intellectual deficits often accompanied by progressive neurological deterioration eliminate their relevance for the mildly retarded population. Multifactorial inheritance refers to the process whereby a disorder or

152 SHONKOFF condition is determined by the synergistic effects of one or more so-called minor genes and environmental factors. Often termed "polygenic," these mechanisms have been postulated by several investigators to explain the increased prevalence of mild retardation among ethnic minorities and lower socioeconomic groups as a result of genetic differences in intelli- gence. Such theorists have argued that poor people and blacks, for exam- ple, have lesser intellectual endowments, which they pass on to their chil- dren in a manner similar to other phenotypic characteristics, such as height or hair color. The polygenic inheritance mechanism is the core around which theories of racial intellectual inferiority have been built. The problem with its application to the study of intelligence is that the methods needed to analyze the relative contributions of biology and environment have not been adequately developed. There is little question that intellec- tual competence is significantly affected by both. How much of the vari- ance is determined by each, however, varies with circumstances. That is to say, in a uniformly optimal environment, heritability accounts for a great deal of the variance; in a wide range of environmental situations, heritabil- ity will explain much less. In summary, there is no evidence that discrete genetic disorders play any role in the incidence of mild mental retardation. The role of genetic factors in the increased frequency of developmental deficits in males appears to be restricted to more severe levels of retardation, but further work is needed to elucidate possible genetic contributions to the apparent developmental vulnerability of boys. Multifactorial inheritance, as it refers to the interac- tion between genetic predisposition and environmental contingencies, is more difficult to assess. As discussed in the remainder of this paper, many biological risk factors that are found disproportionately among ethnic mi- norities and boys have their onset in early life but are not genetic. More- over, even if genetic differences did exist, their influence on outcome for the mildly retarded would be overshadowed by the effects of the subopti- mal environments within which ethnic minorities and the poor reside. Demographic Risk Factors Pregnancies that involve factors that increase the likelihood of perinatal mortality, prematurity, low birth weight, or a wide variety of handicapping conditions, including mental retardation, are called "high risk." It has been estimated that such pregnancies account for more than half of all peri- natal mortality and morbidity (Vaughan et al., 19791. Birch and Gussow (1970) report that "almost every complication of pregnancy, labor, delivery, and the perinatal period which is potentially damaging to children is exces- sively prevalent among economically depressed populations and particu

Biological and Social Factors 153 larly among those further handicapped by ethnic differences" (p. 46~. Ramey and Finkelstein (1978) cataloged a variety of demographic variables found to be associated with "borderline mental retardation," including maternal IQ below 80, family disorganization, poverty, overcrowded hous- ing conditions, parity greater than 5, race, maternal education less than 10 years, illegitimacy, and delayed prenatal care beyond the first trimester (Ramey et al., 19781. Despite the well-publicized value of prenatal care, ap- proximately one quarter of all pregnant women in the United States receive none or only belated medical supervision (Select Panel for the Promotion of Child Health, 1981~. They are more likely to be poor, black, adolescent, un- married, and residing in rural areas. Low birth weight is reported to be three times as likely from such unmonitored pregnancies. The issue of adolescent pregnancy provides a case study in demographic risk. Teenage pregnancies are more common among blacks than whites (Broman, 1980) and are more likely to result in the birth of a low birth weight infant, regardless of social class (American Academy of Pediatrics, 19791. Several investigators have suggested that the increased incidence of small neonates is related to the competition for nutrients between the fetus and the still-growing mother (Naeye, 1981~. Nortman (1974) reported an increased prevalence of handicapped children born to adolescent mothers in Canada, while Baldwin (1976) found that 11 percent of children born to women less than age 16 had IQ scores of less than 70 at age 4, compared with 2.6 percent for the general population. Grant and Heald (1972) sug- gested that risk factors associated with ethnic and socioeconomic status may be the most influential determinants of poor outcome for a teenage pregnancy. This observation is supported by an analysis of data from the Collaborative Perinatal Project of the National Institute of Neurological and Communicative Disorders and Stroke, which found that differences in IQ scores at ages 4 and 7 were more highly correlated with ethnic and so- cioeconomic characteristics than with maternal age (Broman, 19801. In reality, the relationships between discrete demographic variables that predict a high risk for unfavorable pregnancy outcomes and the incidence of specific consequences, such as mild mental retardation, are simply sug- gestive and always tenuous. In a sense a general discussion of demographic factors that increase the risk of mental retardation in a child from a group that is disproportionately represented within the mentally retarded popu- lation is an exercise in circular reasoning. A more careful analysis of the consequences of those specific biological factors that occur with greater frequency among such groups would be more fruitful. The process of de- velopment, however, defies the identification of simple, direct causal rela- tionships. As stated by Birch and Gussow (1970:82~: "When we deal with 'causes' singly, and as simply as the information permits, it is always

154 SHONKOFF within the understood context of a reality in which they are complex and interacting." EARLY PRENATAL INFLUENCES Until the past few decades, the human fetus was believed to be well pro- tected within the mother's womb. Recent research, however, has provided more understanding of the variety of intrauterine factors that can have long-term adverse influences on the organism's ultimate developmental competence. Intrauterine Infections Acute intrauterine infection had long been viewed as a self-limited prob- lem that resulted in either the death of the fetus or complete recovery through elimination of the invading organism by host defense mechanisms (Afford, 19771. The problem of low-grade, chronic, so-called latent infec- tion, however, has become increasingly recognized as an important factor contributing to varying types of long-term sequelae, the dimensions of which are only beginning to be understood. Among the most important or- ganisms in this group are cytomegalovirus (CMV), rubella, toxoplasmosis, and syphilis. They all have a chronic and/or recurrent nature in both mother and fetus and a capacity to adversely affect subsequent cognitive and perceptual development in children. Alford (1977) reported that the susceptibility to infection of women in the childbearing years, as deter- mined by antibody prevalence, is approximately 10 percent for rubella, 15 percent for CMV, and 70 percent for toxoplasmosis. Major variables af- fecting susceptibility include age (younger mothers are more susceptible than older mothers) and socioeconomic status. For reasons that have been inadequately explained, in part because of insufficient data from develop- ing countries, it has been stated that young women and poor women in in- dustrial societies are the most likely to acquire chronic perinatal infec- tions. The overall incidence of maternal infection during pregnancy has been reported as 14 percent, cytomegalovirus being the most common, representing about 13 percent of all infections (Afford, 1981~. According to Alford (1977), between 1 and 7 percent of all infants born in the United States may be infected with one of these chronic organisms, and prevalence is even greater among adolescent women from lower socioeconomic classes. Since CMV is the most common of these infections, it would be instructive to examine its impact in detail. The frequency of congenital cytomegalovirus infection ranges from 0.2 to 8.0 percent of all live births; the average in the United States is 1 per

Biological and Social Factors 155 cent. The highest rates are found in infants born to teenage mothers from lower socioeconomic groups (Hanshaw, 19811. Although approximately 4-5 percent of all women excrete CMV in their urine during pregnancy, most do not have infected infants (Hanshaw et al., 19731. This situation is complicated by the fact that the majority of women who have infections during their pregnancy are asymptomatic and are therefore unaware of the infection. Of the 33,000 infants born in the United States each year with CMV infection, it is estimated that less than 1,500 of them are sympto- matic and therefore easily identifiable in the newborn period. Most of these obviously infected neonates have serious long-term sequelae, includ- ing a high rate of moderate to severe mental retardation. Outcome for the asymptomatic newborn with so-called silent congenital CMV infection is less predictable than for newborns with symptoms but still somewhat worrisome. During the past decade, increasingly sensitive and specific laboratory techniques have facilitated the identification of greater numbers of infected neonates, thus providing an opportunity for prospective studies of both the symptomatic and asymptomatic groups. Results thus far have shown that, although the majority of those with silent infection appear to do well, as many as 10-20 percent develop intellectual or perceptual deficits as well as significant hearing impairment (Hanshaw et al., 1976; Kumar et al., 1973; Melish and Hanshaw, 1973; Reynolds et al., 1974~. As noted by Pass et al. (1980), because of the relatively high fre- quency of asymptomatic congenital CMV (approximately 1 percent of live births), the occurrence of central nervous system damage in even 10 per- cent has significant public health implications. Among those with silent CMV, the influence of socioeconomic status on developmental outcome appears to be important. Hanshaw et al. (1976) screened 8,644 newborns for the IgM antibody against CMV and found 53 children with positive titres. Although only 38 percent of the tested new- borns were born to families in lower socioeconomic groups (Hollingshead groups 4 and 5), 68 percent of the CMV-positive group came from these families. In the study, 44 of the congenitally infected children had IQ tests administered between ages 3.5 and 7.0, and the results were compared with 44 matched and 44 random controls. Although the study sample was small and only 7 children had scores below 79 (all of whom were in the in- fected group), the difference in mean IQ between the CMV-positive and the matched control group was significant with a p value less than .025, after adjustment for social class. No significant IQ differences between the matched and random controls were found when social class differences were taken into account. Further analysis revealed significant differences in IQ scores between CMV-positive and control children from the lower so- cioeconomic families, with no significant IQ differences between those

156 SHONKOFF with and those without congenital infection in the middle-class groups. Predicted school failure, based on an IQ score of less than 90 in association with behavioral, neurological, and auditory evaluations, was not noted among any of the middle- or upper-class CMV-positive children. The lower- class CMV-positive children, on the other hand, had 2.7 times greater predicted school failure than the control children matched for social class. The significant risk for hearing impairment among the infected children (11 percent in this study) clearly contributes additional vulnerability. In summary, the current state of knowledge regarding the influence of asymptomatic congenital cytomegalovirus infection on the prevalence of mild retardation among school children is highly suggestive but far from conclusive. Existing data certainly support the potential adverse effect of silent infection on higher cortical function. The greater prevalence of this condition among children of lower socioeconomic classes appears to be fairly well documented. Controlled studies have shown an effect of the virus on cognitive and perceptual skills independent of social class, yet evi- dence suggests that this may merely represent a subtle biological vulner- ability that can be effectively neutralized by socioeconomic factors (not yet specifically analyzed) in the child-rearing environment. No sex differences have been reported regarding the long-term consequences of these infec- tions. Although a great deal of work obviously remains to be done in this area, available data suggest that congenital infections such as cytomegalo- virus may contribute to the disproportionate number of lower socioeco- nomic class children classified as mentally retarded. Maternal Alcoholism Substance abuse is a major public health problem with considerable at- tendant mortality and morbidity. The special implications of such socio- medical issues (excessive drinking, smoking, drug use, etc.) for the preg- nant woman and her offspring have been the subject of increasing atten- tion. Nevertheless, discrete teratogenic effects attributed to specific chemi- cals or drugs have been well documented in only a very small number of in- stances in comparison to the extensive array of substances that are ingested (both intentionally and inadvertently) by women during their pregnancies. The influences of alcohol on fetal and later childhood development are ex- amined in this section to illustrate some of the problems associated with at- tempting to understand the relationships between such prenatal factors and the later consequence of mental retardation. The association between maternal alcohol consumption during preg- nancy and a constellation of adverse findings in the offspring has been a topic of significant interest and some degree of controversy since the con

Biological and Social Factors 157 cept of the fetal alcohol syndrome was introduced approximately 10 years ago. In its most complete form, this syndrome is characterized by (1) sig- nificant prenatal and postnatal growth deficiency; (2) a combination of characteristic phenotypic abnormalities, including atypical facial features, cardiac defects, and limb anomalies; and (3) central nervous system dys- function with varying degrees of mental retardation (Clarren and Smith, 1978~. A number of serious methodological deficiencies in the existing lit- erature, however, compromise the reliability and validity of the available data. Perhaps the most serious limitations of all studies in this area are related to the problem of the reliability of the amounts of alcohol women claimed they consumed during their pregnancies and the difficulty of establishing uniform criteria for defining such terms as "moderate" and "excessive" intake. Hanson et al. (1978), for example, studied infants born to mothers who reported either an average consumption of one ounce or more of ab- solute alcohol per day or "binges" during the pregnancy with ingestion of 5 or more drinks on a single occasion. Ouellette et al. (1977) calculated total monthly consumption of all alcoholic beverages, divided by 30 to get a daily volume, and defined heavy drinkers as those having more than 5 drinks on occasion with a consistent daily average of more than 45 milliliters of absolute alcohol. In fact, their group of heavy drinkers was found to consume an average of 174 milliliters of absolute alcohol per day. Streissguth et al. (1978) studied 20 individuals, ages 9 months through 21 years, born to chronically alcoholic mothers defined either by "self-report or by reports of social agencies, medical records, and/or family" (p. 364) and reported data that demonstrated a continuum of physical abnormali- ties and mental dysfunctions from severe to mild sequelae. Although careful analysis revealed a relationship between the degree of "dysmor- phogenesis" and the extent of intellectual handicap, a considerable vari- ability of IQ scores among children with similar phenotypic features was found. The hypothesis that the adverse affects of alcohol may in part be dose related is not an unreasonable one, but the methodological limita- tions of the current literature have precluded its evaluation. Moreover, the possible related influences of other ingested substances as well as poor nu- trition have been extremely difficult to analyze. As with all potentially teratogenic substances, the issue of host factors and variable susceptibility also must be addressed. The literature, at this stage of its development, is seriously deficient. Alcoholism is a common problem across a broad ethnic and socioeconomic spectrum. Possible dif- ferences in the vulnerability of pregnant-women based on age, race, in- come, living conditions, general health, and nutritional status have not been adequately examined. Shaywitz et al. (1980) reported the results of a

158 SHONKOFF study of 15 children seen in a learning-disorders unit whose mothers had a history of "alcoholism" (undefined) during their pregnancy. All but one of the children were white and living in private suburban homes. They dem- onstrated a continuum of phenotypic features compatible with the diagno- sis of fetal alcohol syndrome and were all experiencing persistent academic failure in school, yet their full-scale IQ scores ranged from 82 to 115, with a mean of 98. The authors concluded that the concept of the syndrome could be expanded to include more subtle manifestations of central nervous sys- tem dysfunction. Common knowledge suggests that maternal alcohol ingestion can result in a variety of adverse consequences for the fetus, including varying de- grees of mental retardation in later childhood. It is impossible, however, to determine from the available data the relevant variables that contribute to greater or lesser incidence of this syndrome or syndromes. The role of eth- nic or socioeconomic factors has not been well studied, nor have patterns of sex difference been described. Moreover, children whose mothers have chronic drinking problems are obviously a highly vulnerable group from a child-rearing perspective. In view of the high prevalence of alcoholism, this may indeed represent a significant source of biological vulnerability in some groups within the population whose intellectual deficits are unex- plained. At the present time, however, we have no basis for answering this question with much precision. PERINATAL INFLUENCES Perinatal risk factors for subsequent handicaps such as mental retardation have been the focus of extensive investigation, going back as far as Little's (1862) studies of the problem of brain damage related to asphyxia. In 1951, Lilienfeld and Parkhurst introduced the concept of a "continuum of reproductive wastage" to describe the range of possible outcomes, from death to cerebral palsy to varying levels of mental retardation, that were observed to follow difficulties encountered around the time of birth. Pasa- manick and Knobloch (1961) suggested the alternative term "continuum of reproductive casualty" and expanded the spectrum of disorders to in- clude a number of more subtle intellectual and functional deficits. Most recently, Sameroff and Chandler (1975) offered the phrase "continuum of caretaking casualty" to highlight the transaction between biological risk factors and environmental variables that eventually determine develop- mental outcome. Regardless of the phrasing, the central issue relates to the degree to which the brain of a newborn is injured during labor, delivery, or the im- mediate neonatal period. Differential risks regarding the incidence of such

Biological and Social Factors 159 cerebral insults and the degree to which some children are able to recover from a variety of untoward events raise important questions with regard to the prevalence of mild retardation. These issues are examined below within the context of both the general problem of low birth weight and the more specific problems related to discrete insults to the central nervous system. Low birth weight in itself is important as a sign of increased risk for a broad array of pathological conditions that may result in a cerebral injury. Low birth weight as a result of prematurity, which is generally defined as a gestation of less than 37 weeks, is more likely to involve problems with hy- poxia and/or ischemia affecting the cerebral circulation as a result of such disorders as respiratory distress syndrome, hypovolemic shock, and apnea with bradycardia. Additional threats to central nervous system integrity that occur with greater frequency among premature babies include hypo- glycemia, jaundice, infection, postnatal malnutrition, and the increasingly recognized problem of intraventricular hemorrhage. When, however, the newborn's birth weight is significantly low for the expected range given his or her gestational age, the associated problems are different from many of those found in the premature infant. For a "small-for-gestational-age" in- fant, the issue is generally one of intrauterine growth retardation second- ary to such factors as placental insufficiency, maternal malnutrition, in- trauterine infection, or congenital abnormalities. Thus, low birth weight babies comprise a heterogeneous group with a variety of vulnerabilities. The report of the Select Panel for the Promotion of Child Health (1981) states that "it is generally agreed that very low birth weight is among the most significant predictors of later neurological abnormalities and various cognitive and behavioral deficits" (p. 47~. Many of the data regarding cor- relations between birth weight and developmental outcome, however, have been equivocal. Kiely and Paneth (1981) reviewed the methodological dif- ficulties that have characterized these follow-up studies and found them to fall into two broad categories: limitations in study design and problems re- lated to data analysis and reporting. The selection of single hospital sam- ples, for example, has made generalization about the results extremely dif- ficult. The absence of attention to socioeconomic status in the selection of control groups and in the analysis of data is another major shortcoming of many major follow-up studies. With regard to the issue of intelligence test results, studies vary in their reporting mechanisms some neglect to spec- ify the ages at testing or the instruments used, some indicate only mean IQ scores' and others report data on single cutoff points such as 90 or 70. Problems of terminology also have plagued the literature. Caputo and Mandell (1970) noted that many studies used the terms "low birth weight," "immaturity," "prematurity," and "short gestation" interchangeably. In

160 SHONKOFF most of the early reports, birth weight of less than 2,500 grams (5.5 pounds) was generally employed as the sole criterion for determining prematurity; no distinctions were made for infants who were full-term but small for their gestational age. Data from those studies that failed to classify small infants are especially difficult to interpret. An additional problem regarding longitudinal data of this type relates to the rapid rate of technological change in perinatal intensive care. That is to say, by the time school-age follow-up studies are completed, the care techniques for these small neonates have changed so dramatically that it is difficult to assess the validity of the findings for the new generation of tiny newborns. Given these serious limitations in the literature, the difficulties we have in drawing definite conclusions from the existing data are not surprising. Many investigators have reported a high incidence of developmental mor- bidity in these groups. In a prospective study of 241 infants classified by birth weight, gestational age, and sex, Rubin et al. (1973) found that two thirds of low birth weight males and more than half of the total group of former small-for-gestational-age babies of both sexes had problems of suf- ficient magnitude to warrant a wide variety of special educational services (which were not well defined) in the elementary school grades. Analysis of all measures of mental development, language skills, school readiness, and academic achievement from preschool through age 7 revealed lower scores for low birth weight subjects as compared to a random control group. Ranges of scores, however, were not provided and analysis for socioeco- nomic differences was incomplete. Parkinson et al. (1981) studied 45 for- mer full-term, small-for-gestational-age babies between ages 5 and 9 and 19 control children matched for age, sex, birth order, social class, and race. Based on teachers' assessments, the authors found that small-for- gestational-age children may have difficulties at school, the severity of which is related to sex (boys have more problems than girls), social class, and the stage of pregnancy at which slow head growth began. No formal test scores were obtained. Fitzhardinge and Steven (1972) conducted a prospective study of 96 full-term small-for-gestational-age infants and found virtually no major neurological defects (1 percent cerebral palsy, 6 percent seizures) but reported 25 percent diagnosed as having minimal cerebral dysfunction, and one third of the children with IQ scores greater than 100 failing consistently in school. Overall, 50 percent of the boys and 30 percent of the girls had poor school performance, although no analysis for ethnic or socioeconomic status was included. In a large study of prema- tures, Drillien (1964) found a direct relationship between birth weight and psychometric test scores at age 4, with the full-term control group having a

Biological and Social Factors 161 mean IQ of 107 and those with birth weights below 3.5 pounds having a re- ported mean IQ of 80. The literature on the follow-up of asphyxiated newborns also is equivo- cal. An extensive controlled study of several hundred hypoxic newborns followed to school age revealed poor performance on neonatal exams and persistent differences at age 4 on all tests of cognitive function but no sig- nificant IQ differences at 7 years (Corah et al., 1965~. In a review of 20 studies related to perinatal asphyxia, Gottfried (1973) confirmed the im- pression that intellectual deficits were more prominent at younger pre- school ages but noted that early hypoxia may increase the probability of oc- currence of mental retardation in later childhood. Broman (1979) reported that the probability of retardation in asphyxiated groups was increased as much as twelvefold in infancy and sixfold at age 7 but demonstrated that the sequelae of retardation were, in fact, still relatively rare. Sameroff and Chandler (1975) reviewed a considerable body of litera- ture and concluded that socioeconomic and familial factors markedly over- shadowed the effects of perinatal difficulties with respect to long-term de- velopmental outcome. The painstaking longitudinal data collected on the children of Kauai over a 10-year period by Werner et al. (1968) provided one of the most dramatic documentations of the compensatory powers of well-organized families with adequate resources for nurturance. These findings were confirmed by analysis of data on the offspring of over 30,000 pregnancies followed through age 8 in the Collaborative Perinatal Project. According to Broman (1981), birth weight explained only 5-6 percent of the total variance in 8-month Bayley scores and less than 1 percent of the variance in Stanford-Binet IQ scores at age 4. Ethnic identification and maternal education were the best predictors, accounting for 16 percent and 6 percent, respectively, of the variance in IQ at age 4. These data sug- gest that low birth weight by itself is not a major risk factor for cognitive deficit. A related analysis revealed that 10 clinical signs of perinatal as- phyxia explained less than 1 percent of the variance in IQ scores (WISC-R) at age 7 of both white and black children. A more extensive set of perinatal and demographic predictors, however, accounted for 25 percent of the var- iance in IQ scores among whites and 13 percent among blacks. The best predictor was a composite reflecting socioeconomic status, maternal edu- cational level and performance IQ score, head circumference at birth, and, among whites, a clinical diagnosis of brain abnormality in the neona- tal period (Broman, 19811. In this context it is useful to examine the demographic distribution of low birth weight babies. During the 25 years from 1950 to 1976, the pro- portion of low birth weight newborns was consistently higher among non

162 SHONKOFF whites and the difference increased with time. At present, black babies have double the chance of weighing 2,500 grams or less at birth (Select Panel for the Promotion of Child Health, 1981~. According to the National Center for Health Statistics (1980), the overall rate of low birth weight ba- bies in 1976 was 6.1 percent for whites, 13.0 percent for blacks, and 6.9 percent for infants of other races. For babies who are small for their gesta- tional age, the rates are 6.3 percent for blacks and 2.8 percent for whites. It is clear from these data that low birth weight, with its associated in- creased risk of central nervous system insult, is considerably more preva- lent among ethnic minority groups, particularly blacks. Moreover, the likelihood of a poor developmental outcome in a low birth weight or as- phyxiated newborn is significantly increased for children in the lower so- cioeconomic classes. Further evidence suggests that although sex differ- ences in IQ may not be significant, males may have a considerably higher incidence of subtle neurological and perceptual disabilities related to fac- tors of perinatal stress that contribute to learning and behavioral profiles resulting in their disproportionate placement in special educational pro- grams. Further study may provide more elucidation of these speculations. POSTNATAL BIOSOCIAL INFLUENCES Each child begins life with a set of biological strengths and weaknesses, among which is the relative integrity of the brain. Sociocultural and famil- ial factors subsequently play a major role in shaping the ultimate develop- ment of potential abilities and skills. In the nature-nurture debate regard- ing the origins of mental retardation, however, not all sources of organic morbidity are immutably determined by the end of the neonatal period. On the contrary, a variety of biological influences can affect brain function throughout childhood and thereby contribute to the manifestations of mental retardation in its mild or severe forms. Some of the origins of brain insult are obvious severe head trauma or an infectious process such as meningitis or encephalitis. Others are more controversial, especially in their mild to moderate forms. The issues of malnutrition and lead intoxica- tion are reviewed below in some depth as representative of that type of bio- logical influence whose dose-related effects are unclear and whose sociopo- litical overtones are substantial. Malnutrition An extensive body of literature has documented an association between malnutrition in infancy and subsequent intellectual status, especially in developing countries (Cravioto et al., 1966; Hertzig et al., 1972; Stoch and

Biological and Social Factors 163 Smythe, 1963~. In circumstances characterized by severe malnutrition during prenatal life and early childhood, the sequelae of mental retarda- tion and behavioral disorders have generally been substantial and nonre- versible. The effects of moderate or chronic low-grade malnutrition, how- ever, are less well understood. Before considering the available data, we must review current knowledge on the relationship between nutrition and brain growth. The results of extensive animal studies, and to a lesser extent human in- vestigations, have supported the notion that there exists a critical period of "growth spurt" in the immature brain, during which it is most vulnerable with regard to inadequate nutrition. Studies in rats have clearly shown that comparatively mild nutritional restrictions during this sensitive period re- sult in permanent changes in the adult brain associated with behavioral deficits that cannot be reversed by a better diet later. Significant undernu- trition before or after the growth spurt period, however, produced no de- tectable effect that could not be "cured" by dietary rehabilitation (Dobb- ing and Sands, 19711. In humans the sensitive period of rapid brain growth appears to include two important phases: the first extends from mid-preg- nancy until the end of the second year of life and is characterized by an early increase in neuronal and later in glial cell number; the second phase ex- tends well into the third and fourth years and is characterized by rapid my- elination in association with the continuous elaboration of increasingly complex dendritic branching and synaptic connections (Dobbing, 1974~. Thus, current evidence clearly suggests that the period of maximal vul- nerability for brain growth in humans is much more postnatal than pre- viously assumed. This by no means minimizes the critical impact of maternal and therefore fetal nutrition on prenatal brain growth, but it does support the notion that the consequences of intrauterine malnutrition may be re- versible. Animal studies have demonstrated that growth retardation dur- ing only the first part of the sensitive period may not be sufficient to pro- duce permanent deficits (Winick et al., 19681. If the same applies to humans, then adverse sequelae would be substantial only if malnutrition extended from mid-pregnancy through at least the first two years of life or if it were particularly severe over a portion of that period. In any event, Winick (1969) noted that "although the exact timing has yet to be worked out, it would appear that after infancy the brain is much more resistant to the effects of malnutrition" (p. 6771. The available data on the sequelae of significant malnutrition show high rates of intellectual impairment. In a review of seven studies, Chase (1973) reported significant deficits in test performance by malnourished children ages 2-14 in all but one report. Other investigators have noted greater defi- cits in behavioral phenomena such as attentiveness, curiosity, activity, and

164 SHONKOFF social responsiveness, while some have suggested that malnourished child- ren may be particularly susceptible to the stresses and deprivations so fre- quently found in an environment of poverty (Read, 1975~. This latter spec- ulation was reinforced by data collected by Richardson (1976) in Jamaica that showed the consequences of severe malnutrition in infancy for later in- tellectual functioning to be correlated with social background and subse- quent physical growth, rather than with malnutrition itself. Methodological problems in human studies of the effects of malnutri- tion on intellectual development have been monumental. The most obvious relates to the almost universal association of poor nutrition with poverty and its constellation of associated factors that have their own additional negative influence on intellectual development. Cravioto et al. (1966) stated that malnutrition is never remote from impoverishment, and even studies in developed countries have observed a high degree of deprivation (Chase and Martin, 19701. Birch et al. (1971) attempted to control these variables by comparing school-age children with histories of hospitaliza- tions in infancy for kwashiorkor (severe malnutrition) to their siblings who had never experienced malnutrition requiring a hospital admission. Mean IQ (WISC) differences were found to be statistically significant, favoring the nonmalnourished sibling. A few reports have been published of studies involving malnutrition without the complicating factor of socioeconomic deprivation. Lloyd-Still et al. (1974) studied 41 middle-class subjects, ages 2-21, who were substan- tially malnourished in infancy secondary to cystic fibrosis or congenital de- fects of the gastrointestinal tract. Significant differences in scores on the Merrill-Palmer scales were found up to age 5, but no differences were ob- served on the Wechsler scales administered to the older subjects. Klein et al. (1975) reported follow-up data on 50 children, ages 5-14, who had brief periods of starvation in early infancy secondary to pyloric stenosis. Com- parison to siblings and matched controls revealed no significant differ- ences in global measures of intelligence, but significantly lower scores for the index children were found on subtests related to short-term memory and attention. Further studies of malnourished but socially nondisadvan- taged youngsters are clearly needed; nevertheless, the Food and Nutrition Board of the National Research Council (1973) stated that "in spite of many serious methodological shortcomings in the studies that have been made, the weight of evidence seems to indicate that early and severe mal- nutrition is an important factor in later intellectual development, above and beyond the effects of socio-familial influences." The extent to which the kinds of nutritional deficiencies commonly found in the United States may be sufficient to affect intellectual develop- ment in children was considered by Livingston et al. (1975) in their review

Biological and Social Factors 165 of data on pregnant women, infants, and children under age 4 from the Ten State Nutrition Survey (U.S. Department of Health, Education, and Welfare, 1972) and children ages 1-4 from the study of the nutritional sta- tus of preschool children in the United States during 1968-1970 (Owen et al., 1974~. They found that nearly 60 percent of all pregnant women liv- ing below the poverty level were apparently consuming calories at a rate low enough to adversely affect fetal brain development. At two to three times the poverty level, the proportion of vulnerable women was still 44 percent. Moreover, 14 percent of the pregnant women living below the poverty level were in jeopardy for both calories and protein. Of children ages 1-4 living in poverty, 18 percent were in jeopardy for defective brain development ac- cording to the data of the Ten State Nutrition Survey, while the preschool nutrition survey reported a higher frequency of 24 percent at risk. The Se- lect Panel for the Promotion of Child Health (1981) states that approxi- mately one third of all black children in the United States are estimated to suffer some kind of nutritional defect compared with less than 15 percent of white children. No data for Hispanic or other ethnic minorities were re- ported. It was noted that poverty and race are associated with deficiencies in six of eight specific nutrients. Even if these data are all high estimates of the extent of the nutritional problems of poor and ethnic minority children in this country, the potential contribution of this factor to the prevalence of mild retardation could be substantial. Despite some of the inconsistencies and methodological dilemmas noted above, the relationship between prenatal malnutrition and severe or pro- longed early childhood malnutrition and the increased risk of subsequent intellectual impairment is well accepted. Its association with poverty will always be inevitable and never easily separable. As noted by Winick (1969), it "is a self perpetuating problem, a vicious cycle which begins in infancy, condemns a person to a lifetime of perhaps marginal function, making it that much more difficult for him to extricate himself from the existing conditions and to create for his family an environment which will protect his children from the same 'disease' " (p. 677~. Lead Intoxication The increased prevalence of lead intoxication in childhood among ethnic minorities and the poor as well as its association with neurological damage are well known (Byers and Lord, 1943; Lin-Fu, 1972; Perlstein and Attala, 19661. In its most severe form it is characterized by an acute encepha- lopathic process, the long-term sequelae of which frequently include mod- erate to severe mental retardation. Multiple sources of lead in the environment have been identified; the

166 SHONKOFF most prominent include paint chips, household dust and dirt, newsprint, and contaminated air near smelting plants or in congested urban areas with high concentrations of automobile traffic. The urban poor have been consistently identified as the group at highest risk for excessive lead expo- sure. Prevalence rates in low-income, inner-city areas range from 4 to 40 percent positive in community screening programs (Lin-Fu, 19721. More- over, among comparable socioeconomic groups in the population, lead ab- sorption has been reported to be greater for black than for white children (Lin-Fu, 1979~. Although there are no disagreements about the definition or nature of significant lead poisoning, there has been considerable debate about the effects of an increased body burden of lead at low levels. Over the past 15 years, the upper limit of safety for blood levels of lead has gradually been lowered from 60 micrograms per 100 milliliters to a current level of 30 mi- crograms per 100 milliliters. As more attention has been focused on these asymptomatic, subclinical cases of increased lead burden, a great deal of controversy has been generated regarding its consequences. Many studies have suggested that moderate levels of lead intoxication often lead to sig- nificant attentional difficulties, with associated specific learning problems and behavioral disorders (Needleman et al., 1979; Pueschel, 19741. The in- fluences of dysfunctional behavioral sequelae on adaptive performance in the school setting require more careful examination. Whether mental retardation is involved is a far more complex matter. While some investigators have reported an inverse relationship between blood lead determinations and intellectual development (Beattie et al., 1975; de la Burde and Choate, 1975), others have found no consistent rela- tionship between a low level of lead and developmental status (Kotok, 1972~. Two recently published studies addressed the multiple methodolog- ical problems characterizing the literature and have attempted to critically analyze the subtle consequences of a low-level, chronic burden of lead. Needleman et al. (1979) administered an extensive battery of neuropsycho- logical measures to 58 children with high dentine lead levels and 100 with low levels. Although the mean full-scale IQ scores (WISC-R) for each group were normal (102.1 versus 106.6, respectively), the difference be- tween the two groups was statistically significant (p < .03~. No sex differ- ences were noted. Ernhart et al. (1981) studied 30 children comprising a "moderate lead group" and 50 comprising a "low lead group" five years after their initial identification in a summer screening clinic. All of the children were of low socioeconomic status, and the group was approxi- mately half boys and half girls. Preliminary analysis of the scores on sev- eral subtests of the McCarthy Scales of Children's Abilities revealed signif- icant impairment of the high lead group. When parental IQ was included

Biological and Social Factors 167 in the analysis, however, the variance in the children's IQ scores associated with lead substantially decreased. The authors concluded from this work that if there are in fact behavioral and intellectual sequelae of low levels of lead intoxication independent of parental and social influences on develop- ment, they must be minimal. Conclusions regarding this topic are extremely tentative. The ability of lead to damage the brain is well established. The upper limit of safety and the effects of low blood levels on intellectual abilities are being explored. The rate of increased lead exposure is highest among ethnic minority groups and the urban poor. In their follow-up study of 67 children age 7 who had asymptomatic lead exposure between ages 1 and 3, de la Burde and Choate (1975) found that the mean full-scale IQ of the index children was lower than that obtained for a control group (p < .01), and the former included a larger number of children in the borderline and mildly retarded range. Similarly, 67 percent of the control children received "normal scores" on all parts of an extensive test battery, while only 43.3 percent of the lead-exposed group had no failures in the entire series (p < .011. Thus, although the answers are not yet in, one cannot completely disregard the potential contribution of low levels of lead to the preponderance of mild re- tardation among ethnic minorities and poor children. Family Resources, Child-Rearing Practices, and Individual Learning Styles Contemporary conceptualizations of the process of human development place great emphasis on the transaction between biological predispositions and environmental contingencies as mutual determinants of developmen- tal outcome. In this context a consideration of the social characteristics of early childhood and their possible contribution to the increased prevalence of mild retardation among ethnic minority and poor children is most important. The specific characteristics that emanate from a "disadvantaged" socio- cultural milieu can take many forms. Characteristics of a child's home en- vironment, for example, and their relationship to the facilitation of op- timal development have been shown to correlate with performance on standardized tests in the preschool years (Bradley and Caldwell, 1976; Elardo et al., 1975~. In this context a great deal of data have been gener- ated on the nature of the mother-child relationship and its influence on de- velopmental competence. Ramey et al. (1979) reported that within an ap- parently homogeneous social class group, 50-65 percent of the variance in Stanford-Binet scores at age 3 could be accounted for by differences in the mothers' attitudes, behavior, and at-home interactions with their children

168 SHONKOFF during the toddler years. The authors cautioned, however, that these were correlations and were not presumed to reflect direct causal relationships. White (1975) found that "high competence" children generally engaged in more frequent interaction with their mothers, spent more time in "highly intellectual" activities, were the recipients of more maternal '`teaching" and conversation, and received more encouragement than a "low compe- tence" comparison group. Wilton and Barbour (1978) reported similar dif- ferences in the amount of didactic teaching and the frequency of encour- agement of children's activities found in a comparison of high-risk (children with siblings diagnosed as having "cultural-familial retar- dation") versus control lower socioeconomic class children. Zajonc and Markus (1975) analyzed the impact of family size and spacing between births as two of many factors that might influence the degree of intellectual stimulation provided within a family. They reported a large body of data demonstrating an inverse relationship between intellectual performance and the number of children in a family. More marked adverse effects were found for younger siblings with brief intervals between births. Although these findings were noted to be independent of social class variables, Za- jonc (1976) cited 1960 population data reporting larger families with shorter spacing between births among black American families compared with white American families. A more recent analysis of the influence of family configuration on Scholastic Aptitude Test scores, however, suggests that the amount of variance attributable to these factors is negligible (Zajonc and Bargh, 19801. Rutter (1979), in a review of the heterogeneity of so-called maternal deprivation and its consequences for developmental outcome, summarized these concepts by noting that "insofar as depriva- tion is ~ on'~1 factor .~., ,~ ~ `~ ... . . intellectual retardation is a function of a lack of adequate meaningful experiences" (p. 2981. The fact that the overwhelm- ing majority of children from ethnic minority and poor families are not in- tellectually impaired reflects the essential resilience of most children and, more importantly, underlines the fact that we do not have good evidence that dysfunctional family relationships are typical of these groups. It is im- portant to recognize, however, that the stresses of poverty and racism exact a severe price from many families, and the consequences for the emerging competence of young children are often formidable. The concept of motivation and its relationship to developing self-esteem is another factor whose salience requires thoughtful consideration. In im- poverished homes, where the expectations for future success may be blunted, the motivation of a constitutionally competent child to comply with the demands of an achievement-oriented learning situation may be considerably diminished. Zigler and Trickett (1978), emphasizing the po- tential central importance of this issue, charge that a considerable body of

Biological and Social Factors 169 empirical evidence suggests that IQ changes resulting from preschool in- tervention programs reflect motivational changes that affect children's test performances rather than actual differences in their cognitive functioning. The influence of sex differences on child-rearing practices and differ- ences in learning styles is extremely complex. In a study of mother-child in- teraction in lower socioeconomic class preschoolers, Wilton and Barbour (1978) note that many of the dysfunctional encounters found were more pronounced for mothers and their daughters than for mothers and their sons. Although they could not explain the differences on the basis of their data, these investigators suggest that these differences might be related to cultural sex stereotypes reflecting greater efforts to promote the intellec- tual development of boys than girls. An alternative explanation suggests that the behavior characteristics of high-risk boys typically demand more maternal response. Similarly, the greater prevalence of aggressive and po- tentially disruptive behaviors in boys compared with girls may be a major determinant of more frequent referral for special class placement. A com- prehensive analysis of sex differences in constitutional behavioral styles, child-rearing strategies and expectations, and intrafamilial relationships is beyond the scope of this paper. Its contribution to the disproportionate prevalence of school failure among boys, however, may very well be crucial. SUMMARY AND CONCLUSIONS Although its genesis may dig deeply into biological as well as social roots, the phenomenon of mild mental retardation is primarily a cultural con- struct. Its very nature has changed dramatically over time, and its contem- porary definitions are highly influenced by differences among societies. Within the United States in the past 100 years, arbitrary shifts in diagnos- tic criteria have moved children in and out of the mildly retarded popula- tion. Moreover, as society becomes increasingly complex in its technologi- cal demands, new classifications of "defectiveness" will undoubtedly arise. The charge that mild mental retardation is a creation of our system of universal, compulsory education is strongly supported by the consistent age distribution found across virtually all epidemiological studies. It is rarely diagnosed in the preschool years, begins to increase in incidence in the early elementary grades, reaches a peak in the junior high school popu- lation, and then progressively declines in frequency to a relatively low level that persists throughout adult life. The explanation for this inverted U shaped pattern of prevalence can be attributed to the extensive subjection of school-age children to formal testing and the relative inflexibility of school systems regarding the range of abilities and performance that they

170 SHONKOFF will accept without assigning a stigmatizing label. Thus, most of the chil- dren who are classified as mildly retarded during their school careers sub- sequently lose their labels and "disappear" into the general population as independently functioning citizens. Although they may be distinguished by the relatively low level of intellectual demand placed on them by their work and their recreation, they clearly assume the status of "normal" adults. Their early classification of defectiveness reflects a designation that primarily serves the administrative needs of the education system whose achievement criteria are set by the values and needs of the society. Despite the serious methodological problems inherent in epidemiologi- cal studies of mild retardation, the consistency of the disproportionate di- agnosis of children from ethnic minorities and lower socioeconomic groups has been impressive. Because of their high degree of overlap, it has often been difficult to tease out the relative contributions of ethnicity and pov- erty. Whatever the numbers may be, it is clear that children from socially and politically disadvantaged families are more likely to be labeled as mildly retarded in the United States than are white, middle-class children. The data on sex differences are less straightforward. Boys are signifi- cantly more likely than girls to be classified as experiencing school failure, but the bulk of the literature on the prevalence of mental retardation dem- onstrates only a slightly greater proportion of boys among the severely re- tarded population and virtually no significant sex differences among the mildly impaired. This suggests that differences in educational classifica- tion are related to issues that go beyond general intellectual abilities. The list of variables that might explain the observed predominance of boys in special education placements includes differences in the prevalence of specific learning disabilities, attentional deficits, dysfunctional learning styles, and a wide variety of disruptive behavioral disorders as well as cul- tural differences regarding demands and expectations placed on boys and girls. Despite the well-documented greater biological vulnerability of boys, sex differences in the frequency of mild mental retardation have not been consistently found. Ethnic and socioeconomic differences in the prevalence of mild retarda- tion clearly demand a critical analysis. The sociopolitical arguments on this issue are highly persuasive. The debates over the cultural biases of psy- chometric tests, for example, have been well publicized, and their role in discriminatory diagnostic practices has been repeatedly charged. More- over, marked inequities obviously exist in the availability of resources to fa- cilitate optimal intellectual development within those population groups that have been subjected over many generations to the consequences of in- stitutionalized discrimination. For these and related reasons, equalization

Biological and Social Factors 171 of the distribution of mild retardation within the entire population is largely a political task. A word must be said, however, about the distinction between intellec- tual impairment and cognitive differences that contribute to competitive disadvantage within- a specific sociocultural system. IQ tests, with all of their problems, have been shown to do reasonably well in their originally intended function of identifying children who have an increased probabil- ity of failing in school. Thus, although changes in the criteria for making a diagnosis of mild retardation will liberate many children from the stigma of such classification, their performance in a traditional school curriculum is still likely to correspond to their scores on a "standardized" psychomet- ric test. For many children whose life experiences differ from a typical mid- dle-class upbringing, however, discrepancies in test scores and school per- formance may very well reflect a different kind of cognitive ability that does not necessarily imply intelligence. The tyranny of the dominant cul- ture and its power over the standards of educational success will probably continue to undervalue such differences. There is, however, another aspect of this problem, which has its roots in the cultural sphere but extends far into the area of biology-the issue of central nervous system function and brain integrity. Poor and minority children are not the victims of social discrimination alone. A considerable body of data suggests that they also carry a disproportionate burden of bio- logical vulnerability that is largely related to the increased health risks of poverty. Much of the discussion of biological disadvantage among ethnic minorities and lower socioeconomic groups has traditionally focused on the issue of genetic differences in intellectual endowment. Biological dif- ferences in individuals, however, are determined by a great deal more than inherited traits. The developing brain, regardless of its genetic potential, is subjected to a variety of potentially damaging influences throughout its prenatal, perinatal, and postnatal life, which can have adverse effects on its ultimate functioning. Intrauterine factors such as cytomegalovirus and alcohol, complications during the newborn period related to prematurity and/or low birth weight, and early childhood insults such as malnutrition and lead intoxication can all inflict damage on an immature brain result- ing in significant impairment in later intellectual functioning. These threat- ening influences and many others exist with greater frequency among poor and minority populations. The unequal distribution of these risk factors is certainly influenced by social and economic forces, but their existence creates very real, intrinsic biological vulnerabilities in the children who are so afflicted. The ultimate roles of biological and social factors in the etiology of mild

172 SHONKOFF retardation can best be understood in the context of a transactional model of development applied to a basically resilient human organism. The over- whelming majority of poor and minority children are not retarded. Most low birth weight babies do well developmentally. Of those children who were exposed to noxious agents during their prenatal or postnatal life, some will have impairment of their intelligence and others will appear to escape unharmed. Many of those whose brains have been injured will not demonstrate abnormalities on traditional neurological examinations. Ulti- mate developmental outcome for all children appears to be a function of a highly complex series of transactions among a great number of biological and environmental facilitators and constraints. Intelligence is determined by multiple factors, and its impairment rarely has a simple etiology. Some children are extraordinarily resilient and may have well-developed intellectual abilities despite minimal environmental supports. Others are constitutionally limited and will have significant defi- cits in the face of optimally facilitating experiences. Each child's abilities are dependent on the interplay between his or her biological equipment and life circumstances. Few individuals are without vulnerabilities most manage to adapt reasonably well. The distribution of vulnerabilities within the general population, however, is grossly unequal. Poor and minority children bear a proportionately greater burden of them in both a biological and a social sense. In conclusion, it is clear that mild mental retardation is largely a cul- tural invention and not an objective biological property. It reflects a so- ciety's expectations regarding intellectual performance and is subject to modification as values change. Children whose rearing and environmental resources differ from those of the dominant cultural group are at greater risk for having profiles of abilities that may very well be dysfunctional for the demands of the public school system. One must not underestimate, however, the fact that these same "disadvantaged" groups are victimized by a greater frequency of harmful biological factors that can adversely af- fect brain development in early life and later lead to very real intellectual deficits. Poor and minority children have the highest probability of sus- taining injuries through both nature and nurture. Attempts to assign quantitative weighting to the relative contributions of each are thwarted by the limitations of available data. The synergistic effects of cumulative vul- nerabilities in both spheres undoubtedly contribute to the greater preva- lence of mild retardation in these groups. Thus, the ultimate resolution of these inequities will have to go beyond the very important social battles over evaluation and classification procedures, extending into the realm of maternal and child health.

Biological and Social Factors REFERENCES 173 Alford, C. 1977 Prenatal infections and psychosocial development in children born into lower socio- economic settings. Pp. 251-259 in P. Mittler, ea., Research to Practice in Mental Retardation. Vol. III. Baltimore, Md.: University Park Press. 1981 Perinatal infections today and tomorrow. Pp. 81-93 in N. Kretchmer and J. Brasel, eds., Biomedical and Social Bases of Pediatrics. New York: Masson Publishing Co. American Academy of Pediatrics 1979 Statement on teenage pregnancy. Committee on Adolescence. Pediatrics 63:795- 797. Baldwin, W. 1976 Adolescent pregnancy and childbearing-growing concerns for Americans. Popu- lation Bulletin 31:1-34; revised reprint May 1977. Beattie, A., Moore, M., Goldberg, A., Finlayson, M., Mackie, E., Graham, J., Main; J., McLaren, D., Murdoch, R., and Stewart, G. 1975 Role of chronic low-level lead exposure in the etiology of mental retardation. Lan- cet 1:589-592. Berg, J. 1963 Proceedings of the Second International Congress of Mental Retardation-Part I. Basel, Switzerland: S. Karger AG. Birch, H., and Gussow, J. 1970 Disadvantaged Children: Health, Nutrition and School Failure. New York: Grune & Stratton. Birch, H., Pineiro, C., Alcalde, E., Toca, T., and Cravioto, J. 1971 Relation of kwashiorkor in early childhood and intelligence at school age. Pediatric Research 5:579-585. Birch, H., Richardson, S., Baird, D., Borobin, G., and Illsley, R. 1970 Mental Subnormality in the Community: A Clinical and Epidemiologic Study. Bal- timore, Md.: Williams & Wilkins. Blanton, R. 1975 Historical perspectives on classification of mental retardation. Pp. 164-193 in N. Hobbes, ea., Issues in the Classification of Children. Vol. I. San Francisco, Calif.: Jossey-Bass., Inc., Publishers. Bradley, R., and Caldwell, B. 1976 Early home environment and changes in mental test performance in children from 6 to 36 months. Developmental Psychology 12:93-97. Bremer, J. 1951 A social psychiatric investigation of a small community in northern Norway. Acta Psychiatrica Scandinavica, Supplement 62. Broman, S. 1979 Perinatal anoxia and cognitive development in early childhood. Pp. 29-52 in T. Field, ea., The High Risk Newborn. Jamaica, N.Y.: Spectrum Publications. 1980 Long-term development of children born to teenagers. Pp. 195-244 in K. Scott, R. Field, and E. Robertson, eds.,~Teenage Parents and Their Offspring. New York: Grune & Stratton. 1981 Risk factors for deficits in early cognitive development. Pp. 131-137 in G. Berg and H. Maillie, eds., Measurement of Risks. New York: Plenum Press.

174 SHONKOFF Byers, R., and Lord, E. 1943 Late effects of lead poisoning on mental development. American Journal of Dis- eases of Children 66:471-494. Caputo, D., and Mandell, W. 1970 Consequences of low birth weight. Developmental Psychology 3:363-383. Carpenter, R., and Petersdorf, R. 1962 The clinical spectrum of bacterial meningitis. American Journal of Medicine 33: 262-275. Chase, H. 1973 The effects of intrauterine and post-natal undernutrition on normal brain develop- ment. Annals of the New York Academy of Science 205:231-244. Chase, H., and Martin, H. 1970 Undernutrition and child development. New England Journal of Medicine 282: 933-939. Childs, B. 1965 Genetic origin of some sex differences among human beings. Pediatrics 35: 798-81 2. Clarren, S., and Smith, D. 1978 The fetal alcohol syndrome. New England Journal of Medicine 298:1063-1067. Cleland, C. 1979 Mislabeling and replication: methodological caveats. American Journal of Mental Deficiency 83:648-649. Corah, N., Anthony, E., Painter, P., Stern, J., and Thurston, D. 1965 Effects of pert-natal anoxia after seven years. Psychological Monographs 79(3) (whole no. S96): 1-34. Cravioto, J., DeLicardie, E., and Birch, H. 1966 Nutrition, growth and neurointegrative development: an experimental and ecologic study. Pediatrics 38(Suppl.):319-372. de la Burde, B., and Choate, M. 1975 Early asymptomatic lead exposure and development at school age. Journal of Pedi- atrics 87:631-642. Dobbing, J. 1974 The later growth of the brain and its vulnerability. Pediatrics 53:2-6. Dobbing, J., and Sands, J. 1971 Vulnerability of developing brain: IX. The effect of nutritional growth retardation on the timing of the brain growth-spurt. Biologia Neonatorum 19:363. Drillien, C. 1964 The Growth and Development of the Prematurely Born Infant. Baltimore, Md.: Williams & Wilkins. Dunn, L. M. 1968 Special education for the mildly retarded-is much of it justifiable? Exceptional Children 35:5-22. Eisenberg, L., and Earls, F. 1975 Poverty, social depreciation, and child development. Pp. 275-291 in D. Hamburg and H. Brodie, eds., American Handbook of Psychiatry. Vol. 6-New Psychiatric Frontiers. New York: Basic Books. Elardo, R., Bradley, R., and Caldwell, B. 1975 The relation of infants' home environments to mental test performance from six to thirty-six months. Child Development 46: 71-76.

Biological and Social Factors Ernhart, C., Landa, B., and Schell, N. 175 1981 Subclinical levels of lead and developmental deficit-a multivariate follow-up reas- sessment. Pediatrics 67:911-919. Farber, B. 1968 Mental Retardation: Its Social Context and Social Consequences. Boston, Mass.: Houghton-Mifflin. Firkowska, A., Ostrowska, N., Sokolowska, M., Stein, Z., Susser, M., and Wald, I. 1978 Cognitive development and social policy: the contribution of parental occupation and education to mental performance in 11-year-olds in Warsaw. Science 202: 1357-1362. Fitzhardinge, P., and Steven, E. 1972 The small-for-date infant. II. Neurological and intellectual sequelae. Pediatrics 50:50-57. Flor-Henry, P. 1974 Psychosis, neurosis and epilepsy. British Journal of Psychiatry 124:144-150. Gerald, P. 1980 X-linked mental retardation and an x-chromosome marker. New England Journal of Medicine 303:696-697. Goddard, H. 1910 What can the public school do for subnormal children? The Training School 7(5):242-248. Goodman, M., and Tizard, J. 1962 Prevalence of imbecility and idiocy among children. British Medical Journal 1: 216-219. Goodman, M., Gruenberg, E., Downing, J., and Rogot, E. 1956 A prevalence study of mental retardation in a metropolitan area. American Journal of Public Health 46:702-707. Gottfried, A. 1973 Intellectual consequences of perinatal anoxia. Psychological Bulletin 80:231-242. Grant, J., and Heald, F. 1972 Complications of adolescent pregnancy. Clinical Pediatrics 11:567-570. Grossman, H. 1973 Manual on Terminology and Classification in Mental Retardation, 1973 Revision. Special Publication Series No. 2. Washington, D.C.: American Association on Mental Deficiency. Gruenberg, E. 1964 Epidemiology. Pp. 259-306 in H. Stevens and R. Heber, eds., Mental Retarda- tion-A Review of Research. Chicago, III.: University of Chicago Press. Grunewald, K. 1979 Mentally retarded children and young people in Sweden. Integration into society: the progress in the last decade. Acta Paediatrica Scandinavica Supplement 275:75-84. Hanshaw, J. 1981 Cytomegalovirus infections. Pediatrics in Review 2(8):245-251. Hanshaw, J., Schultz, F., Melish, M., and Dudgeon, J. 1973 Congenital cytomegalovirus infection. Ciba Foundation Symposium 10:23. Am- sterdam: Associated Scientific Publishers. Hanshaw, J., Scheiner, A., Moxley, A., Gaev, L., Abel, V., and Scheiner, B. 1976 School failure and deafness after "silent" congenital cytomegalovirus infection. New England Journal of Medicine 295:468-470.

176 SH(:)NKOFF Hanson, J., Streissguth, A., and Smith, D. 1978 The effects of moderate alcohol consumption during pregnancy on fetal growth and morphogenesis. Journal of Pediatrics 92:457-460. Heber, R. 1959 A manual on terminology and classification in mental retardation. American Jour- nal of Mental Deficiency, Monograph Supplement 64(2). Hertzig, M., Birch, H., Richardson, S., and Tizard, J. 1972 Intellectual levels of school children severely malnourished during the first two years of life. Pediatrics 49:814-824. Hutt, C. 1978 Biological bases of psychological sex differences. American Journal of Diseases of Children 132:170-177. Jastak, J., MacPhee, H.j and Whiteman, M. 1963 Mental Retardation: Its Nature and Incidence. Newark, N.J.: University of Dela- ware Press. Kiely, J., and Paneth, N. 1981 Follow-up studies of low-birth weight infants: suggestions for design, analysis and reporting. Developmental Medicine and Child Neurology 23:96-100. Kirk, S., and Weiner, B. 1959 The Onondaga census-fact or artifact. Exceptional Children 25:226-231. Kirman, B., and Bicknell, J. 1975 Mental Handicap. Edinborough, Scotland: Churchill Livingstone. Klein, P., Forbes, G., and Nader, P. 1975 Effects of starvation in infancy (pyloric stenosis) on subsequent learning abilities. Journal of Pediatrics 87:8-15. Kotok, D. 1972 Development of children with elevated blood lead levels: a controlled study. Journal of Pediatrics 80:57-61. Kumar, M., Nankervis, G., and Gold, E. 1973 Inapparent congenital cytomegalovirus infection, a follow-up study. New England Journal of Medicine 288: 1370- 1372. Lake, D., and Bryden, M. 1976 Handedness and sex differences in hemispheric asymmetry. Brain and Language 3:266-282. Lapouse, R., and Weitzner, M. 1970 Epidemiology. Pp. 197-223 in J. Wortis, ea., Mental Retardation-An Annual Review I. New York: Grune & Stratton. I. Lehman, 1959 Rural-urban differences in intelligence. Journal of Educational Research 53:62-68. Lemkau, P., Tietze, C., and Cooper, M. 1941 Mental hygiene problems in an urban district. Mental Hygiene 25:624-646. 1942 Mental hygiene problems in an urban district. Mental Hygiene 26: 100-l 19, 275-288. Leonard, M., Landy, G., Ruddle, F., and Lubs, H. 1974 Early development of children with abnormalities of the sex chromosomes: a pro- spective study. Pediatrics 54:208-212. Lilienfeld, A., and Parkhurst, E. 1951 A study of the association of factors of pregnancy and parturition with the develop- ment of cerebral palsy: a preliminary report. American Journal of Hygiene 53: 262-282.

Biological and Social Factors 177 Lin-Fu, J. 1972 Undue absorption of lead among children-a new look at an old problem. New England Journal of Medicine 286: 702-710. 1979 Lead exposure among children-a reassessment. New England Journal of Medicine 300:731-732. Little, W. 1862 On the influence of abnormal parturition, difficult labor, premature birth, and asphyxia neonatorum on the mental and physical condition of the child, especially in relation to deformities. Transactions of the Obstetrical Society ok London 3:293- 344. Livingston, R., Calloway, D., MacGregor, J., Fisher, G., and Hastings, A. 1975 U. S. poverty impact on brain development. Pp. 377-394 in M. Brazier, ea., Growth and Development of the Brain. New York: Raven Press. Lloyd-Still, J., Hurwitz, I., Wolff, P., and Shwachman, H. 1974 Intellectual development after severe malnutrition in infancy. Pediatrics 54:306- 312. MacMillan, D., Meyers, C., and Morrison, G. 1980 System-identification of mildly mentally retarded children: implications for inter- preting and conducting research. American Journal of Mental Deficiency 85:108- 115. McNemar, Q. 1942 The Revision of the Stanford-Binet Scale. Boston, Mass.: Houghton-Mifflin. Melish, M., and Hanshaw, J. 1973 Congenital cytomegalovirus infection: developmental progress of infants detected by routine screening. American Journal of Diseases of Children 126:190. Menolascino, F. 1977 Challenges in Mental Retardation: Progressive Ideology and Services. New York: Human Sciences Press. Mercer, J. 1971 The meaning of mental retardation. Pp. 23-46 in R. Koch and J. Dobson, eds., The Mentally Retarded Child and His Family: A Multidisciplinary Handbook. New York: Brunner/Mazel. 1973 Labeling the Mentally Retarded: Clinical and Social System Perspectives on Men- tal Retardation. Berkeley, Calif.: University of California Press. Moser, H., and Wolff, P. 1971 The nosology of mental retardation: including the report of a survey of 1378 men- tally retarded individuals at the Walter E. Fernald State School. Birth Defects: Origin al Article Series 7( 1 ) :11 7- 1 34. Naeye, R. 1981 Teenaged and pre-teenaged pregnancies: consequences of the fetal-maternal com- petition for nutrients. Pediatrics 67:146-150. Naeye, R., Burt, L., Wright, D., Blanc, W., and Tatter, D. 1971 Neonatal mortality, the male disadvantage. Pediatrics 48:902-906. National Center for Health Statistics 1980 Factors Associated with Low Birth Weight: United States, 1976. S. Taffel. Publica- tion no. 80-1915. Vital and Health Statistics, Series 21, Number 37. Washington, D.C.: U.S. Department of Health and Human Services. National Research Council 1973 Position paper on The Relationship of Nutrition to Brain Development and Behav- ior. Food and Nutrition Board. National Academy of Sciences, Washington, D.C.

178 SHONKOFF Needleman, H., Gunnoe, C., Leviton, A., Reed, R., Peresie, H., Maher, C., and Barrett, P. 1979 Deficits in psychologic and classroom performance of children with elevated den- tine lead levels. New England Journal of Medicine 300:689-695. New York State Department of Mental Hygiene 1955 A special census of suspected referred mental retardation. Data for Onondaga County, New York, in Technical Report of the Mental Health Research Unit. Syracuse, N.Y.: Syracuse University Press. Nicholson, A., and Hanley, C. 1953 Indices of physiological maturity: derivation and interrelationships. Child Develop- ment 24:3-38. Nortman, D. 1974 Parental age as a factor in pregnancy outcome and child development. Reports on Population and Family Planning No. 16. Ouellette, E., Rosett, H., Rosman, N., and Weiner, L. 1977 Adverse effects on offspring of maternal alcohol abuse during pregnancy. New England Journal of Medicine 297:528-530. Owen, G., Kram, K., Carry, P., Lowe, J., and Lubin, A. 1974 A study of nutritional status of preschool children in the United States, 1968-1970. Pediatrics 53:597-646. Parkinson, C., Wallis, S., and Harvey, D. 1981 School achievement and behavior of children who were small-for-dates at birth. Developmental Medicine and Child Neurology 23:41-50. Pasamanick, B., and Knobloch, H. 1961 Epidemiologic studies on the complications of pregnancy and the birth process. Pp. 74-94 in G. Caplan, ea., Prevention of Mental Disorders in Children. New York: Basic Books. Pass, R., Stagno, S., Myers, G., and Alford, C. 1980 Outcome of symptomatic congenital cytomegalovirus infection: results of long-term longitudinal follow-up. Pediatrics 66:758-762. Perlstein, M., and Attala, R. 1966 Neurologic sequelae of plumbism in children. Clinical Pediatrics 5:292-298. Potts, D. 1970 Which is the weaker sex? Journal of Biosocial Science 2(Suppl.): 147-157. President's Panel on Mental Retardation 1962 Report to the President: A Proposed Program for National Action to Combat Men- tal Retardation. Washington, D.C.: U.S. Government Printing Office. Pueschel, S. 1974 Neurological and psychomotor functions in children with an increased lead burden. Environmental Health Perspectives (May): 13- 16. Ramey, C., and Finkelstein, N. 1978 Psychosocial Mental Retardation: A Biological and Social Coalescence. Paper pre- sented at the Conference on Prevention of Retarded Development in Psychosocially Disadvantaged Children, Madison, Wis. Ramey, C., Farran, D., and Campbell, F. 1979 Predicting I.Q. from mother-infant interactions. Child Development 50:804-814. Ramey, C., Stedman, D., Borders-Patterson, A., and Mengel, W. 1978 Predicting school failure from information available at birth. American Journal of Mental Deficiency 82:525-534. Read, M. 1975 Behavioral correlates of malnutrition. Pp. 335-353 in M. Brazier, ea., Growth and Development of the Brain. New York: Raven Press.

Biological and Social Factors 179 Reschly, D., and Jipson, F. 1976 Ethnicity, geographic locale, age, sex and urban-rural residence as variables in the prevalence of mild retardation. American Journal of Mental Deficiency 81: 154-161. Reynolds, D., Stagno, S., Stubbs, G., Dahle, A., Livingston, M., Saxon, S., and Alford, C. 1974 Inapparent congenital cytomegalovirus infection with elevated cord IgM levels- causal relation with auditory and mental deficiency. New England Journal of Medicine 290:291-296. Richardson, S. 1976 The relation of severe malnutrition in infancy to the intelligence of school children with differing life histories. Pediatric Research 10:57-61. Robinson, N. 1978 Mild mental retardation: does it exist in the People's Republic of China? Mental Retardation 16:295-298. Rubin, R., Rosenblatt, C., and Balow, B. 1973 Psychological and educational sequelae of prematurity. Pediatrics 52:352-363. Rutter, M. 1979 Maternal deprivation, 1972-1978: new findings, new concepts, new approaches. Child Development 50:283-305. Rutter, M., Tizard, J., and Whitmore, K. 1970 Education, Health and Behavior. London: Longman. Ryan, W. 1971 Blaming the Victim. New York: Pantheon. Sameroff, A., and Chandler, M. 1975 Reproductive risk and the continuum of care-taking casualty. Pp. 187-244 in F. Horowitz, ea., Review of Child Development Research, Vol. 4. Chicago, Ill.: University of Chicago Press. Sarason, S. B., and Doris, J. 1979 Educational Handicap, Public Policy, and Social History: A Broadened Perspective on Mental Retardation. New York: Free Press. Select Panel for the Promotion of Child Health 1981 Report to the United States Congress and the Secretary of Health and Human Ser- vices on Better Health for Our Children: A National Strategy. Publication no. 79-55071. Washinton, D.C.: U.S. Department of Health and Human Services. Shaywitz, S., Cohen, D., and Shaywitz, B. 1980 Behavior and learning difficulties in children of normal intelligence born to alcoholic mothers. Journal of Pediatrics 96:978-982. Smith, J., and Pollaway, E. 1979 The dimension of adaptive behavior in mental retardation research: an analysis of recent practices. American Journal of Mental Deficiency 84:203-206. Smith, R., Platou, E., and Good, R. 1956 Septicemia of the newborn. Pediatrics 17:549-575. Stein, Z., and Susser, M. 1969 Mild mental subnormality: social and epidemiological studies. Social Psychiatry Research Publication 47:62. Stevenson, A., and McClarin, R. 1957 Determination of the sex of human abortions by nuclear sexing the cells of the chorionic villi. Nature 180: 198- 199. Stoch, M., and Smythe, P. 1963 Does undernutrition during infancy inhibit brain growth and subsequent intellec- tual development? Archives of Disease in Childhood 38:546.

180 SHONKOFF Streissguth, A., Herman, C., and Smith, D. 1978 Intelligence, behavior, and dysmorphogenesis in the fetal alcohol syndrome: a report on 20 patients. Journal of Pediatrics 92:363-367. Tarjan, G., Wright, S., Eyman, R., and Keeran, C. 1973 Natural history of mental retardation: some aspects of epidemiology. American Journal of Mental Deficiency 77:369-379. Taylor, R. 1980 Use of the AAMD classification system: a review of recent research. American Journal of Mental Deficiency 85:116-119. Tennes, M., Puck, M. Orfanakis, D., and Robinson, A. 1977 The early childhood development of 17 boys with sex chromosome anomalies: a prospective study. Pediatrics 59:574-583. Terman, L. 1916 The Measurement of I'2telligence. Boston, Mass.: Houghton-Mifflin. Turner, G., Brookwell, R., Daniel, A., Selikowitz, M., and Zilibowitz, M. 1980 Heterozygous expression of x-linked mental retardation and x-chromosome marker Fra(x) (927). New England Journal of Medicine 303:662-664. U.S. Department of Health, Education, and Welfare 1972 Ten State Nutrition Survey. Publication no. 72-81334. Washington, D.C.: U.S. Department of Health, Education, and Welfare. Vaughan, V., McKay, R., and Behrman, R., eds. 1979 Nelson s Textbook of Pediatrics. Eleventh Edition. Philadelphia, Pa.: W. B. Saunders. Waber, D. 1976 Sex differences in cognition: a function of maturation rate? Science 192:572-574. Werner, E., Bierman. J., French, F., Simonian, K., Connor, A., Smith, R., and Campbell, M. 1968 Reproductive and environmental casualties: a report of the ten year follow-up of the children of Kauai Pregnancy Study. Pediatrics 42:112-127. Werner, H., and Strauss, A. 1939 Problems and methods of functional analysis in mentally deficient children. Jour- nal of Abnor,,7ul and Social Psychology 34:37-62. White, B. 1975 Critical influences in the origins of competence. Merrill-P(`ln2er Quarterly 21:243- 266. Wilton, D., and Barbour, A. 1978 Mother-child interaction in high-risk and contrast preschoolers of low socioeco- nomic status. Child Development 49: 1 136-1 145. Winick, M. 1969 Malnutrition and brain development. Journal of Pediatrics 74:667-679. Winick M., Fish, I., and Rosso, P. 1968 Cellular recovery in rat tissues after a brief period of neonatal malnutrition. Journal of Nutrition 95:623-626. Winter, S. 1972 The male disadvantage in diseases acquired in childhood. Developn~e'2tal Medicine and Child Neurology 14:517-519. Wishik, S. 1964 Georgia Study of Handicapped Children. Report on a Study of Prevalence, Disability Needs, Resources, and Contributory Factors. Publications for Program Administration and Community Organization. Georgia Department of Public Health, Atlanta.

Biological and Social Factors 181 Witelson, S. 1976 Sex and the single hemisphere: specialization of the right hemisphere for spatial processing. Science 193:425-427. Zajonc, R. B. 1976 Family configuration and intelligence. Science 192:227-236. Zajonc, R. B., and Bargh, J. 1980 Birth order, family size, and decline of SAT scores. American Psychologist 35:662- 668. Zajonc, R. B., and Markus, G. B. 1975 Birth order and intellectual development. Psychological Review 82:74-88. Zigler, E., and Trickett, P. 1978 I.Q., social competence and evaluation of early childhood intervention programs. ,4merican Psychologist 33: 789-798.

Next: Classifying Mentally Retarded Students: A Review of Placement Practices in Special Education »
Placing Children in Special Education: A Strategy for Equity Get This Book
×
Buy Paperback | $125.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!