Supplement
Ethics of Sex Research Involving Children and the Mentally Retarded
by Albert R. Jonsen, Jay Mann
(Excerpt. William H. Masters, Virginia E. Johnson, Robert C. Kolodny, and Sarah M. Weems. Ethical Issues in Sex Therapy. Volume 2, Little Brown & Co., Boston, Massachusetts, 1980, pp. 61-108)
The title of this essay embraces two topics that have been embroiled in ethical controversy: sex research and research with children and the retarded. The ethical controversy surrounding the first is the general subject of this congress. It is our task to report the basic contentions of the ethical controversy surrounding research involving children and retardates and to relate these to sex research. It is our hope that we can illuminate the problems and perhaps even resolve some of them.
Although our report is concerned specifically with children and the retarded, it also raises issues about other groups that might be considered special populations and in need of special protection. Our colleagues on the Ethics Congress task force agreed that children and mental retardates constitute two categories that must be designated as “special” for purposes of participation in sex research. However, some members pointed out that additional groups might merit inclusion under this rubric. It was suggested that the following groups be considered: prisoners, disabled persons (both institutionalized and noninstitutionalized), the mentally ill, homosexuals, and the elderly. The commonality among these groups is that the participation of their members in sex research presumably exposes them to greater risks or takes advantage of a more restricted freedom than is the case with the average adult. We suggest three concepts that might be useful in determining whether a particular individual or class of individuals is disadvantaged in these respects and therefore requires special treatment.
The first concept, diminished capacity, would apply to those individuals whose mental state—for example, senility, immaturity, or mental impairment—prevents their making reasoned judgments about the nature or consequences of their participation. These individuals are presumably incapable of giving consent that is truly informed. Moreover, if their mental state is such that their response to participation is difficult to predict, the investigator is faced with the additional problem of making an accurate risk-benefit assessment.
The second concept, diminished autonomy, applies to children under the control of parents or parent surrogates and to institutionalized subjects. The use of “captive” populations is common in behavioral research. The college student—the traditional “guinea pig” of the psychology researcher—is an example: Although participation in college-based research is ostensibly voluntary, the student may experience covert pressure to participate. In the case of less autonomous groups, such as prisoners or other institutionalized subjects, the pressure may be more intense. A prisoner, for example, may believe cooperation will influence his parole or his treatment. The issue with captive samples is again informed consent. In order to ascertain whether it exists, one would have to determine whether groups whose power to make independent decisions has been restricted are given adequate information about their participation and are under no coercion, however covert, to participate.
The third concept, increased vulnerability, applies to those individuals whose participation in research makes them exceptionally vulnerable to social reprisals. For example, individuals who prefer homosexual partners or who are involved in variant sexual practices such as mate swapping would fall into this category. The major issue here may be confidentiality—a concern in all sex research, but a more crucial consideration in studies dealing with information that, if disclosed, might be damaging to the subjects’ status in the community.
It is clear that other groups share with children and retardates the need for special ethical guidelines. However, we believe that a comprehensive detailing of the issues in research with children and retardates will provide some suggestions for formulating guidelines appropriate to the needs of other groups whose participation in research entails exceptional risks or impairment of the conditions necessary for informed consent.
RESEARCH INVOLVING CHILDREN
Throughout history, children have been subjects of research. A legendary Persian prince is said to have isolated newborn babies from all human speech in hopes of discovering whether language was natural and spontaneous. Queen Caroline of England recruited children from a foundling hospital in order to test Jenner’s vaccination on them before it was applied to her own royal progeny. The modern controversy over research with children and the retarded can be traced to events at Willowbrook State Hospital for the Mentally Retarded where, during the 1950s and 1960s, healthy mentally retarded children were infected with mild hepatitis in order to study the natural history of the disease. That research aroused great public controversy, and its merits and faults are debated even today. It initiated discussion within and without the medical profession over whether any invasion or manipulation of the body or person of a nonconsenting subject could ever be ethically justified apart from expectation of benefit to that very subject [1].
The prevailing codes of ethics governing the use of human subjects for experimentation are curiously ambivalent about children. The progenitor of these codes, the Code of Nuremberg, would—if taken literally—exclude children as subjects, for it requires the voluntary consent of the subject “as absolutely essential,” without any provision for proxy consent [2]. On the other hand, the World Medical Association’s Declaration of Helsinki states that “the consent of the legal guardian should be procured” for nontherapeutic research involving legally incompetent subjects [3]. A similar position has been adopted by the American Medical Association:
“Minors or mentally incompetent persons may be used as subjects only if (i) the nature of the investigation is such that mentally competent adults would not be suitable subjects; (ii) consent, in writing, is given by a legally authorized representative of the subject under circumstances in which an informed and prudent adult would reasonably be expected to volunteer himself or his child as a subject” [4].
This position, permitting nontherapeutic research on the basis of proxy consent of the legally authorized representative, was tacitly accepted by the first DHEW regulations [5].
It is not an unchallenged position. Beecher cited (and disagreed with) the purported British position that any medical intervention on a normal healthy child (or on a sick child without the intent to provide therapy) was ethically and legally unacceptable [6]. A case currently sub judice seeks to establish this position in an American jurisdiction, arguing that parents have no legal right to consent to any intervention not directed to their child’s benefit.* Philosophers and theologians continue to debate the question: Princeton’s Paul Ramsey and Georgetown’s Richard McCormick take opposing views. The former repudiates any ethical justification for research with a normal child; the latter claims that even children bear a certain obligation to benefit society, which responsibility justifies a presumption of their consent to experiments involving minimal risk [7, 8].
* Nielson v. Regents of University of California (Cal. Sup. Ct., Civil Action No. 665-049, filed December 19, 1973).
At the heart of the legal and ethical controversy are two central issues: proxy consent and the relevance of risk-benefit assessment. The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, established by Congress in 1974, has been charged with an investigation of these issues with regard to research involving children and the institutionalized mentally incompetent. The recommendations of the Commission, of which one of the present authors* is a member, have been presented to the Secretary of Health, Education, and Welfare. These recommendations will doubtless influence federal policy governing the conduct and support of research involving children. They will therefore serve as the framework for our analysis of ethical issues in sex research involving children and the retarded.
* A.R.J.
First, we shall survey the history of sex research in order to show how sex research with children fits into the development of the field and to reveal some of the problems that have propelled and restrained research; second, we shall state several modalities of social research in general that pertain to the ethics of research with children in particular; third, we shall detail the areas of research with children that seem particularly important; finally, we shall attempt to fit together the scientific and the ethical imperatives regarding research with children.
We write this essay with an audience in mind. That audience is not one of researchers or ethicists, nor is it one of parents and children. Rather, we have in mind a prospective audience of relative newcomers to the research world: reviewers of protocols. At present and for the foreseeable future, the ethics of research will be judged primarily by those members of the professions and the public who are appointed to institutional review boards wherever biomedical and behavioral research is being done. We must assume that most such persons, while highly motivated and knowledgeable in their own fields, know very little about sex research. We shall try to provide them with a perspective from which they can fairly and intelligently judge proposals to conduct such research. Of course, their judgment will rest on general regulations and on the specific information offered by the researcher, but between these lie many considerations that strongly influence judgments—considerations about the nature of children, about the nature of sexuality, and about the nature of social research. This essay attempts to elucidate some of these considerations.
SEX RESEARCH WITH CHILDREN:
HISTORY AND PROBLEMS
Questions about the ethics of sex research with children should be placed within the context of the history of sex research in general. The ways in which research problems arose and methodologies were formulated to study them follow a pattern that may be instructive when the ethical propriety of involving children and the retarded in sex research is considered. We shall briefly survey the history of sex research and, within that history, view certain proposals to study the sexuality of children and the retarded.
Sigmund Freud propelled sexuality out of prescientific darkness into scientific light. Almost a century of growing interest in the biomedical and behavioral aspects of human sexuality has followed his dramatic and heretical proclamations. It is widely known that the principal proclamation of Freud’s doctrine, around which all else clustered, concerned the sexuality of childhood. He wrote,
“It is part of popular belief about the sexual instinct that it is absent in childhood and that it first appears in the period of life known as puberty. This, though a common error, is serious in its consequences and is chiefly due to our ignorance of the fundamental principles of the sexual life” [9].
Curiously, despite that central proclamation, ignorance of childhood sexuality remains profound. Today, 70 years after Freud wrote those words, John Money labels childhood sexuality “a research frontier, unopened to empirical and operational study” [10]. In another recent article, Warren Johnson notes a dearth of “much needed research to obtain basic scientific knowledge on the subject of childhood sexuality”; he also describes formidable opposition to any effort that might be construed as encouraging childhood sexuality [11]. And Dr. William H. Masters, in response to the question “When will your type of research be done with children?”, had to reply, “Not in my lifetime” [11].
However, the fact that Dr. Masters has in his lifetime seen and effected an immense expansion of the old frontiers of sex research reminds us that sex research in general has proceeded—though with extreme circumspection. At each point in its cautious progress, it encountered obstacles in the form of social attitudes, moral prohibitions, and legal proscriptions that seemed insurmountable at the time. The same discreet exploration may or may not be possible with regard to childhood sexuality. Still, the history of research into adult sexuality shows that it has advanced by virtue of gradual changes in mores and morals as well as the deliberate design of studies and selection of subject matter by the researchers. Public attitudes do not impose the sole constraint upon sex research with children. Other constraints are imposed by the ethics of the researchers themselves, by the ethical standards of the profession, and by the laws and regulations governing the ethical conduct of research. Many potential studies have been shelved before being put to the test of public acceptance because their sponsors were unable to assure themselves that these studies met criteria for ethical research; some studies that meet ethical standards are inevitably repudiated by public attitudes. Perhaps a time will come when public attitudes will be so relaxed as to tolerate even research that thoughtful persons deem ethically inappropriate.
Byrne has noted that scientists initially approached the socially proscribed area of direct investigation of human sexual behavior by studying “animal sex, native sex, and crazy sex” [12]. Studies of the mating habits of species other than man, cross-cultural research into sexual behavior, and investigations of sexual anomalies or deviant and problematic sexual activities provided a more acceptable approach to the questions of sexuality than did direct investigation of human, American, and “normal” sexuality. Indeed, the policies of many funding sources sustain the focus on these areas, being less squeamish about underwriting the study of exotic or problematic sexuality than the study of modal sexual functioning.
Social scientists first undertook systematic surveys outside the clinical setting in the 1920s and 1930s. They usually employed quite indirect and nonintrusive methods and often camouflaged the real topic under study. These surveys differed from earlier investigations in two important ways: They studied samples drawn from general populations; and they were primarily taxonomic, directed toward collection of quantifiable, descriptive data from a sizable sample of subjects. These two features allowed findings to be generalized in the form of statements about normative sexual behaviors.
Sexual Behavior in the Human Male [13] opened a new era in sex research. The study was without precedent in terms of sample size and diversity, the scope of activities studied, the methods of investigation, and data analysis and reporting. In an attempt to remedy defects perceived in earlier studies, Kinsey and his colleagues designed a direct and individualized interaction between interviewer and subject “in the way friends talk to friends” [13]. This technique, a relatively free-flowing conversation recorded in a precise code, presumably yielded more accurate, verifiable, and detailed information than the previous discreet and indirect questionnaires or structured interviews. Kinsey’s pioneering methods initiated the basic form of contemporary sex research: the frank, realistic, intimate encounter between investigator and subject.
Finally, sex research moved from observation and interview to experimentation, the planned manipulation of variables. In the early 1950s, experimenters moved with trepidation, as reflected in their exclusive reliance on male volunteers and the use of low-key stimuli and subjective response measures. Over a period of two decades, typical research stimuli moved from nude or seminude still photos and rather mild erotic literature to explicit film representations of sexual activity; measures of response moved from pen and pencil to the recording of physiological arousal reactions. Human Sexual Response [14] was the astonishing result of these methods and the harbinger of things to come.
To summarize, sex research moved from clinical origins based on reports of pathological cases, through indirect and guarded inquiry among the general population, to direct interrogation of individuals and, finally, experimental methods. This historical survey, while familiar to sex researchers, is essential information for those who must review research. The ethical issues that will be discussed are intimately related to these differences in research methodology.
The history of research into childhood sexuality is much more sparse than that of studies with adults. It follows much the same course, although progressing more slowly. Investigators have made unobtrusive, naturalistic observations of the spontaneous sexual behavior of small children [15, 16]. However, studies involving exposure of children to explicit erotic stimuli, systematic observation, or measurement of sexual response have not yet been performed. Given the many unresolved questions about the effects of intrusive interventions on children’s responses and psychosexual development, researchers may find themselves hard put to predict the consequences of such research. Inability to do so prevents formulation of risk-benefit assessments and informed consent criteria requisite for ethical research. Even if public attitudes about childhood sexuality were to become more permissive, researchers might find themselves unable to devise ethical strategies for highly intrusive studies of children’s sexual responses.
CLINICAL STUDIES
Most of the clinical or quasi-clinical studies in the literature of childhood sexuality report descriptive data as a byproduct of diagnostic and therapeutic procedures necessitated by the children’s condition. Pediatric patients as well as children in psychoanalysis or other forms of psychotherapy have been a major source of data on childhood sexual behavior [17, 18]. Children with problems of gender identity and sexual identity related to endocrine or psychosocial influences have also been studied. These children usually have been brought for treatment to medical clinics where investigators have relied on clinical methods of data collection, interview, and behavioral observation, but physical examinations and laboratory tests such as endocrine assays have also been used [19]. Children brought to clinicians because of sexual involvement with adults have been studied in a similar manner. Typically, these children are interviewed by the clinician; they may also receive a genital examination if trauma is suspected. Often these children have been interrogated by law enforcement authorities seeking evidence for prosecution of the alleged offender [20].
A few researchers have studied individual normal children or small samples of normals, using observational methods similar to those used with clinical subjects. For example, Kleeman reports that he began collecting data on children’s masturbatory activity over 25 years ago, beginning with his own children and children in Edith Jackson’s Rooming-In Project [21, 22]. Kleeman later reported additional single-subject observational studies of the masturbatory activity of normal young children [15, 16]. He cites five major sources of data on masturbation, based on clinical observation [21]: (1) reconstructions from the psychoanalytic treatment of adults and children; (2) case reports of individual children incorporating histories from parents and observations of the child by the therapist; (3) reports from pediatricians [17]; (4) observations of institutionalized children by trained observers, with comparison data from observation of normal children in the home setting [23-25]; and (5) observations of children in clinic or public nursery settings away from their homes [26, 27]. Of necessity, observation is the major quasi-clinical means of collecting data on the behavior of infants and very young children. With older children, additional procedures used in diagnosis and treatment can be employed, including doll play, interview, and projective tests. The use of dolls and projective tests as stimuli can be viewed as quasi-experimental intervention to the extent that these measures manipulate the child’s responses.
Single- or multiple-subject studies in which medical or psychological treatment of the child is a major component raise a number of separate issues in addition to those raised by other types of research. These issues will be discussed in a later section.
TAXONOMIC STUDIES
The transition from studying clinical samples to studying normals and from collecting data on individual cases or small groups of cases to surveying larger samples was cautious and deliberate.
One early study of boys ranging from 8 to 18 years of age was conducted by Merrill in 1918 [28]. Although it differed from the clinical case studies in that it collected interview data from a large group (N = 100) and was performed outside a medical setting, it could legitimately be classified as a quasi-clinical study because it investigated a deviant sample: delinquent boys. The investigator, a probation officer, interviewed boys who were passing through a juvenile court in Seattle. The author’s findings included data on preadolescent masturbation to orgasm and oral-genital homosexual contact. This early study exemplifies the license granted to the investigator reporting on childhood sexual activity when the children in question are labeled as deviant. It is doubtful that an American researcher during the same period would have been able to publish comparable data on nondelinquent normal boys.
Some 30 years later, Ramsey studied a nonclinical group of 291 preadolescent and young adolescent males representing all the boys in a seventh- and eighth-grade group (and some others) in a medium-large city in Illinois [29]. Using personal interviews, the author collected data on masturbation and other sexual activities.
One other early study is also noteworthy because it broke with tradition by surveying a nonclinical sample. Hughes, a state health officer-biologist in North Carolina, used a group-administered questionnaire to survey a cross-section of grade-school and high-school rural and urban boys who were mostly between the ages of 15 and 20 [30]. Because of its predominant focus on boys over 15, this study does not qualify as a study of preadolescent sexual development. However, it serves as an approximation of later nonclinical research with younger children. The study was hardly a model of scientific objectivity. Many items in the questionnaire were embedded in moral judgments (for example, “Has anyone ever tried to give you the mistaken idea that sexual intercourse is necessary for the health of the young man?”). Although the author’s reasons for injecting moral bias into the questionnaire items are a matter of conjecture, these items may have served to counter accusations that Hughes’s questions “put ideas in the heads” of his young subjects.
Kinsey and his colleagues collected detailed interview data from 212 preadolescent boys [13]. Their method as adapted to this age group will be described in a subsequent section. In addition to interviewing preadolescent boys, these investigators collected data on preadolescent sexual activity from retrospective self-report interviews with adults, from reports of parental observations, and from interviews with older subjects who had had sexual contact with preadolescents. The survey reported data on sexual and nonsexual sources of erotic arousal and detailed descriptions of orgasm (based in part on data from the previously cited study by Ramsey [29]) and included observational reports on the speed of reaching orgasm in 1,888 boys, aged 5 months to adolescence, who were timed with a stop watch. In a few cases, observations of young preadolescents were continued over a period of months or years, until the individuals reached an age at which it could be established that true orgasm was involved [13].
Kinsey and co-workers also reported data on 147 preadolescent females ranging in age from 2 to 15 years [31]. These authors mentioned that they had conducted a detailed study of younger children (in particular, children between 2 and 5 years of age) to investigate the sources of their sex education, but that the study needed to be carried further before the results were ready to be reported in detail. Some of the results were subsequently published [32]; the remaining data are recorded on computer tape in the archives of the Institute for Sex Research [33].
STUDIES OF PROCREATIVE INFORMATION
In the past, investigators studying young children’s acquisition of procreative and related sexual information have relied on retrospective interviews with adults as major sources of data. Only a very few researchers have interviewed children directly in order to assess their knowledge and beliefs about sexual anatomy, conception, and procreation. Researchers undertaking such studies must contend with a dispute that has long embroiled sex educators and the general public: a dispute about what children should be taught and what they already know [34]. It is often assumed that asking the child questions about sex and procreation is tantamount to giving the child new information. Researchers sensitive to this issue have generally selected samples of children whose parents approved of their children’s receiving some basic information about sex and procreation. One study, based on interviews with 30 boys and 30 girls aged 3 to 12, reports that, “given the hesitance of many parents to have their children questioned about sexual matters, random sampling was not attempted” [35]; the study sample was recruited from upper middle-class Caucasian families in Berkeley, whose attitudes toward sex education were assumed to be more liberal than those one would encounter in most communities. Another study in which children were interviewed directly about procreative and sexual knowledge used a select sample of boys and girls aged 3 to 5.5 who attended a university nursery school [36]. This study employed a twelve-item questionnaire derived from a similar instrument used by Kreitler and Kreitler [ 37] to study the concepts of sexuality and birth held by Israeli children. The questionnaire covered knowledge of excretion, including names of the functions, organs, and products; the mother’s and father’s respective roles in conception; and the process of birth.
Some interviewers used more circumspect methods in preference to direct questioning. Conn [38] used doll play as a technique in interviewing 100 children aged 4 to 11. Levy [39], anticipating the realistic products currently produced by toy manufacturers, used a doll with a clay penis in order to study young children’s awareness of sex differences. Interviewers at the Institute for Sex Research used another play technique with their young subjects: asking them to draw pictures, which were then used as the basis for questioning by the interviewers about the child’s own behavior [40].
One source of data requiring no interaction of the investigator and the child is content analysis of children’s stories. One study that analyzed the spontaneous stories of a group of boys and girls aged 2 to 5 revealed little sexual content [41]; a similar study using another sample of boys and girls in the same age bracket reported that their spontaneous stories contained a high proportion of violent themes and a low proportion of sexual themes [42].
An ingenious study by the E.G. Brown Center for Family Studies [43] surveyed children’s knowledge of gestation by having classroom teachers in California and Oregon ask young children to draw pictures that answered the questions “Where was I before I was born?” and “What did I look like before I was born?”; the 1,100 drawings received from respondents were then subjected to content analysis.
SEX ROLE STUDIES
The profound influence of sex role socialization on the enactment of one’s sexuality has been described by numerous authors [44]. The study of sex-typed behaviors in young children appears to be a special case in sex research.
Parents often manifest concern about sex-inappropriate behaviors of their children, especially their sons [45]; however, they do not usually translate that concern into fear that sex role research will negatively influence sex role behavior. (In contrast, these fears frequently surface with research focused on genital sexuality or its procreative results.) This suspension of concern has permitted researchers to conduct experiments in sex role socialization without encountering parental opposition, even to the extent of devising experimental situations in which sex-role behaviors are manipulated through modeling or other processes [46]. Experiments in which children’s sexual knowledge or specific “sexual” behaviors (such as genital exploration) were dependent variables would undoubtedly evoke much more opposition from parents than have experiments focused on their socialization—although sex role socialization is at least as crucial to the development and expression of sexuality as is children’s knowledge of procreation or their early interest in one another’s genitals.
Research studies in the area of childhood sexuality begin to lose their immunity to parental censure as they draw closer to genital sexuality. For example, Broderick [47] studied the development of heterosexual socialization in preadolescents, with particular attention to the formation of relationships. He questioned children about their feelings of attraction to other children, about their “girl friends” and “boy friends,” and about dating, kissing, and other dimensions of interpersonal activity prevalent in the age group. Questions about genital sexual behavior were not a part of the study. Nonetheless, the study elicited an irate response from many parents [48].
METHODOLOGY AND MORALITY
This survey of the movement from the clinical to the nonclinical, from observation to interrogation and experimental manipulation, reveals some methodological considerations that are directly relevant to the ethical problems. We will argue that the reactivity of a given study is the key variable in assessing its ethical implications.
The term reactivity refers to the probability that a given method of investigation or exposure to a specific body of content or class of stimuli will influence the subject’s attitudinal, emotional, or behavioral responses. Reactivity sometimes characterizes a random effect inadvertently elicited by a measurement technique: For example, in a study done by one of the present authors [49], the daily checklist that served as an inventory of subjects’ sexual behavior, fantasies, and urges was demonstrably more effective in shaping sexual behavior than were the explicit erotic films used as experimental stimuli. However, we also use the term reactivity with reference to a calculated effect resulting from an experimental manipulation: For example, it is common practice among social psychologists to design experiments so as to have a maximum impact on subjects [50].
Three variables, characteristics of the stimulus, presumably contribute to reactivity. We shall use these as referents in evaluating specific research approaches in terms of the probable impact on the subject. Each variable describes an essentially discrete dimension of research methodology or content; however, there may be a degree of overlap in some instances. The variables are as follows:
1. Intrusiveness. This term describes the degree of investigator-subject interaction dictated by the design of the study. For example, an interview requires more investigator-subject interaction than a questionnaire, which can be mailed out or group-administered; thus, the interview would be relatively high in intrusiveness. At the negative end of the intrusiveness continuum are so-called unobtrusive measures, which require neither cooperation nor interaction from the subject [51]. Content analyses of children’s classroom drawing, cataloguing and describing graffiti, and inspection of library catalogue cards for signs of activity are examples of unobtrusive measures.
2. Inquisitiveness. This term characterizes the nature and extent of the information elicited from subjects. Procedures designed to elicit information about behavior generally deemed private would rate high on the inquisitiveness continuum.
3. Explicitness. This term describes the nature of the experimental manipulations and the content of measuring instruments such as tests, interview schedules, and other protocols in terms of their sexual explicitness. For example, the use in an experiment of films vividly depicting coitus would rate high in explicitness. Although usually applied to sexual material, the term could also be used to describe one dimension of other emotionally charged themes (for example, violence).
It is apparent from our historical survey that sex research has progressed toward methodologies that are higher in reactivity. More recent studies, in contrast to earlier ones, rank higher in intrusiveness, inquisitiveness, and explicitness, and would be expected to elicit more noticeable change in attitudes and behavior. It is precisely this reactivity or tendency to bring about change in attitudes and behavior that constitutes the central ethical problem of sex research with children. Researchers consistently encounter parental opposition when they attempt to recruit children for sex research. This parental opposition is apparently based on fear that the research will effect changes in their children’s sexual attitudes, information, and behavior.
In general, three different beliefs seem to prompt fear in parents. First, they believe that children cannot make critical judgments that would enable them to resist or reject information. Second, they believe that sexual information is emotionally overpowering and will arouse children physically or titillate their curiosity, which may in turn trigger precocious sexual behavior. Finally, parents consider sex education their exclusive province and see themselves as responsible for imparting it to their children in terms of their own cultural and religious convictions. If these beliefs are well-founded, children will presumably be changed by sex research in ways that are beyond their own control and the control of their parents—assuming that sex research is in fact highly reactive. The more intrusive, inquisitive, and explicit the research methodology, the more it will offend traditional beliefs about the psychological, social, and moral dimensions of childhood sexuality.
When researchers propose to act in ways that are beyond the control of their subjects or their subjects’ guardians and when their activities appear likely to effect changes that are considered risky or dangerous, they face the two central ethical issues of research. These two familiar ethical issues—informed consent and risk-benefit assessment—are thus linked to the reactivity of a given research study. Whether or not the dimensions of intrusiveness, inquisitiveness, and explicitness do in fact bear a direct relationship to reactivity in every instance is an empirical question. Some studies high in these three dimensions might conceivably effect less change than other studies that are relatively low in these dimensions. But in general, studies using methodologies characterized by high intrusiveness, inquisitiveness, and explicitness are likely to be viewed as having a high potential for reactivity by adults concerned with the subjects’ welfare.
RECOMMENDATIONS OF THE NATIONAL COMMISSION
FOR THE PROTECTION OF HUMAN SUBJECTS
Before discussing the problems reactivity poses for risk-benefit assessment and consent, we shall summarize the recommendations recently delivered to the Secretary of Health, Education, and Welfare by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, concerning research involving children [52]. The result of extensive review of the problems involved in research with children, these recommendations suggest new avenues for exploring the ethics of research with children.
The Commission defines children as “individuals who have not attained the age of consent for medical care or for research activities in the jurisdiction in which the research will be conducted.” (We shall comment later on this definition, both in itself and in its application to the specific problems of sex research.) The Commission defines research activities as “all procedures performed in a formal investigation designed to contribute to generalizable knowledge; they include, but are not limited to, diagnostic, prophylactic and therapeutic interventions that are the focus of systematic investigation, and other interventions or modifications of procedures, observations, comparisons, measurements, the use of questionnaires and surveys, and the analysis of data, undertaken for the purpose of obtaining specific information” [52] .
The first recommendation is a general commendation of research, stating that research is valuable and necessary for the health and well-being of children, that it can be done ethically, and that it should be supported by the Secretary of Health, Education, and Welfare under certain specific conditions. Those conditions are set forth in the recommendations that follow.*
* Only the salient points are cited here. Each of the ten recommendations is highly nuanced and is followed by explanatory comments. The full text is available (see reference [52]).
The second recommendation states the conditions that must be fulfilled if research is to be judged ethical; scientific soundness; prior animal and adult studies; safest possible procedures; protection of privacy and confidentiality; equitable selection.
The third recommendation states that research “that does not involve greater than minimal risk” may be conducted, provided an institutional review board is satisfied that the general conditions for ethical research are met. (The Commission defines minimal risk as “the probability and magnitude of physical or psychological harm that is normally encountered in the daily lives, or in the routine medical or psychological examination, of healthy children.” Research involving more than minimal risk is justified only by the demonstration that the research offers some direct benefit to the subject or, if it does not, that it represents a “minor increase over minimal risk” and “presents experiences to subjects that are reasonably commensurate with those inherent in their actual or expected medical, psychological or social situations” [Recommendation 5]. In the latter case, the research must promise knowledge “of vital importance for the understanding and amelioration of the subject's disorder or condition.”)
The seventh recommendation states that assent of children should be sought (in general) and must be sought when the children are over the age of seven. The term proxy consent is discarded in favor of “permission of parents or guardians,” which permission should “reflect the collective judgment of the family that an infant or child may participate in research.” In addition, close parental involvement (in some cases even parental presence) is required for certain sorts of research.
The eighth recommendation recognizes that for certain types of research, parental permission may not be a reasonable requirement to protect the subject. The types of research included in this category are (1) research designed to identify factors related to the incidence of certain conditions in adolescents for which in certain jurisdictions they may legally receive treatment without parental consent; (2) research in which the subjects are “mature minors” and the procedures involved entail essentially no more than minimal risk that such individuals might reasonably assume on their own; (3) research designed to understand and meet the needs of neglected or abused children, or children designated by their parents as “in need of supervision.” In such cases, review boards may waive the requirement of parental permission, provided that some alternative mechanism is utilized to protect the children’s rights.
ETHICAL EVALUATION OF RESEARCH
When a protocol is presented, reviewers must first judge whether it affects human subjects and, if it does, whether those subjects are put at risk of detriment to their welfare or rights by the research. Clearly, the probability and magnitude of the risks vary according to the nature of the subjects. Should a researcher propose to use children as subjects in sex research, the characteristics of the prospective population must be carefully delineated in terms of age, gender, status (that is, whether they are patients, institutionalized, school children, juvenile offenders, etc.) and, if possible, social and economic class. One immediate problem with such designation comes from the Commission’s definition of children as “individuals who have not attained the age of consent for medical care or for research activities in the jurisdiction in which the research will be conducted.” Since no jurisdiction has established an age of consent for research, the age of consent for medical treatment would appear to be the most relevant criterion. This varies from jurisdiction to jurisdiction: In some areas there is a statutory age of consent; in other areas there are exceptions to the statutory age for certain types of medical treatment, such as obstetrical care, abortion, contraception, and addiction.
It is interesting to note that these legal exceptions are allowed primarily for matters having to do with sexuality. It might be presumed that research into these matters might be carried out as if the subjects were “mature minors” capable of consent regardless of parental consent. The Commission’s recommendation provides for this:
“If the Institutional Review Board determines that a research protocol is designed for conditions or a subject population for which parental or guardian permission is not a reasonable requirement to protect the subjects, it may waive such requirement provided an appropriate mechanism for protecting the children who will participate as subjects in the research is substituted” [Recommendation 8, p. 17].
In commenting on this recommendation, the Commission notes:
“A number of states have specific legislation permitting minors to consent to treatment for certain conditions (e.g., pregnancy, drug addiction, venereal diseases) without the permission (or knowledge) of their parents. If parental permission were required for research about such conditions, it would be difficult to develop improved methods of prevention and therapy that meet the special needs of adolescents. Therefore, assent of such mature minors should be considered sufficient with respect to research about conditions for which they have legal authority to consent on their own to treatment” [Recommendation 8, p. 18].
Thus, for certain types of research in certain localities, persons otherwise legally defined as children can be considered adults. (However, for purposes of the remainder of this paper, children shall be considered as individuals prior to puberty in whom neither the psychological nor physiological maturity required for the foregoing exception has been attained.)*
* Adolescents (postpubertal individuals under the age of 18) are not viewed as children for the purposes of this paper because their physiological maturity, their potential for sexual activity, their relatively greater autonomy, and their legal status set them apart from younger (prepubescent) children. However, we recognize that adolescents comprise a heterogeneous group whose members are located on a twofold continuum with characteristics ranging from full emancipation, both de facto and de jure, to stringent parental control and from sexual activity and experience to sexual inexperience and naiveté. We suggest that the positions of any individual adolescent on both continua might serve as referents for the considerations applicable to his or her participation in research. Thus, the right of an adolescent to volunteer for research might be assimilated to his or her legal right to medical treatment for sexually related conditions in certain jurisdictions. Beyond legal considerations, researchers and review boards who are sensitive to parental and community response might ensure that selection procedures take into account the individual subject’s level of maturity as a sexual being and his or her degree of autonomy within the family. Anyone who proposes to do either therapy or research involving adolescents must inevitably walk a narrow legal and ethical line between the gradually increasing autonomy of the young person and the gradually diminishing authority of parents.
Several points of importance for sex research involving children emerge from these recommendations. First, the definition of minimal risk as risk not surpassing that encountered in children’s normal daily lives might apply to much sex research. Second, the allowance of more than minimal risk where therapeutic benefit might result would apply to a broad range of sex research associated with treatment of sexually related conditions. Third, the insistence on parental permission, involvement, and participation poses formidable problems for sex research with children. Finally, the categories of research for which parental permission may be waived, mentioned in Recommendation Eight, apply clearly to certain types of sex research, such as inquiries into social or economic circumstances surrounding adolescent pregnancy, investigations of sexual conduct leading to venereal disease, or studies of sexual abuse of children by parents.
In all these recommendations, the determination of risks and benefits and the suitable manner of obtaining assent of children and permission or participation of parents remains largely at the discretion of the institutional review board (IRB), an entity established in 1966 for all institutions where research is supported by the Department of Health, Education, and Welfare. The discretionary power of the IRB poses a major problem to those who wish to conduct research in areas that are either unknown from the scientific viewpoint or controversial in terms of values. Sex research involving children is both. The remainder of this paper will suggest ways in which an institutional review board might interpret the recommendations when judging a protocol for sex research using children as subjects.
The first specific question to be asked about the proposed research is “whether it entails risks or discomforts to the children greater than those normally encountered in their daily lives or in routine medical or psychological evaluations.” The Commission expands on this by referring to “modest changes in diet or schedule, immunization, physical examination, obtaining blood or urine specimens and developmental assessments. Many tools of behavioral research, such as most questionaires, psychological tests or puzzles, may also be considered of minimal risk” [52].
Can the “tools of behavioral research” be so easily considered of minimal risk when the behavior under investigation is sexual behavior? It is important to review the available literature on this controversial point.
Let us first summarize children’s responses to research. There appears to be little risk to the child subject from research involving unobtrusive measures, including discreet observation. Observational data might pose some risk if it were used to embarrass the child or to impose constraints on behavior that might not be in the child’s best interests. One might also argue that individuals of any age are entitled to know when they are being systematically observed and when their behavior is being recorded.
Investigators who have interviewed child subjects or administered questionnaires to children report very few responses indicative of distress. Children in the Moore and Kendall study of reproductive concepts “did not appear anxious in the interviewing situation; some did appear disheartened that they did not have responses to the questions asked. As students in an experimental school, they had all experienced testing previously and many were either sufficiently bright or competitive or both that they wanted to perform well” [36].
The minor distress experienced here appears unrelated to the sexual content of the study. The concept that asking questions will “put ideas in their heads” was not supported by the work of Bernstein and Cowan, whose results “suggest strongly that children actively construct their notions about babies; they don't wait to be told about procreation before they have an idea of how it occurs. What may be taken as misinformation may largely be a product of their own assimilative processes at work on materials with too complex a structure for them to understand” [35]. The consensus of investigators who have worked with prepubescent children of various ages is that children tend to gloss over questions that they do not understand and screen out material that is not relevant to their current interests or compatible with their level of development. Given the extensive use of erotic material in advertising and in the media, including television, the average child is exposed to a spate of stimuli that could conceivably elicit erotic responses. In order to determine whether questions or procedures used in a given study would produce additional stimulation, one would have to have some idea of what the child encounters in everyday experience. The experience of children would vary from one subcultural milieu to another, from one family to another, and as a function of the individual child’s age and level of maturity. In general, it is reasonable for a review committee to consider nonintrusive research as posing minimal risk, even when it involves direct questioning.
What of novel procedures, such as new modes of observation or new levels or types of explicitness? Is it possible to make an assessment of risks of procedures for which there is little or no precedent? One could, of course, adopt the position that no procedures or content be used beyond those already proven innocuous by previous studies. But this would produce a stalemate—new information is needed in order to expand the boundaries of acceptable research, and the boundaries of research must be expanded in order to provide new information. A somewhat more open attitude would be to test new procedures by small increments. For example, if a given questionnaire or procedure has produced no untoward results with a sample of children, subsequent investigators might feel secure in increasing the explicitness, the intrusiveness, or the inquisitiveness of the procedures by some small increment or by one “just noticeable difference.” Another position—which could be combined with the second attitude—would be to avoid procedures that have been demonstrably harmful to child subjects in the short term and to avoid any methodology that incorporates experiences substantially beyond those to which the child would presumably be exposed in the natural environment.
The natural environment varies according to social class and economic status, among other factors; it differs for healthy children and those who suffer chronic disorders. It is strikingly different in different cultures. In order to determine the normal experience of a given group of children, it is necessary to look carefully at the structure of their environment. One can identify natural environments in which there is treatment of children (for example, those with chronic disorders, the institutionalized, and those whose sexual behavior has brought them into contact with law enforcement or judicial agencies). The normal experience of such children might surpass in stressfulness any methodology contemplated by researchers. On the other hand, the experiences of children who have been traumatized by medical investigative procedures can serve as caveats for investigators.
An instance of the use of diagnostic procedures that proved harmful to children with endocrine disorders is reported by Money and Walker [53]. Because of the principle involved, it appears useful to describe this study in some detail. Fifteen girls, initially examined at a Johns Hopkins pediatric endocrinology clinic because of precocious pubertal development, were subsequently referred to a psychohormonal research unit. The girls were interviewed at the research unit, and those old enough to respond were given questionnaires to complete. Both parents of each study subject were also interviewed. The girls, none of whom was younger than age 10 at the time of the last interview, were followed for periods of up to 18 years. During the course of their interviews with these girls, Money and Walker discovered that previous diagnostic procedures had produced severe adverse reactions, which they describe in the following excerpt:
“Reactions to nudity for medical photography and vaginal examinations were explicitly recorded in 6 cases. In the other 9 cases nudity was not an explicit issue. Unconcern over nudity for medical photography was overtly indicated once. In five cases—33% of the sample—reactions to nudity and vaginal examinations ranged from distress to extreme phobic avoidance. One girl, aged 8-4/12, refused to undress in the presence of anyone. Another girl, aged 6-11/12, refused to allow her mother to bathe her or her gynecologist-father to take nude photographs of her. Nude medical photography was a source of embarrassment and severe stress for another girl, aged 7-1/12, who was also “disgusted” over her pediatrician’s attempt to conduct a vaginal examination. Vaginal examinations and vaginal smear tests were responded to by two girls, aged 4-6/12 and 6-7/12, as violently upsetting. The origin of intense phobic reaction to nudity and vaginal examinations, when it occurs, has not been traced. When it does occur, however, it is so powerful that it may result in refusal, defiance, and determination to escape that completely negate the physician’s endeavors on the child’s behalf. Such a reaction is a psychologic danger signal not to be taken lightly. Its resolution requires collaboration of both the mother and the daughter in a program of case management. It may require that the child be examined only by a female physician, and the same one on each return visit. It also requires that the child be helped to comprehend why, whereas the doctor’s words say that she is normal except for growing too fast, his actions convey the message that she is a curious creature, if not a freak, to be poked at, examined, X-rayed, and photographed. Children, like adults, are helped by understanding the rationale of what is done to them, and why, in a medical examination. Otherwise they are too susceptible to overconcern with their physician’s insistent attentiveness to their genitalia“ [53].
Children with psychohormonal abnormalities or gender identity conflict unquestionably encounter childhood experiences quite different from those of their normal peers. If they are diagnosed and receive counseling or treatment, they will necessarily experience some highly reactive procedures. However, they constitute a population for whom relatively intrusive research might prove to be less stressful than the results of being allowed to develop without professional treatment or counseling. Nonetheless, as Money and Walker have stated, it is incumbent on the clinician or researcher to mitigate the adverse effects of his or her procedures insofar as possible. Callous responses by the researcher can exacerbate the distress of a child who is already implicitly labeled “misfit’ or “freak” by parents, peers, and clinicians. In contrast, the concerned and sensitive clinician or researcher can make the investigative environment safer and more supportive than the natural environment. Green, who has worked extensively with children with gender problems, describes some of the measures taken by his group to minimize the possibility of trauma to his subjects [54]:
1. Spell out the risks to the child and to the parents, including the possibility that the study will increase the child’s attention to his or her gender identity conflict and aggravate parent-child dissension.
2. Obtain informed consent of both parents and children (after risk factors have been elucidated).
3. Provide the child with a card enabling him or her to make collect calls to the investigator or directly to the Human Experimentation Committee. (This committee is prepared to handle grievances, if any, from children or parents.)
4. Make therapeutic interventions during interviews, when therapy appears indicated because of the child’s stressful reactions to current life events. (Such interventions should be made when necessary, even at the cost of confounding or losing research data from a subject.)
The foregoing example illustrates the principle that measures for minimizing risks can be incorporated in those studies in which potential benefits must be weighed against a high potential for adverse effects. It also calls attention to an important dictum in using child subjects whose ordinary experience is stressful or atypical as a result of physical, emotional, or sociocultural disadvantages: namely, that the researcher not regard the child’s disadvantaged situation as license to ignore the child’s personal welfare. Rather, within the boundaries imposed by the child’s condition, the investigator should make every effort to avoid procedures that are likely to produce stress, even though comparable stress might be a common element of the child’s ordinary experience.
The reactions of the girls with precocious pubertal development in the study described above [53] provide some basis for assessing the risks one might encounter in performing a study of the genital anatomy and physiology of children (for example, an approximation of Masters and Johnson’s study of adult sexual response [14]). One would expect comparable reactions to any procedure that offended the child’s sense of modesty or propriety or that encroached upon the integrity of the child’s body. However, it is noteworthy that some of the children in Money’s sample did not appear to have reacted negatively to nudity, medical photography, and vaginal examinations (in contrast to those who reported severe distress). It is important to consider individual differences both in the children themselves (and their backgrounds) and in the personal characteristics and research strategies of the investigators. There may be children—perhaps from families with accepting attitudes toward nudity and sexuality—who would not react adversely to genital examinations conducted under conditions designed to minimize anxiety. (Pediatricians do, in fact, examine children’s genitals when such examinations are indicated.) It seems plausible that children who have not been taught to view themselves as freaks would be less likely to register a phobic reaction to such examinations.
Although a few sex educators [11] have suggested that parents might facilitate later adult sexual health by encouraging their children’s sexuality and Money [10] has suggested the value of sexual rehearsal play, it is unlikely that studies in which children are asked to stimulate their own genitals (or allow them to be stimulated by others) would pass public scrutiny. Even if an investigator were to find parents who would agree to their children’s participation and children who were willing to participate, many years would have to elapse before the effects of participation on adult sexual adjustment could be assessed. The risk factor of such a long-term study cannot be evaluated.
Cross-cultural studies permit some evaluation of children’s reactions to sexual stimulation in situations in which it is socially sanctioned [55, 56]. Observations of children in preliterate societies in which sex play is permissible before adolescence indicate that there is nothing inherently traumatic in such activity; rather, trauma occurs when childhood impulses collide with adult disapproval. In many societies there are societally prescribed rituals for initiating prepubescent children into sexual practices, often under the tutelage of adults.
Children in our own society who engage in prepubertal sexual activity meet with powerful cultural taboos. Studies reporting reactions of children who have been sexually involved with adults provide evidence of this [57-59]. Further evidence is found in the Kinsey data collected from adult males who had had sexual contact with prepubescent children [13]. Such reports may provide a basis for assessing the response of children in our own society to sexual stimulation and for estimating the range of variation in children’s capacity to accommodate to research designs that are relatively high in intrusiveness and explicitness.
A moderate proportion of women in our society have had some preadolescent sexual contact with adult males: Kinsey [31] found that 24 percent of his sample of 4,441 women reported such contact. Some children register severe reactions to such contact, but it is unclear whether such reactions follow from the behavior itself or from adult reactions to it [60]. A number of researchers concur in the belief that children with adequate ego strength can undergo potentially damaging sexual experiences, even repeatedly, without sustaining long-term emotional trauma or impaired sexual adjustment in adulthood [61-65]. These researchers concur in identifying extreme adult reactions after the event as a major source of emotional disturbance for the child. Another research finding relevant to the capacity for sexual experience manifested by some children is the observation that children involved sexually with adults may have played a cooperative or initiatory role in the interaction [63, 65, 66]. Gebhard comments on the ethical implications of this finding as follows:
“The “victim” is often a willing sexual partner and even sometimes a seductive one, but if the person is under a certain age our social ethics definitely necessitate legal protection of the “object-victim” in such a situation” [67].
One group of children who usually sustain trauma as a result of sexual contact with adults are those who have been raped or otherwise subjected to force or duress [20].
PERMISSION OF PARENTS
Children can be enrolled as subjects of research with consent of their “legally authorized representatives.” We cited earlier the dispute among ethicists over this point: Some maintain that it applies only to beneficial activities, while others contend that it encompasses activities that are not directly beneficial but involve no risk or minimal risk to the subject. The Commission for the Protection of Human Subjects relies heavily on parents and family as the child subject’s primary protective agents. It departs from the “thin” notion of proxy consent to a richer concept of parental permission and participation, involving not only a detailed knowledge of what the research will entail but also a close involvement and sometimes even physical presence during the conduct of research. However, the Commission recognizes that some children do not dwell within a protective familial unit and that some types of research that might be highly desirable, such as study of child abuse, cannot be carried out with strict adherence to the mandate of parental permission. Such research is allowable only if some effective mechanism can be devised to protect the child’s interests.
The strong insistence on parental permission and participation poses a particular problem for sex research. There is some evidence pointing to a more general acceptance of sexuality as a natural part of childhood [11]. This attitude is accompanied by increased parental concern for teaching children concepts of time, place, and responsibility as means of directing sexual behavior into socially approved channels. The encouraging trend toward parental acceptance and attendant concerns may in time modify an attitude that has seriously hampered research into childhood sexuality: the notion that sexual ignorance preserves the innocence of the child. Despite the weight of evidence that repression of childhood sexual interests and condemnation of early sexual behavior has exacted a high toll in sexual dysfunction among adults [68], many parents still prefer to keep their children uninformed about sex. The parental concerns encountered by Sorensen [69] in his survey of adolescent sexuality reflect the attitudes of many parents toward the presumed innocence of their young and adolescent children. Sorensen’s research team initially selected 839 adolescents aged 13 to 19 for interviews about their sexual behavior. Only 508 parental units gave permission for their children to participate in the study; one might expect that the rate of consent would decline sharply for a comparable sample of younger children. Parents who were opposed to their children’s participation cited a variety of ostensible reasons for demurring (for example, “It wouldn’t matter to me, but she wouldn’t want to”; and “My 14- and 12-year-old are either at school or at home where there is no opportunity for sex play. If they don’t have sex, you don’t need to talk about it with them”). Others were more forthright in expressing their opposition (for example, stating apprehension that the interview “might put some ideas in her head she don’t know”). One Texas mother was reported as saying, “They don’t need to know, talk, study so much about sex. Leave it alone, and they will be better off” [69].
The arguments advanced by parents for withholding permission for sex research with children are familiar ones, echoes of the longstanding controversy over sex education in the schools [34] and of polemics against eroticism in films [70]. The convictions of these parents are sometimes rooted in absolutes acquired from their religious training or from early family teaching; often they are reinforced by strong feelings of anxiety about the erosion of parental control and values and by powerful mistrust and anger directed at the amorphous adversary. These assumptions, because of their tenacious emotional taproots, are unlikely to yield readily to logic or to empirical data. For example, opponents of sex education have remained unconvinced by evidence that promiscuity, unplanned pregnancy, and venereal disease appear to be less prevalent in school districts with sex education programs than in districts without such programs. However, in fairness to parents who hold conservative views on sex education, definitive evaluations of the effects of sex education with various age groups have yet to be made. Detractors have also claimed that the sexual ideologies propagated by many sex researchers contain biases reflecting white middle-class notions about what sex should be, rather than representing the broad spectrum of sexual mores in our society [ 34].
The willingness or unwillingness of parents to give proxy consent for their children to participate in sex research is closely tied to their assumptions about the benefits and risks of providing children with information about sex. There is some evidence that the alternative to providing sexual information is not necessarily blissful ignorance, even in the case of very young children. An information deficit in an area of human experience creates a vacuum that is liable to be filled by myth and superstition more damaging than the facts. For example, Kreitler and Kreitler [37], found that a high percentage of children in their sample, lacking accurate information, had confabulated their own concepts of gestation and birth. They frequently pictured the embryo as frightened and sad in the dreary confines of the womb. Many of the children believed that the baby was liberated from its uterine prison by cutting open the mother’s abdomen. It is difficult to believe that these concepts are more reassuring to the child than accurate information.
Children who are denied factual information often obtain their sex education from the garbled accounts of peers, from locker-room vulgarity, or from pornography. Investigating the sex education of matched samples of adult sex offenders, other offenders, and controls, Goldstein and Kant [71] found pornographic materials encountered during adolescence to be the major source of sex education for a high proportion of their respondents.
The data further suggest that children readily reinterpret any information given to them. Sex researchers, sex educators, and others who provide children with information about sex should attempt to discover whether the message received is the same as the message sent. Review boards might ask investigators to comment on this disparity of information as one of the possible risks of their studies.
Irrespective of the arguments for or against providing the child with sex information, researchers have had little choice but to restrict their sampling to children whose parents were willing for them to participate. As a consequence, studies using biased samples have yielded results that could not readily be generalized to other groups. This is another potential hazard that researchers face when they encounter parental opposition.
Parental opposition, even when based on massive distortion of the actual details of the study, has been known to spread through a community and infect an entire sample. The parents of one child accused Broderick of “Communism” and “Kinseyism” for trying to conduct a relatively low-key survey designed to solicit information from fifth- to twelfth-grade students about dating, falling in love, and kissing [72]. The child’s parents enlisted the support of the local press and launched a successful telephone campaign to discourage other parents in the community from granting permission for their children’s participation. As a result, the heterogeneity of the sample was reduced.
One strategy that is used to allay parental anxieties about their children’s participation is to invite the parent(s) to be present at the interview [13, 31, 36]. Kinsey’s research team interviewed children aged seven and younger in the presence of their parents; this age group appeared more comfortable with their parents present. By contrast, children over seven years of age were interviewed alone because the older children as a group appeared less comfortable with their parents present [73].
The age of the child is an important consideration in the parents’ decision to grant permission. Sorensen found some parents initially reluctant to allow their 13- to 15-year-old children to be interviewed for a survey of adolescent sexuality [69]. Sorensen’s researchers were themselves concerned about the possibility of disturbing the sensitivities or naiveté of subjects in this age group. After much deliberation, the research team decided to include them because of the importance of the information to be gained from younger adolescents.
The nature of the research also may influence parental decisions. Studies incorporating research procedures that could be described as relatively low in intrusiveness, explicitness, and inquisitiveness would probably elicit more parental cooperation than studies high in these dimensions. Additional factors that would be reassuring to parents are the established reputations and academic affiliations of the investigators and the use of an academic or other “respectable” physical setting for the research.
In order to understand the nature of parental concerns, researchers must appreciate the family as a rule-governed system. When the family is viewed in this light, parents may be seen as regarding sexuality as an area over which they wish to retain control. They may regard regulation of their children’s sexual behavior as a mechanism for maintaining homeostasis within the family system. They may recognize, explicitly or implicitly, that rules concerning sexual behavior are closely linked to parental authority in general, to religious values, and to the family’s established life-style. If parents believe research will introduce into this system information antithetical to their own values, they may consider their children’s participation as a threat to family solidarity. In some instances, parents may permit an older child to participate, while refusing the researcher access to younger siblings.
The concerns voiced by parents may also reflect uneasiness lest their own privacy be invaded or their own beliefs challenged by the research. Broderick [72] encountered parental opposition related to the following specific points: (a) parents’ objections to the inclusion of questionnaire items asking for names and parents’ occupation (s); (b) the question of whether students and faculty at a nearby university would have access to the data; and (c) parents’ concerns that the subjects’ responses would consist of either common knowledge or “tall tales,” because “any good parent knows whether his child is involved in any of those things.” Implicit in these concerns is the common assumption that the child is an extension of the parent. In this context, the issue of risk to the child may be subsidiary to parents’ concern for their own well-being, as they perceive it.
If consent from parents is to be truly informed, the parent would need to know the subject matter of the study, the method of investigation, and the possible consequences for the child. In addition, if the child’s right to privacy is to be respected, the parent would require reassurance that the content of a given child’s responses will not be divulged. Although ethically sound, providing parents with advance information about the details of the study may create a methodological dilemma for the investigator; In at least one instance that can be documented [72], advance information resulted in the invalidation of one block of data. In the previously mentioned study, Broderick purposely refrained from informing parents that questionnaires would be distributed in their children’s classrooms, because he feared that parents would bias responses by coaching their children. His apprehensions were realized when students in one classroom provided clear evidence that they had been coached by their teacher; the entire classroom gave an identical response to every projective question: “I do not know. I have not had the experience.” Even the underlining was uniform [72].
An interesting ethical issue would arise should a parent refuse to let his or her child participate in a study involving a group of which the child was a part (for example, a school class). If only a few children are excluded from a group study because of parental unwillingness, they may feel deprived of an experience that has been presented as fun or they may be distressed by their parents’ inexplicable behavior. At the least, they would feel discriminated against as having been excluded from a group activity, perhaps without understanding the reason for their exclusion. Therefore, it appears important for investigators to design selection procedures that would minimize the excluded child’s distress.
When parents acquiesce to the child’s participation, the researcher is faced with the need to obtain informed assent from the child as well. The protocol for obtaining assent will vary according to the age and level of comprehension of the child involved. In order for the child to assess his willingness to cooperate, the child is entitled to know what to expect—what the research participation will entail. The description of the subject matter of the project should be given in terms that the child can understand. The child should also be informed of the provisions for maintaining confidentiality. It is conceivable that in some instances, parents may grant consent for the participation of children who themselves are hesitant or unwilling to participate. Children may also feel coerced by group pressure to participate. In order to preserve the child’s autonomy, the researcher must be alert to any signs of reluctance or ambivalence in the child and must make the child fully aware that nonparticipation is an acceptable option.
The device of project monitor or ombudsman may be employed to deal with many of these problems. Parents may be reassured and more cooperative if they know that someone whom they trust (and who is not otherwise associated with the research) is acting as “overseer” for the project. Children who are reluctant to refuse to participate or embarrassed to withdraw might be encouraged to present their difficulties to this monitor. The monitors themselves might be on guard for signs of distress or lack of cooperation in subjects. They may also assist the researcher in assessing risks and benefits and in soliciting assent from subjects in a manner that is fair, honest, comprehensive, and comprehensible. A monitor should have the authority to withdraw a child from a project and even to recommend to an institutional review board that a project be modified or terminated, for valid reasons.
We have attempted to show that the risk-benefit calculation required for any ethical judgment about research involving children might be converted into an estimate of the reactivity of sex research. The more intrusive, inquisitive, and explicit the research methodology and tools, the higher the risk of effecting changes in attitudes about sex and sexual behavior. Since the possibility of effecting changes over which parents have little or no control is probably the principal deterrent factor in securing parental permission for children’s participation in research, the judgment of investigators and institutional review boards regarding the reactivity of a study should be communicated fully to parents or guardians. For most types of research, parental fears of loss of control can be assuaged or even eliminated by informing them of the index of reactivity of the project and attempting to involve them in the actual conduct of the investigation in ways compatible with the research design. When this is not possible, the active involvement of a project monitor may accomplish the same end. In sum, it is our judgment that most forms of sex research with children can be performed in a manner that meets ethical standards in research.
TREATMENT-RELATED RESEARCH
An examination of the ethics of sex research involving children would be incomplete without some consideration of the implications of research in which treatment of sex-related problems in children is the principal object of study. In some instances, sex-related problems in children may be treated by psychotherapeutic, behavioral, medical, or surgical procedures as part of a formal study; in other instances, treatment of an individual child or a few individuals may result in generalizable knowledge that is subsequently published as research, although the original objective was solely therapeutic change. Thus, whether or not it is so designated at the outset, any case in treatment has the potential for becoming the subject of published research. It is important, therefore, to view treatment not only within its own context but also from the perspective of research.
Children under the age of puberty and postpubescent nonemancipated minors constitute two groups who might be involved in treatment-related research under either of the conditions described above. It is clear from the clinical literature and from unpublished reports of practitioners who treat children and adolescents that psychodynamically oriented psychotherapy, behavior modification, and medical and surgical procedures comprise the major methods used to treat young people for a variety of sex-related problems. Some of the patients (usually the older ones) are self-referred; others are brought to treatment by parents, institutional or agency personnel, or other adults in authority. The problems for which these young patients are being treated include psychohormonal, gender, and sexual identity disorders; sequelae of incest, rape, and other sexual contact with adults; sexual behaviors offensive to adults in authority (for example, public masturbation or sexual advances to peers by institutionalized retardates); venereal and other sex-related diseases; problems associated with contraception and pregnancy; and emotional or behavioral problems diagnosed as symptomatic of sexual conflicts. In addition, there are reports from practitioners at family planning clinics and adolescent clinics that young patients are presenting with complaints of sexual dysfunction—even though the treatability of these problems and the appropriateness of seeking treatment for them has not yet gained full acceptance among adults. It is obvious that contraceptive and pregnancy-related services and treatment for sexual dysfunction would be sought only by those who are already sexually active or who intend to become so; however, pregnancy statistics for girls aged 10 to 13 suggest that the incidence of sexual activity among young children is increasing significantly. As a result, the treatment of newly pubescent and even prepubescent children for sex-related problems may become much more common, accompanied by a corresponding need for research to evaluate the effects of treatment.
The recommendations of the National Commission for the Protection of Human Subjects relevant to research with children have been summarized in a previous section. These recommendations outline suggestions for minimizing risk and criteria for determining whether the potential benefits outweigh potential risks to the child, as well as measures for gaining the informed consent and assent of the child and of parents (or of other adults in loco parentis). These recommendations apply to all types of research involving child subjects, including clinical or treatment-related research. But when treatment is an issue, the interpretation and application of the Commission’s recommendations require more complex decision-making, more attention to the nuances of each individual case, than is necessary in nonclinical research. For example, failure to be accepted as a subject in survey research may have only minor consequences for a child; but the child under consideration for treatment of a severe sex-related disorder has much more at stake. Either decision—to treat or not to treat—has profound implications for the child. The possible consequences of treating the child may be only partially predictable. Treatment is intended to be highly reactive—its objective is change (this is in sharp contrast to survey research, for example, which incorporates measures designed to minimize reactivity). One expects treatment procedures to act upon the child to produce change—although not necessarily the change that was intended. Given the multitude of variables that impinge upon the child within his or her complex ecological system, the individual differences among children, and the paucity of adequately validated treatment procedures, it is not yet possible to predict consistent and precise outcomes.
Even if specific change could be consistently programmed, simply deciding what constitutes therapeutic change poses a formidable dilemma. It is important to remember that the selection of a specific treatment objective for a child with a sex-related problem reflects a specific set of beliefs about what is healthy or normal or “best for the child.” Yet such decisions are being made in a cultural milieu in which sexual mores and practices are themselves changing at an unprecedented rate. It is not unlikely that two sets of experts evaluating a given child’s sexual behavior might find themselves polarized. One group might label the behavior an aberration calling for drastic change; the other group might regard it as a natural and allowable option in a pluralistic society. In view of the high degree of variability in the area of sexual values and beliefs, many conflicts are likely to occur regarding the necessity for treatment of sex-related problems in children. Parents may object to children’s requests for treatment (for example, a young woman who wants to take oral contraceptives); children may resist parents’ attempts to treat them for conditions that the children do not define as problematic; investigators may disagree with child or parents or both about the necessity for treatment.
Given that the risk-benefit ratio may be exceptionally difficult to assess in treatment-related research and that informed consent and assent of both parent (or parent surrogate) and child may be difficult to obtain, what are the implications for investigators and review boards? In summary, the principles to be used in making decisions appear to be identical to those enunciated in previous sections of this paper. However, their implementation is more difficult. In contrast to the basic researcher who may simply abandon a project if he or she cannot make a precise determination of risks, the clinician-researcher cannot resolve a dilemma merely by deciding not to treat. The decision not to treat a patient who may be experiencing severe distress is a decision with serious ethical implications. There are no simple formulas that can be applied to decision-making, either by clinician-researchers or by clinicians in general.
It is important for the clinician-researcher to be able to combine both the clinical and research roles. If, in making clinical decisions, he or she were to function primarily as a clinician, the clinician-researcher would then be justified in doing what clinicians must do routinely: deciding to treat or not to treat on the basis of clinical acumen and the best available data, with full awareness that treatment implies some element of risk. It is important also that clinical needs take precedence over research needs when the well-being of the patient-subject necessitates some deviation from the research design. On the other hand, the role of the researcher would prevail with respect to obtaining informed consent, protecting the privacy of the patient-subject, and instituting all possible measures (such as requesting the appointment of a project monitor) that might alleviate or circumvent potential problems arising from the patient-subject’s participation in the project. Finally, one additional area of responsibility rests with the clinician-researcher: the responsibility of persuading the hesitant or recalcitrant patient-subject or parent when the clinician is convinced that treatment is strongly indicated. In other words, parents and participants must be informed not only of the risks and benefits of participation but also of the risks (and benefits, if any) of nonparticipation.
RESEARCH WITH THE MENTALLY RETARDED
It is not uncommon to link children and the mentally retarded when questions are raised about ethical treatment of the retarded. However, the linkage may be less than helpful. Of course, prepubescent retarded persons are children, and the same ethical considerations would apply to both groups, for the most part. But considerations that would approve some intervention in the life of a normal child on the basis of its anticipated contribution to maturation may not obtain in the case of a retarded child.
Once the retarded individual has entered puberty, the comparison is even less apt. While intelligence may be measured in ways that make the older retarded appear similar to children, many genuine differences separate the two groups. In considering the retarded as a potential research population—in particular, for purposes of sex research—physical maturity, awareness of sexual activities, emotions and attitudes related to sex, and readiness for masturbation and sexual contact with others are all variables that cannot be disregarded (despite the unfortunate tendency to consider the retarded as being sexless). Finally, since many retarded persons live in institutions or under supervision, the effects and influences of institutionalization must be taken into account.
Questions concerning the sexuality of the retarded and, a fortiori, questions related to research about their sexuality, deserve much more ample treatment than a few pages at the end of a long essay. Unfortunately, the National Commission for the Protection of Human Subjects does not provide the same guidance with regard to the retarded as it does in the case of children. Its mandate required it to study the problems of research with the “institutionalized mentally infirm.” This omnibus category—so designated more by de rigueur style of legislative language than by the dictates of psychology or psychiatry—includes the retarded, the senile, and the psychotic, but excludes persons (even persons in these three groups) who do not live under the authority of an institution. This essay, on the other hand, considers only the retarded, both institutionalized and living independently or with family. Thus, the Commission’s recommendations are only partially relevant to our topic. We must conclude that a complete and satisfactory treatment of the subject of sex research with the retarded is hardly possible at this time. We therefore confine ourselves to a statement of some considerations that might serve as elements of a more adequate future treatment of the topic.
A number of studies of the sexual knowledge, attitudes, and behavior of the retarded have been published [74]. However, there remain massive gaps in our empirical knowledge of the sexuality of retarded persons. Doubtless, research proposals to remedy this informational deficit will be forthcoming. Edgerton suggests the following areas as having high priority for research in the field [75]: (1) the relationship between sexual behavior and self-esteem; (2) self-control of sexual behavior among the retarded; (3) the ability of the retardate to use contraceptive devices; (4) the right versus the ability of the retardate to marry and have children; (5) public attitudes toward the sexual behavior of the retarded and their influence on the retardate; and (6) new methods of sex education for and about the retarded.
Mental retardation has been defined as “subaverage general intellectual functioning which originates during the developmental period and is associated with impairment in one or more of the following; (1) maturation, (2) learning, (3) social adjustment. Subaverage refers to performance more than one standard deviation below the population mean” [76].
This definition is very broad. Any diagnosis of mental retardation invokes complex and controversial criteria for intellectual, emotional, and social functioning. Moreover, the mentally retarded are not all alike: Not only are there a variety of forms of retardation based on different medical and psychological criteria, but within these different forms there are individual differences in cognitive ability, emotional stability, social skills, and sexual behavior. In particular, the sexual behavior and attitudes of retarded adults are as heterogeneous as those found among groups of non-retarded adults. By no means do retarded individuals match the two polar stereotypes—sexually naive (or asexual) and promiscuous (or impulse-ridden) [77, 78]. Investigators conducting sex research with children find that parents consistently underestimate the level of sexual knowledge and liberality of attitude of their normal children; the same judgment appears to be true of those charged with care of the retarded. A study of 61 noninstitutionalized mildly and moderately retarded adolescents found subjects to be significantly more liberal in their sexual ethics than their parents predicted (although parents did accurately predict their children’s sex knowledge scores) [79].
It is imperative for those who propose to do sex research involving the retarded (and those who must make judgments about the ethical acceptability of such research) to recognize the equivocal nature of the diagnosis and the great diversity of personal characteristics of persons so diagnosed. The National Commission does this in its recommendations on research involving the institutionalized mentally infirm. The Commission asserts that some persons who fall into this category may be quite capable of giving informed consent. It introduces the concept of assent by acknowledging that some persons, though intellectually or emotionally impaired, have the capacity to understand and to direct their lives in certain ways. The Commission affirms that for research involving no more than minimal risk, the assent of the subject is sufficient [80].
The foregoing recommendation leads to a second consideration relevant to sex research. It is quite difficult to determine what might constitute “no more than minimal risk” for persons designated as mentally retarded. In the case of the institutionalized mentally infirm, the Commission defines minimal risk as “the risk (probability and magnitude of physical or psychological harm or discomfort) that is normally encountered in the daily lives or in the routine medical or psychological examination of normal persons” [Recommendation 2] [80]. This means that the impact of a procedure should be no greater for a person who is mentally infirm than for any normal person undergoing the procedure. However, if we return to the concept of reactivity introduced earlier in this paper, we may wonder whether the impact of sexual information might not be significantly different on a retarded person than on either a normal child or an unimpaired adult. If so, would provision of (or deliberate exposure to) such information constitute more than minimal risk for a retarded person?
The physical maturity of retarded persons who are not chronologically children makes possible overt behavior that is unlikely to occur in children. The provision of sex-related information could be harmful to those whose understanding of self and of social relationships is impaired. One potentially dangerous situation is the possibility of unplanned impregnation and pregnancy. Another risk is the possibility of socially inappropriate behavior, which could even lead to criminal charges: The research might provide information that subjects, because of diminished capacity to interpret it or to integrate it into behavior, could translate into inappropriate activities. (Obviously, inappropriate will mean different things to different judges—but it is clear that an unforeseen and unintended instigation to sexual overtures in public would create more alarm than an unintended inducement to masturbation in private.) Even more serious would be the possibility of sexual exploitation of the retarded person as an unintended consequence of participation in research. In sum, review boards must scrutinize most thoroughly the latent potential for reactivity in proposed research. If the above cautions are observed, persons conducting sex research with the mildly or moderately retarded—particularly those living with parents or parent surrogates—can generally consider their subjects as normal in all respects except those in which they exhibit deficiencies. Invitations to participate in research must be designed to respect those deficiencies, rather than taking advantage of them.
The assessment of risk can lead both the research and the reviewers of research in a vicious circle: One cannot judge the reactivity of providing sex-related information without studying the actual reaction of subjects to whom it is provided; yet, one cannot study their reaction without first estimating the risk. Researchers and reviewers must proceed with caution, scrutinizing the nature of the information, attempting to find analogies, utilizing the technique of the “just noticeable difference” to test new procedures. In general, however, it can be said that when mentally retarded subjects are judged capable of informed consent or assent and when the research involves no more than minimal risk—and when both the capacity of the subject and the risks of the research are evaluated by cautious and prudent persons—the involvement of these subjects in sex research would be permissible.
Clearly, the researchers and the reviewers must make the requisite evaluation; they will presumably be cautious and prudent persons. However, it may sometimes be necessary to request the judgment of someone who is more impartial than the researchers and more intimately acquainted with the subjects than are the reviewers. The Commission suggests that a “consent auditor” be appointed whenever the subject’s capacity to consent or assent is in question; it requires the presence of such a person when the research involves more than minimal risk and does not hold out promise of direct benefit for the subject. Sex research with the retarded is likely to qualify on both counts: doubt about capacity and uncertainty regarding risks. In such situations, the reviewers would be well advised to appoint a consent auditor. This role should be filled by someone who is independent of the research team, responsible only to the institutional review board, and familiar with the physical, psychological, and social needs of the prospective subjects as well as their legal status [Recommendations 2, 3] [80].
A final set of considerations concerns the benefits of sex research to the retarded individuals or to the class of persons designated as retarded. The Commission requires full justification for the use of the retarded population for research. In general, that justification would turn on prospective benefits of the research to the members of that population (whether they be the individuals on whom the research is to be done or other, future members). It would be difficult to offer as justification prospective benefits to normal persons; it would be impossible to accept this reasoning if the research involved more than minimal risk.
For the most part, any proposed research that involves more than minimal risks should hold out promise of therapeutic benefit to the subject. It is conceivable that sex research might include experimental techniques of behavior modification aimed at reducing harmful or unacceptable sexual behavior. In such cases, risks somewhat more than minimal could be tolerated in prospect of individual benefit. Certain other types of research (for example, attempts to discover whether retarded persons can utilize contraceptives) may be intended to benefit the subject, but may also be quite risky—in the example cited, failure to achieve responsible contraception might result in unwanted pregnancy. Such benefit-oriented research should not be undertaken without effective reduction of the risks.
Finally, the relevance of any sex research with the retarded might be questioned. The Commission does not require that all research involving the mentally infirm be “significant” or “relevant,” since it is unlikely that any general definitions or measures of these qualities could be devised to act as criteria. However, it insists that for research with more than minimal risk involving subjects with impaired capacity to consent, “the anticipated knowledge (1) be of vital importance for the understanding or amelioration of the type of disorder or condition of the subjects, or (2) may reasonably be expected to benefit the subjects in the future” [Recommendation 4] [80]. Should sex research involving a retarded population ever require such stringent justification, the involvement of a legally appointed guardian is mandatory. If the risks appear to be more than small increments to the minimum-risk criterion, the research must be reviewed by a national ethical advisory board (which the Commission recommends be established) [Recommendation 5] [80].
In conclusion, one final consideration about the relevance of research must be mentioned, although it involves major value assumptions. Research into the sexuality of children is justified by the intent to understand, correct, and improve sexual development in order to promote the eventual state of satisfying and responsible adult sexuality. In contrast, it might be thought that research into the sexuality of retarded persons, particularly those who are unlikely ever to be able to enter into responsible sexual relationships, has little or no purpose other than satisfying the curiosity of the researcher. Clearly, this objection can be refuted in cases where the research is aimed at aiding persons suffering from behavior disorders harmful to themselves or to others. Response to the broader challenge depends on more thoughtful reflection about the place of sexuality in the lives of those who suffer severe limitations in their intellectual and emotional capabilities. That thoughtful reflection lies beyond the bounds of this essay.
REFERENCES
1. Katz, J. Experimentation with Human Beings. New York: Russell Sage Foundation, 1972. Pp. 1007-1010.
2. Trials of War Criminals before the Nuremberg Military Tribunals. United States v. Karl Brandt. Washington, D.C.: U.S. Government Printing Office, 1949. Vol. 2, p. 181.
3. World Medical Association. Declaration of Helsinki, 1964. In H. Beecher, Research and the Individual: Human Studies. Boston: Little, Brown, 1970. P. 277.
4. American Medical Association. Ethical Guidelines for Clinical Investigations. In H. Beecher, Research and the Individual: Human Studies. Boston: Little, Brown, 1970. P. 223.
5. U.S. Department of Health, Education, and Welfare. The Institutional Guide to DHEW Policy on Protection of Human Subjects. DHEW Publication No. (NIH) 72-102 (Washington, D.C.: U.S. Government Printing Office, 1971). P. 71.
6. Medical Research Council, Great Britain. Responsibility in Investigations on Human Subjects. In H. Beecher, Research and the Individual: Human Studies. Boston: Little, Brown, 1970. Pp. 262-265.
7. Ramsey, P. The enforcement of morals: Non-therapeutic research on children. Hastings Center Report 6(4) :21-30,1976.
8. McCormick, R. Proxy consent in the experimental situation. Perspectives in Biology and Medicine 18:2-20, 1974.
9. Freud, S. Three Essays on Sexuality. In S. Freud, Standard Edition of the Complete Psychological Works. London: Hogarth, 1953. Vol. 7, pp. 135-245.
10. Money, J. Childhood: The last frontier in sex research. The Sciences 16(6) :12-15, 27; 1976.
11. Johnson, W. R. Childhood sexuality: The last of the great taboos? SIECUS Report 5:1-2, 15; 1977.
12. Byrne, D. Social psychology and the study of sexual behavior. Personality and Social Psychology Bulletin 3:3-30; 1977.
13. Kinsey, A. C., Pomeroy, W. B., and Martin, C. E. Sexual Behavior in the Human Male. Philadelphia: Saunders, 1953.
14. Masters, W. H., and Johnson, V. E. Human Sexual Response. Boston; Little, Brown, 1966.
15. Kleeman, J. A Boy Discovers His Penis. In The Psychoanalytic Study of the Child. New York: International Universities Press, 1965. Vol. 20, pp. 239-266.
16. Kleeman, J. A. Genital Self-Discovery During a Boy’s Second Year. In The Psychoanalytic Study of the Child. New York: International Universities Press, 1966. Vol. 21, pp. 358-392.
17. Levine, M. Pediatric Observations on Masturbation in Children. In The Psychoanalytic Study of the Child. New York: International Universities Press, 1951. Vol. 6, pp. 117-124.
18. Furman, R. A. Excerpt from the Analysis of a Prepuberty Boy. In I. M. Marcus and J. J. Francis (Eds.), Masturbation: From Infancy to Senescence. New York: International Universities Press, 1975.
19. Green, R. Sexual Identity Conflict in Children and Adults. New York: Basic Books, 1974. P. 35.
20. Hayman, C. R., Lewis, F. R., Stewart, M. S., and Grant, M. A public health program for sexually assaulted females. Public Health Reports 82:497-504, 1967.
21. Kleeman, J. A. Genital Self-Stimulation in Girls. In I. M. Marcus and J. J. Francis (Eds.), Masturbation: From Infancy to Senescence. New York: International Universities Press, 1975.
22. Olmstead, R. W., Svibergson, R. I., and Kleeman, J. A. The value of rooming-in experience in pediatric training. Pediatrics 2:617-621, 1949.
23. Spitz, R. A., and Wolf, K. M. Autoerotism: Some Empirical Findings and Hypotheses on Three of Its Manifestations in the First Year of Life. In The Psychoanalytic Study of the Child. New York: International Universities Press, 1949. Vols. 3, 4, pp. 85-120.
24. Freud, A., and Burlingham, D. Infants without Families: Reports on the Hampstead Nurseries. In The Writings of Anna Freud, Vol. 2. New York: International Universities Press, 1973.
25. Provence, S., and Lipton, R. Infants in Institutions. New York: International Universities Press, 1962.
26. Roiphe, H. On an Early Genital Phase. In The Psychoanalytic Study of the Child, Vol. 23. New York; International Universities Press, 1968.
27. Sperling, E. Research in Early Genital Arousal: An Overview of Theory and Method. Paper presented at the meeting of the American Psychoanalytic Association, San Francisco, December 18, 1970.
28. Merrill, L. A. A summary of findings in a study of sexualism among a group of one hundred delinquents. Journal of Juvenile Research 3:255-267, 1918.
29. Ramsey, G. V. Factors in the Sex Life of 291 Boys. Madison, N.J.; Published by the author, 1950.
30. Hughes, W. L. Sex experiences of boyhood. Journal of Social Hygiene 12:262-273, 1926.
31. Kinsey, A. C., Pomeroy, W. B., Martin, C. E., and Gebhard, P. H. Sexual Behavior in the Human Female. Philadelphia: Saunders, 1953.
32. Elias, J. E., and Gebhard, P. H. Sexuality and sexual learning in childhood: Research and possible implications for education. Phi Delta Kappan 1:401-405, 1969.
33. Pomeroy, W. B. Personal communication.
34. Breasted, M. Oh, Sex Education. New York: New American Library, 1971.
35. Bernstein, A., and Cowan, P. A. Children’s concepts of how people get babies. Child Development 45:77-91, 1975.
36. Moore, J. E., and Kendall, D. C. Children’s concepts of reproduction. Journal of Sex Research 7:42-61, 1971.
37. Kreitler, H., and Kreitler, S. Children’s concepts of sexuality and birth. Child Development 37:363-378,1966.
38. Conn, J. M. Children’s awareness of the origin of babies. Journal of Child Psychiatry 1:140-176, 1947.
39. Levy, V. M. Studies of children’s responses to differences in genitalia. American Journal of Orthopsychiatry 60:755-762, 1940.
40. Pomeroy, W. B. Personal communication.
41. Pitcher, E. G., and Prelinger, E. Children Tell Stories. New York: International Universities Press, 1963.
42. Ames, L. B. Children’s stories. Genetic Psychology Monographs 73:336-396,1966.
43. Miles, F. F. Children’s Art and Human Beginnings (pamphlet). Eugene, Oregon: E. C. Brown Center for Family Studies, 1971.
44. Safilios-Rothschild, C. Love, Sex and Sex Roles. Englewood Cliffs, N.J.: Prentice-Hall, 1977.
45. Lansky, L. M. The family structure also affects the model: Sex role attitudes in parents of preschool children. Merrill-Paimer Quarterly 13:139-150, 1967.
46. Wolf, T. M. Response consequences to televised, modeled sex-inappropriate play behavior. Journal of Genetic Psychology 127:35-44, 1975.
47. Broderick, C. B. Sexual behavior among pre-adolescents. Journal of Social Issues 22:6-21,1966.
48. Broderick, C. B. Personal communication.
49. Mann, J., Sidman, and Starr, S. Effects of Erotic Films on Sexual Behavior of Married Couples. In Technical Reports of the Commission on Obscenity and Pornography, Vol. 8. Washington, D.C.: U.S. Government Printing Office, 1970.
50. Aronson, E., and Carlsmith, J. M. Experimentation in Social Psychology. In G. Lindzey and E. Aronson (Eds.), The Handbook of Social Psychology (2nd ed.), Vol. 2. Reading, Mass.; Addison-Wesley, 1968.
51. Webb, E. J., Campbell, D. T., Schwartz, R. D., and Sechrest, L. Unobtrusive Measures: Nonreactive Research in the Social Sciences. Chicago: Rand McNally, 1966.
52. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. Report and Recommendations: Research Involving Children. DHEW Publication No. (05) 77-0004 (Washington, D.C.: U.S. Government Printing Office, 1977).
53. Money, J., and Walker, P. A. Psychosexual development, maternalism, nonpromiscuity, and body image in 15 females with precocious puberty. Archives of Sexual Behavior 1:45-60, 1971.
54. Green, R. Personal communication.
55. Malinowski, B. The Sexual Life of Savages in Northwest Melanesia. New York: Harcourt Brace Jovanovich, 1929.
56. Ford, C. S., and Beach, F. A. Patterns of Sexual Behavior. New York; Harper & Row, 1951.
57. Weiss, J., Rogers, E., Darwin, M., and Dutton, C. A study of girl sex victims. Psychiatric Quarterly 29:1-15,1955.
58. Bender, L., and Blau, A. The reaction of children to sexual relations with adults. American Journal of Orthopsychiatry 22:825-837, 1937.
59. Halleck, S. L. Emotional Effects of Victimization. In R. Slovenko (Ed.), Sexual Behavior and the Law. Springfield, Ill.: Charles C. Thomas, 1965.
60. Gagnon, J. H. Female child victims of sex offenses. Social Problems 13:176-192, 1965.
61. Schachter, M. Apropos de pretendus dommages psychiques con-secutifs a une agression sexuelle commise par un adolescent sur la personne d’un petit gargon. Acta Paedopsychiatrica 32:191-195, 1965.
62. Barry, M. J. Incest. In R. Slovenko (Ed.), Sexual Behavior and the Law. Springfield, Ill.: Charles C. Thomas, 1965.
63. Halleck, S. L., and Hersko, M. Homosexual behavior in a correctional institution for adolescent girls. American Journal of Orthopsychiatry 32:911-917, 1962.
64. Finch, E. Sexual activity of children with other children and adults. Clinical Pediatrics 6:1-2,1967.
65. Bender, L. Offended and Offender Children. In R. Slovenko (Ed.), Sexual Behavior and the Law. Springfield, Ill.: Charles C. Thomas, 1965.
66. Gibbens, T. C. N., and Prince, J. Child Victims of Sex Offenses. London: Institute for the Study and Treatment of Delinquency, 1963.
67. Gebhard, P. H., Gagnon, J. H., Pomeroy, W. B., and Christenson, C. V. Sex Offenders. New York: Harper & Row, 1965.
68. Masters, W. H., and Johnson, V. E. Human Sexual Inadequacy. Boston: Little, Brown, 1970.
69. Sorensen, R. C. Adolescent Sexuality in Contemporary America (The Sorensen Report). New York: World Publishing, 1973.
70. Link, W. C. (statement). In The Report of the Commission on Obscenity and Pornography. New York: Bantam, 1970. Pp. 448-451.
71. Goldstein, M. J., and Kant, H. S. Pornography and Sexual Deviance. Berkeley and Los Angeles; University of California Press, 1973.
72. Broderick, C. B. Personal communication.
73. Pomeroy, W. B. Personal communication.
74. Gebhard, P. H. Sexual Behavior of the Mentally Retarded. In F. F. de la Cruz and G. D. La Veck (Eds.), Human Sexuality and the Mentally Retarded. New York: Brunner/Mazel, 1973.
75. Edgerton, R. B. Some Socio-Cultural Research Considerations. In F. F. de la Cruz and G. D. La Veck (Eds.), Human Sexuality and the Mentally Retarded. New York: Brunner/Mazel, 1973.
76. Heber, A. A Manual on Terminology and Classification in Mental Retardation (monograph). American Journal of Retardation (Supp.64),1959.
77. Wha en, R. E., and Whalen, C. K. Sexual Behavior: Research Perspectives. In F. F. de la Cruz and G. D. La Veck (Eds.), Human Sexuality and, the Mentally Retarded. New York: Brunner/ Mazel, 1973.
78. Edgerton, R. B., and Dingman, H. F. Good reasons for bad supervision: Dating in a hospital for the mentally retarded. Psychiatric Quarterly 38 (Supp.): 221-223, 1964,
79. Hale, J. E., Morris, H. L., and Barker, H. R. Sexual knowledge and attitudes of mentally retarded adolescents. American Journal of Mental Deficiency 77:706-709, 1973.
80. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. Report and Recommendations on Research Involving Those Institutionalized as Mentally Infirm: DHEW Publication No. (OS) 78-0006 (Washington, D.C.: U.S. Government Printing Office, 1978). Recommendation 4, p. 16.