Ethical problems in research on risky behaviors and risky populations

The articles by Brooks-Gunn, Fisher, Hoagwood, Liss and Scott-Jones (all in this issue) present a panoply of real-world ethical issues in conducting scientific research on risky behaviors of children, adolescents, and their parents, particularly those from vulnerable populations. The universal, ethical principles of beneficence, justice, and respect for others are always applicable, but they do not resolve issues of child assent, parental consent, legal reporting requirements for illegal behaviors, and the special problems of studying risky behaviors in risky populations. Taken as a group, the articles raise some of the most interesting ethical problems that arise in developmental research. My discussion elaborates some issues and fails to resolve others. I hold the view that both science and ethics can be served by thoughtfully designed and implemented research on important social issues, but that the studies themselves cannot simultaneously solve the many societal problems of participants and be scientifically credible.     

The effects of maternal employment and non-maternal infant care on development at two and four years

Differences in maternal employment during children's first year of life and children's entry into non-maternal care before the age of 1 year failed to predict differences in cognitive and socioemotional development at 2 and 4 years of age, after family background variables were controlled. Two samples were studied: a population sample of 1100 Bermudian children and a smaller subsample of children most of whom were determined to be at risk for developmental problems. To assess the effects of maternal employment, we compared infants with mothers who worked 20 or more hours a week to infants with mothers who worked less than 20 hours a week. To assess the effects of entry into non-maternal care before the age of one, we compared infants who were placed in regular non-maternal care before the age of one versus infants who did not experience regular non-maternal care before the age of one. The results revealed that family background variables frequently predicted many child outcome measures in both the total sample and the smaller research sample. After controlling for family characteristics, no differences were found between children whose mothers worked 20 or more hours a week when they were infants and children with mothers who worked less than 20 hours a week in either sample. In addition, age of entry into non. maternal care before the age of one did not significantly predict any child outcome measures.     

Frequency ratios and the discrimination of pure tone sequences

We examined the effect of frequency ratios on the discrimination of patterns of alternating pure tones (ABABA). Listeners heard a repeating pattern presented in transposition (same frequency ratios between successive tones, different absolute frequencies) and were required to indicate when the pattern changed (different frequency ratios and absolute frequencies). Changes from patterns with simple frequency ratios to those with more complex ratios were more readily detected than were changes from complex ratios to simpler ratios.     

Visual slow brain potentials in children with attention deficit disorder

Using a two-stimulus reaction time paradigm, with two separate reward conditions (contingent and noncontingent), we compared slow wave brain potentials (ERPs) in 144 children with attention deficit disorder (ADD) and 30 normal control children. This article reviews the findings during the 900 msec visual warning stimulus. As we had expected, based on ERP work of Forth and Hare (1989) and Raine, Venables and Williams (1990), and on previous work from our own laboratory, the group differences were found in the negative slow wave portions of the ERP complex during the contingent reward condition but not during the noncontingent condition. Aggressive hyperactive subjects with attention deficit disorder (ADDHA) were discriminated from nonaggressive subjects (including control subjects) during the contingent reward condition in the following ways: (1) greater fronto-central negativity (640–900 msec slow wave) and (2) greater right parietal than left parietal negativity (430–750 msec slow wave). All ADD subgroups, when compared to control (CONTR) subjects, showed greater slow wave negativity (700–900 msec) at the midline occipital electrode site during the contingent reward condition. This could be explained in part as an IQ effect on ERPs reflecting the IQ difference between the ADD subgroups and the controls. These slow wave findings seem to relate to attentional problems of these children. They are discussed in terms of a psychobiological model of inhibition/disinhibition and appetitive activation.     

The influence of parents and family context on children's involvement in household tasks

To determine the relative impact of a number of family variables on children's performance of gender-typed household tasks, this study took account of the gender of the child, the gender of a sibling within the same age range (8–14 years), whether a first or second child and three parental variables: the degree of encouragement given to perform masculineand femininetasks, parental involvement in the same tasks, and parents' general egalitarianism. The sample consisted of 191 white, mainly Anglo Australian two-parent families, with the two oldest children in a boy—boy, girl—girl, boy—girl, or girl—boy sequence. To check on the robustness of effects, measures were taken on two occasions, on average 16 months apart. Among the family context variables, the gender of the child was the strongest contributing variable, with girls doing more feminine tasks than boys and boys doing more masculine tasks than girls. There was limited support for the proposition that first children do more housework than second children of the same gender, while the results for gender of sibling were small and inconsistent. Among the parental variables, encouragement had strong positive effects for feminine tasks (i.e., more encouragement by parents corresponded to more involvement by children). In contrast, parental involvement in the same tasks (modeling) and parental egalitarianism predicted only the performance of masculine tasks, and the direction of the effects was mostly negative (e.g., the more a father was involved in masculine tasks, the less a child did of those tasks). The results point to involvement in gender-typed activities being influenced by multiple factors, with parental encouragement and gender of child being most prominent among these. They also point to the value of sampling on more than one occasion and of considering separately the performance of feminine and masculine tasks. This research was financially supported by the Australian Research Council. We are happy to acknowledge that support, together with the essential assistance of the Department of Education and the many parents and children who participated in the study.     

Influence of the initial projection of the centre of gravity on posturo-kinetic coordination in humans. Application from a horizontal quadrupedal stance during voluntary limb raising

The purpose of the study is to investigate the influence of initial projection of the centre of gravity (CG) on the spatial and temporal organization of the dynamic phenomena accompanying a voluntary movement in humans. From a horizontal quadrupedal stance, ten normal subjects were instructed to raise the right forelimb towards a support target in self paced velocity conditions. Three experimental conditions were tested for which the initial CG projection was either within (C1), at the limits of (C2) or outside the supporting triangle (C3). The results showed that vertical force variations on the four supports always preceded the first vertical displacement of the right wrist (onset of the movement). From C1 to C3, the duration of these anticipatory dynamic phenomena increased enabling an adjustment of the CG position such that equilibrium constraints were fulfilled. For all conditions, the acceleration peak of the CG occurred prior to the onset of movement and from C1 to C3, its magnitude increased in a single direction of the horizontal plane. For each condition, the load transfer was directed to the left forelimb and the right hindlimb (diagonal strategy) and reached a maximal value at the time of lift-off. With respect to this moment, the onset of the movement occurred increasingly early from C1 to C3. These results suggest that contribution of peripheral cues to postural control is more important, when the requirements of horizontal CG displacement increase.     

Is fluency free-operant response-response chaining?

This article briefly reviews behavioral fluency and its 10 products. Fluency development requires three of the four free-operant freedoms: the freedom to present stimuli at the learner's rhythm, the freedom to form the response, and the freedom to speed at the learner's maximum frequency. The article closes with several suggestions that fluent performing is really operant response-response (R-R) chaining, and recommends further controlled laboratory research on free-operant R-R chaining.     

Psychopathic personality traits and temporal perspective: A test of the short time horizon hypothesis

A number of authors have proposed that psychopathic individuals possess an abnormally constricted time horizon (i.e., foreshortened sense of the future). This hypothesis was tested among 101 undergraduates, who were administered a battery of (1) self-report indices of psychopathic personality traits, antisocial behavior, and normal-range personality traits; (2) self-report indices of time perspective; (3) projective tests of time perspective; and (4) laboratory tasks assessing time estimation and capacity for foresight and impulse control. Measures of psychopathy/antisocial behavior tended to be negatively correlated with several self-report indices assessing preoccupation with the future and with a projective task assessing the frequency of thoughts concerning future events, although only one of the correlations with this latter task was significant. In most cases these correlations were not attributable to the variance shared by measures of psychopathy/antisocial behavior and measures of (low) anxiety-proneness, although several correlations decreased substantially after levels of harmavoidance were controlled. In contrast, measures of psychopathy/antisocial behavior were negligibly correlated with laboratory tasks. These results provide mixed support for the short time horizon hypothesis and suggest that further attention to the role of method factors in investigations of future time perspective is warranted.