in young children's lives has been answered in the affirmative and should be put to rest. However, interventions that work are rarely simple, inexpensive, or easy to implement. The critical agenda for early childhood intervention is to advance understanding of what it takes to improve the odds of positive outcomes for the nation's most vulnerable young children and to determine the most cost-effective strategies for achieving well-defined goals.
The environmental change required to alter a child's developmental trajectory is likely to vary for different children in different circumstances. The plasticity of human development works both ways—environments can be both enhancing and debilitating; child change can be for better or worse; and the gains produced by an effective intervention can be maintained by continuing support or lost by the subsequent influences of an impoverished or abusive environment. In the final analysis, all abilities and behaviors unfold within boundaries set by constitutional endowment, and all children (including those with developmental impairments) are primed biologically to seek positive adaptation. Thus, all effective interventions “work ” by supporting those self-righting tendencies. Intervention programs are not panaceas—they simply shift the odds in favor of more desirable outcomes. Nevertheless, on both an individual and a population basis, such probabilistic changes can make a significant difference. This will occur, however, only if the crucial elements of successful strategies are identified and sustained as model demonstration programs are transformed into larger-scale implementation.
The scientific knowledge base guiding early childhood policies and programs is seriously constrained by the relatively limited availability of systematic and rigorous evaluations of program implementation; gaps in the documentation of causal relations between specific interventions and specific outcomes and of the underlying mechanisms of change; and infrequent assessments of program costs and benefits.
Recommendations for policy and practice are ideally based on causal evidence. These include, for example, inferences about the effects of early experience on brain development and behavior, the effects of early behavior on later adolescent and adult functioning, and the impacts of specific interventions (such as enhanced child care and early education, home visiting programs, etc.) on child health and development. The committee's assessment suggests that researchers, policy makers, and practitioners often overestimate the scientific basis for such causal inferences, generally in the service of ensuring continued support for a given policy or practice that is presumed to be effective.
Under such circumstances, some observers have drawn sweeping conclusions from single experiments and have not engaged in a critical examination of the assumptions underlying causal inferences based on nonexperi-