NSF programs in the behavioral and neural sciences

The National Science Foundation’s new Division of Behavioral and Neural Sciences supports basic research and related activities in psychology, ethology, anthropology, linguistics, and the neurosciences. Research equipment and facilities can be provided either through a grant supporting a research project or through a separate grant. The scientific importance and soundness of the research to be conducted, as evaluated by reviewers from the research community, are the main criteria for support. Preparation of proposals is briefly discussed, and sources of further information are noted.     

High-performance computer applications in the behavioral sciences

This symposium revisited the 1985 conference on Advanced Computing for Psychology. That meeting examined the application of new supercomputers in the behavioral sciences. The present symposium reviewed high-performance computing as applied to psychological models, human vision, neuralphysiological processes, and statistical analysis. The recent past and the projected future of high-performance computing in the behavioral sciences were evaluated.     

Algorithms for randomness in the behavioral sciences: A tutorial

Simulations and experiments frequently demand the generation of random numbers that have specific distributions. This article describes which distributions should be used for. the most common problems and gives algorithms to generate the numbers. It is also shown that a commonly used permutation algorithm (Nilsson, 1978) is deficient.     

Ethical and consultation issues in the behavioral sciences and terrorism

This paper briefly examines ethics as a process and as an element of philosophy with a long historical tradition. The paper then reviews elements of the professional ethics of a terrorist or hostage event, issues involved with consulting with an institution, and ethical issues for the clinician who chooses to become personally involved in a hostage event. Although reference is made particularly to psychiatrists, the principles discussed apply in large part to other mental health professionals as well. The paper concludes with the principles articulated by the American Psychiatric Association's Task Force on the Psychiatric Aspects of Terrorism and Its Victims as guidelines for psychiatric intervention in hostage or terrorist situations.     

Mental and physical health of spouse caregivers: the role of personality.

Although personal resources of caregivers, such as coping skills and social support, have been shown to be important in understanding caregiver stress and health outcomes, personality traits have not previously been considered. The purpose of this study was to examine the association between the personality traits of neuroticism and dispositional optimism and mental and physical health outcomes. It was predicted that personality would have direct effects, and indirect effects through perceived stress, on health outcomes. Participants were spouse caregivers of patients diagnosed with Alzheimer's disease. Results showed that neuroticism and optimism were significantly related to mental and physical health. Furthermore, neuroticism had significant direct effects on all of the health outcomes, and substantial indirect effects, through perceived stress, on mental health outcomes. Optimism showed stronger indirect than direct effects on all health outcomes. These findings demonstrate the importance of including personality of the caregiver in theoretical and empirical models of the caregiving process.     

Automated problem solving for the behavioral sciences

In view of the obvious advantages of computers for the behavioral sciences, the question is raised as to how to make more effective use of computing capabilities. One idea is to require that fledgling behavioral scientists receive brief training in flowcharting and algorithm generation rather than a full course in computing skills. Such training would be an important adjunct for those who would go on for further instruction, especially in view of current deficiencies in many computer courses. One recurrent deficiency lies in the lack of specific guidelines for conceptualizing problems for computer implementation. Such guidelines are developed here, along with other suggestions designed to attenuate the difficulty in conceptualization.     

Components of high-performance computing and communication in the behavioral sciences

By its very nature, the spectrum of computational problems approached in the behavioral sciences necessitates the use of almost all elements of high-performance computing and communication. The salient performance issues of various components—networking, disk and tape storage, memoryprocessor configurations, and processing elements—are discussed. Hardware, software, and application issues surrounding parallel computing performance are also discussed.     

Training in law and behavioral sciences: Issues from the criminal justice perspective

Interdisciplinary training in behavioral sciences and the law should be appropriate for a criminal justice education program. However, adopting such an interdisciplinary goal is problematic because the divisions in the types of educational curricula need to be confronted. The relationships of the criminal justice academic community with the profession, and of the behavioral sciences with present criminal justice policies, must also be addressed in the attempt to develop familiarity with behavioral sciences and their legal relevance in shaping the criminal justice system. These factors may hinder extensive implementation of such training. Suggestions are made for incorporating this raining in a criminal justice curriculum.     

Some recurrent issues in the history of behavioral sciences

Many issues seem to have appeared, disappeared, and reappeared in the behavioral sciences during the 20th century. Salient examples discussed in the present paper are consciousness, including the concept itself and consciousness in nonhuman animals; the method of introspection; and cognition, including the interpretation of mental imagery and the role of language in thinking. One possible explanation of the apparent cycles is consistent with a suggestion by John B. Watson: Important issues are found to be intractable and are abandoned, but they recur when newer theories and methods emerge.     

Sexual predator commitment laws: lessons for law and the behavioral sciences

Sex offender commitment laws use a mental health commitment model to lock up the "most dangerous" sex offenders after their prison sentences expire. In Kansas v. Hendricks, the United States Supreme Court rejected the major constitutional challenges to these laws. The Hendricks case clarifies important ambiguities about the use of civil commitment to enforce "police power" interests, as opposed to "parens patriae" interests. Hendricks also clarifies the role of "treatment" in justifying civil commitment. While there remain some important legal issues to be resolved, the future direction of sex offender commitment schemes will turn most significantly on policy decisions. The behavioral sciences can play an important role in shaping these decisions. The most significant questions concern whether expensive commitment programs are the most effective use of scarce treatment and supervision dollars. Additional research should be directed to improving dynamic predictors of recidivism, operationalizing "inability to control" standards, judging the "social meaning" of commitment laws, and assessing the potential treatment disincentives of these laws.     

WolStat: A new statistics package for the behavioral and social sciences

WolStat differs in several important ways from other statistics packages. Data are entered using a text editor, as opposed to a data grid. WolStat performs all possible simple to simple simple simple simple main effects tests and multiple comparison tests for ANOVAs (independent groups, split plot, and randomized block factorial) with up to five factors. More nonparametric statistics can be performed than with most other statistics packages. Latin squares can be randomly generated. Uniformly distributed random integers (with or without replacement) or normally distributed random numbers can be generated. WolStat provides a tutorial and assistance for selecting orthogonal comparisons. WolStat also provides general and specific models for fixed, mixed, and random ANOVAs with one through five factors.