Scientific inquiry in forensic psychiatry.

Forensic psychiatry and psychology, with an extensive history of clinical practice, is only recently emerging as a new scientific specialty. It is confronted with both general scientific-methodological issues, as well as unique, empirically-based psycholegal applications. Aimed towards a broader, more flexible paradigm for understanding the basis of our scientific inquiry, the paper examines various research and methodological issues. As a model for connecting and understanding these complicated and interrelated issues, Holton's explicit and useful model of scientific structure is provided. The methodology, logical and statistical analysis, and themata are explored with regards to their potential impact on scientific inquiry in forensic psychiatry. This paper argues for several considerations in furthering research. From a methodological basis, the prospective researcher has to consider (1) whether the methodology is, in part, predetermining the results; (2) whether the results are a prediction of the inclusion criteria rather than an independent finding; (3) what balance between specificity and generalizability is designed within the research study; and (4) what is the impact of various methodological artifacts on the results and conclusions of forensic psychiatric research (e.g., demand characteristics, evaluation apprehension, experimenter bias, systems influence). Further, the forensic psychiatric researcher is asked to consider the basis of the "validity" of his research findings in comparison with "objective reality" from the perspective of logical and statistical analysis. Scientists within forensic psychiatry are asked to consider (1) whether to employ "best fit" or complementarity in understanding their results; (2) how to make explicit the steps in data transformation and redefinition within their study; (3) what are the implications of exclusive use of null hypothesis testing in establishing research results; and (4) what is the comparative utility of non-parametric and multivariate statistical procedures in studying and understanding experimental variables. Finally, in acknowledging the non-linear and sometimes self-justifying aspect of science, researchers are invited to examine their basic assumptions, and the self-perpetuating and constraining nature of unacknowledged themata, as well as their impact on forensic psychiatry. This paper is conceptualized as a movement towards articulating both general methodological issues and their unique application to forensic psychiatry. The brief exposition of Holton's model and review of illustrative research in forensic psychiatry constitutes one attempt to strengthen the scientific rigor of forensic psychiatric research.