The limits of empirical studies on research ethics

The results of empirical research in psychology and psychiatry are increasingly being used to formulate as well as understand problems at the interface of law and psychiatry. There has been a proliferation of studies, such as the determinants of individual competence or threat to self or others, the results of which are influencing policy and legislative decisions as well as buttressing holdings in court cases. In this article, I explore the issues of interpretation of epidemiological studies, particularly the role of ideological positions on the design and results of empirical findings, the importance of the way data are interpreted, and the role of ideologies in the way research findings are presented to provide support for policy positions. Two levels of analysis are involved in determining the validity of a study. The first addresses the questions of whether the study meets the statistical and epidemiological requirements for reliable results. These include considerations such as the appropriateness of the study design and methods for gathering and interpreting data. The second focuses on the underlying framework of the study. This involves factors such as the perspectives and values of those conducting the study, the explicit and implicit dominating ideologies where they operate, and the extent to which the study is constructed to reaffirm specific ideologies. This level of analysis is essential for disclosing the influences of ideologies on the results of studies and the way in which data are interpreted. In this article, I try to demonstrate through critiques of selected studies that the first stage of analysis is insufficient without an examination of underlying preconceived values to establish the meaningfulness of results.     

Extraction of relief from visual motion

We quantified the ability of human subjects to discriminate the relative distance of two points from a slanted plane when viewing the projected velocities of this scene (orthographic projection). The relative distance from a plane (called relief) is a 3-D property that is invariant under linear (affine) transformations. As such, relief canin principle be extracted from the instantaneous projected velocity field; a metric representation, which requires the extraction of visual acceleration, is not required. The stimulus consisted of a slanted planeP (specified by three points) and two pointsQ ₁ andQ ₂ that are non-coplanar withP. This configuration of points oscillated rigidly around the vertical axis. We have measured thesystematic error andaccuracy with which human subjects estimate the relative distance of pointsQ ₁ andQ ₂ from planeP as a function of the slant of_P. The systematic error varies with slant: it is low for small slant values, reaches a maximum for medium slant values, and drops again for high slant values. The accuracy covaries with the systematic error and is thus high for small and large slant values and low for medium slant values. These results are successfully modeled by a simple relief-from-motion computation based on local estimates of projected velocities. The data are well predicted by assuming (1) a measurement error in velocity estimation that varies proportionally to velocity (Weber’s law) and (2) an eccentricity-dependent underestimation of velocity.     

Controlled attention sharing influences time estimation

A seminal attentional model of time estimation predicts that subjective duration will be positively correlated to the amount of attention given to temporal processing. This prediction holds under prospective conditions, in which one is forewarned that judgments of time will be asked, in contrast to retrospective conditions, in which such judgments are required after the relevant period without any prior warning. In three experiments, an attention-sharing method was used. Subjects were asked to control the amount of attention that they devoted to one or the other component of a dual-task paradigm. The first experiment involved word categorization and reproduction of duration. The following experiments, based on signal detection theory, required discrimination of both the duration and the intensity of a single stimulus, in the visual (Experiment 2) or the auditory (Experiment 3) modality. The results indicate that when the attention is directly controlled by the subject, the subjective duration shortens as the amount of attention devoted to the temporal task diminishes. The implications of these results for the possible existence of an internal timer are considered.     

The dynamics of Prägnanz

Summary There is quite wide-spread agreement about the relevance of pattern Pragnanz (Koffka, 1935) with respect to the human interpretation of visual patterns. There is less agreement about whether pattern Pragnanz is based solely on pattern information (static) or also on the history of the perceiver (dynamic). In Van Leeuwen and Van den Hof (1991), experimental data concerning serial patterns are presented within the framework of the dynamic-network approach initiated by Buffart (1986, 1987). These experimental data are claimed to give evidence against the static-coding approach initiated by Leeuwenberg (1969, 1971). In the present paper, however, I show first that Buffart's theoretical basis is incorrect, and that in fact Leeuwenberg's static-coding approach is the basis for the dynamic-network approach. Second, I show that those experimental data rather give evidence in favor of the static-coding approach, by using those same data for a test of the most recent static-coding model (Van der Helm & Leeuwenberg, 1991; Van der Helm, Van Lier, & Leeuwenberg, 1992). Finally, I propose a reconciliation between the two approaches, in the sense that the dynamic-network model could be shaped in such a way that it yields a simulation, and maybe even an enrichment, of the static-coding model.     

A compensatory approach to optimal selection with mastery scores

A Bayesian approach for simultaneous optimization of test-based decisions is presented using the example of a selection decision for a treatment followed by a mastery decision. A distinction is made between weak and strong rules where, as opposed to strong rules, weak rules use prior test scores as collateral data. Conditions for monotonicity of optimal weak and strong rules are presented. It is shown that under mild conditions on the test score distributions and utility functions, weak rules are always compensatory by nature. The authors are indebted to Wilbert Kallenberg for his valuable comments and to Jan Gulmans for providing the data for the empirical example. The names of the authors are alphabetical; they are equally responsible for the contents of this paper.     

A Grammatical Specific Language Impairment in Children: An Autosomal Dominant Inheritance?

The aim of this study is to provide further characterization of a subgroup of so-called “Grammatical specific language-impaired (SLI)” children. The Grammatical SLI children have a persistent and disproportionate impairment in grammatical comprehension and expression of language. Previous research has indicated that their language impairment may be characterized by a domain-specific and modular language deficit. This study provides an initial investigation as to whether there is a genetic basis underlying their disorder as has been found for other forms of SLI and for SLI in general. The incidence of familial aggregation of language impairment was investigated in 12 Grammatical SLI children (aged 9:3 to 12:10). A familial language impairment (LI) history was classified as positive if one or more of the probands' relatives had a history of a speech/language or reading/writing problem which required speech therapy or any other form of remedial help. Case history information provided an initial indication that the Grammatical SLI children had a significantly higher incidence of a positive familial LI history than could be expected by chance. A questionnaire provided evidence of a positive LI history in the first-degree relatives of the SLI probands and 49 normally developing control probands. The SLI probands had a clearly and significantly higher incidence of a positive familial LI history than the control probands (77.8 vs. 28.5%, respectively). The results are consistent with a genetic basis underlying Grammatical SLI. The pattern of impairment in the SLI probands' relatives is consistent with an autosomal dominant genetic inheritance. In contrast to the control probands, the SLI probands' impaired relatives did not show a male gender bias. Thus, the gene does not appear to be sex-linked. The data indicate that further research is warranted to investigate the nature of the LI in the relatives of the Grammatical SLI probands and the genetic characteristics of this subgroup. The implications for the biological, domain-specific, and modular bases to language are discussed.     

Effects of diagnostic classification systems on clinical hypothesis-generation

In this study it was determined whether (a) classification as opposed to absence of classification has an effect on the quality of clinical hypotheses (b) the DSM-III-R and the CBCL have a different effect on the quality of clinical hypotheses, and (c) the potential difference between the DSM-III-R and the CBCL is moderated by the different number of syndromes identified by these systems. To investigate these questions, an experiment was conducted in which 86 clinicians generated hypotheses for six cases. The clinicians were divided into a DSM-III-R, a CBCL, and a control group. Of the six cases, two were classified by both classification systems as one syndrome, two were classified as one syndrome by the DSM-III-R but as two syndromes by the CBCL, and two were classified as two syndromes by the DSM-III-R but as one syndrome by the CBCL. The quality of the hypotheses was determined by means of four dependent variables selected from an overview of qualitative criteria: explanatory value, redundancy, possibility of operationalization, and specificity. No differences between the CBCL and the control groups were found. The DSM group performed better than the control group regarding explanatory value and redundancy. The DSM-III-R group also scored better than the CBCL group regarding explanatory value, particularly when the number of identified syndromes was two for the CBCL and one for the DSM-III-R.