On working through: a model from artificial intelligence

Working through is centrally important to clinical psychoanalysis. It is inadequately explained in analytic theory. An artificial intelligence model of the process is proposed. Models of problem solving show that the complexity of necessary computation is an important determinant of how a problem is solved. Not optimal, but only good enough solutions are usually found. The quality of solutions depends on the time and resources available. Generally it is far easier to use existing methods than to develop new approaches. When problems must be solved in an emergency fashion, as in trauma, poor solutions are likely to emerge. In studying the annealing of metals and other complex optimization problems, a process, the Boltzman algorithm, was discovered, which continues the search for better solutions while gradually developing a coherent structure of the overall solution. The algorithm provides a model both for psychoanalytic working through and for the normally ongoing process of psychological development and reworking whose deficiency is characteristic of much psychopathology. Working through in the analytic situation is the reactivation of this normal process, and a good analytic outcome is achieved when the process can continue without the analyst. Properties of the Boltzman algorithm clarify such concepts as "optimal" frustration and anxiety which correspond to working in the area where the stable but not rigid structures emerge in the algorithms operation. These studies are an example of how computer science and artificial intelligence are a potentially rich source for psychoanalytic theory.     

Validity of the Millon Clinical Multiaxial Inventory in the assessment of affective disorders

The Millon Clinical Multiaxial Inventory (MCMI) was administered to 270 adult outpatients with major affective disorders at the same time that a semistructured, diagnostic interview was conducted by a clinician. The mood of the patient was then rated, and the clinician completed the Hamilton Depression Rating Scale and the Mania Rating Scale. A consensual diagnosis was arrived at by the team of investigators using DSM-III criteria. Significant correlations were found between four MCMI affective scales and the global mood state of the patient. Analysis of covariance indicated that the MCMI affective scales are significantly related to DSM-III affective disorders even after the effect of the current mood of the patient is partialled out. The clinical usefulness of each of the scales is discussed.     

A comparison of two measures of egocentrism

This study examined the relationship between two prominent measures of egocentrism, Enright's Adolescent Egocentrism Scale (AES) and Elkind's Imaginary Audience Scale (IAS), in a sample of 458 adolescents, between 10 and 15 years old. Both correlational and factor analyses indicated that the two measures assess distinctly different phenomena. As predicted, IAS scores were correlated with self-reported levels of shyness, nervousness, and social skills. These results provide partial evidence for the validity of the IAS as a measure of self-consciousness, though not necessarily egocentrism per se. The validity of the AES was not examined.     

Gender-bound definitions of mental health

Fifteen male and 15 female American therapists-in-training (clinical and counseling psychology graduate students) were asked to take the Minnesota Multiphasic Personality Inventory (MMPI) under each of two instructional sets. In one set, they were instructed to respond to the items as a healthy male would respond, and in the other, as a healthy female would respond. The MMPI profiles obtained from male and female subjects were not significantly different, indicating that these male and female therapists-in-training did not differ in their perceptions of healthy men and women. When the data for male and female subjects were combined, however, healthy women were perceived differently than healthy men on several scales, although the MMPI profiles obtained under both instructional sets were well within normal limits.     

Comparison of response formats for Multidimensional Health Locus of Control Scales: six levels versus two levels

Two formats of the Multidimensional Health Locus of Control (MHLC) Scales were administered to 54 college students. Each subject completed the MHLC Scales in the standard 6-level response format (ranging from strongly disagree to strongly agree) and in a revised 2-level format (ranging from disagree to agree). Comparisons of internal consistency measures, principal components, and classification of subjects into groups indicate that the 2-level response format yields comparable data to those obtained with the 6-level format, particularly when classification of subjects is the goal.     

The MMPI and clinical diagnosis: a comparison of classification system outcomes with discharge diagnoses

This study examined the agreement or congruence rate between clinical-discharge diagnoses rendered by a psychiatrist, and admission and discharge MMPI-derived diagnoses from four diagnostic classification systems that have been developed for the MMPI. The four classification systems included a simple high-point code based on the most elevated clinical scale in the profile, the Henrichs revision of the Meehl-Dahlstrom rules, the Goldberg equations, and a system developed by Lachar. Subjects consisted of 150 patients selected from a larger pool of patients who had completed a 9-week adult residential treatment program. Overall, this study yielded modest hit rates between 26% and 34% for MMPI-derived diagnoses and psychiatric diagnoses across the various classification systems. In addition, stability of MMPI-based diagnoses from admission to discharge assessments ranged from 48% to 51% depending on the classification system employed. Findings are discussed in terms of their implications for the use of the MMPI in patient diagnosis. It is recommended that the MMPI be used in conjunction with other sources of clinical and test information in deriving clinical diagnoses.