Criticizing previous research, especially K. Goldstein's work with schizophrenic and brain-damaged patients, for methodology reflecting a mechanistic view of mental functioning, the author presents cogent ideas that are the basis for his ongoing research. Artificial dichotomies that lead to stilted, rigid thinking and thought experimentation are questioned and found lacking. Concrete-abstract, perceptual-conceptual, and projective-cognitive dichotomies imply an either-or approach which loses sight of the observation psychoanalysts make daily in their clinical work: that particular private associations can always interfere to some extent in the selection of essential attributes. The impact of personality must be considered in any study of concept formation, and a systematic investigation of the hierarchy of attributes is needed, especially with regard to normal subjects. An elucidation of how hierarchies are organized—i.e., the shifting orders and classes used as bases for abstraction of attributes of objects—would touch upon a forefront of psychoanalytic ego psychology, the psychology of thinking.