IN MEMORIAM

IN MEMORIAM

Kenneth Mark Colby

Professor of Psychiatry, Emeritus

Los Angeles

1920—2001

Fifty years ago there was only one psychiatrist thinking about the ways in which computers could contribute to the understanding of mental illness. Kenneth Mark Colby was that psychiatrist. Thus began a project that lasted until his death in 2001.

 

Kenneth Colby was born in Waterbury, Connecticut. He graduated from Yale University in 1941 and from Yale Medical School in 1943. He practiced psychoanalysis for the first several decades of his career, and was clinical associate at the San Francisco Institute of Psychoanalysis in 1951 when he published A Primer for Psychotherapists, a small book of elementary principles which many still regard as the finest introduction to psychodynamic psychotherapy ever published. But Colby became disenchanted with psychoanalysis because, in his view, it failed to satisfy the most fundamental requirement of a science, that being the generation of reliable data: “Reports of clinical findings are mixtures of facts, fabulations, and fictives so intermingled that one cannot tell where one begins and the other leaves off. …we never know how the reports are connected to the events that actually happened in the treatment sessions, and so they fail to qualify as acceptable scientific data.” In 1961 he spent a year as a Fellow at the Center for Advanced Study in the Behavioral Sciences, where he developed several of the ideas that were to inform the rest of his career. Among these was the conviction that computer models of the mind promised a more scientific approach to the study of cognitive processes and their aberrations. Following this conviction, he joined the Department of Computer Science at Stanford University in the early sixties, and soon became a pioneer in the emerging field of artificial intelligence. In 1967 the National Institute of Mental Health recognized his research potential when he was awarded a Career Research Scientist Award. At the Stanford Artificial Intelligence Laboratory, Colby created a natural language program called “Parry” that simulated the thinking of a paranoid individual. This thinking entails the consistent misinterpretation of others motives – others must be up to no good, they must have concealed motives that are dangerous, and their inquiries into certain areas must be deflected - which Parry achieved via a complex system of assumptions, attributions, and “emotional responses” triggered by shifting weights assigned to verbal inputs. This program was the first to pass the “Turing Test” (named for the British mathematician Alan Turing, who defined as “intelligent” any computer that could successfully impersonate a human in a typed “conversation”). Parry did so in the early seventies, when human interrogators, interacting with the program via remote keyboard, were unable with more than random accuracy to distinguish Parry from an actual paranoid individual.

 

Professor Colby came to UCLA as a professor of psychiatry in 1974, at the invitation of Jolly West, M.D., then chair of the Department of Psychiatry and director of the Neuropsychiatric Institute. He was jointly appointed professor in the Department of Computer Science a few years later, and continued to work on the theory and application of artificial intelligence in neuropsychiatry. One project entailed combining a Cromemco computer with a Votrax voice synthesizer into Intelligent Speech Prosthesis. This system, which ultimately utilized an extremely early “notebook computer”(built at UCLA), allowed individuals suffering from aphasia to “speak” by helping them search for and articulate words using whatever phonemic or semantic clues they were able to generate. During his tenure at UCLA Colby also published important theoretical works on the unreliability of diagnosis in psychiatry (Colby & Spar, 1983) and on cognitive science and psychoanalysis (Colby & Stoller, 1988). He was one of the first to appreciate the possibilities of computer-assisted psychotherapy, and with his son Peter created a program called “Overcoming Depression,” which included a natural language component that administered a version of “cognitive-behavioral” therapy for depression. He and Peter formed Malibu Artificial Intelligence Works in 1989, soon after Professor Colby retired from full-time academic life, and they continued to develop and market “Overcoming Depression” until Professor Colby’s death at age 81, on April 20, 2001.

 

Colby was a rigorous scientist and a creative and original thinker, but one who always recognized his debt to his predecessors, “We are always aided and burdened by our predecessors’ knowledge. Thinking is furthering the thinking of others.” He also was an excellent chess player, and in 1979 published a well-received book titled Secrets of a Grandpatzer, which was recommended by candidate master Daniel Waite for “pattern recognition and having fun” (The Chess Journal, February, 2001).

 

Besides his son, Peter, of Malibu, he is survived by his wife of 43 years, Maxine Hansbold Colby, a daughter, Erin Johnson of Santa Maria, California, and two grandsons.

 

James E. Spar

Michael T. McGuire