In Science and Sanity Korzybski wrote:
"Thus Doctors Wilder D. Bancroft, J. Holmes Richter, H. Beckett Lang, John A. Paterson, Walter Freeman, and others demonstrated that it is possible to find a correlation between the functional psychoses and the state of dispersion of the nerve colloids. For instance, in dementia praecox the nervous system appears in a state of colloidal over-dispersion; and in manic depressive psychoses, in a state of decreased dispersion." (Science and Sanity, p511)
Bancroft's work turned out to be an example of how strong beliefs among scientists can lead them to find proof of their beliefs that ultimately turns out to be illusory. We see what we believe and then confuse the orders of abstraction. Within a year after the publication of Science and Sanity more rigorous experimental studies were published that discredited Bancroft's theory and supporters such as Lang and Freeman abandoned it. In the New York Journal of Medicine Lang wrote:
(From The Rise and Decline of Colloid Science in North America by Andrew Ede, 2007, p.163)
In his 1937 book "The Mentally Ill in America" historian Albert Deutsch wrote:
Not many years ago Professor Wilder D. Bancroft of Cornell University formulated a theory that all functional psychoses are due to disturbances of the brain proteins, and that the specific type of psychosis produced depends upon whether the proteins are "over-coagulated" or "over-dispersed." (Coagulation is the process of thickening a colloid; dispersal is the process of thinning.) Bancroft was inspired mainly by experiments designed to test a theory advanced in 1875 by the famous physiologist, Claude Bernard, to the effect that "anaesthesia is due to the reversible coagulation of some of the proteins of the brain and of the sensory nerves. Unconsciousness results from the blocking of the centers of consciousness, and insensitivity to pain from the blocking of the sensory nervous system." It is Bancroft's belief that mental disorders may be cured by sending coagulation or dispersing agents to the brain through the blood stream. Coagulation of the brain proteins causes excitement, he maintains, while dispersion of the brain proteins has a sedative effect. Bancroft and his associates divide the psychoses into two classes; one type is improved by administering sodium amytal (a coagulating agent) and aggravated by sodium rhodonate (a dispersing agent), while the process is exactly reversed in the other type of psychosis. Experiments made by the United States Public Health Service in 1932 to test Bancroft's theory produced negative results, and it has not found a place in general psychiatric thought and practice. (2nd edition, 1946, p476)
The foreword to Deutsch's book was written by none other than Korzybski's psychiatric mentor, William Alanson White, superintendent of St. Elizabeth's Hospital, Washington D.C., and was republished in 1946 with the same negative conclusion regarding psychosis and colloidal dispersion. Nevertheless as late as his December 1948 seminar Korzybski continued to assert unequivocally that dementia praecox was a disorder of "over-dispersed colloids." (Lecture VI, 29 Dec. 1948, Tape 11B)
By the 21st century drug therapy for schizophrenia had advanced considerably. Explanation of the mechanism of action of these drugs had shifted from ascribing their effects on the dispersion of brain colloids to their interaction with specific brain colloids, especially membrane protein receptors for neurotransmitters such as dopamine. The latest theories of the genesis of schizophrenia posit that it results from overpruning of synaptic connections in the brain during late adolescence and early adulthood.