Samuel Gershon: Events and Memories
9. Model Psychoses: Amphetamine

I came to NYU in early 1963 from St. Louis and quickly saw that Bellevue Hospital had a very different mix of patient populations. The most conspicuous groups were drug users and abusers in large numbers (Angrist & Gershon 1967 a, b).

The variety of agents was almost limitless and hallucinogenic properties were attributed to many strange products. So we had a dilemma; what substances produced abnormal behavioral effects and were they due to an active substance or group belief and individual expectations?

In about 1964, a young psychiatrist came to see me after training with Don Klein. In his introduction, Burton Angrist told me he had learned a great deal of psychopharmacology from Dr. Klein but was looking for an area of study that was different and more in line with the clinical populations he was interested in. This turned out to be the psychopharmacology of psychotomimetics and drugs of abuse, including the fads that changed from year to year in the addict population. So we mutually agreed to this area for study, believing it was both important and could contribute to scientific knowledge.

We started with a simple question; what was myth and reality among drug abusers? Burt noted a popular practice in the culture at that time was smoking the baked inner white layer of banana peel. That was a real conundrum to begin with. Surveys of users recorded its purported hallucinogenic effects so he had to find colleagues who could determine the chemical nature of the constituents. At NYU that was Arnold Friedhoff, Schweitzer and me. We were very good at chromatographic analysis and the sacrifice of many bananas was the clear answer. There was no chemically active product and what we found was mostly carbon, a cheap and safe pharmacologically inactive substance (Angrist, Schweitzer, Friedhoff, Gershon 1967).

This small experiment taught us a lot about mass drug abuse and the reliability of reported effects. In this case, they were the expectations of guru users passed on to their acolytes. The sobering results caused Burt to examine compounds with a better provenance.

Burt’s next step was to examine the pattern and effects of amphetamine abuse in New York City (Angrist and Gershon 1969). A cross sectional picture revealed its use was widespread and in some cases devastating. We heard about many psychotic features upon interviewing those who mixed drugs with alcohol. Many were unemployed, including some physicians who had lost their license. This group with their impairments formed the population for the next phase of our work, including carefully planned clinical and chemical studies that yielded over 30 publications, mostly in five years (1967-1972), among them clinical findings, (Angrist and Gershon 1970;  Angrist, Schweitzer, Friedhoff, Gershon, Hekimian, Floyd 1969; Angrist, Schweitzer, Gershon, Friedhoff 1970; Angrist, Shopsin, Gershon and  Friedhoff, 1970) and basic science studies in animals predominantly examining amphetamine, its stereoisomers (Angrist and Gershon, 1971; Angrist, Shopsin and Gershon 1978; Wallach, Friedman and Gershon 1972) and its metabolites (Angrist, Schweitzer, Friedhoff and Gershon 1970; Angrist, Shopsin, Gershon  and Wilk, 1972; Schweitzer, Friedhoff, Angrist and Gershon 1971).

This body of work demonstrates what a principal investigator with a clear concept can accomplish in making a truly translational impact on neuroscience and clinical research.

Background Events and Findings

It was well known for many years that abuse of amphetamines could produce psychotic episodes (Angrist and Gershon 1967a, b). This was demonstrated more dramatically during the Second World War when the Japanese military used amphetamine for increasing alertness and working longer hours. After the war ended, we gained considerable evidence that psychoses sometimes resulted and most recovered fully. In the English literature, there were reports that some psychoses resembled schizophrenia, while others insisted they were more like a delirium or a disorganized hyperactive state, akin to mania.

This was a primary question in the first stage of our studies. Burt began by admitting such patients to our research ward and recording their symptoms and behavioral profile until aberrant features subsided. He was a superb, sensitive and highly observant psychiatrist, who carefully noted the changing clinical state over the duration of their stay. We found the patterns fitted well with the different clinical pictures observed and reported in schizophrenia. We decided to test this hypothesis by comparing the profile we saw in nature with that we obtained in a research setting (Angrist and Gershon 1970). These experiments were essential because many investigators held that the symptoms of schizophrenia were different from the amphetamine-induced behaviors observed in addicts. After experiments with different doses of amphetamine over varying durations, Angrist and I concluded that, compared with the spectrum of naturally occurring schizophrenic disorders, administration of amphetamine to normal subjects produced pictures that were indistinguishable.

Once we had demonstrated this finding, we decided it would be profitable to examine the underlying pharmacological effects of amphetamine.

Amphetamine blocks the reuptake of dopamine and to a lesser extent norepinephrine and serotonin. Our data suggested that antipsychotic drugs, which also block dopamine receptors, would antagonize amphetamine-induced psychotic states. We used haloperidol to mitigate or terminate schizophrenic symptoms in our amphetamine induced psychotic subjects; the findings demonstrated a similar antagonistic effect but the blocking effect was more rapid and greater in degree for drug-induced cases compared to that seen in chronic schizophrenic patients. Discussion persists as to whether you need a predisposed substrate for amphetamine to induce a psychosis or if it is being produced de novo with a restituto ad integrum on full elimination of the drug. Our data supports the latter assumption. Patients with a sub-syndromal profile of schizophrenia seem to be sensitive to sub-threshold doses that produce a psychosis.  

Another use of this research was to employ amphetamine induced effects in animals as a screen for psychoactive drugs. Amphetamines produce motor hyperactivity in rats and mice and this behavior can be used to screen for antipsychotic and anti-manic agents, although this does not apply reliably to all compounds. For example, it is not modified by lithium or valproate, only by most antipsychotics and sedatives. Therefore, we extended our studies to amphetamine-induced stereotopy in rodents to dogs (Angrist and Gershon 1972; Rotrosen, Wallach, Angrist and Gershon 1972; Wallach, Angrist and Gershon 1971) and cats (Wallach and Gershon 1972) where many antipsychotics antagonized this behavior. Clozapine, an atypical antipsychotic, antagonized stereotypy but not hyperactivity. This discrepancy, together with differences on dopamine effects, led others to suggest clozapine was not an effective antipsychotic. In our own early clinical studies on clozapine, we found it a very interesting and effective antipsychotic. It was also unique with a very low incidence of EPS and a therapeutic action on treating tardive dyskinesia caused by other antipsychotics. Clozapine use has suffered from the dispute over its different effects on dopamine neurotransmission but this has been shown irrelevant to its therapeutic activity.

All the studies in this broad arena were created by a talented group of scientists working collaboratively in a supportive academic setting. This form of interaction is less frequent now and I believe science has suffered a loss.


Angrist BM, Gershon S. Drug-induced psychoses. Part 1. Hospital Practice1967; 11: 36-39.

Angrist BM, Gershon S. Drug-induced psychoses. Part 2. Hospital Practice 1967; 11: 50-53.

Angrist BM, Schweitzer J, Friedhoff AJ, Gershon S. Banana smoking: Chromatographic analysis of baked skins. NY State J.Med 1967; 15: 2893-2895.

Angrist BM, Gershon S. Amphetamine Abuse in New York City 1966-1968. Seminars in Psychiatry 1969; 196-207.

Angrist BM, Schweitzer J, Friedhoff AJ, Gershon S, Hekiminian L, Floyd A. The clinical symptomology of amphetamine psychosis and its relationship to amphetamine levels in urine. Int. J. Pharmacopsychiat. 1969; 2: 125-139.

Angrist BM, Gershon S. The phenomenology of experimentally induced amphetamine psychosis; preliminary observations. Biol. Psychiat. 1970; 2: 95-107.

Angrist BM, Schweitzer J, Gershon S, Friedhoff AJ. Mephenteramine psychosis; misuse of the Wyamine inhaler. Am. J. Psychiat. 1970; 126: 1315-1317.

Angrist BM, Schweitzer J, Friedhoff AJ, Gershon S. Investigation of P-methoxyamphetamine excretion in amphetamine induced psychoses. Nature 1970; 225: 651-652.

Angrist BM, Gershon S. A pilot study of pathogenic mechanisms in amphetamine psychosis utilizing differential effects of D and L amphetamine. Pharmacopsychiatrie Neuro Psychopharmacologie 1971; 4: 65.

Angrist BM, Shopsin B, Gershon S, Wilk S, Metabolites of monoamines in urine and CSF after large dose amphetamine administration. Psychopharmacologia 1972; 26: 1-9.

Rotrosen J, Wallach MB, Angrist BM, Gershon S. Antagonism of apomorphine induced stereotypy and emesis in dogs by thioridazine, haloperidol, and pimozide. Psychopharmacologia  26: 1972; 185-194.

Schweitzer J, Friedhoff AJ, Angrist BM, Gershon S. Excretion of P-methoxyamphetamine administered to humans. Nature 1971; 229: 133- 134.

Wallach MB, Angrist BM, Gershon. The comparison of stereotyped behavior-inducing effects of D and L amphetamine in dogs. Comm. Behav. Biol. 1971; 6: 93-96.

Wallach MB, Friedman E, Gershon S. 2,5-dimethoxy-4-amphetamine (DOM), a neuropharacological examination. J.Pharmacol. Exp. Therap. 1972; 182: 145-154.

Wallach MB, Gershon S. The induction and antagonism of CNS stimulant-induced stereotyped behavior in the cat. Eur. J. Pharmacol. 1972; 18: 22-26.


Samuel Gershon
December 3, 2015