Imprimir Republish

Good practices

Academic legacy in dispute

King's College finds work by one of its most renowned researchers "unsafe"

Veridiana Scarpelli

Hans Jürgen Eysenck (1916–1997), a German psychologist who spent most of his career in the United Kingdom, is known for his work on intelligence and personality, conducted over a period of 28 years as a researcher at the Institute of Psychiatry at King’s College London (KCL). Having written 80 books and more than 1,000 scientific articles, his reputation survived a series of controversies while he was alive, including when he openly sought research funding from the tobacco industry, and when he presented data suggesting that black people and immigrants are less intelligent than white people in the USA. In a list of the 100 most prominent psychologists of the twentieth century, published in the journal Review of General Psychology in 2002 and based on how often their articles and books were cited, he was ranked 24th—the list was topped by B.F. Skinner, Jean Piaget, and Sigmund Freud.

But now, 22 years after Eysenck’s death, a controversial part of his scientific legacy has been disqualified in a report produced by a KCL investigative committee that looked at 25 papers published between 1988 and 2000 by the psychologist and his colleague Ronald Grossarth-Maticek of Heidelberg University, Germany. According to the report, published on the Retraction Watch website in October, the results of the articles are not reliable. The papers were linked to a research program that investigated whether specific personality traits made individuals more prone to cancer and cardiovascular disease, and presented treatments designed to reduce this risk. The KCL committee was unable to obtain the underlying raw data, which was supposedly extracted from population studies that tracked more than 30,000 individuals over a period of 15–20 years in Yugoslavia in the 1960s and 1970s, and in Heidelberg in the 1970s and 1980s. It also found no references to the existence of ethics committee protocols or reports.

It therefore focused its analysis on attempts by German psychologist Manfred Amelang, from Heidelberg University, to replicate the results from the 1990s onwards with investment from German funding agency Deutsche Forschungsgemeinschaft (DFG). For 10 years, Amelang monitored 5,000 Heidelberg residents, tracking psychosocial factors that could increase or reduce the incidence of cardiovascular disease. Among the six personality types classified by Eysenck, no correlation to any disease was found. A variation was detected in one, but later discarded due to a lack of statistical significance.

The central argument of the report, however, is that the results produced by Eysenck and Grossarth-Maticek are not plausible, which was suggested at the time too—to the extent that the findings were strongly contested. The initial criticisms and suspicions were summarized in an article published in the British Medical Journal in 1992 by two psychiatrists: Anthony Pelosi, a professor at the University of Glasgow, Scotland, and Louis Appleby, former national director of mental health for the UK government. The article highlighted omissions and inconsistencies, including the vague description of the methodologies and data selection bias, showing that some results were highly unlikely, such as a 100% greater risk of cancer death among those classed as “emotionally repressed” compared to a control group, and an 80% reduction of this risk when individuals more prone to the disease underwent bibliotherapy, which uses reading to alleviate stress and promote behavioral changes. Despite the criticism, none of Eysenck’s papers were retracted. “These findings are incompatible with modern clinical science and the understanding of disease processes,” says KCL’s report, which was sent to the 11 journals that printed the articles, including Psychological Reports, published by Sage, and Behavior Research and Therapy, published by Elsevier.

The university’s decision to review the work of the eminent researcher came earlier this year after Anthony Pelosi, the psychiatrist who highlighted the inconsistencies in 1992, wrote an article in the Journal of Health Psychology in which he revisited the problems and presented unpublished documents showing how the tobacco industry sponsored several studies by Eysenck and funded Grossarth-Maticek’s population studies in Heidelberg. In 1965, Eysenck questioned early evidence that cigarettes cause lung cancer. Pelosi suggests that there was a union of interests between Eysenck’s theories about how personality traits influence health and the tobacco industry’s efforts to soften the evidence that smoking causes cancer.

An editorial in the same issue of the Journal of Health Psychology by David Marks, a retired professor at City University of London, urged King’s College and the British Psychological Society to investigate Eysenck’s scientific output. According to Marks, the data he obtained is so far from a normal statistical distribution that it can only be the result of error. “The two researchers could rightfully be canonized as ‘Saint Hans’ and ‘Saint Ronald’ for working such miracles if only their claims could ever be proven, which will never happen. To his eternal shame, the attempts by Hans Eysenck to discredit the well-established causal links between tobacco smoking and cancer while in receipt of large sums from the tobacco industry is one of the most shameful deceits committed by any scientist in the twentieth century.” Ronald Grossarth-Maticek, now retired from Heidelberg University, has announced plans to sue Pelosi and Marks for slander. He claims that no proper attempt was made to reproduce the results he obtained with Eysenk, because subsequent experiments did not adopt identical methods.