Intended for healthcare professionals

Rapid response to:

Editorials

Hans Eysenck: controversialist or worse?

BMJ 2019; 365 doi: https://doi.org/10.1136/bmj.l1897 (Published 29 April 2019) Cite this as: BMJ 2019;365:l1897

Rapid Response:

Re: Hans Eysenck: controversialist or worse?

This cautionary tale is as much about hubris as hypothesis.

There is a widely-held belief that hypothesis-based research is superior to the 'fishing expeditions' of hypothesis-free research.

Actually, neither is better than the other.

The problem with hypothesis-based research is that the originator of the hypothesis can all too easily fall into the trap of excessive self-belief even to the point of faking their results. This problem leads directly to the bigger problem that one rotten apple brings the barrel into disrepute.

The problem with hypothesis-free research is that it often in practise isn't hypothesis free. For example, genome-wide association studies and their precursor, linkage studies, are deemed to be hypothesis-free. In the early days this caused them to be turned down by hypothesis-driven research funders.

But of course such proposals do have an unstated hypothesis - that there is some DNA sequence related variation or other that is associated with or causes a particular transmissible phenotype. And if that hypothesis is confirmed to the delight of the grantholder, he or she is terribly likely to fall into the trap of genetic determinism: I have found an association therefore I have found the cause. This too is a case of excessive self-belief.

Good research requires self-belief tempered with self-doubt, a handful or more of dispassionate collaborators and funding structures that do not put excessive pressure on researchers for quick results and that do not make value judgements about the presence or absence of hypothesis.

Competing interests: No competing interests

14 October 2019
Martin A. Yuille
Honorary Reader
The University of Manchester
The University of Manchester