Monday, 6 March 2017

"Logical fallacies in animal model research"

A paper which is a bit 'out of left field' is presented for your reading pleasure today and how one should be rather careful about how animal research - "with focus on animal models of mental illness" - is translated into relevance to humans [1].

The paper by Espen Sjoberg is pertinent to various diagnostic labels including depression and schizophrenia. I would perhaps disagree with the author including autism under the specific heading of 'mental illness' (bearing in mind various mental health diagnoses can follow an autism diagnosis), but the discussions on how we should all be a little cautious about translating findings from animal research to real people ring as true for autism as they do for the other conditions discussed.

The paper is open-access but a few choice phrases are worth highlighting from the 'what can we do about this ' section of the Sjoberg paper in these days of animal models of [insert condition name here].

So: "Avoid making inferences about the animal’s thoughts, feelings, inner motivation, or understanding of the situation. We can report what the animals did, and what this means in the context of our hypothesis, but take care not to make assumptions of the inner workings of the animal." I note the words 'theory of mind' (ToM) are mentioned in another section of the paper which is interesting and potentially relevant to autism research history...

Onwards: "No matter how validated an animal model is, we cannot be certain that a newly observed effect also applies to humans." The valproate model of autism is specifically mentioned in the Sjoberg text - where prenatal exposure to valproate may have implications for offspring development - but is not the only rodent model of autism (see here). The implication is that one has to be be mindful that just because findings might talk about intestinal inflammation associated with the valproate mouse model of autism for example (see here), this does not make them directly applicable to real people without looking at real people (who were prenatally exposed to valproate). The caveat being: "Remember that the strength of an animal model is to generate new knowledge and hypotheses relevant to the target group, including the assessment of potentially useful treatments, but that these new possibilities are only hypothetical once they are discovered."

Finally: "Replicating an experiment in order to establish interval validity and reliability of an animal model is essential." Replication is a cornerstone of reliable science and so, where possible, when one finds and reports on a specific aspect of an animal model of some label or other, the experiment should be [independently] replicated. The author even goes further, bringing in another couple of 'R' words: reproduction and reconstruction ("A reconstruction involves redesigning an experiment, while maintaining the original hypothesis, in order to accommodate different species").

The bottom line: don't over-hype your [animal research] findings if you've only looked at animal models and are presenting novel findings. In particular, make sure you write any press release mindful of what you were studying and how applicable the results may/may not be (outside of animals)...

Music to close: an old favourite of this blog, Weezer and The Fonz...

----------

[1] Sjoberg EA. Logical fallacies in animal model research. Behavioral and Brain Functions. 2017; 13: 3.

----------

ResearchBlogging.org Sjoberg EA (2017). Logical fallacies in animal model research Behavioral and Brain Functions : 10.1186/s12993-017-0121-8