What’s the link between obesity and dishonesty?
If that question seems preposterous on its face, you’re probably among the critics of a 2020 paper in Scientific Reports which claimed to find that obese people were more deceptive than thinner folk.
The researchers, led by Eugenia Polizzi di Sorrentino, of the Institute of Cognitive Science and Technologies at the National Research Center in Rome:
explore[d] the link between energy, obesity and dishonesty by comparing the behaviour of obese and lean subjects when hungry or sated while playing an anonymous die-under-cup task.
They found that:
only lean females lie less when sated. By contrast, obese subjects lie more than lean subjects in both conditions, and they lie more to avoid the lowest payoff than to get the highest payoff. Our findings suggest that the observed patterns are more likely mediated by factors associated with obesity than by short term energy dynamics, and call for a better integration of the psychological, economic and biological drivers of moral behaviour.
The paper earned a few derisive comments on the journal website. One stated:
The golden age of phrenology feels positively enlightened compared to this.
Another made a deeper cut:
Unfortunately, while the author’s work is an interesting exercise, there is a fundamental flaw in their methods that render the work, and their conclusions, completely invalid. While subjects were screened for “severe illnesses (such as diabetes), psychological disorders, pregnancy and dieting” – a detailed sleep history does not appear to have been obtained. As up to 80% of patients suffering from obstructive sleep apnea may go undiagnosed (https://www.ncbi.nlm.nih.go…, the likelihood that a significant percentage of the obese subject group in this study had an underlying sleep pathology is a virtual certainty. All of the cognitive changes described in this paper, including increased impulsivity leading to the “increased dishonesty” as measured, are well-known effects of the chronic sleep deprivation created by sleep-disordered breathing.
While it is not surprising that the authors made such an error of omission in their subject screening (sadly a common occurrence), given that said omission completely invalidates the work, it is disappointing the reviewers and editors didn’t catch it. Under the circumstances the paper should be retracted.
The article also drew a blizzard of criticism on PubPeer. As Gideon Meyerowitz-Katz, an Australian epidemiologist who Tweets as “Health Nerd,” posted, the paper suffers from multiple flaws:
In fact, it appears as if the main finding of this paper completely contradicts the results of this analysis. There does not appear to be any statistically significant differences in the reported values when comparing lean and obese people at all!
Or, as Yoni Freedhoff, a family medicine physician in Canada, put it:
I realize I’m swearing a lot today, but how the fucking fuck did this clear ethics? Or get published?
We emailed Polizzi for comment but have yet to hear back.
We mentioned the paper in a post on August 6. On August 10, the paper was amended with the following notice:
Readers are alerted that the methods and conclusions of this paper are subject to criticisms that are being considered by editors. Further editorial action may be taken as appropriate once the investigation into the concerns is complete and all parties have been given an opportunity to respond in full.
On August 11, the Obesity Action Coalition wrote to the editors of the journal, a Springer Nature publication, requesting that the paper be retracted.
A spokesperson for Scientific Reports told us:
We are continuing to investigate the concerns raised regarding this paper.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].
I’m so glad that you two are so “woke”. It’s really important that science not hurt anyone’s feelings with its horrible results. Keep up the good work, and good luck with your own publishing careers too! I know that neither of you have had many papers of your own published, but keep trying: eventually your hard work will pay off!
I review lots of papers on obesity. I notice that when the paper shows that something or other (anything) is “bad” about obesity, fellow reviewers are extremely lenient, as if they are ready to believe anything that goes along with their negative stereotypes, and fail to notice major problems. Some such ‘anti-obesity’ papers with egregious errors have made their way into prestigious journals.
My late professor’s (a great statistician) advice: When strange results appear, first question your data, your methodology and your own mind.
Due to the magic of numbers and ease of software it is easy to connect any two dots these days. One can also empirically show that existence of this site leads to bad humor, but is that really correct? People who enter science have to learn logic, argumentation and causality.
The “sleep apnea” hypothesis doesn’t “completely invalidate” the findings; it just presents a potential mediator that may explain the relation between obesity and the outcome. It may mean the current study is incomplete (as are all studies), but it doesn’t mean the results are wrong.
There may be other issues with the analysis and interpretation of course, but I don’t find this argument useful.
I find the entire discussion more sentimental than scientific. The association of many behavioral traits with obesity has been published earlier. The behavior addressed here has a moral implication and that appears to be the cause of the unrest. However, the experimental results only show association and not causation. The authors use a language that has a causal connotation which should have been avoided. Also human behavior is extremely contextual and we need not expect the same behavior in every context. So while we should respect the experimental results, over-interpretation should be avoided. The experiments only show that a given sample, in a given context demonstrated association of a behavioral trait with obesity. This does not mean that obese people are always dishonest, or obesity causes dishonesty. Rather than rejection or retraction, a “weird” result should stimulate more attempts to test the reproducibility and study the context dependence of the results. Also rather than over-interpreting to generalize the inference or prematurely inferring causation, attempts to investigate the causal relations should be the correct response to the provocating results.
I just want to state for the record that to make the conclusion that “obese people lie more”, it is irrelevant whether or not obese people are more likely to lack sleep. That only matters when you are trying to get at causality. But if you are strictly interested in correlation, then it is irrelevant and you don’t need to control for it.
The real point is that in fact the data do not show any significant results between lean and obese anyway. Read the PubPeer comments for details. It seems fairly likely that the authors are reporting a chance observation (not statistically significant anyway) that wasn’t the original purpose of the study. This supports the idea that this was not carefully reviewed to begin with.