Home // Critical thinking // Coffee and Hallucinations … RLY?

Coffee and Hallucinations … RLY?

I have a few things to blog about, but will do my best to pace myself. Today I’ll have a whinge about a “study” (scare quotes intentional) from La Trobe University in Melbourne that purports to show that a high intake of coffee increases a person’s propensity to experience hallucinations. This has of course been taken up by the news media. It’s the perfect medical scare story: everyone’s favourite drug makes you go crazy.

Pleh, I say. And also bah.

First red flag: this is not yet published, but Professor Crowe is giving press releases and interviews to the general media. Not an especially auspicious sign.

So, I can’t check out the details of what was done, including any statistics. I can however comment on what they’ve said in their press release about the experiment:

“The participants were assigned to either a high or a low stress condition and a high or a low caffeine condition on the basis of self-report.”

It gets worse. They didn’t administer caffeine, or even measure intake or serum caffeine levels. Perhaps they asked how many coffees the participants drank, but different types of coffee (brewed, espresso, instant…) differ in their caffeine content, and even with the same sort of coffee, different baristas will give different results (for example, I’d lay good odds there’s a lot more caffeine in one shot from me than in a double shot from almost any cafe in which I’ve observed the barista). There are no details about how they determined whether the person was subject to a high or low stress condition – so we can basically almost discount the stated difference between the groups in the “study”, before we even really begin.

“The participants were then asked to listen to white noise and to report each time they heard Bing Crosby’s rendition of “White Christmas” during the white noise. The song was never played. The results indicated that the interaction of stress and caffeine had a significant effect on the reported frequency of hearing “White Christmas”. The participants with high levels of stress or consumed high levels of caffeine were more likely to hear the song.”

So. Maybe interesting. Not hallucinations though. Not without a lot more detail. At most I suppose we could call them secondary hallucinations (an hallucination triggered by a real perception), but my money’s on illusions/misperceptions. This was done under the guise of a hearing test, and the subjects were led to think White Christmas would be played. Therefore they were listening for it. It is perhaps interesting that (possibly) more highly stressed or caffeinated individuals were more likely to think they heard something, but it expressly does not mean…

“This study also helped to explain the mechanism by which stress may facilitate the symptoms of schizophrenia in non-clinical samples. Caffeine has only recently been reported to increase proneness to hallucinate. ‘The results also support both the diathesis-stress model and the continuum theory of schizophrenia in that stress plays a role in the symptoms of schizophrenia and that everyone, to some degree, can experience these symptoms. This was demonstrated by a significant effect of stress on the occurrence of hallucinatory experiences, or hearing the song.”

Absolute. Rot.

There is no basis – from what has been reported (and only in their press release and news media) – on which to extrapolate illusions when primed to expect a particular experience, to schizophrenia. None. Zip. Nada.

Certainly, stress plays a part in psychotic illness. Certainly it’s plausible to think stimulating the brain might not be the best thing in the world to do, especially if that brain is vulnerable. This wretched (or at best, wretchedly reported) excuse for a study takes us no further than those two basically common-sense statements.

So, go forth and sup your coffee. ;)


Posted in Critical thinking, Medicine and psychiatry, Midweek Medicine and tagged as , , , , , , ,


Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Statistical data collected by Statpress SEOlution (blogcraft).