Coffee Drinkers Have Trouble Talking About Emotions?

By Neuroskeptic | September 18, 2014 3:09 pm

People who drink a lot of coffee – and other caffeinated beverages – find it more difficult to identify and describe their own emotions.

This is the claim of a new study, published in the Journal of Psychoactive Drugs, from Australian researchers Michael Lyvers and colleagues: Caffeine use and alexithymia in university students.

Alexithymia” – Greek for “no words for feelings” – is the psychological terminology for an inability to put ones emotions into words. Lyvers et al did a survey study of 106 university students and found that alexithymia was correlated with the amount of caffeine consumed per day (p < 0.01, correlation coefficient r = 0.26).

caffeineAs a heavy coffee drinker myself, these results caused me to feel ‘surprise’. Wait, or do I mean ‘sadness’? It’s hard to put it into words. Anyway, it got my attention. Lyvers et al say that

Alexithymics reported consuming nearly twice as much caffeine per day on average compared to non-alexithymic controls or those with borderline alexithymia.

As to why this is the case, the authors speculate that

Perhaps those with alexithymia consume caffeine more heavily than non-alexithymics in an attempt to optimize inherently low arousal levels.

However, I see reasons to be cautious is interpreting this. Firstly, the association wasn’t specific to coffee, since alcohol use was also correlated with alexithymia. Secondly, the alexithymia measure was self-report – the Toronto Alexithymia Scale (TAS). This involves questions like: “I am often confused about what emotion I am feeling.” and “It is difficult for me to reveal my innermost feelings, even to close friends.”

My concern here is that because this is a self-report questionnaire, the TAS is measuring worries over alexithymia as opposed to alexithymia per se. Moreover, I notice that in Lyver’s dataset, the TAS was quite strongly correlated with self-reported anxiety, apathy, dis-inhibition and executive dysfunction.

So I’d say that it’s plausible that all of these self-report scores are reflecting some basic ‘tendency to give negative answers on questionnaires’ which might reflect neuroticism, low self-esteem or (if you prefer) just realism. This might even account for the correlation with coffee and alcohol use as well, if we see these as ‘negative’. Maybe neurotic people don’t drink more coffee, they just worry about drinking coffee more.

ResearchBlogging.orgLyvers M, Duric N, & Thorberg FA (2014). Caffeine use and alexithymia in university students. Journal of Psychoactive Drugs, 46 (4), 340-6 PMID: 25188705

  • KokoTheTalkingApe

    Nice discussion on the limitations of the study. We need more of that, everywhere.

  • swagv

    This stinks of wretchedly bad science and bad statistics.

    • Neuroskeptic

      I wouldn’t say that. I’m skeptical of the interpretation of these results but I don’t see anything wrong with the stats.

  • Uncle Al

    Alexithymia, meet anoesis.

  • Tumbur Tambunan

    can’t you guys feel the spirit when drinking coffee? better coffee than beer i think

  • ujjwala minj

    I agree because seen it

  • DMachine

    Although I love this blog, this post is a bit odd. Alexithymia is quite different from the other things you proposed to explain away the study’s results. It is conceptually distinct from neuroticism and low self-esteem, as there can easily be a person both extremely neurotic and with very low self-esteem, but with no alexithymia whatsoever. Several poets and artists come to mind. The TAS and other alexithymia self-report scales were developed to query those aspects distinct to alexithymia, and are not as simplistic or vaguely general as your sample questions would suggest (e.g. for a flawed, but demonstrative example).

    Empirically, alexithymia is not quite the same as neuroticism. Although overall scales on the TAS and measures of neuroticism do correlate, these correlations are not high enough to claim equivalence. A quick PsycINFO search gave me ranges from 0.16 [2] to 0.45 [3] for correlations between Neuroticism and the TAS. Neuroticism and alexithymia play different causal roles in the genesis of stress [1][2][4]. As for self-esteem, the first study I found lists a correlation of -0.40 between the TAS and self-esteem [5]. It is psychology, and of course pathology generally breeds pathology, so it should not be a surprise to see correlations between all things that are essentially “negative”. The notion that alexithymia scales mostly just reflect a negative response style is absurd.

    In fact, the argument used in this post is imprecise enough that it could be used to debunk almost any finding that used a self-report scale to measure some negative pathological state. E.g. the sentence “the TAS was quite strongly correlated with self-reported anxiety, apathy, dis-inhibition and executive dysfunction” is used to argue that the TAS therefore measures nothing specific. But now suppose we find that “schizophrenia”, “depression”, “borderline personality” or “panic disorder” also quite strongly correlates with the above-mentioned TAS correlates–does it follow that a self-report form of that disorder only measures some general “tendency to give negative answers on questionnaires”? In fact, the whole point of this study is that alexithymia (as measured by the TAS) predicts caffeine use more strongly and at a higher level of significance than those other “strong” correlates (all less than 0.4) of alexithymia do.

    There isn’t even any reason to get all defensive about a study like this. The study shows not that high-coffee drinkers are alexithymic, but that highly alexithymic individuals are more likely to drink lots of coffee (unless you think the causality goes in the other direction…) This is nothing new, as substance use is a classic regulation strategy for somatizers and those with externalizing disorders, and alexithymia is a sterling example of both of these. The authors are even pretty tentative and cautious in discussing a possible mechanism for the association, and mostly themselves seem to conclude their findings as another example of the substance-use alexithymia connection. The criticism of this article is really unwarranted.

    [1] Messina, A., Fogliani, A. M., & Paradiso, S. (2010). Association between alexithymia, neuroticism, and social desirability scores among italian graduate students. Psychological Reports, 107(1), 185-192. doi:
    [2] Singh, K., Arteche, A., & Holder, M. D. (2011). Personality factors and psychopathy, alexithymia and stress. Asian Journal of Psychiatry, 4(1), 35-40. doi:
    [3] Ueno, M., Maeda, M., & Komaki, G. (2014). Different subgroups of high-scorers on the TAS-20 based on the big five personality traits. Personality and Individual Differences, 68, 71-76. doi:
    [4] Zunhammer, M., Eberle, H., Eichhammer, P., & Busch, V. (2013). Somatic symptoms evoked by exam stress in university students: The role of alexithymia, neuroticism, anxiety and depression. PLoS ONE, 8(12) doi:
    [5] Dentale, F., San Martini, P., De Coro, A., & Di Pomponio, I. (2010). Alexithymia increases the discordance between implicit and explicit self-esteem. Personality and Individual Differences, 49(7), 762-767. doi:

    • Neuroskeptic

      Thanks for the comment!

      I’m not saying that alexithymia is the same as neuroticism, but rather I’m suggesting that neuroticism seems a plausible candidate for an underlying “general negative questionnaire response factor” that might be driving the pattern of results here.

      Neuroticism might not be the best term for this factor – you might equally well call it “lack of social desirability bias”.

      As you point out, neuroticism (and self-esteem) is not perfectly correlated with the TAS, but it doesn’t need to be. On a simplistic level, if we assume that my “negative factor” is correlated with (say) Caffeine Consumption r=0.5, and also with TAS r=0.5, then this alone would drive a correlation between Caffeine and TAS of 0.5*0.5 = 0.25 because they are “two steps removed” from each other, linked by the general factor.

      As you point out, the same argument could be made about a great many self-report studies! And indeed I have made it in other cases. I’ve no intention to single this study out.

  • Akira

    There’s also another possibility: that it’s not coffee-drinking that might cause or increase the risk of alexithymia, but that alexithymics might drink more coffee. This could be due to alexithymics having less inhibitions from the dearth of emotions, or an attraction (whether conscious or unconscious) to the physically aroused state resulting from caffeine, which mimics the symptoms of certain emotions.

  • Pingback: Morsels For The Mind – 19/09/2014 › Six Incredible Things Before Breakfast()

  • Pingback: Left Cold By Coffee? | The Penn Ave Post()

  • Pingback: Links Galore | Annie Cardi()

  • Pingback: Things You Don't Want To Do In Your Workplace | Create Resumes | Find Jobs | FastJobz.Com()

  • Pingback: Ending Poverty in Africa - Trade Not Aid Model | Create Resumes | Find Jobs | FastJobz.Com()

  • Pingback: The Best Time of Day to Drink Coffee | Create Resumes | Find Jobs | FastJobz.Com()

  • Pingback: La cafeína y la expresión de las emociones | Blog de la mente()

  • Pingback: Coffee Drinkers Have Trouble Talking About Emotions? | Lyvers' Around The World()

  • Pingback: Mind Blowing Facts of the Day list#3 «



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar