Ah, coffee, the beautiful stimulant without which about 90% of science and medicine would instantly retreat to the dark ages. Seriously, who hasn't needed that extra cup of joe (or two, or three) to make it through another 14 hour experimental day, a 10 hour surgery, or, you know, both?


But our favorite adenosine antagonist has a downside. Too much caffeine and you can get a racing heartbeat, nerves, and a certain amount of palm sweat, not to mention problems when you try to sleep at night.

But all of that is just yourself. Does caffeine affect the way you view other people? Well, if you're getting a little anxious, yes, it does.

Smith et al. "Storm in a coffee cup: caffeine modifies brain activation to social signals of threat" SCAN, 2011.

Caffeine does act the way other stimulants do. Instead of acting on neurotransmitters like dopamine or norepinephrine, caffeine acts on adenosine. Adenosine is another neurotransmitter, this one associated with things like sleepiness. High levels of adenosine promote sleepiness, and low levels promote wakefulness. Caffeine acts on the adenosine receptor as an antagonist, meaning that it blocks the effects of adenosine, promoting wakefulness.

But wakefulness and sleepiness are not all that adenosine does. Adenosine and adenosine receptors have been implicated in things like anxiety disorders. This could be particularly important when these receptors are expressed in the amygdala, medial prefrontal cortex, and periaqueductal grey, areas of the brain associated with things like fear processing and anxiety. And of course, if adenosine receptors are there, and caffeine is there, too, you might be getting some effects of caffeine on processing in these areas.

The authors of this study wanted to know if caffeine affected the processing of threat-related signals (people making angry faces), and where in the brain this might be taking place. They recruited 14 relatively low caffeine users (it's hard to get people who don't have any exposure, but I do wish the n had been about double at least), and gave them 250 mg of caffeine, about 1 cup of coffee (too bad it was only one dose! I'd love to see a dose response curve). They then put them in an fMRI machine to measure the blood oxygenation levels in the brain as people looked at threatening pictures.

What they got wasn't particularly surprising. Caffeine increased the diastolic blood pressure in all the patients, as well as increasing wakefulness. Caffeine also increased self-reported measures of anxiety. It was in the fMRI signals, however, where they found something new.

What you can see here is a composite signal in response to caffeine, focused on the amygdala. The authors found a significant increase in amydala activity when caffeine was combined with threatening faces. They also found an increase in activity in the periaqueductal grey. And in the amygdala, the increase in response was negatively correlated with the amount of caffeine the subject usually consumed, which means that this effect might be subject to tolerance.

So it looks like some of the effects of caffeine on things like anxiety might occur by changing the response of areas of the brain like the amygdala. I think to really prove this point I'd need to see a study in rats, say, with fear related responses (like fear conditioning), which was then reduced by the local injection of adenosine, and increased by local injections of caffeine. But it's nice to see the study in humans, and interesting to see that caffeine can change how we perceive things like threatening stimuli.

But the best news? It appears to be subject to tolerance. So at least if you have your daily cup of coffee, you know the coffee's probably not to blame. That guy really is probably looking at you funny.

Jessica E. Smith, Andrew D. Lawrence, Ana Diukova, Richard G. Wise, and Peter J. Rogers (2011). Storm in a coffee cup: caffeine modifies brain activation to social signals of threa SCAN DOI: 10.1093/scan/nsr058