Why our brains wear out at the end of the day


The transcript has been edited for clarity.

Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I’m Dr. F. Perry Wilson of the Yale School of Medicine.

Once again, we’re doing an informal journal club to talk about a really interesting study, “A Neuro-metabolic Account of Why Daylong Cognitive Work Alters the Control of Economic Decisions,” that just came out. It tries to answer the question of why our brains wear out. I’m going to put myself in the corner here. Let’s walk through this study, which appears in Current Biology, by lead author Antonius Wiehler from Paris.

The big question is what’s going on with cognitive fatigue. If you look at chess players who are exerting a lot of cognitive effort, it’s well documented that over hours of play, they get worse and make more mistakes. It takes them longer to make decisions. The question is, why?

Why does your brain get tired?

To date, it’s been a little bit hard to tease that out. Now, there is some suggestion of what is responsible for this. The cognitive control center of the brain is probably somewhere in the left lateral prefrontal cortex (LLPC).

The prefrontal cortex is responsible for higher-level thinking. It’s what causes you to be inhibited. It gets shut off by alcohol and leads to impulsive behaviors. The LLPC, according to functional MRI studies, has reduced activity as people become more and more cognitively fatigued. The LLPC helps you think through choices. As you become more fatigued, this area of the brain isn’t working as well. But why would it not work as well? What is going on in that particular part of the brain? It doesn’t seem to be something simple, like glucose levels; that’s been investigated and glucose levels are pretty constant throughout the brain, regardless of cognitive task. This paper seeks to tease out what is actually going on in the LLPC when you are becoming cognitively tired.

They did an experiment where they induced cognitive fatigue, and it sounds like a painful experiment. For more than 6 hours, volunteers completed sessions during which they had to perform cognitive switching tasks. Investigators showed participants a letter, in either red or green, and the participant would respond with whether it was a vowel or a consonant or whether it was a capital or lowercase letter, based on the color. If it’s red, say whether it’s a consonant or vowel. If it’s green, say whether it’s upper- or lowercase.

It’s hard, and doing it for 6 hours is likely to induce a lot of cognitive fatigue. They had a control group as well, which is really important here. The control group also did a task like this for 6 hours, but for them, investigators didn’t change the color as often – perhaps only once per session. For the study group, they were switching colors back and forth quite a lot. They also incorporated a memory challenge that worked in a similar way.

So, what are the readouts of this study? They had a group who went through the hard cognitive challenge and a group who went through the easy cognitive challenge. They looked at a variety of metrics. I’ll describe a few.

The first is performance decrement. Did they get it wrong? What percentage of the time did the participant say “consonant” when they should have said “lowercase?”

You can see here that the hard group did a little bit worse overall. It was harder, so they don’t do as well. That makes sense. But both groups kind of waned over time a little bit. It’s not as though the hard group declines much more. The slopes of those lines are pretty similar. So, not very robust findings there.

What about subjective fatigue? They asked the participants how exhausted they were from doing the tasks.

Both groups were worn out. It was a long day. There was a suggestion that the hard group became worn out a little bit sooner, but I don’t think this achieves statistical significance. Everyone was getting tired by hour 6 here.

What about response time? How quickly could the participant say “consonant,” “vowel,” “lowercase,” or “uppercase?”

The hard group took longer to respond because it was a harder task. But over time, the response times were pretty flat.

So far there isn’t a robust readout that would make us say, oh, yeah, that is a good marker of cognitive fatigue. That’s how you measure cognitive fatigue. It’s not what people say. It’s not how quick they are. It’s not even how accurate they are.

But then the investigators got a little bit clever. Participants were asked to play a “would you rather” game, a reward game. Here are two examples.

Would you rather:

  • Have a 25% chance of earning $50 OR a 95% chance of earning $17.30?
  • Earn $50, but your next task session will be hard or earn $40 and your next task session will be easy?

Participants had to figure out the better odds – what should they be choosing here? They had to tease out whether they preferred lower cost lower-risk choices – when they are cognitively fatigued, which has been shown in prior studies.

This showed a pretty dramatic difference between the groups in terms of the low-cost bias – how much more likely they were to pick the low-cost, easier choice as they became more and more cognitively fatigued. The hard group participants were more likely to pick the easy thing rather than the potentially more lucrative thing, which is really interesting when we think about how our own cognitive fatigue happens at the end of a difficult workday, how you may just be likely to go with the flow and do something easy because you just don’t have that much decision-making power left.

It would be nice to have some objective physiologic measurements for this, and they do. This is pupil dilation.

When you’re paying attention to something, your pupils dilate a little bit. They were able to show that as the hard group became more and more fatigued, pupil dilation sort of went away. In fact, if anything, their pupils constricted a little bit. But basically there was a significant difference here. The easy group’s pupils were still fine; they were still dilating. The hard group’s pupils got more sluggish. This is a physiologic correlate of what’s going on.

But again, these are all downstream of whatever is happening in the LLPC. So the real meat of this study is a functional MRI analysis, and the way they did this is pretty clever. They were looking for metabolites in the various parts of the brain using a labeled hydrogen MRI, which is even fancier than a functional MRI. It’s like MRI spectroscopy, and it can measure the levels of certain chemicals in the brain. They hypothesized that if there is a chemical that builds up when you are tired, it should build up preferentially in the LLPC.

Whereas in the rest of the brain, there shouldn’t be that much difference because we know the action is happening in the LLPC. The control part of the brain is a section called V1. They looked at a variety of metabolites, but the only one that behaved the way they expected was glutamate and glutamic acid (glutamate metabolites). In the hard group, the glutamate is building up over time, so there is a higher concentration of glutamate in the LLPC but not the rest of the brain. There is also a greater diffusion of glutamate from the intracellular to the extracellular space, which suggests that it’s kind of leaking out of the cells.

So the signal here is that the thing that’s impacting that part of the brain is this buildup of glutamate. To tie this together, they showed in the scatterplot the relationship between the increase in glutamate and the low-cost bias from the decision fatigue example.

It’s not the strongest correlation, but it is statistically significant that the more glutamate in your LLPC, the more likely you are to just take the easy decision as opposed to really thinking things through. That is pretty powerful. It’s telling us that your brain making you fatigued, and making you less likely to continue to use your LLPC, may be a self-defense mechanism against a buildup of glutamate, which may be neurotoxic. And that’s a fascinating bit of homeostasis.

Of course, it makes you wonder how we might adjust glutamate levels in the brain, although maybe we should let the brain be tired if the brain wants to be tired. It reminds me of that old Far Side cartoon where the guy is raising his hand and asking: “Can I be excused? My brain is full.” That is essentially what’s happening. This part of your brain is becoming taxed and building up glutamate. There’s some kind of negative feedback loop. The authors don’t know what the receptor pathway is that down-regulates that part of the brain based on the glutamate buildup, but some kind of negative feedback loop is saying, okay, give this part of the brain a rest. Things have gone on too far here.

It’s a fascinating study, although it’s not clear what we can do with this information. It’s not clear whether we can manipulate glutamate levels in this particular part of the brain or not. But it’s nice to see some biologic correlates of a psychological phenomenon that is incredibly well described – the phenomenon of decision fatigue. I think we all feel it at the end of a hard workday. If you’ve been doing a lot of cognitively intensive tasks, you just don’t have it in you anymore. And maybe the act of a good night’s sleep is clearing out some of that glutamate in the LLPC, which lets you start over and make some good decisions again. So I hope you all make some good decisions and keep your glutamate levels low. And I’ll see you next time.

For Medscape, I’m Perry Wilson.

Dr. Wilson is an associate professor of medicine and director of the Clinical and Translational Research Accelerator at Yale University, New Haven, Conn. He reported no relevant conflicts of interest.

A version of this article first appeared on

Next Article: