Last week, I argued that energy management [1], not time management, is the key to productivity. It’s far easier to make a schedule than to do the work. Available energy being more limiting than time helps explain why we so often fall short of our productive ideals.
But what, exactly, is energy?
It sounds obvious: We work, we get tired, and then it’s harder to work more. You had energy, you used it up. Now you’re running on empty, and work is difficult. Pretty straightforward, right?
Except, it’s not so simple. The science behind this “obvious” idea is surprisingly complex, involving biological, psychological and sociological factors. Even a basic question, like whether it’s harder to do an effortful task after doing another effortful task or an easier one, leads straight into one of the most infamous scientific controversies of the last two decades.
So today, I’m going to dive into some of this complexity. I know my appetite for esoteric social scientific debates is higher than average, and many people are simply interested in how to feel more energized and able to do their work.
But to have any hope of managing our energy, first we must understand what energy even is. And in order to do that, we need to grapple with that complexity.
Let’s dive in. I promise it will be worth it.
Is Energy a Resource?
A basic idea, built right into the idea of energy itself, is that it is some sort of resource: a metaphorical battery that is depleted and refilled. For a while, this was the scientific consensus. In the 1990s, psychologist Roy Baumeister and colleagues proposed the theory of ego depletion [2] that worked off of this premise.
Self-control and, by extension, mentally effortful tasks tap a universal mental “resource.” As with any good scientific theory, it made a falsifiable prediction: people would be less successful at exhibiting self-control after a “depleting” task than after a neutral control task. Energy would be used up, and they would be more likely to succumb to impulse or temptation.
Ego depletion also had an ancillary hypothesis: Not only was energy like a battery, depleted with use and recharged with rest, but the overall capacity could grow or shrink with use, much like a muscle. By exercising self-control regularly, we could become more disciplined.
Both the battery and muscle analogies have a certain commonsense appeal. And, for a time, it appeared they had solid scientific support as well. To date, over 600 published studies in the literature have found support for ego depletion, and a 2010 meta-analysis [3] by Martin Hagger and colleagues found that not only was the effect statistically significant, but it was practically significant with an effect size about 50% larger than typically found in social psychology.
As evidence accumulated, ego depletion researchers looked for a physical property in the brain that corresponded to the behavioral effects. And many believed they found it: glucose.
The brain is a hungry organ. Despite accounting for only 2% of the body’s weight, it consumes nearly 20% of our daily calories. Thinking, it turns out, is a costly business, and that price is paid in the currency of glucose.
Once again, support [4] began to build not only for the behavioral reality of ego depletion, but its correlation to brain glucose levels. Drinking a sugar-sweetened beverage could temporarily boost glucose and stave off the energy-depleting effects of mental effort, whereas a placebo drink sweetened with an artificial sweetener would not.
It was a textbook case of science done right: a commonsense observation was translated into an experimental hypothesis, the hypothesis was rigorously tested in controlled experiments, and, finally, research found the physical mechanism mediating the effect. Credits roll, end of story.
Cracks in the Ego Depletion Story
Except, that’s not what happened. Instead, research on ego depletion imploded, calling into question not only this theory, but the entire edifice of social science.
The first cracks in the simple sugar-powered battery analogy came from an interesting 2010 experiment [5] by growth-mindset researchers Job, Dweck and Walton.
In their experiment, they found that a person’s beliefs about willpower moderated the ego depletion effect. If a person believed willpower was a like a battery that gets used up, they were more depleted in the follow-up task than if they believed willpower was unlimited.
But if ego depletion is drawing on a physical property of the brain, like glucose levels, how could it mere beliefs about willpower itself influence the results?
Other researchers found that incentives could influence depletion. Small rewards could eliminate the effect of depletion altogether [6]. This was another strike against a straightforward reading of the ego depletion theory. After all, if your car is out of gas and stranded on the highway, it’s not as if throwing some cash on the dashboard will unlock a secret fuel tank.
Attacks mounted against glucose as a biological mediator of the ego depletion effect. While the brain does consume a lot of glucose, any additional amount consumed owing to self-control is negligible. The amount consumed by the visual cortex is much greater [7], but we rarely feel fatigued from simply looking at stuff.
In light of these findings, other researchers proposed alternative accounts: perhaps ego depletion was better understood as a decline in motivation, not a resource, and so could be influenced by beliefs or incentives. Maybe effort is a perception of opportunity costs [8]? Or a kind of affective state [9]?
All of these attacks would have been part of the normal back-and-forth of social science, the theory/counter-theory jabs academics lob all the time, had it not been for a bombshell paper that came out in 2016.
Ego Depletion and the Replication Crisis
By this point in the story, rumors were already circulating that some psychological results were not to be trusted. The field of social priming, where brief (sometimes subliminal) exposure to stimuli was thought to have large effects on behavior, had trouble replicating some of their classic experiments [10]. Science, if it is to have any meaning, has to be reliable. An effect that exists on Monday can’t disappear on Tuesday when a different scientist runs the experiment.
Researchers were coming to realize that practices like failing to publish null results, or tweaking an experiment or analysis until a significant effect appeared, weren’t as innocent as they had thought. To quote one set of researchers [11], “Everyone knew it was wrong, but they thought it was wrong the way it’s wrong to jaywalk. [But simulations revealed] it was wrong the way it’s wrong to rob a bank.”
After correcting for unpublished null findings, one meta-analysis of ego depletion effects came up with much smaller effect sizes [12] than Hagger’s original 2010 meta-analysis. Suddenly, hundreds of studies all pointing in the same direction felt more suspicious than confirmatory.
To quell doubts, Hagger himself led a preregistered replication attempt. This asked many labs, all following standardized protocols with no possible p-hacking, to re-run ego depletion experiments. Published in 2016, the aggregate statistics found no statistically significant effect [13].
Run by one of the major ego depletion researchers, the 2016 study failing to replicate findings had a catastrophic effect on the field. Ego depletion as a theory was dead, a cautionary tale into the dangers of unrigorous science.
Ego Depletion: Back from the Dead?
Except, of course, you knew it wouldn’t be so simple.
Ego depletion was wounded, and many of its early studies were fatally flawed, but it’s still far from dead.
The theory adapted [14] in the face of some of its challenges. For instance, the new theory suggests that, while ego depletion is real, we rarely find ourselves truly “on empty.” Instead we conserve energy for future use when it is running low.
To use a new analogy, think of it like spending money. After a pricey holiday season, you may feel a little overspent and decide to be more frugal with your spending in January to compensate. But it’s not as if you’ve literally spent your very last dollar—if an emergency (or significant opportunity) came up, you’d probably find a bit more money to spend.
Beliefs and motivations can also be seen as inputs to our energy system [15], rather than viewing things through the overly simplistic lens of a single limited resource governing all behavior.
Defenders of ego depletion argue that many of the failed replications failed to fully test the theory.
For one, there’s the issue of dose. To coordinate many different labs using different tasks, many of the large-scale preregistered ego depletion experiments used very short self-control tasks to “deplete” participants. These “depletion tasks” may only have been only 10 to 15 minutes in duration, which is likely too short to meaningfully fatigue the participants. Thus, the lack of significant effects could be due to the studies being underpowered rather than the effect itself being unreal.
Second, there’s the issue of the selected control task. Many experimental designs used boring tasks as the “neutral” condition. However, sticking to a boring task may itself deplete our mental energies, making the control and depletion conditions more similar than they should be.
Third, there’s the issue of whether the depletion task itself was properly validated. Many experiments used letter-crossing tasks, where participants were asked to read a short text and cross out certain letters, such as “cross out any e next to a vowel.” For theoretical reasons, this type of task was assumed to deplete self-control. However, researchers have pointed out that crossing out letters may be tedious, but it doesn’t involve the type of motivational conflict that typifies self-control problems, such as choosing to eat broccoli versus cheesecake.
Proponents argue that, when taking these into account, ego depletion is still real [16], albeit weaker than previously thought, and more dependent on contextual factors.
Even the biophysical basis is being revised. While the theory that glucose is the mediator of ego depletion is definitively dead, recent neuroscience work using brain wave monitors has found elevated levels of delta-wave activity (the kind normally seen in deep sleep) in the regions of the brain associated with self-control after a longer depleting task [17].
It may be that, unlike a fuel that gets burned up, ego depletion is more like garbage that builds up and needs to be collected, with metabolic by-products of neural activity increasing the incentive to take a mental break.
What’s the State of the Understanding in 2026?
It’s clear that, despite the scientific roller-coaster ride, the consensus on energy is far from settled. It’s completely reasonable to have skepticism about ego depletion given its tarnished history. I know I certainly do.
But, despite my enthusiastic promotion of an early alternative theory in terms of opportunity costs [18], the evidence hasn’t clearly aligned in support of an obvious successor. Instead, perhaps unfortunately, reality is simply messier than the original ego depletion theory permitted. The phenomenon of feeling drained after working hard on something is decidedly real, but the actual mechanisms through which it happens may be a mixture of depletion, motivation, attention and beliefs.
There are important practical consequences of this messy picture as well. It means there isn’t just a single factor, like glucose, that mediates the ease with which we do hard things—we can’t improve our energy just by drinking a soda anymore than we can make a car go faster by dousing it with gasoline.
But that complexity is also an opportunity. If energy comes not from a single resource but from multiple factors, there are more levers we can pull when trying to get more energy out of ourselves and our work.
The story of ego depletion is a twisting one, but it’s just one aspect in the fascinating science of what makes us feel alive and energized. Next, I’ll shift away from controversy to discuss some science with much more stable footing: how stress impacts our health and energy levels.