It’s 9:18pm and you’re staring at your notes, and you’ve read the same paragraph about the Krebs cycle — or price elasticity, or the Treaty of Westphalia — four times without retaining any of it. You have a Red Bull sweating onto your desk. You have two tabs open with YouTube videos you told yourself were “study help.” Your exam is at 8am. You are trying very hard to feel ready, and instead you feel like you might throw up.
If this is the specific version of exam anxiety you’ve googled tonight, the most useful thing I can tell you is this: the cramming you’re doing right now is costing you more than it’s giving you. Not as a motivational line. As a real, measurable effect documented in research going back decades.
The real problem isn’t that you don’t know enough
Here’s what’s happening in your brain right now. You’re in a high-stakes, low-control situation. Your nervous system has registered this as a threat. And once that threat response is active, your working memory — the mental scratch pad you use to actually think through problems, connect ideas, hold multiple concepts at once — starts leaking capacity.
This is what psychologist Irwin Sarason documented in 1984 in his landmark paper on stress, anxiety, and cognitive interference (Journal of Personality and Social Psychology, 46, 929–938). Sarason’s key finding: test anxiety doesn’t just make you feel bad. It actively crowds out cognitive space with “worry” and “task-irrelevant thinking” — your brain narrating catastrophes while you’re trying to recall facts. You’re not forgetful tonight. You’re pre-occupied.
The same mechanism appears in Sian Beilock’s research on choking under pressure (Psychological Science, 2005): high working memory students — the ones who perform best under normal conditions — are the most likely to choke when anxiety consumes the cognitive resources they rely on. Anxiety isn’t an equalizer. It’s a tax on the students who could otherwise do the best.
So those two hours of notes you’re about to grind through at 10pm, 11pm, 12am? Your anxious brain isn’t encoding them the way your rested brain does. You’re writing information into a filing system that won’t be accessible tomorrow morning. The study session feels like doing something. It is doing something. It’s just not doing what you need.
There’s another thing worth naming: the material you’re most tempted to review tonight is usually the material you’re least confident about. Which means you’re most likely to spend the next two hours staring at the gap between where you are and where you wish you were, generating more dread rather than actual capability. The part of the course you know well — and there’s always some of it — doesn’t feel urgent enough to review, so you skip it. You end the session feeling less prepared than when you started, which is almost the worst possible way to go into sleep before an exam.
What the research says about sleep and memory consolidation
Sleep isn’t the absence of studying. It’s a core part of how memory works.
During slow-wave sleep, your brain replays the day’s learning — specifically, it transfers information from the hippocampus (short-term storage) into the neocortex (long-term storage). Cutting that process short by two hours for a 10pm-to-midnight cramming session means the material you learned this afternoon never gets fully committed. You arrive at the exam having spent more time awake and less time consolidating.
There’s a specific type of insomnia that exam anxiety reliably produces: you get into bed exhausted, your mind starts running through everything you might have forgotten, and then you can’t sleep — which generates more anxiety, which makes you less likely to sleep. By 1am you’re checking your notes on your phone “just to make sure,” and now the blue light and the renewed cortisol spike have reset the cycle completely.
The honest personal truth: the all-nighter is almost never about preparation. It’s about the feeling of doing something when you feel helpless. The act of studying at midnight is, in large part, an anxiety-management behavior disguised as academic behavior. It doesn’t work for either purpose.
One more wrinkle: caffeine, which most students use to power through the late session, has a half-life of around five to six hours. The coffee or energy drink you have at 9pm is still at half-strength at 2am, still raising your cortisol, still delaying sleep onset. The stimulant you took to help you study is now the main thing between you and the sleep that would actually help you perform. It’s a loop with no good exit once you’re deep enough into it.
What to do tonight instead
Close the books at 9pm. Seriously — not 9:30, not “one more chapter.” 9pm.
Here’s what that actually looks like:
At 9pm: Put the notes away. Not in your bag where you can see them. In a drawer, under your bed, somewhere that requires a deliberate action to retrieve. The point is to remove the option. You don’t trust 9:15pm-you to not open them “for a second.”
At 9:15pm: Get outside for ten minutes. Walk around the block. Not to exercise, not to clear your head in some transformative way — just to change the physical environment your anxiety has been building in. Your dorm room or apartment has become, over the last few hours, a place associated with “I’m not ready.” Your nervous system needs a break from that room.
At 9:30pm: Eat something if you haven’t. Not another coffee. Actual food, ideally with some carbohydrates — there’s evidence that carbohydrates support serotonin production, and serotonin supports sleep onset. This is not the time for your performance nutrition phase. It is the time for a bowl of cereal or some toast.
10pm to 10:30pm: Wind down. This means something low-stimulation — not a show with a cliffhanger, not a conversation that’ll get your adrenaline going. Read something unrelated to the exam. Take a shower. Do the boring thing.
In bed by 10:30pm. Even if you don’t fall asleep immediately, lying in the dark with your eyes closed is better than sitting at a desk with your notes. Your body will do what it can.
One thing that genuinely helps with the racing-mind problem: a brain dump. Before you get into bed, take three minutes and write out everything that’s worrying you about tomorrow — on paper, not your phone. Not to solve it, not to study it, just to physically externalize it so your brain stops trying to hold it. Research on worry postponement (Borkovec, 1983) found that scheduled, bounded worry time reduces intrusive thoughts during sleep. The paper version does something similar: it signals to your nervous system that the thoughts have been noted, and you don’t have to keep rehearsing them.
On the question of whether you know enough: The catastrophizing version of this night runs on the assumption that there’s something critical you don’t know that you could learn between now and 8am. That’s almost never true. The things you don’t know tonight — the gaps that genuinely exist — are not going to be filled by two more hours of notes in this state. But the things you do know, the connections you’ve already made, the patterns you’ve spotted in lecture? Those are in your brain. Sleep is what makes them accessible. Anxiety is what makes them unavailable. That’s not a metaphor — it’s the actual mechanism of how memory retrieval works under cortisol load.
One thing to repeat to yourself: The version of you that walks into that exam tomorrow having slept six-and-a-half hours will outperform the version of you that crammed until 1am and slept four and a half. Not because sleep is magic. Because your anxious, sleep-deprived brain cannot access what it already knows.
You already know more than you think you do. Let sleep help you prove it.
If the mind-racing problem persists past lights-out, the grounding technique in this piece tends to work better than generic deep-breathing advice for actual spiraling. And if this is a pattern for you — anxiety that spikes in the days before any high-stakes moment — breathing before a presentation covers the physiology in more detail.