At 3 a.m. I did not want an answer. I wanted company.

I used to sleep like a log. Lights out, gone in seconds, living whole other lives behind my eyelids, back by morning as if nothing had happened. Then the insomnia came, and one night I was awake with a question that would not leave me alone. Was I the only one curious about dreams? And why do we wake up homesick for a world we have never visited?

I asked ChatGPT. I got an answer about dreams, I felt better, and I stopped. That is why I am writing this. A question that could have kept me company for weeks got closed in a few minutes, and I was relieved about it.

So who killed it? I have three suspects.

Suspect one: Explanation. The case against it is old. You notice something strange, the world widens, someone hands you a label, and the widening stops. I believed a version of that myself, that night. Tania Lombrozo, a psychologist, argues the opposite. Wanting to understand something is itself a kind of curiosity, and the relief of understanding is what the curiosity was reaching for. I did not enjoy hearing it, since I had been busy treating my insomnia as a philosophical crisis. Blaming explanation for killing curiosity is like blaming the destination for ending the road trip. Suspect one walks.

Suspect two: Alexander Fleming. He is the hero of the popular case. A window left open, a stray mould spore, a petri dish, penicillin. Wonder rewarded, tidiness punished. A 2002 letter in the Medical Journal of Australia puts the likelier source of the spore in a dumb-waiter shaft connected to a mycology laboratory below. It also records what Fleming did next. Blood serum inactivated penicillin, which ended his hopes of using it inside the body. In 1929 he used it on an assistant’s eye infection, and it worked. In 1934 he stopped clinical work on it. Florey and Chain at Oxford later started from the culture he had sent them.

I will not guess what Fleming thought. I only read what he did. That makes it a story about a question that survived because other people kept it open. Suspect two walks as well.

Suspect three: Speed. Ellen Langer calls it premature cognitive commitment. You accept an explanation the first time you meet it, without turning it over, and it hardens into a frame that resists better evidence later. Then a feeling seals it. Liquin and Lombrozo, in a 2022 paper in Cognitive Psychology, found that the satisfaction of “ah, now I get it” is only weakly tied to whether you actually learned anything. You can feel closure and walk away no wiser. The feeling is a poor witness, and it was the only witness I consulted.

The verdict

The accused was in the room all along. It was me, with ChatGPT open at 3 a.m., grabbing the first thing that made the uneasy feeling stop.

I still do not know why I dream what I dream. Whole other lives, worlds I have never set foot in, gone by morning. The insomnia did not get solved that night. It got explained, and I mistook the relief for the end of the question. Everyone involved will be dust sooner or later, the mould spore included, and I was in a hurry.

Curiosity was never going to die from getting an answer. Mine died the moment I stopped checking whether the answer was true and just kept the feeling of being done. Get the answer, by all means. Then ask once more whether it holds.

Who Killed My Curiosity at 3 a.m. was originally published in Creative Loop on Medium, where people are continuing the conversation by highlighting and responding to this story.