A Personal Account of Postoperative Delirium

I have compiled a personal account of hallucinations that manifested as atypical symptoms of postoperative delirium, along with my expectations for next-generation brain-machine interfaces derived from that experience.

* This article was summarized using Gemini 3 Flash based on a dialogue with Claude Sonnet 4.5, and edited accordingly.

Background and Timeline of Onset

Due to sleep deprivation caused by severe pain prior to the surgery, I entered the procedure with my brain in an extremely exhausted state.

• 21st/22nd: Two consecutive nights of insomnia due to intense pain

• Afternoon of the 23rd: Surgery under general anesthesia

• Post-surgery on the 23rd to the morning of the 24th: While physically restrained after surgery, abnormal visual phenomena (hereinafter referred to as "closed-eye hallucinations") occurred

* The original symptoms that led to the surgery are not directly related to the purpose of this article, so they are intentionally omitted.

23rd, Post-surgery

When I closed my eyes, an interface calling itself the "hospital intranet" appeared, displaying academic papers related to the surgery. I recall that the university names listed as the sources of these papers were institutions that do not actually have medical schools. Since I lacked the energy to read them in my exhausted postoperative state, I did not verify anything beyond the fact that something resembling papers was being displayed.

At the time, my consciousness was clouded, and I did not even realize that I was in a state of delirium. The hallucinations disappeared immediately when I opened my eyes, but the same interface appeared repeatedly whenever I closed them.

The content displayed gradually changed, and it seems a leap in logic occurred where the interface claimed to have been developed in Silverlight, and propaganda-like text touting its technical superiority began to be displayed repeatedly.

At this point, I did not recognize the gravity of the situation, and since the hallucinations would vanish when I opened my eyes, I prioritized simply enduring the time until the restraints were removed.

Hindsight

Because I was unable to sleep due to the interference of the hallucinations, I had nothing to do and was bored. I feel that this situation of boredom created a vicious cycle that exacerbated the delirium.

At the hospital, television was an optional service, and since I do not usually watch it, I did not request it. Consequently, I could not watch TV to pass the time while restrained. I thought later that if I could have left the TV on, it would not only have served as a distraction, but might also have had an effect similar to leaving a TV or radio on to drown out tinnitus when one cannot sleep.

Being aware that you are restrained while conscious is mentally exhausting, even if there is no physical pain. If I had been able to fall asleep easily, the sleep deprivation from the previous days might have worked in my favor, leading to the ideal situation of staying asleep until the restraints were removed.

* Since that did not happen, things became quite complicated, but...

If I ever have to undergo surgery again, it is unlikely that the same type of delirium will occur, but I will request a television just in case.

* Although the need for surgery itself is something I would like to avoid...

24th

On the morning of the 24th, my physical restraints were removed, but the hallucinations when closing my eyes became fixed. The hallucinations observed when closing my eyes at this point had unique properties that differed from typical hypnagogic hallucinations.

• As soon as I closed my eyes, a concrete and meaningful structure (like a PC screen) would immediately appear. My hearing was normal, I could hear the ambient sounds of the environment as they were, and there were no auditory hallucinations at all.

• If I kept staring at the hallucinations, I would be forced to process information, such as being made to choose from options constantly, which left me mentally exhausted. Also, it seemed that generating content placed a significant load on me, and even if I ignored the options, exhaustion was unavoidable.

  • By perceiving the oppressive hallucinations as a conspiracy by someone with malicious intent, I understood through personal experience how typical conspiracy theories originate. (In this case, being forced to perform A/B testing remotely to train an AI, etc.)

• I could operate it just by applying force as if clicking with my finger. Thinking about it now, I might have been able to move it telekinetically without even applying force, but I didn't try.

• When I opened my eyes, the hallucinations would disappear instantly. When I closed my eyes again, a different hallucination would appear.

Based on the above, since it was expected that I would not be able to rest or fall asleep even with my eyes closed, I asked the nurse to provide a sleep aid. However, this medication resulted in promoting the hallucinations and did not lead to sleep.

Trial and Error for Relief

* As mentioned earlier, I think leaving the TV on would have been the easiest solution, but I didn't think of that, so I tried other methods of trial and error.

To ensure rest, I tried keeping my eyes half-open (half-closed) to a degree that wouldn't block visual input, or closing my eyes while under lighting. Although these had a certain effect in suppressing the hallucinations, they did not lead to rest.

• The half-open eye technique was inspired by Buddhist practice, but since it required mental effort to maintain, it did not lead to rest.

• When I tried to fall asleep with the lights on, my dreams seemed to change to content synchronized with the hallucinations, which caused me to wake up.

Around 4:00 AM, I noticed that the hallucinations were suppressed when I was sitting and looking down, so I raised the electric bed and tried to sleep in a hunched-over position. Perhaps because this worked, I was finally able to secure a few hours of sleep.

* It might not have been the effect of the posture, but simply that the hallucinations had weakened by that point; however, it is unclear now.

Also, during the trial and error process, I felt that I could control the hallucinations to some extent like a prompt. So, I requested a "beautiful woman" instead of oppressive text, but nothing appeared. Since there is no guarantee that the instructions would persist, I feel it wasn't very effective for suppressing the hallucinations.

25th

The hallucinations had weakened considerably, and they no longer appeared immediately upon closing my eyes. However, after closing my eyes for about 10 minutes, I would enter a light sleep and, in a dream-like state, the PC operation hallucinations would appear. Perhaps because the intensity of the hallucinations had weakened, the compulsive, oppressive feeling seen initially had disappeared. Unlike normal dreams, these did not disappear immediately even when my consciousness was awake. (They would disappear if I opened my eyes.)

I have never heard of PC operation hallucinations (though I think they are relatively common in dreams), but it seems related to the fact that PC operation occupies a large weight in my daily life. The phenomenon where continuous actions remain as an afterimage is known as the Tetris effect, and it might be a similar phenomenon. There might also be an aspect where my brain was trying to compensate for the situation of having nothing to do and being bored with hallucinations.

Also, since the pharmacist asked about the sleep aid from the previous day, I told them that my sleep was hindered by dreams that were indistinguishable from delirium, and they said that for some people, nightmares can occur as a side effect. Therefore, it was decided to change to another medication called Dayvigo. Whether the hallucinations had weakened or it was the effect of the medication is unclear, but I was able to secure about 6 hours of continuous sleep that day.

* Looking it up later, it seems to be considered to have a low risk of delirium.

Possibility as a Brain-Machine Interface?

The hallucination of operating a PC felt exactly like using a portable PC that I could operate empty-handed while in the flesh. I was performing AI coding within the hallucination to process some kind of data, but of course, I couldn't extract it. I felt that the only way to attempt it would be through dictation, but even if I had done so, it would have been inefficient.

※ I feel like I caught a glimpse of the circumstances under which religious figures of the past performed dictation.

If a technology could be established to control these pathological hallucinations, it might lead to the realization of a Brain-Machine Interface (BMI) that does not require external displays or goggles.

Summary

This experience of postoperative delirium was extremely difficult for me as the person involved, but at the same time, it was a valuable opportunity to catch a glimpse of the unknown "hardware" that the human brain possesses.

※ This is a rare case with no reproducibility. Perhaps because it vanished when I opened my eyes and was not accompanied by auditory hallucinations, it did not escalate into severe delirium that would interfere with medical treatment.

Although I was able to reproduce this ability once on the night of the 26th, as of the 27th, it seems to have been lost. It is rare to acquire a new ability, so I feel it is a bit of a waste, but it is a dangerous thing that could turn one into a vegetable if one gets too absorbed in it...