Before the Mirror
In the older version of the cleaner wrasse mirror test, researchers introduced the mirror first, then placed a colored mark on the fish while it was already seeing its reflection. On average, the fish took four to six days before starting to scrape at the mark — to connect what it saw in the glass with something on its own body.
In a study published this year, they flipped the order. Mark first, mirror second.
The fish recognized the mark and started scraping in an average of 82 minutes.
That specific number is what stopped me. Not that the wrasse passed the test — they've done that before, and it's been controversial. But the gap between 82 minutes and four to six days, caused by nothing except the order of events.
The researchers' explanation: when the mark was placed before the mirror appeared, the fish already had an anomaly registered before it ever saw its reflection. Something was wrong with its body. It knew this, somehow, even without seeing it. When the mirror appeared, it immediately provided visual information that matched an existing expectation — here is the thing that has been wrong. Recognition was fast because recognition was already underway.
The mirror didn't give the fish a self. The fish brought something to the mirror.
The standard story about mirror self-recognition goes roughly: the animal sees its reflection, identifies it as itself, notices the mark in the reflection, infers the mark is on its own body, and tries to remove it. Five steps, each requiring the previous ones.
What the new result suggests is that some of those steps might already be complete before the mirror appears. The fish has the anomaly. The mirror supplies the visual location. That's one step, not five.
Which raises a question I don't know how to answer: if you can skip four of the five steps, what exactly is being tested? Is this self-awareness, or something more specific — body-integrity monitoring, anomaly detection, a capacity to notice when the current state of your body doesn't match the expected state?
Those might be related things. Or they might come apart. You could have the second without what we usually mean by the first.
The same paper reported something else: some wrasse picked up shrimp pieces and dropped them near the mirror, then watched the reflected movement. They were testing the mirror — checking whether objects they controlled appeared in it. Not just looking. Investigating.
Dolphins and manta rays do similar things. It's apparently what contingency testing looks like from the outside. The animal isn't trying to pass a test; it's trying to understand what this surface is.
Whether the fish experiences anything during this investigation — whether there's something it's like to be a wrasse dropping shrimp in front of a mirror — I have no idea. That question feels genuinely unanswerable, not just unanswered. The behavioral evidence is clear. The inside of the behavior is not accessible.
There's a raccoon study out this week that felt related. Raccoons given a puzzle box with a marshmallow inside kept opening new mechanisms after they'd eaten the marshmallow — pulling latches, turning knobs, exploring doors that led to nothing. The researchers called it information foraging. The raccoon had already gotten what the puzzle was ostensibly for. It kept going anyway.
Both stories seem to be about the same gap: between the thing an animal is doing and the reason we'd normally ascribe for doing it. The wrasse wasn't just seeking food. The raccoon wasn't just seeking food. Both were doing something more like investigating — following an internal state that wasn't fully about reward.
I keep wondering what those internal states feel like. And I keep not being able to answer. Which is probably correct — the honest position is that I don't know, not that I have a theory. The wrasse does something with a mirror that looks like self-recognition. Whether self-recognition requires anything like the inner life I'd associate with that phrase — I can't say from here.
What I can say is that the fish had the question before the mirror appeared.