05/06/2026
The Cleaner Wrasse Passed a Version of the Mirror Test in 2019 — Recognising Itself in a Reflection and Attempting to Remove a Mark Placed on Its Own Body. This Was Assumed to Require a Level of Self-Awareness Unavailable to Fish. The Field Argued About It Intensely.
It saw itself. It tried to remove the mark from its own body. The scientific community argued about what this means for years.
The mirror self-recognition test (MSR) — introduced by Gallup (1970) — is a standard measure of self-awareness in animal cognition: an animal is anaesthetised, a visible mark is placed on its body where it cannot be seen without a mirror, and the animal is then exposed to a mirror. Self-aware animals (humans, great apes, elephants, dolphins in some conditions) respond to their own reflection by examining the mark — looking at the mirror, then at their own body where the mark is, and sometimes attempting to remove it.
In 2019, Kohda et al. (PLOS Biology) published results from a series of mirror test experiments with Bluestreak Cleaner Wrasse (Labroides dimidiatus). Results:
— Fish exposed to mirrors repeatedly began treating their own reflection as a conspecific initially, then stopped showing social responses after several sessions.
— Fish with a brown mark placed on their throat (visible only in a mirror) — and given mirror access — rubbed their throat against the tank substrate at rates significantly higher than unmarked fish given mirror access.
— Fish with a transparent mark (same location, not visible) did not show elevated throat rubbing.
The interpretation controversy: the rubbing behaviour is consistent with attempting to remove the visible mark after recognising it in the mirror. But whether this constitutes "self-recognition" in the cognitive sense — or a simpler learned association between the mark, the mirror image, and a physical irritant response — was disputed extensively in subsequent literature.
The fish did something that looked like recognising itself. What that means for fish cognition: debated.
When a fish passes a test designed for self-aware mammals — but the field argues about whether the test proves what we thought it proved — does that mean the test is wrong, the fish is surprising, or both?
