LAUSR

We welcome the comments by Morin, who offers a philosopher’s perspective on our research. First, Morin comments on our claim that cleaner fish can have a mental image of the self [“autoscopic image”] (1). Using multiple experiments, we document compelling evidence that cleaner fish recognize the self in photographs via self-face recognition, with the most likely mechanism being a mental image of the self, not by kinesthetic visual matching (2). Morin states, “There still could be other unknown underlying mechanisms no one thought about yet.” This is a claim that could be made against any experimental study, and we await a genuine alternative hypothesis explaining our results that we can test. Second, Morin claims, “Even if fish have an autoscopic image, they would have one single, tiny aspect of private self-awareness.” As stated in the paper and elsewhere, cleaner fish likely consider the mark an ectoparasite, which they are motivated to remove, with no other probable intention underlying the observed throat-scraping behaviors (2 Movie S1, 3). Indeed, cleaner fish immediately look at their mirror reflection or photograph image after scraping their throat, to see if their aim of removing the mark has been achieved. We consider this recognition that they need to check their reflection or image to see whether their intention has been achieved or not as an example of meta self-awareness, which we are further exploring in currently unpublished research. Cleaner fish have demonstrated motivations, aims, intentions, and mental images of self (i.e., a broad private self-awareness), which are likely organized by episodic-like memories (4). Third, a comment on the role of inner speech in self-awareness. We know in humans that loss of verbal speech and possible disruptions to inner speech, for example due to aphasia or stroke, do not result in a loss of self-awareness (5–7). We disagree with Morin’s assumptions here (8) and consider visual thinking, or nonlanguage thinking, capable of including thoughts about the self (9). Cleaner fish are obviously not capable of speech, yet after achieving mirror self-recognition, can readily distinguish between faces in photographs of the self, familiarand unknown individuals, and adjust their social reactions accordingly [e.g., friendly or aggressively; (2) p. 6]. That nonlinguistic animals are capable of high-level thinking and concepts is not a new or controversial idea, e.g., ref. 10, and we clearly demonstrate that cleaner fish have a concept of the self and of others. Lastly, Morin claims our results are not applicable in the wild, a common criticism of captive-based behavioral experiments. All fish were sourced from the wild, kept under standardized conditions, and exposed to biologically relevant stimuli. Furthermore, we carefully prevented associative or incidental learning from occurring during the mark and photograph tests. Most critically, we demonstrate that nonlinguistic animals have mental states, which is irrelevant to whether they are held under captive or wild conditions.