Fleming and Michel assert that aquatic animals, such as fish that evolved prior to the evolution of land animals, have limited ability to see long distances underwater. Consequently, they argue that such animals only use vision for non-conscious perception in the service of action, a function that is subserved by the dorsal visual stream of mammals (Goodale & Milner, Reference Goodale and Milner1992). Even in the case of limited visual distance in fish, the argument that this precludes conscious recognition ignores the possibility of repeated or even prolonged glimpses of objects or other animals and the use of working memory. Human infants acquire these abilities in the first year of life to deal with occlusions and the presence of multiple objects (Kibbe, Reference Kibbe2015), and these abilities are strongly associated with conscious shape processing in the visual ventral stream of mammals (DiCarlo, Zoccolan, & Rust, Reference DiCarlo, Zoccolan and Rust2012; Goodale & Milner, Reference Goodale and Milner1992). In the case of fish, a predator might sense prey multiple times while tracking it before striking, even as it passes behind occlusions or is masked by other nearby objects, and prey might similarly sense a predator multiple times while trying to evade it.

There is indeed experimental evidence that fish can recognize objects even when the objects are occluded (Sovrano & Bisazza, Reference Sovrano and Bisazza2008), and both predator and prey fish also show evidence of illusory contour perception, which likely indicates similar visual shape perception as humans (for a review, see Brown, Reference Brown2015). Furthermore, in fish that swim in schools or closely to the reef, there is likely to be prolonged up-close visual experience of surrounding animals and objects compared to that in the open ocean, which could lead to conscious perception, perhaps enhanced by unique adaptations they have, such as sensitivity to ultraviolet and polarized light (Brown, Reference Brown2015). Of course, all of this could occur unconsciously in fish, but this seems less parsimonious than to assume that similar visual abilities in fish and mammals are similarly conscious.

Furthermore, the target article’s focus on the limitations of underwater vision ignores other senses that fish likely use for perception at a distance, which the article’s own logic might lead to the conclusion of conscious non-visual perception. This includes the auditory system, which various fish species use for segmentation, recognition, and communication (Fay, Reference Fay1998; Ladich, Reference Ladich2019, Reference Ladich2023; Radford, Kerridge, & Simpson, Reference Radford, Kerridge and Simpson2014), abilities that are subserved at least in part in mammals by auditory ventral stream processes that are strongly associated with conscious perception (Dykstra, Cariani, & Gutschalk, Reference Dykstra, Cariani and Gutschalk2017; Snyder & Elhilali, Reference Snyder and Elhilali2017). Fish can also use other senses, such as smell, mechanosensation, and electroreception, to perceive other fish (Newton, Gill, & Kajiura, Reference Newton, Gill and Kajiura2019; Pohlmann, Grasso, & Breithaupt, Reference Pohlmann, Grasso and Breithaupt2001; Ylönen et al., Reference Ylönen, Kortet, Myntti and Vainikka2007). One might argue that such abilities could occur with unconscious sensory processing mechanisms, but again, this seems like a less-than-parsimonious conclusion.

It is also worth noting that some invertebrate aquatic animals, such as octopuses, cuttlefish, and squid, have very similar eyes to humans and may have haptic perception that similarly allows for object recognition, including cross-modal recognition (Kawashima & Ikeda, Reference Kawashima and Ikeda2025). This, along with evidence for problem-solving in octopuses (Dissegna et al., Reference Dissegna, Borrelli, Ponte, Chiandetti and Fiorito2023) and delayed gratification in cuttlefish (Schnell et al., Reference Schnell, Boeckle, Rivera, Clayton and Hanlon2021), provides suggestive evidence that these underwater animals are likely to be conscious of their surroundings, despite limited sensory horizons. Of course, octopuses and cuttlefish may be an exception to the rule that underwater species are not conscious, but the evidence seems more consistent with conscious perception in at least some fish and cephalopod species.

In addition to conscious perception in fish, it seems likely that they use brain circuits shared with mammals for certain kinds of conscious memory and navigation. For example, weakly electric fish engage in active electric sensing and navigation that is accompanied by activity in a brain area that may be homologous to the CA3 region of the hippocampus (Fotowat et al., Reference Fotowat, Lee, Jun and Maler2019), which connects bidirectionally to various cortical areas and is strongly associated with conscious spatial and non-spatial memory (Eichenbaum, Reference Eichenbaum2017). Goldfish (and turtles) show evidence of spatial learning (Rodríguez et al., Reference Rodríguez, López, Vargas, Gómez, Broglio and Salas2002), which in mammals is similarly dependent on the hippocampus. The hippocampus and the construction of cognitive models is interesting to consider here. Such model-based mechanisms that the target article would deny to fish and cephalopods seem especially useful in cases of limited sensory horizons, enabling avoidance of dangers and the seeking of foods that are not continuously detectable with the senses.

Before concluding, I acknowledge my slight bias in favor of the notion that sensory pathways are often sufficient for conscious perception. In contrast, higher-order theories and global workspace theories propose that frontal and parietal areas must receive information that forms the contents of conscious perception (see review of theories, Seth & Bayne, Reference Seth and Bayne2022); these theories would, therefore, deny conscious perception to fish and cephalopods even if they have functionally analogous sensory pathways as mammals. On the other hand, it is possible that frontal-parietal theories of consciousness will need to yield more nuanced mechanisms if they are to explain wider ranges of experience such as conscious emotional feelings and urges to move to music, which could arise from higher-order processing in areas such as the insula and premotor cortex (LeDoux & Brown, Reference LeDoux and Brown2017; Snyder, Gordon, & Hannon, Reference Snyder, Gordon and Hannon2024). Likewise, it seems unlikely that animals that lack frontal and parietal cortex – but otherwise have similar peripheral and central nervous systems to mammals – have no evolutionary path to conscious perception, despite the potential survival benefits of conscious processing (Cleermans & Tallon-Baudry, Reference Cleeremans and Tallon-Baudry2022).