In our comment, we suggest that the notion of “sensory horizon” risks overlooking the multisensory nature of animal minds. According to the authors, “an organism’s sensory horizon is the spatial limits of a particular sensory modality” (p. 6; emphasis added). This might be fine as an abstract definition, but such a unimodal notion flies in the face of actual aquatic dynamics.

To begin with, many aquatic vertebrates, especially teleost fish, possess sophisticated auditory systems that are well adapted for underwater sound propagation (Popper & Fay, Reference Popper, Fay, Fay and Popper1999). These auditory systems support vital behaviours such as predator evasion, prey detection, communication, and navigation – all of which imply a significant degree of sensory integration. The lateral line system, which is present in most aquatic vertebrates, exemplifies a specialised mechanosensory modality that detects water movements and vibrations; it plays a crucial role in spatial orientation and environmental perception (Coombs, Görner, & Münz, Reference Coombs, Görner and Münz1989). Moreover, there is strong evidence that aquatic animals are capable of visuo-auditory integration. For example, goldfish (Carassius auratus) exhibits enhanced startle responses when exposed simultaneously to visual and auditory stimuli, demonstrating multisensory enhancement effects that parallel those observed in mammals and birds (Fay, Reference Fay, Popper and Fay2005). In zebrafish (Danio rerio), calcium imaging studies have revealed that brain regions such as the optic tectum and torus semicircularis respond to both visual and auditory inputs, showing clear cross-modal modulation (Truszkowski et al., Reference Truszkowski, Carrillo, Bleier, Ramirez-Vizcarrondo, Felch, McQuillan, Truszkowski, Khakhalin and Engert2017). These brain structures functionally resemble the mammalian superior colliculus, suggesting a conserved neural architecture supporting multisensory integration across vertebrate species.

Evolutionary evidence also points to the direction that multisensory integration likely began in aquatic environments. Clack (Reference Clack2012) emphasises the gradual modification of aquatic auditory systems into structures adapted for terrestrial hearing, highlighting an evolutionary continuity rather than a sudden innovation. Auditory localisation, a crucial ability for spatial cognition, is well documented not only in aquatic but also in amphibious species (Edds-Walton & Fay, Reference Edds-Walton and Fay2008), posing a challenge to the view that vision alone enabled unconscious multisensory integration. From a more ecologically grounded perspective, cross-modal integration may have evolved incrementally, shaped by the sensory demands of aquatic organisms. Ghazanfar and Schroeder (Reference Ghazanfar and Schroeder2006) argue that multisensory integration is a fundamental feature of neural processing, reflecting the variety of ecological pressures animals face. While terrestrial life certainly expanded the functional role and complexity of vision, it may not have been the original source of conscious vision, as the target article seems to suggest.

Even without multisensory integration, although aquatic animals vary in their reliance on vision and hearing for prey detection and predator avoidance depending on species and environmental conditions, many species rely more heavily on auditory cues, particularly sound localisation and echolocation, to detect prey and evade predators, especially in environments where visibility is limited (Au, Reference Au1993; Fay & Popper, Reference Fay and Popper2000). Hearing provides aquatic animals with 360-degree environmental information and can penetrate visual blind spots. Since sound travels farther underwater, it enables these animals to effectively detect environmental sounds, prey movements, and predator locations. Given all these, the authors’ divide-and-conquer strategy beginning from vision seems all the more inapt.

More positively, we urge that multisensory horizon should be in place right at the beginning of the analysis. The main reason is that most (if not all) creatures that can perceive at all do have more than one sensory modality. In the authors’ account, it works as if those aquatic animals first anchor their sensory horizons with vision only, even if they have other senses, and then broaden their sensory horizons after landing, as it were, again solely due to vision. This seems ecologically and evolutionarily unsound. Just to give more examples, in addition to audition, aquatic animals (especially fish and crustaceans) possess highly developed olfactory and tactile senses crucial for survival, foraging, predator avoidance, and reproduction. Most fish have sensitive olfactory organs located at the snout that detect dissolved chemical molecules in water, enabling behaviours such as food search, predator detection, mate localisation, and homing, exemplified by salmon’s precise return to natal streams and sharks’ ability to detect prey odours from kilometres away using binaural olfactory cues – hammerhead sharks’ widely spaced nostrils enhance directional tracking (Hara, Reference Hara1994). Tactile sensation is mediated through the skin and specialised structures like barbels, scales, and fin rays; benthic fish (e.g., carp, catfish) and crustaceans (e.g., shrimps, crabs) heavily rely on sensitive tactile receptors on barbels and body surfaces to explore environments, forage, and recognise obstacles. These tactile systems are highly responsive to water flow, contact stimuli, and fine particulate matter, serving as primary detection mechanisms underwater (Bleckmann, Reference Bleckmann1994). In turbid waters or nocturnal conditions, olfaction and touch compensate for limited vision, with benthic species often displaying reduced visual sensitivity but enhanced tactile and olfactory capabilities (Jelle, Fay, & Tavolga, Reference Jelle, Fay and Tavolga1988). Some aquatic animals also possess taste receptors distributed on body surfaces or specialised organs (e.g., octopus suckers), where combined chemosensory inputs further augment detection abilities (Derby & Weissburg, Reference Derby, Weissburg, Derby and Thiel2014). Note that the sensitivity of olfactory and tactile systems varies with species, habitat zone (benthic vs. pelagic), and life history. Chemical signal transmission underwater differs from air due to solubility and flow dynamics, necessitating sensory adaptations (Nevitt, Reference Nevitt1999). Non-vertebrate aquatic taxa like crustaceans and molluscs also exhibit sophisticated olfactory and tactile functions – for example, shrimps’ antennules serve dual chemosensory and mechanosensory roles (Derby & Weissburg, Reference Derby, Weissburg, Derby and Thiel2014). Overall, olfaction and touch are indispensable sensory modalities for many aquatic animals, especially where vision is limited.

The visuocentric tendency of the target article can probably be fixed easily. For example, when the authors mention “a limited sensory horizon” in the introduction, they could consider that the horizon in question is not as limited given its multisensory nature. They could also rethink what should be said when they mention “visual horizon” (introduction) and “visual sensory volume” (figure 3). More generally, postdiction in other sensory modalities could also be taken into consideration (Stiles, Tanguay, & Shimojo, Reference Stiles, Tanguay and Shimojo2022). Overall, evolutionary stories have to be speculative, and we have no objection to that. However, such speculations should be based on realistic descriptions, in this case, that of the aquatic world.