I consider most relevant the challenge that Morsella et al.'s proposal poses to the traditional account of consciousness in terms of higher cognitive/executive processes, which is based mainly on visual research data. Among the various issues their target article raises, the focus on the olfactory system merits attention, given the general misconception about the low cognitive level of chemosensory modalities such as olfaction and taste. Because a similar picture is found with respect to taste, I consider it relevant to draw the authors' attention to the gustatory modality as a valuable case for the theory, because taste perception is closely linked to action. Except in laboratory settings where taste solutions can be applied to unconscious subjects, tasting under normal circumstances requires voluntary ingestion. Thus, taste research in rodents can shed light on the following issues.

First, the role of memory in modifying the percept which leads to action seems to be dismissed in the theory. When speaking about motivation, Morsella et al. mention the role of memory as part of general higher-order cognitive processes with little influence over perceptual processes. In fact, they state that “the [conscious] field itself has no memory … it only presents, for action systems, the outputs of dedicated memory systems” (sect. 6, para. 2). However, even in the “creature in the cave” metaphor the conflict arises from memory, because the smell of smoke has necessarily been previously associated with danger in order to induce the urge to exit. It can be said that memory is in the field of consciousness even if it is not conscious. Thus, memory has presumably modified the smell of smoke at the lower level of sensory processing. Likewise, the gustatory system is modified by learning and memory even at the lower sensory relay levels in the brainstem (Yamamoto & Yasoshima Reference Yamamoto, Yasoshima and Bermudez-Rattoni2007). Thereafter, it can be proposed that the conscious content of a food item necessarily includes its hedonic value that in turn depends on memory. A taste is novel, safe, or aversive but never neutral, because tasting implies ingesting and this is followed by unavoidable consequences. The outcome can be either positive, such as to relieve thirst and/or hunger, or negative, such as to experience visceral malaise (see Gallo & Rolls [Reference Gallo and Rolls2012] for a special issue on the topic). According to the authors' proposal, for the conscious field to be complete and unambiguous during action selection, it would not be adaptive for a taste to be represented without its hedonic features, just as it would not be adaptive for an object to be represented in the wrong spatial location.

Second, animal research on taste memory is consistent with the authors' proposal, which locates consciousness at a lower level of the nervous function than previously proposed. For example, in the anatomically based dichotomy between declarative and non-declarative long-term memory (Squire Reference Squire2004), consciousness has been traditionally associated with declarative memory (also termed explicit and cognitive memory). This conscious memory is thought to be mediated by a medial temporal lobe system involving the hippocampus and the adjacent perirhinal, entorhinal, and parahippocampal cortex, as well as connections with other cortical areas. Visual and taste recognition memory tasks in rats have been conventionally viewed as paradigms of declarative and non-declarative memory, respectively. The primitive phylogenetic and ontogenetic origin of the gustatory system might have contributed to this state of affairs. Taste recognition memory represents a basic adaptive mechanism. The term refers to the ability to assess the familiarity of a taste that was not followed by negative consequences in previous encounters. Concomitantly with the taste being classified as safe, consumption increases, thus showing attenuation of neophobia. Research evidence using recognition memory tasks suggests that taste and visual memories share an anatomical basis. In fact, consolidation of safe taste memories has been associated with the hippocampus and the perirhinal cortex (De la Cruz et al. Reference De la Cruz, Rodriguez-Ortiz, Balderas and Bermudez-Rattoni2008). Also, we have assessed Fos-like immunoreactivity to demonstrate amygdala-dependent changes in the activity pattern of the perirhinal cortex during the formation of taste memory (Gómez-Chacón et al. Reference Gómez-Chacón, Gámiz and Gallo2012; Reference Gómez-Chacón, Morillas and Gallo2015). This indicates that taste recognition memory involves some of the areas previously related to visual recognition memory. Therefore, it seems that the strict anatomically based dichotomy of declarative versus non-declarative memory applied to visual and taste recognition memory may need to be updated when exploring memory maintenance mechanisms because independent circuits do not function in isolation from each other and may share common components.

Furthermore, in visual recognition memory, there are data supporting the relevance of neural circuits that involve lower brain areas, such as the thalamus. In fact, recollection and familiarity, proposed as different components of recognition memory, seem to depend on dissociable brain circuits. Recollection involves remembering specific contextual details about a prior episode, whereas familiarity involves simply knowing that an item was presented, without having available any additional information about the learning episode (Squire et al. Reference Squire, Wixted and Clark2007). In addition to the functional dissociation of the temporal lobe circuits that attributes to the hippocampus and the perirhinal cortex selective roles in recollection and familiarity, respectively (Aggleton & Brown Reference Aggleton and Brown2006), the neural circuits proposed include, among other subcortical areas, the anterior and the mediodorsal thalamic nuclei, respectively (Carlesimo et al. Reference Carlesimo, Lombardi, Caltagirone and Barban2015).

In all, it can be proposed that the authors' theory would benefit from reconsidering the role of memory, whether available to consciousness or not, because it modifies the encapsulated perceptual contents themselves. Taste memory is a good model to be taken into account.