The neural requirements for phenomenal consciousness were stipulated in Conclusion 4.33. I concluded that only mammals could be shown to meet these criteria. I then discussed doubts expressed about some of these requirements by some neurologists, and concluded that the evidence presented against the "majority view" did nothing to advance the case for phenomenal consciousness in birds and reptiles. In this section, I discuss three alternative arguments for consciousness in non-mammals: the argument from homology and two arguments from analogy.
The argument from homology: Which other vertebrates have brains that support consciousness?
The main reason why non-mammals could not be shown to satisfy the neural requirements for phenomenal consciousness identified above was that their brains lack a true neocortex, with its six-layered structure which allows a high degree of neural intra- and inter-connectivity. However, a very good case for the occurrence of consciousness in these creatures can be made if it can be shown that:
(a) their brains possess structures that are homologous (i.e. structures which have originated from the same structure in a common ancestor) to those that support consciousness in mammals;
(b) the structures also play an analogous causal role in regulating these creatures' behaviour; and
(c) their behaviour either:
(i) fulfils the criteria for primary consciousness (Baars, 2001; Rose, 2002) or
(ii) is of a level of complexity comparable with that of mammals.
Despite large differences in form and function, it is possible to identify homologous structures - which have originated from the same structure in a common ancestor - in the brains of mammals and vertebrate non-mammals.
According to Prescott, Redgrave and Gurney (1999) all vertebrates have brains that can be divided into a hindbrain (with a medulla and cerebellum), midbrain (with a tectum and tegmentum) and forebrain (subdivided into a diencephalon and telencephalon). A thalamus, hypothalamus and pituitary gland can be found in the diencephalon of all vertebrate brains, while the telencephalon contains a pallium with three major subdivisions in all vertebrates, as well as other structures (amygdala, septum and striatum).
In mammals, where the pallium is termed the cerebral cortex, most of the dorsal pallium forms the neocortex, while the medial pallium forms a group of structures that includes the hippocampus, and the lateral pallium forms the olfactory cortex. The extent to which each of these areas in mammals is homologous with like-named pallial areas in non-mammals is, however, only partially resolved (1999, p. 107, italics mine).
For the purposes of our inquiry, the neocortex, or isocortex, is the region that matters most, as neural events occurring below this level are inaccessible to consciousness (Roth, 2003, pp. 36-38). However, disagreements remain as to which part of the reptilian and avian telencephalon correspond to the neocortex. Currently there is no single criterion by which homology between structures can eb established. Commonly used criteria include: identical patterns of connectivity to other brain parts, neurochemistry; and embryonic orgins. However, these approaches yield inconsistent results (Aboitiz, Morales and Montiel, 2000).
Conclusion 4.46: Arguments for animal consciousness based on homology cannot be applied to non-mammals, as the mammalian neocortex has no clear counterpart in other vertebrates.
A causal argument analogical argument for consciousness in non-mammals
Despite the failure of the argument from homology, a strong case for the occurrence of consciousness in non-mammals can still be made if it can be shown that:
(a) their brains possess structures that are comparable in their degree of complexity to those that support consciousness in mammals;
(b) the structures also play an analogous causal role in regulating these creatures' behaviour; and
(c) their behaviour either:
(i) fulfils the criteria for primary consciousness (Baars, 2001; Rose, 2002) or
(ii) is of a level of complexity comparable with that of mammals.
In reptiles and birds, the dorsal ventricular ridge serves as a principal integratory centre and exhibits a pattern of auditory and visual connections with sensory centres and the thalamus which is broadly similar to that of the sensory neocortex in mammals. Fish and amphibians lack this structure (Russell, 1999; Aboitiz, Morales and Montiel, 2000). In birds, the dorsal ventricular ridge includes two areas: the hyperstriatum ventrale and neostriatum (Medina, 2002). There is good evidence that the mammalian neocortex and the neostriatum-hyperstriatum ventrale complex in birds have similar integrative roles. Interestingly, the relative size of the hyperstriatum ventrale in different species is the best predictor of their feeding innovation rate (Timmermans et al., 2000).
The ventricular ridges of birds are well-developed, but not laminated (Kavanau, 1997, p. 258).
However, even though largely non-laminated, the avian telencephalon [forebrain] can generate visual performances of a complexity rivaling and even exceeding those of mammals, previously thought to have been correlated uniquely with cortical lamination... The mechanisms of visual information processing in the brains of birds are ... at least as efficient as those in the mammalian striate cortex (Kavanau, 1997, p. 257).
Tool-making ability in different bird species has also been shown to correlate with the size of their neo- and hyper-striatum ventrale (Chappell and Kacelnik, 2004). The neostriatum caudolaterale is a structure believed to correspond to the frontal cortex in mammals, which is involved in planning of movement (Lissek and Gunturkun, 2003).
In this connection, Edelman also notes that "[a] good number of avian species migrate mind boggling distances with incredible fidelity year after year" (personal email, 19 July 2004).
The foregoing information suggests that the brains of birds have a neural complexity that is currently under-estimated.
Birds also clearly satisfy the second of the neural requirements for consciousness stipulated in Conclusion 4.33, as their waking and sleeping EEG patterns are similar to those of mammals. For other vertebrates, the situation is different:
Among vertebrates, true sleep, involving a shift from fast to slow waves in the forebrain, appears to be limited to mammals and birds, though there are hints of it in some reptiles (Cartmill, 2000).
It is not currently known if birds meet the first requirement for consciousness in Conclusion 4.33, as an analogue to the reentrant pattern of interaction in the mammalian thalamocortical system has yet to be identified (Kavanau, 1997, p. 257; Cartmill, 2000; Edelman, personal email, 19 July 2004).
On neurological grounds, the case for primary consciousness in birds is still weak but suggestive. However, this case is strengthened by the following facts:
(i) none of the scientists and researchers whom I contacted by email were able to nominate any kind of complex behaviour which could serve to identify conscious states or feelings, and which is found only in mammals; and
(ii) birds and mammals do, however, share a number of distinctive complex behaviours that have not been found in other vertebrates:
Interestingly, there are really only six animal groups capable of vocal learning: humans, cetaceans, bats, parrots, songbirds, and hummingbirds (see Jarvis et al., 2000). Parrots, in particular, represent an interesting case. The studies by Pepperberg and her colleagues (see Pepperberg, 1998; Pepperberg & Shive, 2001; Pepperberg & Wilcox, 2000) have shown that parrots are capable of acquiring a level of word comprehension rivalled in nonhuman species only by pygmy chimpanzees (see Savage-Rumbaugh, McDonald, Sevcik, Hopkins, & Rubert, 1986; Savage-Rumbaugh, Rumbaugh, & McDonald, 1985; Savage-Rumbaugh, Sevcik, Rumbaugh, & Rubert, 1985) (Edelman, personal email, 19 July 2004).
I acknowledge that we can conceive of the above activities occurring in the absence of phenomenal consciousness, but as I have argued earlier in this chapter, such thought experiments prove nothing. Rather, what I am advancing here is an argument from analogy:
(i) we already have very strong neurological grounds for believing that mammals are conscious;
(ii) birds have some neurological features that are analogous to those of mammals;
(iii) additionally, the complex behaviours exhibited by birds, which are the behaviours that are most likely to require phenomenal consciousness, are analogous to those of mammals, matching the corresponding behaviours of even the most advanced non-human mammals in their complexity;
(iv) there are no behaviours that would be expected of a phenomenally conscious animal that birds are known to lack, therefore
(v) it is likely that birds too are phenomenally conscious.
Conclusion 4.47: There is a powerful analogical argument, based on neurology and behaviour, for the occurrence of phenomenal consciousness in birds.
On the other hand, there are certain complex behaviours that are arguably pertinent to phenomenal consciousness and are found in mammals but not in reptiles, amphibians or fish:
It seems that a snake does not have a central representation of a mouse but relies solely on transduced information. The snake exploits three different sensory systems in relation to prey, like a mouse. To strike the mouse, the snake uses its visual system (or thermal sensors). When struck, the mouse normally does not die immediately, but runs away for some distance. To locate the mouse, once the prey has been struck, the snake uses its sense of smell. The search behavior is exclusively wired to this modality. Even if the mouse happens to die right in front of the eyes of the snake, it will still follow the smell trace of the mouse in order to find it. This unimodality is particularly evident in snakes like boas and pythons, where the prey often is held fast in the coils of the snake's body, when it e.g. hangs from a branch. Despite the fact that the snake must have ample proprioceptory information about the location of the prey it holds, it searches stochastically for it, all around, only with the help of the olfactory sense organs (Sjolander, 1993, p. 3).
Finally, after the mouse has been located, the snake must find its head in order to swallow it. This could obviously be done with the aid of smell or sight, but in snakes this process uses only tactile information. Thus the snake uses three separate modalities to catch and eat a mouse (Dennett, 1995b, p. 691).
A snake has no ability to anticipate that a mouse running behind a rock will reappear. Cats and other predatory mammals are able to anticipate that the prey will reappear (Grandin, 1998).
I have not been able to verify that birds possess these three abilities, but it is likely that they do: the last ability, in particular, requires high-level visual skills, which birds certainly possess.
As I suggested earlier in this chapter, the kind of consciousness presupposed here might be perhaps best thought of as an integrative consciousness. As this integrative consciousness seems to involve more than just attention and intentional agency, it does not appear reducible to access consciousness.
If some animals have a special kind of integrative consciousness, we have to ask how it is related to phenomenal consciousness. There may well be a nomic connection between the former and the latter. Could phenomenal consciousness exist without this integrative consciousness? Sjolander (1993) argued that the lack of these abilities he described should count as evidence against an animal's being phenomenally conscious, and I am inclined to agree with him, if the lack is permanent and life-long: a conscious newborn baby lacks the concept of object permanence, and an autistic person suffering from sensory overload is unable to integrate information (Grandin, 1998).
There remains the possibility, however, that reptiles may possess a very primitive affective consciousness of the kind described by Panksepp (1998, 2003), even in the absence of integrative consciousness. Later in this chapter, I will examine evidence cited in support of the view that reptiles possess conscious emotions, while fish and amphibians do not (Cabanac, 1999, 2003).
Just as we proposed a causal analogical argument in support of phenomenal consciousness in birds, we can now propose a similar argument against its occurrence in reptiles, fish and amphibians:
(i) the brains of fish and amphibians satisfy none of the neural requirements of consciousness described earlier in this chapter. The EEGs of reptiles share a few points of commonality with those of mammals and birds, they also fall far short of meeting the requirements;
(ii) the brains of fish and amphibians have no brain structure whose causal role is analogous to the mammalian neocortex, while the structure in reptiles whose causal role is analogous to that of the neocortex is very rudimentary;
(iii) additionally, reptiles, amphibians and fish appear to lack certain behavioural capacities that one might reasonably expect of a phenomenally conscious animal, therefore
(iv) it is likely that these animals are not phenomenally conscious.
Conclusion 4.48: There is a powerful analogical argument, based on neurology and behaviour, against the occurrence of phenomenal consciousness in reptiles, and an overwhelmingly strong argument against its occurrence in amphians and fish.
An inferential analogical argument for animal consciousness
For invertebrates, whose brains are too unlike those of mammals to permit a comparison of their brains with ours, it may be impossible even to find structures whose causal role is analogous to that of the mammalian neocortex. We are thus forced to rely on a more indirect, inferential approach: first, look for complex, widespread activity in the brain that might be analogous to that associated with mammalian conscious states; and second, identify behaviour patterns that are better explained in terms of a first-person account than by a third-person account.
Could we make a case for phenomenal consciousness based on behavioural analogies alone? We can certainly conceive of such a possibility, but it seems to be a law of nature in the real world that phenomenal consciousness is impossible without exceptionally high interconnectivity of the components of the conscious system (see e.g. Rose, 2002, p. 24).
On both neurological and behavioural grounds, honeybees and cephalopods (squid and octopuses) are the most likely candidates for phenomenal consciousness in invertebrates. The brain of an octopus, with its 300 million neurons (Chudler, 2003), is considered by David Edelman (2004) to be "an interesting challenge", although its brain has not yet been described in sufficient detail to suggest "the presence of the kind of complex, widespread activity that might be analogous to that associated with mammalian conscious states" (personal email, 19 July 2004). Additionally, its range of behaviour includes prey manipulation, tool use and possibly play (Mather and Anderson, 2000).
The 1 cubic millimeter brain of a honey bee, which only contains about 960,000 neurons (Giurfa, 2003), sounds far less promising, and David Edelman contends that "it is not likely that the interaction of ... a million neurons in, say a honeybee, would yield something we would call consciousness" (personal email, 19 July 2004). However, a honeybee's brain contains higher-order integration centres, which allow stimuli of different modalities to be associated during learning and recall (Giurfa, 2003, pp. 10-11). Additionally, honeybees are capable of some impressive behavioural feats:
Honey bees exhibit complex, non-elementary forms of learning that are akin to higher-order cognitive performances such as contextual learning, categorization and learning of abstract rules, performances that were usually attributed exclusively to invertebrates (Giurfa, 2003, pp. 15-16).
Conclusion 4.49: The vast majority of invertebrates are likely to lack consciousness. The two best invertebrate candidates for consciousness are octopuses and honeybees.