A Neurological Excursus: Can feelings occur in animals lacking a neocortex?
The neurological arguments against the possibility of feelings in non-mammals are formidable. Anyone who wishes to argue for the possibility of feelings in non-mammals (as I intend to) needs to confront these arguments squarely. However, a few neuroscientists (Panksepp, Liotti, Denton, Parsons and Koch), have publicly argued for the possibility of consciousness in animals lacking a neocortex. I propose to evaluate their arguments and discuss the implications for animal consciousness. In this section, I argue that although these authors raise some valid questions concerning the neural requirements for consciousness, the evidence they raise turns out to be irrelevant to the question of whether birds and reptiles are phenomenally conscious. In the following section, shall develop an alternative line of argument that birds are phenomenally conscious.
I wish to highlight five points in the following exposition.
First, despite our uncertainty about why consciousness evolved and what the minimum neural requirements for consciousness are, there are good reasons to believe that it requires a high level of neural intra- and inter-connectivity, such as can be found in the neocortex.
Second, there is good evidence for the view that we possess two forms of consciousness, which are really and not merely conceptually distinct. One form of consciousness corresponds to feelings and crude, low-level processing of sensory inputs; the other is associated with more detailed, fine-grained processing of sensory inputs. Some authors refer to these as affective and cognitive consciousness - a misleading terminology, in my view, as emotions require at least minimal cognition: typically, our feelings are directed at things in our environment, about which we have beliefs. There is some basis for the popular view that animals may be capable of having phenomenally conscious feelings despite having a low level of cognitive sophistication.
Third, the roots of our emotions go back a long way, as they are controlled by ancient structures in the brain, below the level of the cortex. (Deep emotions can even inhibit cognitive activity in the neocortex.) Animals had emotions long before they developed cognitive consciousness.
Fourth, the part of the brain's limbic system which is most strongly associated with conscious emotional feelings is the anterior cingulate cortex. (Processing in other regions of the limbic system appears to take place below the level of consciousness.) Although the anterior cingulate cortex, which borders the neocortex, has long been regarded as part of the limbic system and hence more primitive than the neocortex, it actually possesses an extremely complex laminar structure, and recent research (Allman et al., 2001) suggests that it should be regarded as part of the neocortex. Moreover, it is unique to mammals. While emotions have a long pedigree, it is thus likely that they were subconscious for most of their evolutionary history and became conscious feelings relatively recently. Most animals that have emotions probably lack phenomenal consciousness.
Fifth, behavioural studies of human beings and animals that lack a cerebral cortex but appear to be conscious are inconclusive. The possibility that the emotional behaviours displayed may be unconscious cannot be excluded. Moreover, for animals whose cerebral cortex was removed during infancy, the trauma of decerebration may have affected the neural development of their brain stem, effectively "corticising" it so that some parts were able to take over some of the functionality normally handled by a cerebral cortex. Something similar may have also happened to human children born with hydrancephaly. Also, the degree of brain damage varies widely among these children: not all completely lack a cerebral cortex.
The hypothesis that animals can experience conscious emotions that are completely independent of both the neocortex and anterior cingulate cortex therefore remains unproven, from both a neurological and a behavioural perspective. The data cited below, while intriguing, fails to establish the possibility of conscious feelings in reptiles and birds.
The role of the brain in consciousness: what neurologists are still not certain about
I would like to begin my review of the literature by citing the response of one neuroscientist, Jaak Panksepp, to an email query of mine: "Is there currently a 'consensus' view among neurologists as to the minimum requirements for consciousness?" Panksepp's forthright reply was: "No there is not" (personal email, 15 June 2004).
The current controversy among neurologists regarding primary consciousness centres on the following four areas:
Does consciousness require a high degree of neural connectivity?
The view that consciousness is made possible by the high degree of connectivity between the brain's neurons, is a common but by no means universal one. On this view, consciousness is a property not of individual neurons but only of a very large collection of interacting neurons. Koch and Crick question this assumption:
An alternative hypothesis is that there are special sets of "consciousness" neurons distributed throughout cortex and associated systems. Such neurons represent the ultimate neuronal correlate of consciousness, in the sense that the relevant activity of an appropriate subset of them is both necessary and sufficient to give rise to an appropriate conscious experience or percept (Koch and Crick, 2001).If consciousness does not require a high degree of inter-connectivity and can reside in small sets of neurons, then the argument that it can only reside in the highly inter-connected regions of the cerebral cortex is undermined. It might then be the case that animals with a very primitive cortex - or none at all - are phenomenally conscious.
On the other hand, scientists who question the view that connectivity is what makes consciousness possible owe us an alternative explanation of recent studies (Laureys et al., 2002; Baars, 2003) showing that patients in a persistent vegetative state, none of whom were suffering from brainstem damage that was known to cause a loss of consciousness, displayed a marked reduction in connectivity between different regions of their cortex. Moreover, recovery in PVS patients is correlated with a return of function of the associative cortical areas (Rose, personal email, 22 July 2004). As far as I am aware, no alternative model has been put forward which would account for these phenomena.
Conclusion 4.36 There is good prima facie evidence that primary consciousness requires a high degree of neural inter-connectivity.
What is the minimum number of neurons required to generate consciousness in an animal's brain?
Edelman and Tononi (1999) list sufficient (nonsensory) cortical mass as one of the criteria required to support primary consciousness in animals (see Rose, 2002, p. 24). Koch (2003) is one prominent dissenter from this majority view:
Why be a cortical chauvinist? Do we really know that the cerebral cortex and its satellites are necessary for consciousness? Why not squids? Or bees? Endowed with one million neurons, bees can perform complicated actions, including amazing feats of visual pattern matching. For all I know, a couple of thousand neurons and their associated machinery may be sufficient to see, to smell, to feel pain and to enjoy life! Maybe even fruitflies are conscious, to a very limited extent. Today we just don't know (Koch, 2003, p. 320).
As far as I know, Koch is the only neuroscientist to have suggested that insects may possess primary or phenomenal consciousness. and the view of David Edelman, that "it is not likely that the interaction of a couple of thousand, or even a million neurons in, say a honeybee, would yield something we would call consciousness" (personal email, 19 July 2004), is a fairly representative one.
Conclusion 4.37: The minimum neural requirements for consciousness remain unknown.
Why did consciousness originate, and what biological function does it serve?
Consciousness, in most models that have been developed to date, is supposed to have arisen in order to "produce the best current interpretation of the environment - in the light of past experiences - and to make it available, for a sufficient time, to the parts of the brain which contemplate, plan and execute voluntary motor outputs (including language)" (Koch and Crick, 2001). Edelman and Tononi (2000) propose a similar account, in which consciousness arose as an ability to categorise incoming perceptual stimuli and create a "scene" in the mind's eye. In both accounts, consciousness developed to cope with the brain's sensory processing requirements, and serves as a way of responding to external events.
According to very different, motor-based account by Cotterill (1997, 2001, 2002), consciousness arose as a way of coping with unexpected events for animals that engaged in self-paced probings of their surroundings, when the ancestors of today's mammals (and possibly birds) acquired the capacity to: (i) detect and remember over a short span any deviations from the expected bodily feedback from their surroundings, (ii) veto (in real time) routine movements that might place them in danger, and (iii) learn new sequences of muscle movements (context-specific reflexes) as a way of dealing with similar situations in future.
What Cotterill's kinesthetic account shares in common with the foregoing "visual" accounts is a focus on external events as an explanation for consciousness. However, other neuroscientists (Denton et al., 1996; Denton et al., 1999; Egan et al., 2003; Liotti et al., 2001; Parsons et al., 2000; Parsons et al., 2001) have recently developed an alternative, "internalist" model, in which consciousness arose as an emotional response to sensations within the body - such as extreme thirst or hunger for air, pain, hunger for food and extreme temperature changes - which compel an animal to pay attention and respond rapidly because its existence is immediately threatened. On this accounts, primary consciousness is considered to have arisen within the brain's basic vegetative systems. Primal emotions are thus "affections of the mind arising from bodily states" (Liotti et al, 2001, p. 2038, citing a definition in the Oxford Dictionary).
According to Liotti and Panksepp (2003) behaviours such as thirst, air hunger, hunger and impeded micturition (an urge to urinate) are likely to have triggered the development of consciousness:
All these basic and phylogenetically old behaviours share an affective appraisal of danger, likely signaling and prompting the rest of the brain onto a course of action that helps promote survival. Because such responses are present throughout phylogeny, subjective awareness of changes in internal states may represent the earliest glimmer of consciousness.
Egan et al. (2003) use the phenomenal experience of satiation to illustrate the advantage of consciousness for a thirsty animal. Animals experience a feeling of satiation soon after drinking - long before the water they have drunk has replenished their bodies. Egan et al. (2003, p. 15246) argue that the ability to experience this rapid feeling of satiation would confer a survival benefit on an animal that needed to make a quick exit from a water hole, where predators might be waiting.
Some philosophers might object to the apparent conceptual confusion here between two kinds of consciousness - access consciousness and phenomenal consciousness. They will argue that phenomenology is redundant here: to make a quick exit from a water-hole, an animal does not need to feel sated; all it needs is to be able to access its internal bodily states. However, this objection overlooks the possibility that there may well be a law of nature linking this kind of access consciousness to the feeling of satiety. In that case, it would still be (counterfactually) true that without this feeling, an animal would not be able to make a quick get-away.
A more telling objection is that the "internalist" theory described above account for only some of the distinguishing features of consciousness. In particular, it is unable to explain the perceptal unity of consciousness (Rose, personal email, 22 July 2004), or its dynamic nature - consciousness is more aptly described as a process than a state, and its most likely source is therefore the "dynamic core" of interactions between the cortex, thalamus and basal ganglia (David Edelman, personal email, 19 July 2004).
However, defenders of this "internalist" account commonly posit the existence of not one but two forms of consciousness in the brain. The account is therefore not intended as a global account of consciousness, but as an explanation of how consciousness might have originated, in its most primitive form.
Conclusion 4.38: There is no consensus among neurologists about why consciousness evolved. There is a much greater consensus, however, about how it is produced.
Is there a more primitive form of consciousness in the brainstem, independent of the cerebral cortex?
Panksepp (1998, 2001, 2003f) has proposed that we possess two distinct kinds of consciousness: cognitive consciousness, which includes perceptions, thoughts and higher-level thoughts about thoughts and requires a cortex, and affective consciousness, which relates to our feelings and arises within the brain's limbic system. Of course, these two kinds of awareness interact continually, but according to Panksepp, the latter is the more ancient, and it is controlled below the level of the cortex, in the brainstem. Panksepp has suggested (1998, p. 314) that this affective consciousness first arose in a region of the mid-brain, known as the peri-acqueductal gray (PAG).
If this alternative view is correct, even animals with a primitive cerebral cortex - or none at all - may have primitive feelings which are consciously experienced (Panksepp 1998, 2001, 2003f; Denton et al. 1999; Parsons et al., 2000; Parsons et al., 2001; Liotti et al. 2001; Cabanac, 2002, 2004; Liotti and Panksepp, 2003). According to this account, affective consciousness is phylogenetically very ancient: it is certainly shared by all mammals and possibly reptiles and birds as well. Panksepp proposes that some human beings in a permanent vegetative state, as well as some anencephalic infants, may also possess it (personal email, 15 June 2004).
The following table summarises the main arguments adduced by Panksepp (2003) for a real distinction between two forms of consciousness in the brain:
Table 4.2 Contrasts between Affective and Cognitive Consciousness
| Affective awareness | Cognitive consciousness |
| Intrinsically valenced - characterised by positive or negative feelings. | Not valenced. |
| Largely subcortical. Emotional responses and many basic affective tendencies survive many forms of cortical brain damage that severely impair cognitions. | Cognitions are largely cortical and are impaired by cortical brain damage. |
| Affects are more powerful and easier to induce in the young. Children are very "emotionally alive". | Sophisticated cognitive activities prevail among adults. |
| Feelings are easily activated by direct brain stimulation. Affects may be generated more by analog types of neurohumoral processes. | Conscious cognitions are difficult to activate by direct brain stimulation. Cognitions may be generated more by digital-type computations. |
| Emotions generate spontaneous, trans-cultural, facial and bodily expressions as well as prosodic vocal changes. | Cognitions do not generate this kind of behaviour. |
| In general, our right cerebral hemisphere tends to be more emotionally deep and perhaps negativistic (or realistic). When the more emotionally introspective right hemisphere is damaged, the linguistically proficient left hemisphere commonly carries on as if nothing very serious has transpired and chooses to repress negative emotions. At its most extreme, right hemisphere damaged patients often deny that their left side is even paralysed when it clearly is, from an objective standpoint. | The left hemisphere tends to be more cognitively skilled and positively valenced in comparison. Thus left hemisphere-damaged individuals are very much aware of their post-stroke plight. |
Of the contrasts listed here between emotion and cognition, those of special significance are the fact that the two forms of consciousness are controlled by separate regions of the brain and the fact that one can be badly damaged without having much effect on the other. These facts indicate that the distinction between the two forms of consciousness is real rather than merely conceptual.
However, the distinction between these two forms of consciousness should not be exaggerated: in normal individuals, they are highly integrated. Recent studies (Allman et al., 2001) suggesting that the anterior cingulate cortex of the brain has an important role both in regulating the emotions and in rational problem-solving, illustrate this point.
I would also like to observe that the terminology used is philosophically misleading, as it suggests that affective consciousness is completely independent of cognition. It has been argued above that emotions can only be attributed to organisms that are capable of having beliefs. If this is the case, then affective consciousness must also contain a cognitive element - even if the sensory processing involved is very low-level.
The distinction between cognitive and emotional processes within the brain is generally accepted by neuroscientists (see LeDoux, 1998, p. 161). However, the notion that there are two forms of consciousness - in the brain is much more contentious. The key point at issue is the role of affect (phenomenally conscious feeling) in emotions. LeDoux, a notable adversary of the "affective neuroscience" perspective approach championed by Panksepp, questions its relevance on the grounds that conscious feeling is not an essential feature of emotion - the icing on the cake is how he describes it (1998, p. 302) - and is likely to have emerged only recently in evolutionary history.
There are two ways in which Panksepp and his colleagues could buttress their claim that animal consciousness originally arose in the brainstem. First, brain imaging studies of human patients experiencing powerful emotions should reveal an activiation of regions in the brain stem. Second, affective consciousness should remain intact in human beings or animals lacking a cerebral cortex.
It is generally agreed that human beings and animals that lack a cerebral cortex cannot report their feelings - in other words, they do not possess primary consciousness. Neuroscientists choosing to use the second method of validating affective consciousness are therefore bound to propose alternative diagnostic criteria for conscious. I will discuss these issues below.
Conclusion 4.39: There is good prima facie evidence for the existence of two forms of consciousness in humans and other mammals. One form of consciousness is more "affective" and is associated with feelings and crude, low-level processing of sensory inputs; the other is more "cognitive", and is linked with more detailed, fine-grained processing of sensory inputs.
Do brain imaging studies reveal an affective consciousness that is independent of the cerebral cortex?
Is the cerebellum conscious?
One region of the hind-brain which has recently attracted a great deal of scientific attention is the cerebellum, a phylogenetically ancient structure which is found in all vertebrates. Parsons et al. (2000, 2001) cite recent brain imaging studies indicating that the cerebellum, long considered strictly a motor control structure, plays and important role in sensation, cognition, language processing, affect, thirst and hunger, even in the absence of motor behaviour on the subject's part:
Neuroimaging and neurological studies ... suggest cerebellar involvement in the generation of words according to a semantic rule, timing of events, solving perceptual and spatial reasoning problems, mental rotation, visual information processing, cutaneous and tactile discrimination, kinesthetic sensation, and working memory, among other processes... It has been proposed that the lateral cerebellum may be activated during several motor, perceptual, and cognitive processes specifically because of the requirement to monitor and adjust the acquisition of sensory data... Furthermore, there are reports suggesting the involvement of posterior vermal cerebellum in affect (Parsons et al., 2000, p. 2332, italics mine).
Curiously, the size of an individual's cerebellum has also been shown to correlate well with his or her IQ, which is equal to or even higher than any other correlation between IQ and the brain. In rats, the volume of the cerebellum's outer molecular layer correlates with curiosity (Skoyles, 1999).
Neuroimaging studies show that the cerebellum also plays a vital role in air hunger, thirst and hunger for food - all of which stimulate the cerebellum in a similar manner (Parsons et al., 2001). The cerebellum has reciprocal ancient connections to the hypothalamus, a structure important in vegetative functions. The authors suggest four possible roles for the cerebellum in air hunger - it may (i) subserve implicit intentions to breathe, (ii) provide internal models to predict the consequences of inhaling carbon dioxide, (iii) modulate emotional responses, or (iv) monitor sensory data. In a related article, Parsons et al. discuss the role of the cerebellum in thirst and speculate that "cerebellar involvement in thirst may be related to the intention to drink, inextricably interwoven in the subjective state of thirst, together with a conscious state oriented toward satiation of a desire" (2000, p. 2334), but acknowledge that this is merely a hypothesis, and that more work needs to be done.
Additionally, lesions of the cerebellum in monkeys appear to affect their emotional dispositions, making them more docile (Parsons et al., 2000).
These results are highly suggestive. For the time being, all we can say is that the cerebellum plays a causal role in primal emotions and some cognitive tasks. However, the studies do not show whether activity in the cerebellum is an enabling factor, a modulating factor or part-and-parcel of the experiences themselves.
Although the cerebellum has a connection with higher cognitive functions, the most likely reason for this connection is that it is linked to higher association areas of the cerebral cortex, particularly the prefrontal cortex. Generally, the role of the cerebellum appears to be one of tracking inputs used by a wide range of processes, performed subcortically and cortically (Skoyles, 1999).
As we noted earlier, the fact that ablation of the cerebellum in animals destroys their motor co-ordination but leaves their sensations intact, constitutes one powerful argument that this region plays a minimal role in consciousness as such (Tononi, 2004; Panksepp, 1998, p. 314).
Although the cerebellum contains a large number of neurons and possesses high internal connectivity, the connectivity does not apply across disparate regions: "Individual patches of cerebellar cortex tend to be activated independently of one another, with little interaction between distant patches" (Tononi, 2004).
Conclusion 4.40: While there is increasing evidence that the cerebellum plays a significant role in various forms of conscious activity, its neural features, coupled with data from ablation studies, suggest that it probably could not support consciousness in the cerebral cortex were absent.
The role of the brainstem in affective states and vegetative functions: an overview of results
Summarising the data from animal studies, Liotti and Panksepp (2003) claim that at least seven core emtional systems have been provisionally identified in mammals, and that these emotional operating systems are regulated by subcortical regions of the mammalian brain. These systems were described in the previous chapter.
Liotti et al. (2001) also show that vegetative functions such as hunger for air, thirst, and hunger for food all activate similar regions of the brain: "There is evidence of commonality in structures involving these primal emotions" (2001, p. 2039). They conclude (2001, p. 2040) that "primary consciousness ... involves essential elements in the rhombencephalon [hind-brain], mesencephalon [mid-brain], hypothalamus, thalamus [upper brain stem], amygdala", and other areas, including the anterior cingulate, which borders the cerebral cortex.
Three comments are in order here. First, it is important not to over-state the evidence for conscious processing below the level of the cerebral cortex. A variety of regions of the brain are activated during thirst and air hunger, and some of these are likely to be merely enabling or modulating factors for consciousness. However, they do not show that consciousness can occur in the absence of a cortex.
Second, the fact that conscious feelings are controlled by core systems in the brainstem does not imply that these subcortical systems are sufficient for consciousness - a point acknowledged by Liotti and Panksepp (2003).
Third, it is the anterior cingulate cortex whose activity appears to correlate most consistently with so-called "primal" forms of consciousness such as air hunger and thirst (Liotti et al., 2001, Egan et al., 2003, Liotti and Panksepp, 2003). This region of the brain with a complex layered structure which borders the cerebral cortex:
A meta-analysis of neuroimaging studies involving basic drives (thirst, air hunger, hunger, impeded micturition) shows an overlap of activations in the regions of the dorsal anterior and middle cingulate gyrus (Liotti and Panksepp, 2003).
Conclusion 4.41: Conscious feelings and vegetative functions are controlled by core systems below the level of the cortex. The subcortical region whose activity appears to correlate most consistently with so-called "primal" forms of consciousness is the anterior cingulate cortex, a structure in the brain's limbic system.
Is the anterior cingulate complex subcortical?
Liotti and Panksepp (2003) propose the following interpretation of the results to date:
Several lines of evidence suggest that peri-conscious, affective processing of emotion takes place sub-cortically, in areas such as brainstem, hypothalamus and amygdala, while cognitive appraisal of emotions takes place in prefrontal cortex and anterior cingulate cortex (Liotti and Panksepp, 2003).But does the term "peri-conscious" signify phenomenal experience, or some state underlying it? The terminology introduced at this point is unclear.
Liotti and Panksepp suggest that the anterior cingulate cortex serves as an alarm centre, "alerting the organism that immediate conscious action needs to be taken in order to remove the threat" (2003).
The authors seem to share the assumption that because the anterior cingulate cortex is part of the "limbic system" of the brain, it is a subcortical structure that predates the cerebral cortex. However Allman et al. (2001) have recently challenged the commonly held assumption that the anterior cingulate cortex has a more primitive laminar structure than neocortex and preceded it in evolution:
[W]e propose that the anterior cingulate cortex is a specialization of neocortex rather than a more primitive stage of cortical evolution. The evidence from single neuron recording, electrical stimulation, EEG, PET, fMRI and lesion studies indicate that the anterior cingulate cortex has an important role in emotional self-control as well as focused problem solving, error recognition, and adaptive response to changing conditions (Allman, Hakeem, Erwin, Nimchinsky and Hof, 2001).It should also be pointed out that the anterior cingulate cortex is not found in birds or reptiles.
Conclusion 4.42: As the anterior cingulate cortex is unique to mammals, and may in fact be part of the neocortex, the discovery that this structure may support so-called "primal" forms of consciousness, lends no support whatsoever to the case for phenomenal consciousness in birds and reptiles.
Can affective consciousness occur in individuals lacking a cerebral cortex?
Brain imaging studies have failed to establish that consciousness can occur in the absence of a cerebral cortex. However, the identification of consciousness in individuals lacking a cerebral cortex would certainly clinch the case. I propose to examine the evidence for consciousness in human children suffering from anencephaly and from animals.
Evidence from anencephalic children
Merker (2003) defines the condition of hydranencephaly as follows:
Basically hydranencephaly indicates that a child is missing much or most of their cerebral hemispheres, that is, the two masses of folded brain tissue (cortex) that surround the brain stem. Literally "anencephaly" means "without brain", but this is technically incorrect as a term for the cases to which it is applied, which almost invariably have a brain stem (Merker, 2003).
Despite a current consensus among neurologists that "[p]ain and suffering are attributes requiring cerebral cortical functioning" (American Academy of Neurology, 1989) and that "[i]nfants with anencephaly, lacking functioning cerebral cortex, are permanently unconscious" (Medical Task Force on Anencephaly, 1990), Panksepp and "quite a few other neurologists" consider that the behaviours exhibited by anencephalic infants and decorticate human beings, "have a high probability of reflecting affective experience" (personal email, 15 June 2004).
As evidence, Panksepp cites an article by Shewmon (1999) describing three cases of children born with hydranencephaly (a rare condition in which the brain's cerebral hemispheres are absent and replaced by sacs filled with cerebrospinal fluid) and another child born with a similar condition. "Each of these children defied a prognosis of permanent vegetative state, rendered with absolute certainty by multiple physicians, including pediatric neurologists and neurosurgeons" (Shewmon, 1999, p. 370). Despite the total or near-total absence of a cerebral cortex in these children, they proved to be capable of a variety of forms of discriminative awareness, including "distinguishing familiar from unfamiliar people and environments, social interaction, functional vision, orienting, musical preferences, appropriate affective responses, and associative learning" (Shewmon, 1999, p. 364). The two children with rudimentary limbic structures were more affective and sociable, and possessed more motor function (Shewmon, 1999, p. 371).
The following account by Barb Aleman (2003) on behalf of the International Hydranencephaly Support Group, of her hydrancephalic deceased daughter Kayda's behaviour gives examples of some prima facie conscious affective behaviours where more research needs to be done:
During her last couple of years she would only sleep if she had her husky dog under her left arm and her bunny under her right. If either was missing or not in the "right" arm she would not sleep. I tested this many times, always with the same result.She would only sleep listening to a story not music. Again, we tested it numerous times.
If you put on a book that had more than one tape she would stay awake to listen to the whole book.
If she didn't want a particular toy or stuffed animal you'd given her she'd push or toss it onto the floor (http://hydranencephaly.com/communication.htm).
A recent survey of 81 parents of children with hydranencephaly provides further reason to query the prevailing view among neurologists that these children are not conscious. Although the survey was not scientifically conducted and has not been published in any medical journal, it was consistent with earlier research findings. The following table lists selected results relating to movement, awareness and affective consciousness:
Table 4.3 Selected behaviours in a survey of 81 children with hydranencephaly, which constitute prima facie evidence of consciousness
| Survey Question | Response by percentage (absolute figures in brackets) |
| Can your child move his/her arms? | Yes: 60.49% (49) No: 2.46% (2) A little: 32.1% (26) Used to: 0 |
| Can your child move his/her legs? | Yes: 61.72% (50) No: 3.7% (3) A little: 28.39 (23) Used to: 0 |
| Can your child give hugs or kisses? | Yes: 19.73% (16) No: 66.66% (54) Sometimes: 11.11% (9) |
| Can your child cry? | Yes: 91.35% (74) No: 6.17% (5) |
| What is your child's general mood? | Happy: 56.79% (46) Irritable: 6.17% (5) Fussy: 13.58% (11) Quiet: 18.5% (15) Can't tell: 2.46% (2) |
| What was your child's mood as a baby? | Happy: 22% (18) Irritable: 46.91% (38) Fussy: 11.11% (9) Quiet: 14.8% (12) Couldn't tell: 0 |
| Is your child aware of his/her surroundings? | Yes: 74% (60) No: 2.46% (2) Sometimes: 14.8% (12) Used to be: 0 Can't tell: 6.17% (5) |
| Is your child aware of objects? | Yes: 40.74% (33) No: 17.28% (14) Sometimes: 38.27% (31) Used to be: 0 |
| Does your child have a favorite toy? | Yes: 48.14% (39) No: 46.91% (38) |
| Does your child have a security item? | Yes: 29.6% (24) No: 62.96% (51) |
| Can your child hear? | Yes: 92.59% (75) No: 1.23% (1) |
| Can your child make sounds? | Yes: 96.29% (78) |
| Does your child use any words meaningfully? | Yes: 9.87% (8) No: 83.95% (68) |
| Will your child echo you? | Yes: 23.45% (19) No: 65.43% (53) |
| Can your child see? | Yes: 27% (22) No: 16% (13) Sometimes: 12.34% (10) A little: 20.98% (17) Not sure: 19.73% (16) |
If we adhere to the methodology proposed for this thesis, then the appropriate way of assessing whether these children are indeed phenomenally conscious is to look for features of their behaviour that are readily explained by a "first-person" account, but can only accounted for with difficulty by a "third-person" account. For a scientist proposing a "third-person" account of these children's behaviour, the children's emotional responses (smiling, giggling and vocalisations) to familiar people and favourite objects, likings for favourite songs and dislikes for certain kinds of music, are certainly startling, unexpected behaviours that constitute a good prima facie case for affective consciousness without a functioning cerebral cortex.
Despite these promising results, there are three reasons for caution. First, some of the behaviours cited can be explained by non-conscious mechanisms. We have already concluded that associative learning does not require mental states, and Rose has suggested that the emotional responses to music and to familiar faces or voices emotional conditioning (personal email, 22 July 2004).
Second, the usefulness of these results in arguing for consciousness without a cerebral cortex is limited by the fact that some children suffering from hydranencephaly still possess residual cerebral cortex:
[M]any children have some of their cerebral hemispheres so can use these and learn to do more than would be expected by this diagnosis. Just as all children are different, all children with hydranencephaly are different as well (Merker, 2003).
The fact that hydranencephaly varies in degree of severity from one individual to another may account in part for some of the abilities displayed.
Finally, Shewmon (1999, p. 372) has raised the possibility that vertical plasticity (the ability of subcortical regions of the brain to take over functions that the cortical regions normally handle) may explain these children's emotional behaviour. If this is the case, then it could still be argued that certain neural features that are normally possessed only by the cerebral cortex (e.g. high intra- and inter-connectivity) are required for phenomenal consciousness.
Conclusion 4.43: There is a good prima facie case that children with hydranencephaly are phenomenally conscious. However, the emotional behaviours displayed by these children does not conclusively establish that a functioning cerebral cortex is not required for consciousness in normal human beings, as: (i) some of the emotional behaviours may be non-conscious,(ii) hydranencephaly varies considerably in degree of severity from one individual to another, and (iii) vertical plasticity (whereby other regions of their brains become "corticised") may account for some of their emotional behaviour. More research is needed to correlate the emotional capacities of children with hydranencephaly with their degree of cerebral damage, and to determine which behaviours are most likely to manifest conscious feelings.
It should also be noted that a list of defining criteria for "affective consciousness" has yet to be developed. I would like to propose the following indicators as most promising indicators for identifying phenomenal consciousness: musical preferences and singular preferences for individuals and objects.
Affective states in decorticate mammals
Decorticate mammals may behave in a way that is difficult to distinguish from their "normal" counterparts, suggesting that conscious feelings are also present in these animals. To quote Panksepp (2004):
In the '70s I did the following class experiment with 16 students in an advanced behavioral neuroscience class: Each student received two adult rats, one which had had all neocortex removed at day 3 of birth, and the other had sham surgery. Each student had a lab session to study their rat in whatever way they wished, and then to decide who was missing 25% of their brain (the conscious "thinking cap" so to speak). Surprisingly, 12 of 16 selected the decorticate as being the normal animal. Why this mistake that was statistically significant? Largely, because the decorticates were more emotionally active it seems (personal email, 24 June 2004).
On the other hand, Shewmon (1999) has proposed that vertical plasticity in the brains of juvenile rats can account for their recovery of function. On this interpretation, the program controlling neural development is more flexible in juvenile animals, and adapts to the loss of the cerebral cortex by enhancing the capacities of lower-level structures in the brain, to partially compensate for the loss. This would explain why "adult cats bilaterally hemispherectomized [subjected to a surgical removal of both cerebral hemispheres - V.T.] as kittens behave nearly indistinguishably from normal, in marked contrast to cats hemispherectomized as adults (which are severely disabled)" (1999, p. 372).
If Shewmon is right, subcortical regions in the brains of decerebrated juvenile animals may acquire neural features that are normally found only in the cerebral cortex - e.g. a high degree of intra- and inter-connectivity. In effect, these animals' brains become "corticised". The neural features that normally distinguish the cerebral cortex would still be required for consciousness.
The identification of affective consciousness in animals whose cerebral hemispheres have been removed as adults would certainly clinch the case for an affective consciousness that is independent of the cerebral cortex. The evidence to date is not promising:
[I]t has been demonstrated that removing the cerebral cortex from a cat or a rat ... leaves intact the ability to generate motivated behavior. The animal still searches for food, eats to maintain body weight, shivers when cold, fights or escapes when attacked, and so on. These behaviors appear awkward and clumsy when compared to controls and are often poorly adjusted to circumstances (Prescott et al., 1999).
The behaviour of these unfortunate animals certainly qualifies as emotional, but there is no reason to consider it phenomenally conscious. However, the case for affective consciousness would be much strengthened if it could be shown that these animals exhibited singular preferences such as are found in children with hydranencephaly, as these are unexpected phenomena on a "third person" account.
Conclusion 4.44: The apparently conscious behaviour shown by mammals whose cerebral hemispheres have been removed while very young, is compatible with the hypothesis that subcortical regions of their brains modified their development to compensate for the loss (vertical plasticity). This in no way weakens the generally accepted theory that certain neural features of the cerebral cortex (e.g. high intra- and inter-connectivity) are required for phenomenal consciousness.
Implications for the occurrence of conscious feelings in animals
There is much that we still do not know about the brain, but the data cited above appears to be of limited evidential value in establishing the occurrence of conscious feelings in non-mammals:
Conclusion 4.45: Despite some tentative evidence for phenomenal consciousness in decerebrated animals, neurological studies performed on mammals to date give us no good reason to believe that conscious feelings can occur in reptiles and birds.