Evolution of consciousness

articles

Mashour and Alkire: The evolution of consciousness

Introduction

Are other animals besides humans conscious? If so, when did consciousness evolve? Darwin considered these related questions. A paper written by G. A. Mashour and M. T. Alkire (2013) draws on consciousness studies, evolutionary neurobiology, animal psychology and anaesthesiology in an attempt to provide some answers. The paper discusses the significance of recent experiments on subjects awakening from anaesthesia, where the neural activity revealed by scans, as first a primitive and then a greater level of consciousness come about, casts light on the phylogeny and evolution of consciousness. Identifying the neural structures involved in emerging consciousness and their phylogenetic lineage, Mashour and Alkire use parallels between phylogeny and evolution to make inferences about the evolution of these complex systems. Ontogeny also gives useful insights. A short sketch of the paper is presented here.

Distinctions

Evolutionary biology is now a fundamental pillar of the life sciences, yet the emergence of consciousness during evolution remains unclear. It probably occurred long before Homo sapiens came on the scene. Part of the difficulty lies in the need to make careful distinctions when talking about consciousness, in particular, the distinctions between phenomenal and access consciousness, and between wakefulness and awareness.

Phenomenal consciousness is that in which its possessor has subjective, personal experiences; these include sensations. Access consciousness is that in which content is available for reasoning, for acting or reporting on. It is difficult for scientists investigating consciousness in (non-human) animals, because their subjects cannot provide verbal reports. For humans, subjective experiences may often precede the ability to report on them (babies probably develop phenomenal consciousness before access consciousness). Scientific progress in this field requires reproducible experiments, where:

Consciousness emerges from unconsciousness at a measurable point.
Phenomenal and access consciousness emerge close together.
Neural structure and function can be assessed.

These requirements can be satisfied by observing the return of consciousness from general anaesthesia, using scans and noting behavioural responses.

In considering the neural correlates of consciousness, the distinction between awareness and wakefulness is important. Awareness [my definition] covers a range of states which combine receiving of information with an experiential quality, whereas wakefulness is about levels of consciousness in the sense of arousal states ranging from comatose, anaesthetised, asleep, drowsy, relaxed, paying attention, concentrating on a task, to fully alert. Some current theories propose the cortex as the main site for awareness, with subcortical, midline brain structures as the source of arousal signals.

Top-down approach

The following evidence supports the role of the neocortex in consciousness:

Electroencephalograms (EEGs) of almost all mammals and birds in the awake state show desynchronised, high frequency and low amplitude activity. In depressed levels of consciousness (anaesthetised, sleeping, minimally consciousness) there is low frequency, high amplitude activity.
Consciousness is linked with neural activity in the thalamocortical system. Midline regions of the brainstem and midbrain keep the cortex in an aroused state. Specific cortical regions serve as cognitive modules contributing to conscious content. The thalamus with its connections to all parts of the cortex integrates and modulates the output from the cognitive modules. This creates something more than the sum of its parts: the greater the integration and adjustment, the greater the consciousness.
Widespread brain activity is correlated with consciousness. Sensory input spreads quickly from the the sensory cortex to the parietal, temporal and pre-frontal areas. This is accompanied by recurrent local feedback, followed by long-range feedback in the anterior (rostral) to posterior (caudal) direction. There is strong evidence to support the association of the frontal and parietal cortex with awareness across multiple sensory modalities. The lateral frontoparietal network mediates consciousness of the environment, whereas the medial frontoparietal network mediates dreaming and internally directed attention. A caudal to rostral information flow is concerned with sensory processing; this can occur even in the absence of consciousness, for example, when anaesthetised. The information flow in the reverse direction is thought to be associated with experience itself and this is inhibited by general anaesthesia.

A precursor of the neocortex was present in the earliest evolving vertebrates. The basic structure of hindbrain, midbrain and forebrain was not reinvented as each new species evolved. Rather, as different ecological niches were exploited at different times, different brain regions, best suited for survival, were enhanced within that basic structure.

Bottom-up approach

Located in the brainstem, the ascending reticular activating system was discovered in the late 1940s. Arousal involves not just this system, but also specific nuclei and cell types in the brainstem, midbrain, basal forebrain and diencephalon from which long-range axons innervate the cortex. Arousal signals transmitted by these axons create the neurochemical environment capable of supporting consciousness.

Darwin, in his later years, studied similarities and differences in the emotional expressions of humans and other animals. He reasoned that if animals show emotion through behavioural expression, then since humans are animals they must share some of the neurobiological evolutionary path with those animals capable of similar expression of emotions. Before behaviourism became fashionable, Darwin understood that, as a consequence of shared evolution, commonalities in emotional expression probably reflected similar states of mind. Consciousness may have emerged not so much in order to make an internal representation of the world as to serve basic emotional drives, such as the need for air, water, food and sex. Behaviour is driven fundamentally by a requirement to maintain homeostasis. Basic behaviours around homeostasis were evident in the first multicellular organisms that needed a vascular system to supply cells not directly exposed to the environment.

The brain structures needed for generating arousal and primitive emotional responses are generally located in the brainstem, midbrain and limbic system and are as old as the vertebrate radiation itself (ibid: 10359).

More complex brains, having a more developed neocortex, gave animals a greater grasp of their surroundings. For example, a dehydrated frog, placed in full sunlight but next to a water source, will usually die without moving, unless some accident makes it aware of the water. The frog has only limited visual capacity. If by chance it finds the water, the frog will drink, which shows that its interoceptors still function. Lizards have more evolved visual systems, so a lizard placed in the same situation will see the water and drink, suggesting that it associates its exteroceptor-derived perceptions with its interoceptor-detected needs. This coupling of an internally based need system with an externally based situational-awareness system is likely to be the foundation for the emergence of consciousness (ibid: 10359).

The brainstem arousal structures are near the cranial nerves which innervate the face and neck; these nerves give an animal the ability to orient itself to its surroundings. They also enable the facial expression of emotion, a feature found in almost all vertebrates. At its most primitive, the open mouth of a predator reveals appetitive drives, which, if successful, satisfies a basic need for food. For the prey, this elicits heightened arousal, fear and a survival response.

Consciousness in non-human species

If consciousness evolved with cephalic development of the CNS, then it should be possible to identify roughly where it emerged on the evolutionary tree. This requires consideration of the common features associated with this emergence, rather than supposed uniquely human features like self-reflection. A candidate feature is motility arising from sensation. The single-celled paramoecium is covered with thousands of cilia, which both sense the environment and provide locomotion. More compelling is the sea-squirt, which, in its larval stage, has neural ganglia and sensory receptors. These are used to find a suitable attachment, but once located, the neural tissue is digested and the sea-squirt remains sessile from this time.

The neural core of consciousness

Light on evolutionary development can be gained from identifying the neural correlate of human consciousness at its most primitive. This can be done by observing the regaining of consciousness from general anaesthesia. Positron emission tomography (PET) scans revealed activity in the brainstem (locus coeruleus), hypothalamus, thalamus and anterior cingulate (media pre-frontal area). There was only limited neocortical activity, involving frontal-parietal communication, for this level of primitive consciousness. Apart from this limited cortical involvement, “the core of human consciousness appears to be associated with phylogenetically ancient structures mediating arousal and activated by primitive emotions” (ibid: 10360).

Consciousness of the world

This finding from anaesthesia experiments, of the activity of primordial structures with a limited cortical contribution, tempts us to conclude that primitive consciousness emerged just as our mammalian ancestors evolved beyond reptiles with their predominately subcortical brains (ibid: 10361). Considering the precursor to this level of consciousness, we can go back further into the past, about 315 million years ago, when the synapsid line, from which the mammals evolved, and the sauropsid line, from which reptiles and birds evolved, both diverged from the anapsid line. Besides mammals, birds also display evidence of consciousness, specifically, episodic memory recall and attribution of mental events to another being. They, like mammals, show a marked increase in brain-body ratios compared to reptiles, extended parental care of offspring, distinct sleep stages and complex social interactions. The avian pallium, although histologically different, shares many network characteristics with the cortex. Birds possess a similar brainstem and diencephalic bodies like the thalamus and hypothalamus. This suggests that consciousness emerged on two distinct lineages which diverged 315 million years ago.

Even if ontogeny does not serve as a reliable recapitulation of phylogeny, it still provides useful insights into neural development. In humans, a rudimentary precursor of the thalamus appears as early as day 22 or 23 after conception. Peripheral sensory receptors develop from about week 20 of gestation, thalamocortical connections by week 26, electrical activity changes to more continuous patterns from weeks 25 to 29, and sleep-wake distinctions are discernable by week 30. Thus the structural and functional pre-requisites for consciousness are present by the third trimester; from this stage the foetus may experience pain. During the third trimester, the proportion of time spent in REM sleep reaches a maximum; it will not be exceeded subsequently through the rest of the individual’s lifespan. This internally driven neuronal activity may be preparing the cortex for the influx of stimuli after birth. I would add here that a great many cortical connections are made after birth, especially in the first two years.

Consciousness of the self

One aspect of consciousness that seems linked to higher cognitive ability is self-awareness. G.G. Gallup in 1970 found that chimpanzees, but not monkeys, passed the mirror self-recognition (MSR) test. In Gallup’s controlled experiment, the animals were given ample time to become accustomed to the mirror, so that their social responses to their reflected images were much diminished. The number of social and self-directed responses were both measured before each animal had a mark placed on its forehead or ear while under anaesthetic. After recovering from the anaesthesia, the chimpanzees but not the monkeys exhibited mark-directed responses on seeing their mirror image, such as touching the mark or examining their fingers after touching.

Orangutans and bonobos also pass the test; gorillas, though, had more equivocal results. The inference that self-recognition in a mirror indicates possession of a concept of self has been challenged by some, but it does at least indicate a definite divide in self-awareness. In primates, the higher consciousness associated with this self-awareness may have evolved around five million years ago when the great apes split away from the lesser apes.

Other large-brained mammals, like dolphins, may be capable of passing the test. Even some birds, such as the magpie, belonging to the genus Corvidae, exhibit behaviour consistent with MSR; they possess a relatively large pre-frontal pallium. This suggests that consciousness of both the world and the self has ancient roots.

The uniqueness of human consciousness

It would seem, then, that a wide number of species possess primitive consciousness. However, by the integrated information theory of consciousness, the evolution of more complex brain networks enabled the synthesis of outputs from more functionally diverse modules, resulting in a higher capacity for consciousness. Testing with artificial agents supports the view that integration of information correlates positively with fitness. Is the higher capacity for consciousness one of level or quality? It is unlikely that a sedentary human has a higher level of consciousness (arousal) than a cheetah pursuing its prey, which suggests it is the quality, the richness of experience, that makes the difference. Alternatively, it may be that advanced symbolic processing in human cognition eclipses subjective experience. For human consciousness, with its capacity for reflection, forward planning and self-consciousness, seems to be unique. Whatever is most particular to human consciousness may be associated with the evolution of the frontal cortex. Although the relative size of the frontal lobe to the whole neocortex is approximately the same in humans and the great apes, richer connectivity may be the key, in particular, anterior to posterior pathways. Such is the information flow from the frontal to parietal regions, it might be that human consciousness is more internally directed. To varying extents, this could be a feature of animals capable of passing the MSR test. One theory posits human consciousness as a dream-like phenomenon, modulated by environmental input.

Conclusion

The basic structure of the vertebrate central nervous system is evolutionarily ancient and highly conserved across species. Emergence from anaesthesia may provide a practical and reproducible model to study the real-time evolution of the neural correlates required for consciousness of the world and of the self. The activity of the arousal centres in the human brainstem and diencephalon, together with limited neocortical activity which includes recurrent processing, “can result in primitive phenomenal consciousness” (ibid: 10363). Applying “reverse engineering” to these results, Mashour and Alkire postulate that other mammals and even birds with similar neurological structures are capable of this type of consciousness; further, the underlying mechanisms of consciousness in humans (and other animals) were present in the earliest stages of vertebrate brain evolution. The unique richness of human experience is likely due to the greater complexity and connectivity of networks in the human brain, with its functionally dominant prefrontal cortex (ibid: 10363). All this suggests that the differences across species in ability to experience the world are, as Darwin surmised, differences of degree rather than kind (ibid: 10357).

Bibliography

Mashour, G.A. & Alkire, M.T. (2013) Evolution of Consciousness: Phylogeny, ontogeny and emergence from general anesthesia, PNAS, www.pnas.org/cgi/doi/10.1073/pnas.1301188110