A blog from the Centre for Research Ethics & Bioethics (CRB)

Tag: consciousness (Page 1 of 4)

A new project will explore the prospect of artificial awareness

The neuroethics group at CRB has just started its work as part of a new European research project about artificial awareness. The project is called “Counterfactual Assessment and Valuation for Awareness Architecture” (CAVAA), and is funded for a duration of four years. The consortium is composed of 10 institutions, coordinated by the Radboud University in the Netherlands.

The goal of CAVAA is “to realize a theory of awareness instantiated as an integrated computational architecture…, to explain awareness in biological systems and engineer it in technological ones.” Different specific objectives derive from this general goal. First, CAVAA has a robust theoretical component: it relies on a strong theoretical framework. Conceptual reflection on awareness, including its definition and the identification of features that allow its attribution to either biological organisms or artificial systems, is an explicit task of the project. Second, CAVAA is interested in exploring the connection between awareness in biological organisms and its possible replication in artificial systems. The project thus gives much attention to the connection between neuroscience and AI. Third, against this background, CAVAA aims at replicating awareness in artificial settings. Importantly, the project also has a clear ethical responsibility, more specifically about anticipating the potential societal and ethical impact of aware artificial systems.

There are several reasons why a scientific project with a strong engineering and computer science component also has philosophers on board. We are asked to contribute to developing a strong and consistent theoretical account of awareness, including the conceptual conceivability and the technical feasibility of its artificial replication. This is not straightforward, not only because there are many content-related challenges, but also because there are logical traps to avoid. For instance, we should avoid the temptation to validate an empirical statement on the basis of our own theory: this would possibly be tautological or circular.

In addition to this theoretical contribution, we will also collaborate in identifying indicators of awareness and benchmarks for validating the cognitive architecture that will be developed. Finally, we will collaborate in the ethical analysis concerning potential future scenarios related to artificial awareness, such as the possibility of developing artificial moral agents or the need to extend moral rights also to artificial systems.

In the end, there are several potential contributions that philosophy can provide to the scientific attempt to replicate biological awareness in artificial systems. Part of this possible collaboration is the fundamental and provoking question: why should we try to develop artificial awareness at all? What is the expected benefit, should we succeed? This is definitely an open question, with possible arguments for and against attempting such a grand accomplishment.

There is also another question of equal importance, which may justify the effort to identify the necessary and sufficient conditions for artificial systems to become aware, and how to recognize them as such. What if we will inadvertently create (or worse: have already created) forms of artificial awareness, but do not recognize this and treat them as if they were unaware? Such scenarios also confront us with serious ethical issues. So, regardless of our background beliefs about artificial awareness, it is worth investing in thinking about it.

Stay tuned to hear more from CAVAA!

Written by…

Michele Farisco, Postdoc Researcher at Centre for Research Ethics & Bioethics, working in the EU Flagship Human Brain Project.

Part of international collaborations

A charming idea about consciousness

Some ideas can have such a charm that you only need to hear them once to immediately feel that they are probably true: “there must be some grain of truth in it.” Conspiracy theories and urban myths probably spread in part because of how they manage to charm susceptible human minds by ringing true. It is said that even some states of illness are spread because the idea of ​​the illness has such a strong impact on many of us. In some cases, we only need to hear about the diagnosis to start showing the symptoms and maybe we also receive the treatment. But even the idea of diseases spread by ideas has charm, so we should be on our guard.

The temptation to fall for the charm of certain ideas naturally also exists in academia. At the same time, philosophy and science are characterized by self-critical examination of ideas that may sound so attractive that we do not notice the lack of examination. As long as the ideas are limited hypotheses that can in principle be tested, it is relatively easy to correct one’s hasty belief in them. But sometimes these charming ideas consist of grand hypotheses about elusive phenomena that no one knows how to test. People can be so convinced by such ideas that they predict that future science just needs to fill in the details. A dangerous rhetoric to get caught up in, which also has its charm.

Last year I wrote a blog post about a theory at the border between science and philosophy that I would like to characterize as both grand and charming. This is not to say that the theory must be false, just that in our time it may sound immediately convincing. The theory is an attempt to explain an elusive “phenomenon” that perplexes science, namely the nature of consciousness. Many feel that if we could explain consciousness on purely scientific grounds, it would be an enormously significant achievement.

The theory claims that consciousness is a certain mathematically defined form of information processing. Associating consciousness with information is timely, we are immediately inclined to listen. What type of information processing would consciousness be? The theory states that consciousness is integrated information. Integration here refers not only to information being stored as in computers, but to all this diversified information being interconnected and forming an organized whole, where all parts are effectively available globally. If I understand the matter correctly, you can say that the integrated information of a system is the amount of generated information that exceeds the information generated by the parts. The more information a system manages to integrate, the more consciousness the system has.

What, then, is so charming about the idea that ​​consciousness is integrated information? Well, the idea might seem to fit with how we experience our conscious lives. At this moment you are experiencing multitudes of different sensory impressions, filled with details of various kinds. Visual impressions are mixed with impressions from the other senses. At the same time, however, these sensory impressions are integrated into a unified experience from a single viewpoint, your own. The mathematical theory of information processing where diversification is combined with integration of information may therefore sound attractive as a theory of consciousness. We may be inclined to think: Perhaps it is because the brain processes information in this integrative way that our conscious lives are characterized by a personal viewpoint and all impressions are organized as an ego-centred subjective whole. Consciousness is integrated information!

It becomes even more enticing when it turns out that the theory, called Integrated Information Theory (IIT), contains a calculable measure (Phi) of the amount of integrated information. If the theory is correct, then one would be able to quantify consciousness and give different systems different Phi for the amount of consciousness. Here the idea becomes charming in yet another way. Because if you want to explain consciousness scientifically, it sounds like a virtue if the theory enables the quantification of how much consciousness a system generates. The desire to explain consciousness scientifically can make us extra receptive to the idea, which is a bit deceptive.

In an article in Behavioral and Brain Sciences, Björn Merker, Kenneth Williford and David Rudrauf examine the theory of consciousness as integrated information. The review is detailed and comprehensive. It is followed up by comments from other researchers, and ends with the authors’ response. What the three authors try to show in the article is that even if the brain does integrate information in the sense of the theory, the identification of consciousness with integrated information is mistaken. What the theory describes is efficient network organization, rather than consciousness. Phi is a measure of network efficiency, not of consciousness. What the authors examine in particular is that charming feature I just mentioned: the theory can seem to “fit” with how we experience our conscious lives from a unified ego-centric viewpoint. It is true that integrated information constitutes a “unity” in the sense that many things are joined in a functionally organized way. But that “unity” is hardly the same “unity” that characterizes consciousness, where the unity is your own point of view on your experiences. Effective networks can hardly be said to have a “viewpoint” from a subjective “ego-centre” just because they integrate information. The identification of features of our conscious lives with the basic concepts of the theory is thus hasty, tempting though it may be.

The authors do not deny that the brain integrates information in accordance with the theory. The theory mathematically describes an efficient way to process information in networks with limited energy resources, something that characterizes the brain, the authors point out. But if consciousness is identified with integrated information, then many other systems that process information in the same efficient way would also be conscious. Not only other biological systems besides the brain, but also artifacts such as some large-scale electrical power grids and social networks. Proponents of the theory seem to accept this, but we have no independent reason to suppose that systems other than the brain would have consciousness. Why then insist that other systems are also conscious? Well, perhaps because one is already attracted by the association between the basic concepts of the theory and the organization of our conscious experiences, as well as by the possibility of quantifying consciousness in different systems. The latter may sound like a scientific virtue. But if the identification is false from the beginning, then the virtue appears rather as a departure from science. The theory might flood the universe with consciousness. At least that is how I understand the gist of ​​the article.

Anyone who feels the allure of the theory that consciousness is integrated information should read the careful examination of the idea: The integrated information theory of consciousness: A case of mistaken identity.

The last word has certainly not been said and even charming ideas can turn out to be true. The problem is that the charm easily becomes the evidence when we are under the influence of the idea. Therefore, I believe that the careful discussion of the theory of consciousness as integrated information is urgent. The article is an excellent example of the importance of self-critical examination in philosophy and science.

Pär Segerdahl

Written by…

Pär Segerdahl, Associate Professor at the Centre for Research Ethics & Bioethics and editor of the Ethics Blog.

Merker, B., Williford, K., & Rudrauf, D. (2022). The integrated information theory of consciousness: A case of mistaken identity. Behavioral and Brain Sciences, 45, E41. doi:10.1017/S0140525X21000881

This post in Swedish

We like critical thinking

How can we detect consciousness in brain-damaged patients?

Detecting consciousness in brain-damaged patients can be a huge challenge and the results are often uncertain or misinterpreted. In a previous post on this blog I described six indicators of consciousness that I introduced together with a neuroscientist and another philosopher. Those indicators were originally elaborated targeting animals and AI systems. Our question was: what capacities (deducible from behavior and performance or relevant cerebral underpinnings) make it reasonable to attribute consciousness to these non-human agents? In the same post, I mentioned that we were engaged in a multidisciplinary exploration of the clinical relevance of selected indicators, specifically for testing them on patients with Disorders of Consciousness (DoCs, for instance, Vegetative State/Unresponsive Wakefulness Syndrome, Minimally Conscious State, Cognitive-Motor Dissociation). While this multidisciplinary work is still in progress, we recently published an ethical reflection on the clinical relevance of the indicators of consciousness, taking DoCs as a case study.

To recapitulate, indicators of consciousness are conceived as particular capacities that can be deduced from the behavior or cognitive performance of a subject and that serve as a basis for a reasonable inference about the level of consciousness of the subject in question. Importantly, also the neural correlates of the relevant behavior or cognitive performance may make possible deducing the indicators of consciousness.  This implies the relevance of the indicators to patients with DoCs, who are often unable to behave or to communicate overtly. Responses in the brain can be used to deduce the indicators of consciousness in these patients.

On the basis of this relevance, we illustrate how the different indicators of consciousness might be applied to patients with DoCs with the final goal of contributing to improve the assessment of their residual conscious activity. In fact, a still astonishing rate of misdiagnosis affects this clinical population. It is estimated that up to 40 % of patients with DoCs are wrongly diagnosed as being in Vegetative State/Unresponsive Wakefulness Syndrome, while they are actually in a Minimally Conscious State. The difference of these diagnoses is not minimal, since they have importantly different prognostic implications, which raises a huge ethical problem.

We also argue for the need to recognize and explore the specific quality of the consciousness possibly retained by patients with DoCs. Because of the devastating damages of their brain, it is likely that their residual consciousness is very different from that of healthy subjects, usually assumed as a reference standard in diagnostic classification. To illustrate, while consciousness in healthy subjects is characterized by several distinct sensory modalities (for example, seeing, hearing and smelling), it is possible that in patients with DoCs, conscious contents (if any) are very limited in sensory modalities. These limitations may be evaluated based on the extent of the brain damage and on the patients’ residual behaviors (for instance, sniffing for smelling). Also, consciousness in healthy subjects is characterized by both dynamics and stability: it includes both dynamic changes and short-term stabilization of contents. Again, in the case of patients with DoCs, it is likely that their residual consciousness is very unstable and flickering, without any capacity for stabilization. If we approach patients with DoCs without acknowledging that consciousness is like a spectrum that accommodates different possible shapes and grades, we exclude a priori the possibility of recognizing the peculiarity of consciousness possibly retained by these patients.

The indicators of consciousness we introduced offer a potential help to identify the specific conscious abilities of these patients. While in this paper we argue for the rationale behind the clinical use of these indicators, and for their relevance to patients with DoCs, we also acknowledge that they open up new lines of research with concrete application to patients with DoCs. As already mentioned, this more applied work is in progress and we are confident of being able to present relevant results in the weeks to come.

Written by…

Michele Farisco, Postdoc Researcher at Centre for Research Ethics & Bioethics, working in the EU Flagship Human Brain Project.

Farisco, M., Pennartz, C., Annen, J. et al. Indicators and criteria of consciousness: ethical implications for the care of behaviourally unresponsive patients. BMC Med Ethics 2330 (2022). https://doi.org/10.1186/s12910-022-00770-3

We have a clinical perspective

Can subjectivity be explained objectively?

The notion of a conscious universe, animated by unobservable experiences, is today presented almost as a scientific hypothesis. How is that possible? Do cosmologists’ hypotheses that the universe is filled with dark matter and dark energy contribute to making the idea of ​​a universe filled with “dark consciousness” almost credible?

I ask the question because I myself am amazed at how the notion that elementary particles have elementary experiences suddenly has become academically credible. The idea that consciousness permeates reality is usually called panpsychism and is considered to have been represented by several philosophers in history. The alleged scientific status of panpsychism is justified today by emphasizing two classic philosophical failures to explain consciousness. Materialism has not succeeded in explaining how consciousness can arise from non-conscious physical matter. Dualism has failed to explain how consciousness, if it is separate from matter, can interact with physical reality.

Against this discouraging background, panpsychism is presented as an attractive, even elegant solution to the problem of consciousness. The hypothesis is that consciousness is hidden in the universe as a fundamental non-observable property of matter. Proponents of this elegant solution suggest that this “dark consciousness,” which permeates the universe, is extremely modest. Consciousness is present in every elementary particle in the form of unimaginably simple elementary experiences. These insignificant experiences are united and strengthened in the brain’s nervous system, giving rise to what we are familiar with as our powerful human consciousness, with its stormy feelings and thoughts.

However, this justification of panpsychism as an elegant solution to a big scientific problem presupposes that there really is a big scientific problem to “explain consciousness.” Is not the starting point a bit peculiar, that even subjectivity must be explained as an objective phenomenon? Even dualism tends to objectify consciousness, since it presents consciousness as a parallel universe to our physical universe.

The alternative explanations are thus all equally objectifying. Either subjectivity is reduced to purely material processes, or subjectivity is explained as a mental parallel universe, or subjectivity is hypostasized as “dark consciousness” that pervades the universe: as elementary experiential qualities of matter. Can we not let subjectivity be subjectivity and objectivity be objectivity?

Once upon a time there was a philosopher named Immanuel Kant. He saw how our constantly objectifying subjectivity turns into an intellectual trap, when it tries to understand itself without limiting its own objectifying approach to all questions. We then resemble cats that hopelessly chase their own tails: either by spinning to the right or by spinning to the left. Both directions are equally frustrating. Is there an elegant solution to the spinning cat’s problem? Now, I do not want to claim that Kant definitely exposed the “hard problem” of consciousness as an intellectual trap, but he pointed out the importance of self-critically examining our projective, objectifying way of functioning. If we lived as expansively as we explain everything objectively, we would soon exhaust the entire planet… is not that exactly what we do?

During a philosophy lecture, I tried to show the students how we can be trapped by apparent problems, by pseudo-problems that of course are not scientific problems, since they make us resemble cats chasing their own tails without realizing the unrealizability of the task. One student did not like what she perceived as an arbitrary limitation of the enormous achievements of science, so she objected: “But if it is the task of science to explain all big problems, then it must attempt to explain these riddles as well.” The objection is similar to the motivation of panpsychism, where it is assumed that it is the task of science to explain everything objectively, even subjectivity, no matter how hopelessly the questions spin in our heads.

The spinning cat’s problem has a simple solution: stop chasing the tail. Humans, on the other hand, need to clearly see the hopelessness of their spinning in order to stop it. Therefore, humans need to philosophize in order to live well on this planet.

Pär Segerdahl

Written by…

Pär Segerdahl, Associate Professor at the Centre for Research Ethics & Bioethics and editor of the Ethics Blog.

If you want to read more about panpsychism, here are two links:

Does consciousness pervade the universe?

The idea that everything from spoons to stones is conscious is gaining academic credibility

This post in Swedish

We challenge habits of thought

Consciousness and complexity: theoretical challenges for a practically useful idea

Contemporary research on consciousness is ambiguous, like the double-faced god Janus. On the one hand, it has achieved impressive practical results. We can today detect conscious activity in the brain for a number of purposes, including better therapeutic approaches to people affected by disorders of consciousness such as coma, vegetative state and minimally conscious state. On the other hand, the field is marked by a deep controversy about methodology and basic definitions. As a result, we still lack an overarching theory of consciousness, that is to say, a theoretical account that scholars agree upon.

Developing a common theoretical framework is recognized as increasingly crucial to understanding consciousness and assessing related issues, such as emerging ethical issues. The challenge is to find a common ground among the various experimental and theoretical approaches. A strong candidate that is achieving increasing consensus is the notion of complexity. The basic idea is that consciousness can be explained as a particular kind of neural information processing. The idea of associating consciousness with complexity was originally suggested by Giulio Tononi and Gerald Edelman in a 1998 paper titled Consciousness and Complexity. Since then, several papers have been exploring its potential as the key for a common understanding of consciousness.

Despite the increasing popularity of the notion, there are some theoretical challenges that need to be faced, particularly concerning the supposed explanatory role of complexity. These challenges are not only philosophically relevant. They might also affect the scientific reliability of complexity and the legitimacy of invoking this concept in the interpretation of emerging data and in the elaboration of scientific explanations. In addition, the theoretical challenges have a direct ethical impact, because an unreliable conceptual assumption may lead to misplaced ethical choices. For example, we might wrongly assume that a patient with low complexity is not conscious, or vice-versa, eventually making medical decisions that are inappropriate to the actual clinical condition.

The claimed explanatory power of complexity is challenged in two main ways: semantically and logically. Let us take a quick look at both.

Semantic challenges arise from the fact that complexity is such a general and open-ended concept. It lacks a shared definition among different people and different disciplines. This open-ended generality and lack of definition can be a barrier to a common scientific use of the term, which may impact its explanatory value in relation to consciousness. In the landmark paper by Tononi and Edelman, complexity is defined as the sum of integration (conscious experience is unified) and differentiation (we can experience a large number of different states). It is important to recognise that this technical definition of complexity refers only to the stateof consciousness, not to its contents. This means that complexity-related measures can give us relevant information about the levelof consciousness, yet they remain silent about the corresponding contentsandtheirphenomenology. This is an ethically salient point, since the dimensions of consciousness that appear most relevant to making ethical decisions are those related to subjective positive and negative experiences. For instance, while it is generally considered as ethically neutral how we treat a machine, it is considered ethically wrong to cause negative experiences to other humans or to animals.

Logical challenges arise about the justification for referring to complexity in explaining consciousness. This justification usually takes one of two alternative forms. The justification is either bottom-up (from data to theory) or top-down (from phenomenology to physical structure). Both raise specific issues.

Bottom-up: Starting from empirical data indicating that particular brain structures or functions correlate to particular conscious states, relevant theoretical conclusions are inferred. More specifically, since the brains of subjects that are manifestly conscious exhibit complex patterns (integrated and differentiated patterns), we are supposed to be justified to infer that complexity indexes consciousness. This conclusion is a sound inference to the best explanation, but the fact that a conscious state correlates with a complex brain pattern in healthy subjects does not justify its generalisation to all possible conditions (for example, disorders of consciousness), and it does not logically imply that complexity is a necessary and/or sufficient condition for consciousness.

Top-down: Starting from certain characteristics of personal experience, we are supposed to be justified to infer corresponding characteristics of the underlying physical brain structure. More specifically, if some conscious experience is complex in the technical sense of being both integrated and differentiated, we are supposed to be justified to infer that the correlated brain structures must be complex in the same technical sense. This conclusion does not seem logically justified unless we start from the assumption that consciousness and corresponding physical brain structures must be similarly structured. Otherwise it is logically possible that conscious experience is complex while the corresponding brain structure is not, and vice versa. In other words, it does not appear justified to infer that since our conscious experience is integrated and differentiated, the corresponding brain structure must be integrated and differentiated. This is a possibility, but not a necessity.

The abovementioned theoretical challenges do not deny the practical utility of complexity as a relevant measure in specific clinical contexts, for example, to quantify residual consciousness in patients with disorders of consciousness. What is at stake is the explanatory status of the notion. Even if we question complexity as a key factor in explaining consciousness, we can still acknowledge that complexity is practically relevant and useful, for example, in the clinic. In other words, while complexity as an explanatory category raises serious conceptual challenges that remain to be faced, complexity represents at the practical level one of the most promising tools that we have to date for improving the detection of consciousness and for implementing effective therapeutic strategies.

I assume that Giulio Tononi and Gerald Edelman were hoping that their theory about the connection between consciousness and complexity finally would erase the embarrassing ambiguity of consciousness research, but the deep theoretical challenges suggest that we have to live with the resemblance to the double-faced god Janus for a while longer.

Written by…

Michele Farisco, Postdoc Researcher at Centre for Research Ethics & Bioethics, working in the EU Flagship Human Brain Project.

Tononi, G. and G. M. Edelman. 1998. Consciousness and complexity. Science 282(5395): 1846-1851.

We like critical thinking

Can AI be conscious? Let us think about the question

Artificial Intelligence (AI) has achieved remarkable results in recent decades, especially thanks to the refinement of an old and for a long time neglected technology called Deep Learning (DL), a class of machine learning algorithms. Some achievements of DL had a significant impact on public opinion thanks to important media coverage, like the cases of the program AlphaGo and its successor AlphaGo Zero, which both defeated the Go World Champion, Lee Sedol.

This triumph of AlphaGo was a kind of profane consecration of AI’s operational superiority in an increasing number of tasks. This manifest superiority of AI gave rise to mixed feelings in human observers: the pride of being its creator; the admiration of what it was able to do; the fear of what it might eventually learn to do.

AI research has generated a linguistic and conceptual process of re-thinking traditionally human features, stretching their meaning or even reinventing their semantics in order to attribute these traits also to machines. Think of how learning, experience, training, prediction, to name just a few, are attributed to AI. Even if they have a specific technical meaning among AI specialists, lay people tend to interpret them within an anthropomorphic view of AI.

One human feature in particular is considered the Holy Grail when AI is interpreted according to an anthropomorphic pattern: consciousness. The question is: can AI be conscious? It seems to me that we can answer this question only after considering a number of preliminary issues.

First we should clarify what we mean by consciousness. In philosophy and in cognitive science, there is a useful distinction, originally introduced by Ned Block, between access consciousness and phenomenal consciousness. The first refers to the interaction between different mental states, particularly the availability of one state’s content for use in reasoning and rationally guiding speech and action. In other words, access consciousness refers to the possibility of using what I am conscious of. Phenomenal consciousness refers to the subjective feeling of a particular experience, “what it is like to be” in a particular state, to use the words of Thomas Nagel. So, in what sense of the word “consciousness” are we asking if AI can be conscious?

To illustrate how the sense in which we choose to talk about consciousness makes a difference in the assessment of the possibility of conscious AI, let us take a look at an interesting article written by Stanislas Dehaene, Hakwan Lau and Sid Koudier. They frame the question of AI consciousness within the Global Neuronal Workspace Theory, one of the leading contemporary theories of consciousness. As the authors write, according to this theory, conscious access corresponds to the selection, amplification, and global broadcasting of particular information, selected for its salience or relevance to current goals, to many distant areas. More specifically, Dehaene and colleagues explore the question of conscious AI along two lines within an overall computational framework:

  1. Global availability of information (the ability to select, access, and report information)
  2. Metacognition (the capacity for self-monitoring and confidence estimation).

Their conclusion is that AI might implement the first meaning of consciousness, while it currently lacks the necessary architecture for the second one.

As mentioned, the premise of their analysis is a computational view of consciousness. In other words, they choose to reduce consciousness to specific types of information-processing computations. We can legitimately ask whether such a choice covers the richness of consciousness, particularly whether a computational view can account for the experiential dimension of consciousness.

This shows how the main obstacle in assessing the question whether AI can be conscious is a lack of agreement about a theory of consciousness in the first place. For this reason, rather than asking whether AI can be conscious, maybe it is better to ask what might indicate that AI is conscious. This brings us back to the indicators of consciousness that I wrote about in a blog post some months ago.

Another important preliminary issue to consider, if we want to seriously address the possibility of conscious AI, is whether we can use the same term, “consciousness,” to refer to a different kind of entity: a machine instead of a living being. Should we expand our definition to include machines, or should we rather create a new term to denote it? I personally think that the term “consciousness” is too charged, from several different perspectives, including ethical, social, and legal perspectives, to be extended to machines. Using the term to qualify AI risks extending it so far that it eventually becomes meaningless.

If we create AI that manifests abilities that are similar to those that we see as expressions of consciousness in humans, I believe we need a new language to denote and think about it. Otherwise, important preliminary philosophical questions risk being dismissed or lost sight of behind a conceptual veil of possibly superficial linguistic analogies.

Written by…

Michele Farisco, Postdoc Researcher at Centre for Research Ethics & Bioethics, working in the EU Flagship Human Brain Project.

We want solid foundations

An unusually big question

Sometimes the intellectual claims on science are so big that they risk obscuring the actual research. This seems to happen not least when the claims are associated with some great prestigious question, such as the origin of life or the nature of consciousness. By emphasizing the big question, one often wants to show that modern science is better suited than older human traditions to answer the riddles of life. Better than philosophy, for example.

I think of this when I read a short article about such a riddle: “What is consciousness? Scientists are beginning to unravel a mystery that has long vexed philosophers.” The article by Christof Koch gives the impression that it is only a matter of time before science determines not only where in the brain consciousness arises (one already seems have a suspect), but also the specific neural mechanisms that give rise to – everything you have ever experienced. At least if one is to believe one of the fundamental theories about the matter.

Reading about the discoveries behind the identification of where in the brain consciousness arises is as exciting as reading a whodunit. It is obvious that important research is being done here on the effects that loss or stimulation of different parts of the brain can have on people’s experiences, mental abilities and personalities. The description of a new technology and mathematical algorithm for determining whether patients are conscious or not is also exciting and indicates that research is making fascinating progress, which can have important uses in healthcare. But when mathematical symbolism is used to suggest a possible fundamental explanation for everything you have ever experienced, the article becomes as difficult to understand as the most obscure philosophical text from times gone by.

Since even representatives of science sometimes make philosophical claims, namely, when they want to answer prestigious riddles, it is perhaps wiser to be open to philosophy than to compete with it. Philosophy is not just about speculating about big questions. Philosophy is also about humbly clarifying the questions, which otherwise tend to grow beyond all reasonable limits. Such openness to philosophy flourishes in the Human Brain Project, where some of my philosophical colleagues at CRB collaborate with neuroscientists to conceptually clarify questions about consciousness and the brain.

Something I myself wondered about when reading the scientifically exciting but at the same time philosophically ambitious article, is the idea that consciousness is everything we experience: “It is the tune stuck in your head, the sweetness of chocolate mousse, the throbbing pain of a toothache, the fierce love for your child and the bitter knowledge that eventually all feelings will end.” What does it mean to take such an all-encompassing claim seriously? What is not consciousness? If everything we can experience is consciousness, from the taste of chocolate mousse to the sight of the stars in the sky and our human bodies with their various organs, where is the objective reality to which science wants to relate consciousness? Is it in consciousness?

If consciousness is our inevitable vantage point, if everything we experience as real is consciousness, it becomes unclear how we can treat consciousness as an objective phenomenon in the world along with the body and other objects. Of course, I am not talking here about actual scientific research about the brain and consciousness, but about the limitless intellectual claim that scientists sooner or later will discover the neural mechanisms that give rise to everything we can ever experience.

Pär Segerdahl

Written by…

Pär Segerdahl, Associate Professor at the Centre for Research Ethics & Bioethics and editor of the Ethics Blog.

Christof Koch, What Is Consciousness? Scientists are beginning to unravel a mystery that has long vexed philosophers, Nature 557, S8-S12 (2018) https://doi.org/10.1038/d41586-018-05097-x

This post in Swedish

We transcend disciplinary borders

The hard problem of consciousness: please handle with care!

We face challenges every day. Some are more demanding than others, but it seems that there is not a day without some problem to handle. Unless they are too big to manage, problems are like the engines of our lives. They push us to always go beyond wherever we are and whatever we do, to look for new possibilities, to build new opportunities. In other words: problems make us stay alive.

The same is true for science and philosophy. There is a constant need to face new challenges. Consciousness research is no exception. There are, of course, several problems in the investigation of consciousness. However, one problem has emerged as the big problem, which the Australian philosopher David Chalmers baptised “the hard problem of consciousness.” This classical problem (discussed even before Chalmers coined this expression, actually since the early days of neuropsychology, notably by Alexander Luria and collaborators) refers to the enigma of subjective experience. To adapt a formulation by the philosopher Thomas Nagel, the basic question is: why do we have experiences of what it is like to be conscious, for example, why do we experience that pain and hunger feel the way they do?

The hard problem has a double nature. On the one hand, it refers to what Joseph Levine had qualified as an explanatory gap. The strategy to identify psychological experiences with physical features of the brain is in the end unable to explain why experiences are related to physical phenomena at all. On the other hand, the hard problem also refers to the question if subjective experience can be explained causally or if it is intrinsic to the world, that is to say: fundamentally there, from the beginning, rather than caused by something more primary.

This double nature of the problem has been a stumbling block in the attempt to explain consciousness. Yet in recent years, the hardness of the problem has been increasingly questioned. Among the arguments that appear relevant in order to soften the problem, there is one that I think merits specific attention. This argument describes consciousness as a cultural concept, meaning that both the way we conceive it and the way we experience it depend on our culture. There are different versions of this argument: some reduce consciousness as such to a cultural construction, while other, less radical arguments stress that consciousness has a neurological substrate that is importantly shaped by culture. The relevant point is that by characterising consciousness as a cultural construction, with reference both to how we conceptualise it and how we are conscious, this argument ultimately questions the hardness of the hard problem.

To illustrate, consider anthropological and neuroscientific arguments that appear to go in the direction of explaining away the hard problem of consciousness. Anthropological explanations give a crucial role to culture and its relationship with consciousness. Humans have an arguably unique capacity of symbolisation, which enables us to create an immaterial world both through the symbolisation of the actual world and through the construction of immaterial realities that are not experienced through the senses. This human symbolic capacity can be applied not only to the external world, but also to brain activity, resulting in the conceptual construction of notions like consciousness. We symbolise our brain activity, hypostatise our conscious activities, and infer supposedly immaterial causes behind them.

There are also neuroscientific and neuropsychological attempts to explain how consciousness and our understanding of it evolved, which ultimately appear to potentially explain away the hard problem. Attention Schema Theory, for instance, assumes that people tend to “attribute a mysterious consciousness to themselves and to others because of an inherently inaccurate model of mind, and especially a model of attention.” The origin of the attribution of this mysterious consciousness is in culture and in folk-psychological beliefs, for instance, ideas about “an energy-like substance inhabiting the body.” In other words, culturally based mistaken beliefs derived from implicit social-cognitive models affect and eventually distort our view of consciousness. Ultimately, consciousness does not really exist as a distinct property, and its appearance as a non-physical property is a kind of illusion. Thus, the hard problem does not originate from real objective features of the world, but rather from implicit subjective beliefs derived from internalised socio-cultural models, specifically from the intuition that mind is an invisible essence generated within an agent.

While I do not want to conceptually challenge the arguments above, I here only suggest potential ethical issues that might arise if we assume the validity of those arguments. What are the potential neuroethical implications of these ideas of consciousness as culturally constructed? Since the concept of consciousness traditionally played an important role in ethical reasoning, for example, in the notion of a person, questioning the objective status of conscious experience may have important ethical implications that should be adequately investigated. For instance, if consciousness depends on culture, then any definition of altered states of consciousness is culturally relative and context-dependent. This might have an impact on, for example, the ethical evaluation of the use of psychotropic substances, which for some cultures, as history tells us, can be considered legitimate and positive. Why should we limit the range of states of consciousness that are allowed to be experienced? What makes it legitimate for a culture to assert its own behavioural standards? To what extent can individuals justify their behaviour by appealing to their culture? 

In addition, if consciousness (i.e., the way we are conscious, what we are conscious of, and our understanding of consciousness) is dependent on culture, then some conscious experiences might be considered more or less valuable in different cultural contexts, which could affect, for example, end-of-life decisions. If the concept of consciousness, and thus its ethical relevance and value, depends on culture, then consciousness no longer offers a solid foundation for ethical deliberation. Softening the hard problem of consciousness might also soften the foundation of what I defined elsewhere as the consciousness-centred ethics of disorders of consciousness (vegetative states, unresponsive wakefulness states, minimally conscious states, and cognitive-motor dissociation).

Although a cultural approach to consciousness can soften the hard problem conceptually, it creates hard ethical problems that require specific attention. It seems that any attempt to challenge the hard problem of consciousness results in a situation similar to that of having a blanket that is too short: if you pull it to one side (in the direction of the conceptual problem), you leave the other side uncovered (ethical issues based on the notion of consciousness). It seems that we cannot soften the hard problem of consciousness without the risk of relativizing ethics.

Written by…

Michele Farisco, Postdoc Researcher at Centre for Research Ethics & Bioethics, working in the EU Flagship Human Brain Project.

We like challenging questions

Are you conscious? Looking for reliable indicators

How can we be sure that a person in front of us is conscious? This might seem like a naïve question, but it actually resulted in one of the trickiest and most intriguing philosophical problems, classically known as “the other minds problem.”

Yet this is more than just a philosophical game: reliable detection of conscious activity is among the main neuroscientific and technological enterprises today. Moreover, it is a problem that touches our daily lives. Think, for instance, of animals: we are (at least today) inclined to attribute a certain level of consciousness to animals, depending on the behavioural complexity they exhibit. Or think of Artificial Intelligence, which exhibits astonishing practical abilities, even superior to humans in some specific contexts.

Both examples above raise a fundamental question: can we rely on behaviour alone in order to attribute consciousness? Is that sufficient?

It is now clear that it is not. The case of patients with devastating neurological impairments, like disorders of consciousness (unresponsive wakefulness syndrome, minimally conscious state, and cognitive-motor dissociation) is highly illustrative. A number of these patients might retain residual conscious abilities although they are unable to show them behaviourally. In addition, subjects with locked-in syndrome have a fully conscious mind even if they do not exhibit any behaviours other than blinking.

We can conclude that absence of behavioural evidence for consciousness is not evidence for the absence of consciousness. If so, what other indicators can we rely on in order to attribute consciousness?

The identification of indicators of consciousness is necessarily a conceptual and an empirical task: we need a clear idea of what to look for in order to define appropriate empirical strategies. Accordingly, we (a group of two philosophers and one neuroscientist) conducted joint research eventually publishing a list of six indicators of consciousness.  These indicators do not rely only on behaviour, but can be assessed also through technological and clinical approaches:

  1. Goal directed behaviour (GDB) and model-based learning. In GDB I am driven by expected consequences of my action, and I know that my action is causal for obtaining a desirable outcome. Model-based learning depends on my ability to have an explicit model of myself and the world surrounding me.
  2. Brain anatomy and physiology. Since the consciousness of mammals depends on the integrity of particular cerebral systems (i.e., thalamocortical systems), it is reasonable to think that similar structures indicate the presence of consciousness.
  3. Psychometrics and meta-cognitive judgement. If I can detect and discriminate stimuli, and can make some meta-cognitive judgements about perceived stimuli, I am probably conscious.
  4. Episodic memory. If I can remember events (“what”) I experienced at a particular place (“where”) and time (“when”), I am probably conscious.
  5. Acting out one’s subjective, situational survey: illusion and multistable perception. If I am susceptible to illusions and perceptual ambiguity, I am probably conscious.
  6. Acting out one’s subjective, situational survey: visuospatial behaviour. Our last proposed indicator of consciousness is the ability to perceive objects as stably positioned, even when I move in my environment and scan it with my eyes.

This list is conceived to be provisional and heuristic but also operational: it is not a definitive answer to the problem, but it is sufficiently concrete to help identify consciousness in others.

The second step in our task is to explore the clinical relevance of the indicators and their ethical implications. For this reason, we selected disorders of consciousness as a case study. We are now working together with cognitive and clinical neuroscientists, as well as computer scientists and modellers, in order to explore the potential of the indicators to quantify to what extent consciousness is present in affected patients, and eventually improve diagnostic and prognostic accuracy. The results of this research will be published in what the Human Brain Project Simulation Platform defines as a “live paper,” which is an interactive paper that allows readers to download, visualize or simulate the presented results.

Written by…

Michele Farisco, Postdoc Researcher at Centre for Research Ethics & Bioethics, working in the EU Flagship Human Brain Project.

Pennartz CMA, Farisco M and Evers K (2019) Indicators and Criteria of Consciousness in Animals and Intelligent Machines: An Inside-Out Approach. Front. Syst. Neurosci. 13:25. doi: 10.3389/fnsys.2019.00025

We transcend disciplinary borders

Anthropomorphism in AI can limit scientific and technological development

Anthropomorphism almost seems inscribed in research on artificial intelligence (AI). Ever since the beginning of the field, machines have been portrayed in terms that normally describe human abilities, such as understanding and learning. The emphasis is on similarities between humans and machines, while differences are downplayed. Like when it is claimed that machines can perform the same psychological tasks that humans perform, such as making decisions and solving problems, with the supposedly insignificant difference that machines do it “automated.”

You can read more about this in an enlightening discussion of anthropomorphism in and around AI, written by Arleen Salles, Kathinka Evers and Michele Farisco, all at CRB and the Human Brain Project. The article is published in AJOB Neuroscience.

The article draws particular attention to so-called brain-inspired AI research, where technology development draws inspiration from what we know about the functioning of the brain. Here, close relationships are emphasized between AI and neuroscience: bonds that are considered to be decisive for developments in both fields of research. Neuroscience needs inspiration from AI research it is claimed, just as AI research needs inspiration from brain research.

The article warns that this idea of ​​a close relationship between the two fields presupposes an anthropomorphic interpretation of AI. In fact, brain-inspired AI multiplies the conceptual double exposures by projecting not only psychological but also neuroscientific concepts onto machines. AI researchers talk about artificial neurons, synapses and neural networks in computers, as if they incorporated artificial brain tissue into the machines.

An overlooked risk of anthropomorphism in AI, according to the authors, is that it can conceal essential characteristics of the technology that make it fundamentally different from human intelligence. In fact, anthropomorphism risks limiting scientific and technological development in AI, since it binds AI to the human brain as privileged source of inspiration. Anthropomorphism can also entice brain research to uncritically use AI as a model for how the brain works.

Of course, the authors do not deny that AI and neuroscience mutually support each other and should cooperate. However, in order for cooperation to work well, and not limit scientific and technological development, philosophical thinking is also needed. We need to clarify conceptual differences between humans and machines, brains and computers. We need to free ourselves from the tendency to exaggerate similarities, which can be more verbal than real. We also need to pay attention to deep-rooted differences between humans and machines, and learn from the differences.

Anthropomorphism in AI risks encouraging irresponsible research communication, the authors further write. This is because exaggerated hopes (hype) seem intrinsic to the anthropomorphic language. By talking about computers in psychological and neurological terms, it sounds as if these machines already essentially functioned as human brains. The authors speak of an anthropomorphic hype around neural network algorithms.

Philosophy can thus also contribute to responsible research communication about artificial intelligence. Such communication draws attention to exaggerated claims and hopes inscribed in the anthropomorphic language of the field. It counteracts the tendency to exaggerate similarities between humans and machines, which rarely go as deep as the projected words make it sound.

In short, differences can be as important and instructive as similarities. Not only in philosophy, but also in science, technology and responsible research communication.

Pär Segerdahl

Written by…

Pär Segerdahl, Associate Professor at the Centre for Research Ethics & Bioethics and editor of the Ethics Blog.

Arleen Salles, Kathinka Evers & Michele Farisco (2020) Anthropomorphism in AI, AJOB Neuroscience, 11:2, 88-95, DOI: 10.1080/21507740.2020.1740350

We recommend readings

This post in Swedish

« Older posts