With infectious enthusiasm and great generosity, Cleeremans gives us a sense of the field and where it is going
Axel Cleeremans is a leader in the science of consciousness. Cleeremans is Professor of Cognitive Psychology at Université Libre de Bruxelles, where he heads the Consciousness, Cognition & Computation Group and the Center for Research in Cognition & Neurosciences. In addition, Cleeremans is Editor-in-Chief of Frontiers in Psychology, which has an excellent Consciousness Research section, and soon-to-be President of the Association for the Scientific Study of Consciousness (ASSC). His biographical sketch on his website concludes with this entry:
2047 Death by stroke right after having solved consciousness
Clearly he has a sense of humor. I recently spoke to Cleeremans to get a better sense of the field and where it is going.
Driven by questions
How does consciousness work? And why are we conscious at all? Is there an evolutionary reason to make it so that we are aware of our mental states in ways that many other organisms are probably not? These are some of the questions Cleeremans likes to spend his days thinking about.
Cleeremans first encountered consciousness research as its own object of investigation at the inaugural ASSC conference in 1996. He said, “Meeting all these people interested in consciousness itself was extremely stimulating and engaging. It’s a great community because it is one of the rare societies where there is true interdisciplinary work.” That interdisciplinary work is hard won. Cleeremans said, “A genuine interdisciplinary dialog takes a lot of time, at least ten years.”
When asked to define consciousness, Cleeremans referenced the hard problem. Philosopher David Chalmers first wrote about the hard problem, which is the strange truth that there is something it is like to be you, in 1996. What it is like is your personal, phenomenal experience. How can we understand the felt qualities of subjective experience? “In my view the problem stays intact today,” Cleeremans said. “Specifically the problem of phenomenal experience. Why is it the case that it feels like anything to be us? Nobody has a good theory of how or why the biological activity of the brain produces our mental states and the subjective qualities that are associated with wakefulness.”
When asked whether those in the field have a clear and consistent definition of consciousness, Cleeremans laughed. “No, certainly not,” he said. “Consciousness is a mongrel concept, right. It has many aspects. So sometimes it’s difficult to figure out exactly what people mean.”
Cleeremans went on to describe some of these aspects, including self-awareness, awareness of other people’s mental states, and metacognition. To complicate things further, he said, these aspects can dissociate. You can have self awareness without theory of mind, or self awareness without phenomenal awareness of vision (aka blindsight), for example.
One concept Cleeremans brought up multiple times is the distinction between levels and contents of consciousness. Some researchers are interested in understanding the differences between sleep, anesthesia, wakefulness, and a variety of altered states such as meditation or hallucinogen use. These are levels of consciousness. Others are interested in understanding how and why a specific perception is experienced consciously while the vast majority of sensory inputs are processed unconsciously. These perceptions are the contents of consciousness.
Another angle some researchers consider is to study the purpose or benefits of consciousness. Cleeremans points out that this line of research does not address the hard problem at all.
In summary, Cleeremans said, “All of these issues get tangled up in different ways, because we don’t really know what are the essential properties and what are the accessory properties.”
Measures of consciousness
Consciousness researchers largely determine “what it is like” for an individual to experience something by asking questions. Cleeremans said, “You can measure phenomenal experience. I can ask you what you are feeling and you can say something about it. Is that an exhaustive measure? Certainly not. But we can talk about our mental states. We have many words to characterize our mental states and communicate them to other people.”
When talking about methods in use, Cleeremans again divided the discussion into approaches being used to study levels vs contents of consciousness.
When studying levels of consciousness, there is the standard qualitative approach, a Q&A to assess the extent of responsiveness. There is also an effort to develop a quantitative measure, with promising results combining brain stimulation with recording to measure the effect of the stimulus on brain activity. For example, you can use transcranial magnetic stimulation and then measure the response with electroencephalogram (EEG), both non-invasive approaches. Researchers can then compare brain responses to stimulation at different levels of consciousness.
When studying contents of consciousness, whether or not a sensory stimulus has been consciously perceived can be determined by asking the subject to choose between a set of coded responses. For example, a dim visual stimulus will flash, and the researcher will ask, “Did you see anything?” Depending on the study, subjects may be asked to give a simple yes/no response or graded response, and maybe also to indicate their confidence in their response (see the perceptual awareness scale, for example, in “When Studying Subjective Experience in Humans, Just Ask”).
Simultaneously measured neural activity can then be analyzed for brain regions that are active when subjects report being aware of the stimulus versus when they were exposed to a stimulus but were not aware of it. The goal of such studies is to identify the neural correlates of consciousness.
This reporting approach may be coming to an end, in preference for the new “no-report paradigm.” Rather than asking subjects to report on perceptions, researchers measure a behavioral proxy, such as eye movement or change in pupil size (see this paper for discussion). One advantage of this approach is that the subjects do not need to be alerted that they are awaiting a sensory stimulus at all, so they are not actively expecting anything. Importantly, researchers identify different neural correlates of consciousness using the no report paradigm than they do when subjects are asked to report on their perceptions.
Cleeremans, who has published on these epistemological issues, said, “The problem of measure in and of itself is not investigated enough in my opinion.” Though he thinks more discussion about best methods would be good for the field, he did not seem concerned about any limitations. He said, “It’s a rich landscape of possibilities there, with people disagreeing on what the best measure is.”
The possibilities are only expanding. When asked if there is anything he would like to be able to measure that is currently inaccessible, Cleeremans thought for a moment, then essentially said that he wants to be able to read people’s minds. “We would all love to know what goes on in other people’s minds,” he said. “So of course that is what you want to measure — the contents of your mental states based on the biological activity of your brain.”
In the field of cognitive neuroscience, ‘mind reading’ is the phrase science news outlets use to report on the nascent field of decoding. In essence, decoding can reconstruct a sensory stimulus such as an image the research subject was viewing based on recordings of neural activity (or indirect measures of neural activity, like fMRI). Decoding can also be used to get at internal mental processes, such as decision making.
Cleeremans pointed to exciting work by Hakwan Lau, who is developing decoding methods for consciousness research. One recent paper from Lau’s lab, by Vincent Taschereau-Dumouchel et al., used decoding to identify representations of feared animals in the brain (view the paper in PNAS). As other groups have been finding, there is a remarkable similarity between individuals, such that the representation of a feared animal in one individual’s brain can be aligned and mapped onto the brain of a second individual. Cleeremans said, “If we could reconstruct mental states through decoding, that opens up all sorts of possibilities.”
Major theories of consciousness
Cleeremans described the current state of the field as being in stasis. He said, “For the past five or ten years it felt like there were these big ideas out there and then people defending these big ideas, doing and designing experiments meant to basically support the theory. So you have these mounting piles of data supporting this theory or that theory but never any genuine effort at making the two interact.”
He pointed to the Templeton World Charity Foundation initiative to accelerate research on consciousness as a good development that may help the field move on. The Templeton initiative funds adversarial collaborations between holders of different theoretical positions, so they must work together to design experiments to test their theories.
There are four major theories being considered: Global Workspace Theory, Integrated Information Theory, First-Order Theory, and Higher-Order Theory. You can learn more about these theories and find some recommended reading on the Awake & Alive Mind Fundamentals page.
Cleeremans said, “If you had to pick one, many people would pick Global Workspace Theory as the best account we have, at least of the functional aspects of consciousness. It still doesn’t explain, or address even, the hard problem, but it’s the most solid theory in terms of empirical evidence and empirical predictions.”
Getting started in consciousness studies
I asked a number of questions aiming to get at advice for those just entering the field. As is true with any scientific field, going to conferences can be a good way to get oriented.
There are two major conferences in the field. Just a few years before the first ASSC conference, the Center for Consciousness Studies at the University of Arizona in Tucson held its first conference. Both organizations continue to hold annual conferences, with the ASSC focused on scientific approaches, while the Tucson conferences have a wider scope, including spiritually driven approaches. You can find more info and registration pages here and here.
I asked what kind of math is most useful to know. Cleeremans seemed a bit surprised at the question, but said statistics. Specifically Bayesian statistics. He noted that many in the field, and in cognitive neuroscience generally, apply the paradigm of predictive processing, which is rooted in Bayesian statistics (see this recent primer by Karl Friston in PLOS).
True to the earlier statement that the science of consciousness is an interdisciplinary field, Cleeremans said that studying consciousness requires deep understanding of different disciplines. “I think you have to be aware of the ideas of philosophers, relevant neuroscience findings, and relevant psychological studies,” he said. “People with a strong neuroscience background have the most potential, but we are so far from understanding how the biological mechanisms of the brain produce mental states, it’s unclear whether that would give anyone an edge. Cristof Koch, who is an outstanding neuroscientist, has been doing this for twenty years and there is still no answer from him,” Cleeremans concluded with a laugh. You can read what Koch has learned so far in his latest book, The Feeling of Life Itself.
When asked whether he had any advice for people starting out in the field, Cleeremans said, “My advice is be patient. It’s a very complicated problem. I think a lot of the action is in thinking of good experimental paradigms. That requires deep knowledge of psychology, of course, but also clear understanding, and that only comes from philosophers I think, of what you are looking for. I am driven by good questions. So that would be another piece of advice. Find a good question to ask. If we have a good question, then maybe we can find an answer.”