This post will be extended; for now, this new piece of research by James Elliot provides evidence for partial awareness (Kouider), i.e. that we don’t necessarily see the whole, but can access parts of the whole. This differs somewhat from earlier “gradedness” accounts which focussed on the “gradedness” of subjective reports (Overgaard). A couple of years ago, another paper showed how gradedness depended on the abstractness of the level that had to be reported (Windey & Cleeremans).
More goodies from Hakwan Lau’s lab — this article by Peters and Lau (2015) is of direct interest to Dr Bert Timmermans‘ lecture on measures of consciousness. It is highly probable that some processing of the outside world goes on outside of awareness, either by unconscious perception as shown by priming studies (but Sid Kouider would argue that we just perceive the stimulus at a different level, partially), or simply by us not always being able to identify what influences our behaviour (the highly controversial notion of unconscious behavioural priming). But what constitutes “unconscious perception”?
Researchers have been using both the objective and subjective threshold model. In the first, a stimulus is considered conscious as soon as a direct measure (decision task: did you see a word or not) shows above-zero sensitivity. Even if the person still claims they haven’t seen anything, subjectively. And unconscious influences can only be observed with indirect measures (such as the influence of an unseen primed word “table” on recognition of the subsequently presented word “chair”). Still, many researchers, especially in the field of implicit learning*, have been using a subjective threshold. A subjective threshold holds that as consciousness is inherently subjective, something is unconscious as soon as people can’t report it. So an above-zero sensitivity in a direct (decision) task in the absence of reported subjective awareness is proof of unconscious processing.
This paper by Peters & Lau suggest however that there may not be any unconscious perception as defined by the subjective threshold: when people report no subjective awareness (i.e. are at subjective threshold), they also show zero sensitivity in the decision task (i.e. are at objective threshold). So, objective and subjective thresholds coincide! In their words “This surprising finding suggests that the thresholds for subjective awareness and objective discrimination are effectively the same: if objective task performance is above chance, there is likely conscious experience.” Indeed, at least for briefly presented stimuli, this is also something we observed some while ago (Sandberg, Timmermans, Overgaard & Cleeremans, 2010), and which gives hope, in that it suggests that when done correctly, simply asking people what they saw is a sensitive enough measure. It also means that there may not be such a thing as unconscious perception at the subjective threshold (even if our 2010 results suggest that unconscious information may just start playing a role once conscious information seeps though). Whether the both objectively and subjectively “unseen” information would still leave enough of an activation trace to influence behaviour on a subsequent indirect task remains… to be seen!
Update: keep in mind that there is a caveat, in that it might just be that confidence ratings (which is the subjective measure they use) simply do not tap into subjective awareness, and that it is very much linked to the system’s metacognitive knowledge about how well it is doing, a function that can be executed both consciously and unconsciously. In other words, confidence ratings may not be exclusive, i.e. may be reflecting the also the presence or absence of Unconscious knowledge, like when you rate your confidence in a task based on a “hunch”. Also, it may be that over the course of the experiment, confidence ratings thus come to tap into the same unconscious knowledge as task performance, in that the system/person has learned the task, and also the distribution of their performance accuracy… unconsciously!
[* In IL, something is only conscious if a person reports on it. Now, for implicit learning this is understandable, because you’re trying to exclude knowledge of rules (of an artificial grammar for instance) while showing people’s above-zero sensitivity to rules they never knew were in the material, and an objective threshold would mean that you would have no proof of learning whatsoever and hence would assume all learning is conscious.]
This episode explores the question of how the brain gives rise to our thoughts, emotions, memories and personality. We see how the process of becoming ‘you’ starts at birth. The brain of a newborn baby is not yet fully developed, instead it grows and shapes itself around life experience.
Wiring up begins immediately, and rapidly, as the child’s brain starts to adapt to whatever situation – culture, habitat, language – it’s born into. This allows humans to flourish in any stimulating environment, but as the story of three Romanian orphans reveals, a lack of social contact and stimulation can result in permanent brain damage as the brain fails to make vital connections in those critical early years.
Tracing the development of the brain – the ‘making of you’ – through a lifetime, Dr Eagleman tests the social stress levels of teenagers as their brains go through profound changes, meets London cabbies whose intense training to memorize street maps physically alters the shape of their hippocampus, and joins a group of elderly nuns who are defying the symptoms of Alzheimer’s by keeping their brains active and building up ‘cognitive reserve’.
As we make new memories, learn new skills and have life experiences, the brain is constantly and dynamically rewiring itself. It never stops. Nor do we – the human brain is always changing, and therefore so are we. From cradle to grave, we are works in progress.
Heuristic use of perceptual evidence leads to dissociation between performance and metacognitive sensitivity
And more on the brain as an accumulator of evidence for behavioural decisions and/or awareness: Brian Maniscalco, Megan Peters and Hakwan Lau (at Lau’s UCLA lab) show in this psychophysics paper that sometimes an increase in decision accuracy can be accompanied by a decrease in metacognitive sensitivity. In particular, they explain how the “consciousness” information stream, as expressed in the level of confidence, is driven mainly by the response-congruent evidence, and is largely insensitive to the level of response-incongruent evidence. The perceptual decision itself, however, is driven equally by response-congruent and response-incongruent evidence. In other words, your behavioural decisions may be good, but your conscious awareness tends to look only at evidence confirming that decision, and has “done away” with any disconfirming information that the “lower” system might have had.
The 2015 William James prize, which the Association for the Scientific Study of Consciousness annually awards to the best piece of research in the field of consciousness, was given to Alexandra Vlassova, who, in two rather phenomenal papers, shows how “evidence accumulation” works not just for conscious awareness, but also for unconscious information: accumulated perpetual evidence will bear on decision accuracy but not on confidence rating (a measure for awareness).
These are the two papers:
- Vlassova, A., Donkin, C. & Pearson, J. (2014). Unconscious information changes decision accuracy but not confidence. Proceedings of the National Academy of Sciences, 111 (42), 16214–16218. “One of the more intriguing but controversial ideas in psychology is that unconscious information can influence our decisions without us even knowing it. Here, we explicitly tested these controversial ideas with a novel behavioral task and computational models of decision-making. We report that unconscious information can be accumulated in a similar manner but less effectively than conscious information. However, unlike conscious information, unconscious information does not seem to boost decision confidence. Our findings cannot be accounted for using existing models of priming or adaptation.”
- Vlassova, A. & Pearson, J. (2013). Look before you leap: sensory memory improves decision-making. Psychological Science. 24(9), 1635-1643. “[…] Further, memory and perception showed equivalent rates of evidence accumulation, suggesting a high-capacity memory store. We propose an account of continued evidence accumulation by sequential sampling from a simultaneously decaying memory trace. Memories typically decay with time, hence immediate inquiry trumps later recall from memory. However, the results we report here show the inverse: Inspecting a memory trumps viewing the actual object.”
GO has been notoriously difficult to crack, because unlike chess, it has a limited number of rules but many more possibilities, yet using deep neural networks, a computer has now won against a champion. The fascinating/frightening thing is that the underlying algorithm/structure is relatively “simple” in AI terms — comparable algorithms have been used in recent years in computer vision: check out MIT’s algorithm giving impressive descriptions of any of your own or online’s pictures. Flickr used a similar algorithm, but had to take it down because people were offended by some mismatches (which basically happen because it almost exclusively works on pattern association principles).
The Material Soul — Anil Seth explores what neuroscience can –and can’t– tell us about consciousness
In this article in January’s issue of Sp!ked, Anil Seth, co-director of the Sackler Centre for Consciousness Science (professor of cognitive and computational neuroscience) at the University of Sussex, talks about the neuroscience of consciousness, with a particular focus on the self and the role of free will in that.
“There’s a frequent misunderstanding that a scientific explanation of a phenomenon first needs a definition everyone agrees is the right one. Almost any talk I give, people will always ask, ‘can you define consciousness?’. But if you look at the history of science, definitions continually evolve. Life is a good example. Definitions of life have evolved along with our understanding of it. The gene wasn’t really defined until scientists had figured out that there was something that was a hereditary basis for organism characteristics.”
“A big part of being a self is our experience of determining, or being in control of, our behaviour. We do X but we could have done Y, and we do so in a way that seems not completely constrained by our environment. So we do have this experience of being the origin of voluntary actions. But we have it selectively. When you quickly move your hand quickly away from a hot stove, you don’t experience any sense of volition. It’s a reflex action. But when making very difficult decisions, having to balance multiple factors, we can have a very strong and actually quite aversive experience of deciding this way or that way, and being responsible for our actions.”
In this interesting piece, Michael Graziano, the researcher who wrote Consciousness and the Social Brain (in between children’s books and musical pieces), expands on how consciousness is essentially the result of a process in which the brain re-describes itself to itself, kind of the brain representing itself. Graziano, together with his wife Sabine Kastner, puts forward the idea that the origin of this capacity lies in being a social species that has Theory of Mind. So, rather than assuming we have introspective capabilities that we can turn to others, the idea is that over the course of evolution we developed ways of predicting what others will do, using some sort of model of what the other is, and eventually we learned to turn that model on ourselves, the brain thus learning to inspect its own processes.
Philosopher Peter Carruthers made a philosophical case about this some time ago in an interesting paper (which however relies a lot on the notion of representationalism), and Axel Cleeremans, Antoine Pasquali, and I put forward similar ideas suggesting that there are multiple levels in which a brain can represent its content to itself, modelling how it might do so (digest general article; theory article 1; theory article 2; neural network modelling article)
For the mathematically inclined: Measuring consciousness based on Tononi’s Integrated Information Theory
Max Tegmark (MIT physics dept.) : “Integrated Information Theory of consciousness (IIT) [. It] says that information being processed is conscious if a mathematical quantity called Φ (“Phi”) is large. Phi quantifies integration, the extent to which information is interconnected into a unified whole rather than split into disconnected parts. The theory has generated interest from the neuroscience community, but also controversy, including recent critique from Scott Aaronson.
I want to see the question of whether IIT is correct or not resolved by experimental tests. Unfortunately, Tononi’s proposed measure of integration is too slow to compute in practice from state-of-the-art patient data, requiring longer than the age of our universe, let alone the lifetime of the patient. I’ve therefore worked hard over the last year in search of a faster way to compute integration, and I’m happy report that I’ve found one — in fact, several. In a paper I just posted, I explore and classify existing and novel integration measures by various desirable properties, finding that although there at first seem to be a few hundred options, there are in fact only a handful if attractive ones. I was happy to discover that there’s an approximation based on graph theory that let’s you dramatically speed up the exact formulas, so that they can be applied to real-world data from laboratory experiments without posing unreasonable computational demands. Let’s try to shed more light on the fascinating questions and theories about consciousness my making experimental tests!”
Libet’s interpretations of his famous fee will-experiment have remained controversial, but this hasn’t stopped scientists carrying out variations of his experiment. Among other things, this has revealed that people with Tourette’s syndrome, who have uncontrollable tics, experience a shorter veto window than people without the condition, as do those with schizophrenia. Emilie Caspar and Axel Cleeremans of the Free University of Brussels (ULB) in Belgium decided to see if the same was true in impulsive, but otherwise healthy, people.