Dreaming update

Note: Sorry this is short and late. I am having computer problems. Encephalon is back – #81 is at http://cepholove.southernfriedscience.com/

Christof Koch has written an article on dreaming for the Scientific American ( here ). It discusses common ‘facts’ about dreaming that have been shown to be in error. This put a different light on the relationship between dreaming and consciousness.

  1. There is not an absolute mapping between REM sleep and dreaming. Non-REM dreams are more static snapshots then narratives in the first person – but not always. Sleepwalking dreams are during non-REM sleep (not too surprising as REM sleep includes a sort of paralysis.

  2. The pons in the brainstem is required for REM sleep but not for dreams. Areas in the temporoparietal-occipital junction in the neocortex are required for dreaming.

Further, we do not smell during dreams, like we see, hear, touch. This fits with our low awareness of smell in consciousness. And there are many more interesting observations in the article, so I recommend reading it.

“they (dreams) bear witness that the brain alone is sufficient to generate consciousness. We dream with eyes shut in the dark, disconnected from the outside world. The brain regions responsible for basic sensory perception are deactivated. Nor is behavior necessary, as we are motionless except for our breathing and eye movements”

Hacker on consciousness

There is an interview by James Garvey of the philosopher Peter Hacker posted on TMP magazine site (here) about his ideas on consciousness.

“Philosophy does not contribute to our knowledge of the world we live in after the manner of any of the natural sciences. You can ask any scientist to show you the achievements of science over the past millennium, and they have much to show: libraries full of well-established facts and well-confirmed theories. If you ask a philosopher to produce a handbook of well-established and unchallengeable philosophical truths, there’s nothing to show. I think that is because philosophy is not a quest for knowledge about the world, but rather a quest for understanding the conceptual scheme in terms of which we conceive of the knowledge we achieve about the world. One of the rewards of doing philosophy is a clearer understanding of the way we think about ourselves and about the world we live in, not fresh facts about reality. … The world doesn’t have scaffolding. Rather, in doing philosophy, we come to realise the character of the grammatical and linguistic scaffolding from which we describe the world, not the scaffolding of the world.”

Hacker is not happy with the confusion that neuroscience has produced in the hands of popular writers.

“Merely replacing Cartesian ethereal stuff with glutinous gray matter and leaving everything else the same will not solve any problems. On the current neuroscientist’s view, it’s the brain that thinks and reasons and calculates and believes and fears and hopes. In fact, it’s human beings who do all these things, not their brains and not their minds. I don’t think it makes any sense to talk about the brain engaging in psychological or mental operations. … you will find that the operations of the brain thus conceived are being advanced as explanations for human behaviour, for our thinking, believing, seeing, hoping and fearing. That’s wrong, because it’s no explanation. If someone wants to know why poor old Snodgrass, as the result of some lesion, can’t do something that normal people can do, and you say that his brain can’t do it, you haven’t advanced any explanation at all. … There is no such thing as a brain’s thinking, wanting, reasoning, believing or hypothesizing.”

I must say I have some difficulty with this. But it is reasonable to point out that ‘brain does it’ is not an explanation. What is needed is how the brain does it. And it is reasonable not to separate the brain from the person it is part of. He goes on with something that I do really like. He takes on Nagel’s idea, “an organism has conscious mental states if and only if there is something that it is like to be that organism – something it is like for the organism” and rejects it for three reasons.

(1)“… the answer to ‘What was it like for you to do it?’ isn’t ‘It was like wonderful’ – unless we’re in California – but rather ‘It was wonderful’. So it is a plain confusion to think that for any given experience of a conscious creature, there is something that it is like for the creature to have that experience.”

(2)“Another kind of mistake is a systematic confusion between the qualities of an experience and the qualities of the objects of an experience. The question ‘What was it like for you to love Daisey?’ can be given an answer by specifying the hedonic character of the experience of being in love with her. It may have been wonderful, or heart-breaking. The question ‘What is it like to see something red?’ has no such answer. Seeing a red button, for example, is neither wonderful nor heart-breaking, neither exciting nor boring – it simply lacks any hedonic quality.”

(3)“It looks mysterious because it’s modeled on ‘what’s it like for an X to be a Y?’ “Daddy, what was it like for you to be a soldier in The Second World War? All sorts of stories follow. Daddy, what’s it like for you to be a human being? What on Earth is this question? The question is illegitimate. It’s stepped over the bounds of sense. The requirement is that there be a difference between the X and the Y. What’s it like for a woman to be a surgeon? Perfectly decent question. What’s it like for a woman to be a woman? … as opposed to what? A woman who isn’t a woman?”

Is that not refreshing? But here is the really good stuff.

… understand how consciousness could arise from boring old matter?

“Slow down and start thinking of consciousness as we should, not in terms of ‘what it’s like’. Start with elementary sentience.” Hacker tells a story of a creature hundreds of millions of years ago, a creature with light sensitive cells, avoiding predators, reproducing, and all the while the mechanisms of evolution whir away. Light sensitive cells develop into eyes. Eyes give a creature the ability to see. Creatures that can see can be conscious of something moving in the underbrush over there, and so on. He’s trying to take the mystery out of the equation. He bangs a fist on a table, “How could this stuff be conscious? – It couldn’t! How could consciousness arise from mere matter? – It can’t. Consciousness ‘arises’ from the evolution of living organisms.”

Not convincing

ScienceDaily has an item on research by G.Kuhn and others on whether autism would effect the perception of illusions. (here) I find this research unconvincing as reported by ScienceDaily. As the original paper is not free on line, I have not been able to read it.

Magicians rely on misdirection — drawing attention to one place while they’re carrying out their tricky business somewhere else. It seems like people with autism should be less susceptible to such social manipulation. But a new study in the U.K. finds that people with autism spectrum disorder are actually more likely to be taken in by the vanishing ball trick, where a magician pretends to throw a ball in the air but actually hides it in his hand.

There appears to be an assumption that all magic tricks rely on social manipulation. Manipulation yes, but not always social manipulation. This illusion has nothing to do with social clues and it can be done with an obviously mechanical machine. The illusion is due to the prediction of a motion in the consciousness. Everyone who has a conscious awareness of the event will see the ball leave the hand as it did on several previous movements of the hand. This type of illusion is well known and well studied.

I am sure that it is a good idea to use illusions to understand conditions like autism, but it would be a good course to first understand why the illusion works for normal people.

The researchers keep ‘digging’ when they try to explain the results (at least in the ScienceDaily summary). Conscious experience is not a simple, straight forward phenomenon that can just be taken at face value especially when looking at illusions. All magician’s tricks are not based on just social manipulation.

Hearing shapes

Perception of shapes is possible by touch and by sight. Kim and Zatorre have been using a coding of shape information into sound information to examine the nature of shape perception. They use boards with 2D drawings on them have textured surfaces giving visual and tactile targets. These are coded to give matching ‘soundscapes’ where one dimension is coded by frequency and the other by stereo panning. With current and previous experiments, they show that subjects can be trained to identify both visual and tactile targets from soundscapes. Further subjects trained to match sound and touch can do the matching of sight and sound without further training.

 

Here is the abstract:

Shape is an inherent property of objects existing in both vision and touch but not audition. Can shape then be represented by sound artificially? It has previously been shown that sound can convey visual information by means of image-to-sound coding, by whether sound can code tactile information is not clear. Blindfolded sighted individuals were trained to recognize tactile spatial information using sounds mapped from abstract shapes. After training, subjects were able to match auditory input to novel auditory-tactile pairings. Furthermore, they showed complete transfer to novel visual shapes, despite the fact that training did not involve any visual exposure. In addition, we found enhanced tactile acuity specific to the training stimuli. The present study demonstrates that as long as tactile space is coded in a systematic way, shape can be conveyed via a medium that is not spatial, suggesting a metamodal representation.

 

Not mentioned in the abstract is their theory of what these experiments say about the perception of shapes.

This transfer of crossmodal learning further supports our hypothesis that shape can be represented at a highly abstract level in a form independent of the sensory modality in which it is learned. This amodal, abstract representation of shape is closely associated with the findings of human imaging studies that identified the lateral occipital region as a common brain region involved in shape recognition by both vision and touch and by audition using the same type of sound transformation used in the current study.

 

It is interesting to think of shape perception as a link between two aspect of perception: the housing of our perceptions in a three dimensional space and the separation of our perceptions into distinct objects. Both of these seem hardwired and at the foundation of the form our conscious awareness takes.

ResearchBlogging.org
Kim, J., & Zatorre, R. (2010). Can you hear shapes you touch? Experimental Brain Research, 202 (4), 747-754 DOI: 10.1007/s00221-010-2178-6

Flexibility of the senses

There were two recent items in ScienceDaily reporting two sides of the brain’s flexibility. In one paper S. Lomber and others reported that in cats born deaf, vision is reorganized to use the neural space that is usually used by hearing. This gives better localization in peripheral vision and better motion detection. (here) In the other paper J. Rauschecker and others found that people blind from birth show fMRI activity in visual modules from sound and touch stimuli. These modules analyzed spatial location, patterns and motion. (here)

There are some interesting aspects of this plasticity. First, the ‘born’ in born deaf and born blind is probably very important. If the eyes are not signaling then the visual cortex is not undergoing the developmental stage that ties it to vision. If the ears are not signaling then the auditory cortex is not tied to hearing. This lack of the normal connection probably allows unusual connections to be made more easily.

Second, parts of the cortex that do not get normal stimulation take on stimulation from neighbouring parts of the cortex – rather than being idle or dieing.

Third, aspects of perception like the position of something in three dimensional space can involve more than one sense mode. I can tell where something is by the sound it makes or if trained by the echoes from it; I can tell from the position of my body when I touch it where it is; and I can see where the light is reflected off the object. Location and movement in space is so important to perception that we can think of it as the pre-existing framework that objects are placed in. Location is not so much a product of vision as vision is a tool of location and not its only tool.

Fourth, consciousness has this spatial framework no matter what sense is feeding it. Thus born blind can ‘see’ what they hear and the born deaf can ‘hear’ what they see. Seeing really has two meanings, one that means sensory information that comes via the eye and one that means the usual conscious awareness of the world from a viewpoint similar to the eye’s. Hearing also has two meanings. Hearing is what we get from our ears and it can also mean the part of our consciousness that we know as sound.

The where, when, how and why

A recent review article by Friedemann Pulvermuller looks at what is known about the neurobiology of language. He uses the question of what recent progress has been in the where, when, how and why of language processing in the brain. He does a masterful job and yet I am, personally, disappointed. In what way I am disappointed comes later. First comes Pulvermuller’s insights.

Where: We have our old friends Wernicke’s and Broca’s areas and their surroundings, referred to as the left-perisylvian language cortex. They are very heavily connected to one another. But that is not the only where: widespread areas in both hemispheres can be involved depending on the meaning of words. For example the word “kick” causes activity in areas dealing with the legs.

When: Processing is not serial; phonological, lexical, syntactic, semantic and pragnatic processing are simultaneous. Basic understanding is gained within 250 msec of hearing/seeing even if the utterance is not attended to. Robust activity at about 500 msec depend on a combination of strong (loud) stimulus, attention to it or need for re-analysis.

How: It appears that there is not separation between perception and action but a complex interaction between them involving prediction. Bottom-up sensory activity produces a hypothesis, then a top-down action-like synthesis produces a prediction to be matched with further input.

Why: Here we have a number of important features to explain but very few answers. Basically, the brain needs to do its language work with speed, flexibility and ease of learning.

Now for my disappointment. First, there is a lot here that is similar to non-language processing, especially the predictive testing and monitoring, and this is hardly mentioned. The gulf between studying preception, cognition and action in the language sphere and in other activities is not necessary and hinders progress in both. Secondly, there is my focus of interest, consciousness, which is also hardly mentioned. In the section on timing, it would have been reasonable to notice that event-related-activity that dies out before about 250 msec does not reach consciousness. Those that remain active passed about 300 msec do reach consciousness. This is probably the nature of Pulvermuller’s early and late activity but he does not make the connection. Reaching consciousness allows the use of working memory, which would likely be essential to re-analysis an utterance. Further, consciousness and attention are usually closely linked. I would like to find hints to why language seems more likely to reach consciousness then many other activities and I found none in this article.

Aside from my perhaps unreasonable obsession with consciousness, I hope that many read the article because it brings together the bio and the linguistics in the Bioliguistics Journal.
ResearchBlogging.org
Friedemann Pulvermuller (2010). Brain-Language Research: Where is the Progress Biolinguistics, 4 (2), 255-288

Arriving at where we started

“And the end of all our exploring

Will be to arrive where we started

And know the place for the first time.” – T S Eliot

There is a bit of this sentiment in embodied cognition theory. We can see this in the way our language is riddled with heart metaphors implying that the heart is the seat of some emotions. The idioms show that once it was natural to include more than the brain in our feeling/knowing/aware selves. “He has a broken heart”; “my heart is not in it”; “she is all heart”; “they are hard hearted”, “my heart tells me to do it”. These saying hark back to a time when we did not separate ourselves at the neck, but thought of ourselves as an indivisible whole.

One way to look at the connection of the brain with the rest of us, is to notice, that our conscious awarenesses know our emotional state through its effects on our bodies. We are conscious of our excitement because our heart rate rises. We are conscious of our displeasure because we feel our brow frown. We are conscious of our fear by our breathing becoming weak. The easiest route to consciousness is through the senses and that includes our sensing our own bodies.

Another way to view the connection is to look at where our conscious thoughts come from. We can think of everything having a predecessor. We start with what can be felt by a newborn and assume that growing up provides layer upon layers of elaboration of the things a baby can know. We use nested metaphors and analogies. So the feeling of social comfort can be tapped by feelings or warmth or closeness. Physical pain and social rejection are related.

Embodied cognition research has found many connections. Recently there was an article in the Scientific American by Adam Waytz that elaborated on the heart-brain connection. (here)

Psychology’s recognition of the body’s influence on the mind coincides with a recent focus on the role of the heart in our social psychology. It turns out that the heart is not only critical for survival, but also for how people related to one another. In particular, heart rate variability (HRV), variation in the heart’s beat-to-beat interval, plays a key role in social behaviors ranging from decision-making, regulating one’s emotions, coping with stress, and even academic engagement. Decreased HRV appears to be related to depression and autism and may be linked to thinking about information deliberately. Increased HRV, on the other hand, is associated with greater social skills such as recognizing other people’s emotions and helps people cope with socially stressful situations, such as thinking about giving a public speech or being evaluated by someone of another race. This diverse array of findings reflects a burgeoning interest across clinical psychology, neuroscience, social psychology, and developmental psychology in studying the role of the heart in social life.

We are returning to ways of knowing ourselves that are very old, but, hopefully, we are returning with more understanding.

Friston and Freud

I have liked Friston’s ideas for some time, so what a shock it is to find him defending Freudian ideas. Naïve me, I thought that Freud’s model was dead in the water. Why? It is untested, does not fit with current evidence and, further, is probably untestable therefore not good science. It fails the Occam’s razor test and so is not even good philosophy. It relies heavily on introspection which is unreliable as a source of information about anything except introspection itself. Friston does not seem to put forward any good reason for trying to raise the dead.

In terms of theoretical and computational neuroscience, we will focus of Helmholtz’s suggestion that the brain is an inference machine; this idea is now a fundamental premise in neurobiology. Key examples of this include the Bayesian brain, predictive-coding and the free-energy principle. This framework assumes that the brain uses internal hierarchical models to predict its sensory input and suggests that neuronal activity (and synaptic connections) try to minimize the ensuing prediction-error or (Helmholtz) free-energy. This free-energy is a measure of surprise and is essentially the amount of prediction-error. It is an information theory quantity that, mathematically, plays the same role as free-energy in statistical thermodynamics. Free-energy is not an abstract concept; it can be quantified easily and is used routinely in modelling empirical data and in neuronal simulations of perception and action.

We have an analogy, a metaphor, here. Parts of this framework are accepted by various neuroscientists but not necessarily literally. In the context of the brain, free-energy is abstract and is metaphoric.

When the term energy is used by Freud, it is also a metaphor. Freud’s mind is divided into id, ego and superego. The id, home of primitive drives and pleasure seeking, is bubbling with free libido energy. The ego tries to control the id to conform to reality, social rules, perception, judgement by reducing that energy from the id.

This Freudian use of the word ‘energy’ has nothing to do the thermodynamics use and neither has anything to do with use in predictive error reduction. Freud and thermodynamics share a word; thermodynamics and Friston’s version of error reduction share equations, a mathematical structure. The flow of electricity and of water share equations, that does not make them physically similar. Optics and weight share the word ‘light’ but that does not make them similar.

Friston carries on and does some damage to the idea of hierarchy in brain processes and the default network in order to make them fit with the primary and secondary Freudian processes.

Would it not be better to allow Freud’s terrible theory to just fade away? Do we really need to have people struggle with the myth that they have a sex craved monster living in the cellar of their minds?

 

ResearchBlogging.org
Carhart-Harris, R., & Friston, K. (2010). The default-mode, ego-functions and free-energy: a neurobiological account of Freudian ideas Brain, 133 (4), 1265-1283 DOI: 10.1093/brain/awq010

Watch out for modules

A posting on the Neuroskeptic blog (here) reminds us that various scanning methods create an image of the brain that we know is misleading. The picture we must avoid slipping into without good evidence is one where particular functions/processes in the brain are localized and somewhat isolated from others. The lovely coloured areas on scans are somewhat artificial. Research on the connections between neurons shows that cells that are quite distant from each other are connected and work together (see previous posting). Other research shows that quite distant neurons can fire in unison. The paper that Neuroskeptic writes about is one that looks at the prefrontal cortex and shows that is not as modular as it is assumed to be. The same networks within the prefrontal cortex can do various different work.

As far as our subject, consciousness, is concerned, it is clear that the production of consciousness involves much of the cortex and some lower brain areas engaging in synchronous activity. It is about as global and all inclusive an event as we can currently see in the brain. We do not know what is happening in consciousness but we do know that it is not confined to a particular region of the cortex. No diagram that labels a location for consciousness, a consciousness module, should be trusted.

Of course knowing where there is the most neural activity is a step towards understanding mental processes in general. At present that information is very coarse-grained in space and in time. It is only a step. It adds some clarity, but if it is taken too literally it can also make things less clear, especially for the general public.

Neurons are not magic

The answer to any question these days seems to be “mirror neurons” as if magic was an acceptable explanation if the magician was a neuron. They are used to explain imitation, language, empathy and theory of mind effects. There are cells that are active if a particular motor act is performed or if it is seen to be performed by another. They have been given the name mirror neurons because presumably the mirror neurons for a particular action would be firing in both the actor and observer of the action at the same time.

We can think of them as cells that fire when the concept of a particular motor act is an appropriate concept. There is nothing magic in that. Some have taken the role of mirror neurons further and have made them the central component of a theory of mind-reading – because the observer and the actor have same mirror neuron activity, the observer can understand the intention of the actor. It is true that we are fairly good at understanding the intentions of others, but I have assumed that this was essentially good guessing using a complex cognitive process.

In a paper, Pierre Jacob examines the original theory of Gallese and Goldman. Jacob characterizes their theory as two ideas. First that the mirror neurons in the agent and observer resonate because the observer replicates the mirror neuron activity of the agent. Second, the activity of mirror neurons in the observer are instrumental in deciding what intentions the agent had, the intentional cause of the action.

Jacob feels both ideas are flawed. Mirror neurons have some aspects of intention in their functioning and therefore cannot be the sole source of knowledge about intention. For example, a reach with the hand elicits different mirror neurons depending on whether there is a target around. Further a motor act can have many intentions and so cannot be the sole source of knowledge about the agent’s intention. For example, a reach with the hand can be to grasp something or just touch it.

Jacob proposes that mirror neurons along with sensory data are used to predict the agent’s future intentions and actions. This becomes a complex cognitive process rather than an almighty mind-reading leap.

MNs do not compute a representation of the agent’s intention from a representation of her motor command, in accordance with an internal forward model. Instead, they compute a representation of the agent’s motor command from a prior representation of the agent’s intention, in accordance with an internal inverse model. Thus, the present account does not detract from the significance of MNs for primates’ social cognition, since it emphasizes their contribution to an observer’s ability to predict a conspecific’s next motor act.

 

ResearchBlogging.org
JACOB, P. (2008). What Do Mirror Neurons Contribute to Human Social Cognition? Mind & Language, 23 (2), 190-223 DOI: 10.1111/j.1468-0017.2007.00337.x

Gallese, V. and Goldman, A.I. (1998). Mirror neurons and the simulation theory of mindreading Trends in Cognitive Sciences (2), 493-501