Jim Beran's Talk about Real-time Plasticity at Toward a Science of Consciousness 2012, Tucson, Arizona

Description of Content
This article describes the talk Jim Beran gave on April 10, 2012 at Toward a Science of Consciousness 2012 at Tucson, Arizona. For each slide, there is a paraphrase of what Jim said next to it.

Slide 1
Good evening!

I ’ m going to talk about protozoans, microtubules, plasticity, evolution, and consciousness. That’s a lot to cover in 20 minutes, so I’ll touch some points very briefly, but I’d be happy to discuss those with you in more detail later.

Before I start, a few words of thanks--

Slide 2
I would especially like to thank the organizers for including me in such an interesting session.

The last item might remind some of you of a talk I gave in Stockholm last spring about a different topic—change in conscious experience due to inductive coupling of transient electrical currents in microtubules. Around that time, an abstract I submitted to another conference had been rejected, but one of the anonymous reviewers who rejected it made a comment that helped me organize my Stockholm talk. I have found that comment useful again today.

Slide 3
Here’s the reviewer’s comment, together with a caricature of what the reviewer might look like--I don’t know his or her identity, so I have no idea whether this image is even close.

Slide 4
Neuroscientists frequently refer to “neural circuits”. Contrary to the anonymous reviewer’s position, the consensus view in neuroscience is apparently that cytoskeleton, which would include microtubules, actin filaments, intermediate filaments, etc., is generically involved in constructing and operating “neural circuits”, including those that underlie consciousness.

But what are “neural circuits”?

Slide 5
Here is the best explanation I’ve found of the term “neural circuits”, from the 2008 fourth edition of a textbook by Purves et al. entitled Neuroscience.

Two features of particular interest: --First, a neural circuit must be an “ensemble” of neurons that do not function in isolation--in other words, a neural circuit has two or more neurons and they somehow function together; and --Second, a neural circuit must process information; Purves et al. go on to explain that the direction of information flow is a preeminent feature of a neural circuit, a feature that is essential to understanding the circuit’s purpose.

Slide 6
Going back to the previous image, note that this image is very schematic, illustrating just a few of the important roles of cytoskeleton in neural circuits.

Cytoskeleton plays an important part in a growth cone at the leading edge of a growing axon, through which the axon finds its target. Cytoskeleton also plays an important part in synapses. And cytoskeleton is also involved in transport of neurotransmitter vesicles down an axon to a synapse.

In these and other ways, then, neural circuits depend on microtubules and other cytoskeletal components for their construction and operation. And neural circuits underlying consciousness appear to be constructed and operated the same way as other neural circuits, to the best of our present knowledge.

Slide 7
Here’s where I am especially grateful to that anonymous reviewer--in my imagination, the conversation would continue with his (or her) response.

This slide also shows my reasons for suspecting that the anonymous reviewer is not a neuroscientist.

So what did the reviewer really mean?

Slide 8
He (or she) probably didn’t mean that microtubules and other cytoskeletal components don’t have ANY involvement in consciousness in the generic ways I've just described.

But the reviewer might have meant that there is no persuasive evidence that cytoskeleton has features specific to consciousness.

What kind of evidence might be available?

It again seems clear that we can’t get direct eyewitness evidence: No one can look at an image of cytoskeleton and observe features specific to consciousness.

But we might be able to obtain “circumstantial evidence”. In particular, we might discover ways that cytoskeleton would evolve or develop features specific to consciousness.

Slide 9
At this point, our friend, the reviewer might also point out that evolution and development of consciousness and of features specific to consciousness should somehow correspond.

But previous studies haven’t found evidence of a correspondence like that.

So we try a different tack, going back to earlier forms of life for hints about how consciousness evolved.

Slide 10
This slide goes way back, to a time when there were only single-celled organisms, each without a nucleus and without microtubules. The nucleus and microtubules later arose before the first eukaryotic organisms, which were initially single-celled protists.

Let’s look at some examples of protists.

Slide 11
Here we see a small sample of the tremendous variety of now-living organisms that are called protists, including the protozoans.

[Note: The image of protists is omitted pending permission. Meanwhile, see, e.g., Wikipedia for interesting images of protists.]

Much of the variation between protists results from microtubules and other cytoskeletal components. In general, protists are capable of transducing sensory stimuli, especially mechanical stimuli, into meaningful responses, such as contraction or motility.

Slide 12
This slide simply emphasizes differences between protists and neurons.

For example, as suggested by the dashed lines in this image, a protist does not have an axon, dendrites, dendritic spines, or synapses.

Slide 13
The protists led to multi-celled organisms of various kinds, with other features resulting from cytoskeletal evolution. Sponges, for example, have no neurons, but do have simple forms of communication between cells.

And then came neurons and synapses. Jellyfish do not have brains, but they have networks of neurons, so their cytoskeletons are able to construct and operate neural circuits.

Finally came brains, with neurons having cytoskeletons that construct and operate very complex neural circuits.

The evolutionary path shown here hints that cytoskeletal evolution is a key to the ability to change conscious experience.

Slide 14
So where have we gotten?--Thinking about protists and cytoskeleton has brought us to these questions.

And we’re not alone!

Slide 15
Our reviewer is looking over our shoulder and getting impatient. And maybe a little tired of being depicted like this.

Slide 16
This slide simply explains why we want to find that DNA.

Because previous work hasn’t found DNA with mutations that jointly affect cytoskeletal proteins and consciousness, we propose a new strategy.

Slide 17
Here is our strategy--

Slide 18
Our first step is based on having plausible assumptions that allow us to identify different types of DNA mutations. I propose a set of three “working assumptions” that seem promising.

Slide 19
The first working assumption is that microtubules and other cytoskeletal/synaptic components interact in concert with changes in conscious experience.

I propose to call this type of interaction "real-time plasticity" or "RTP".

Slide 20
The second working assumption ties the first assumption to neural circuits, and more specifically to neural circuits that transduce neural signals to changes in conscious experience. As shown here, we assume that real-time plasticity is what gives these transducer circuits the ability to perform N-to-C transduction.

Slide 21
Finally, our third working assumption is that a transducer circuit performs a relatively stable mapping from its neural input signals to changes in conscious experience, providing a time sequence of experienced changes.

We propose that the transducer’s information flow can therefore be modeled as if the transducer were a channel. I’ve put in a question mark because this equation for channel capacity is at best approximate and probably unrealistically simple.

Now let’s put these assumptions to work.

Slide 22
This diagram shows two different types of mutations. Our working assumptions suggest that both types occur during joint evolution of cytoskeleton and consciousness.

The first type of mutation, shown at right in the diagram, is dubbed the “Cyto-RTP” mutation. In a Cyto-RTP mutation, some DNA changes, and this affects the real-time plasticity of one or more neurons in a transducer.

The other type of mutation, shown at left, is called a “neural input” mutation, because it affects a neural input signal to a transducer circuit that has RTP.

So now we’ve completed the first step of our strategy and can check it off our list; let’s see how far we can get with the second step.

Slide 23
Our second step uses the mutation types identified in the first step: Here, we develop hypothetical accounts for evolution and/or development of experiences people have.

Slide 24
We illustrate our hypothetical accounts in terms of a transducer’s capacity and range.

In this and following slides, the green line will represent a transducer circuit’s actual capacity, which might initially be zero, but could increase as it develops increased capacity, presumably reaching a plateau when it is fully developed.

The dashed red lines will represent a transducer’s range, which includes the changes in conscious experience it has produced. The blue lines with arrows are transitions the transducer has made in response to neural input signals--when the transducer produces a change it has not produced before, its range also increases, as shown here in two places.

Slide 25
We also have ways of showing other events that affect a transducer’s capacity and range--this slide shows one way of representing the two types of mutations. Note that a Cyto-RTP mutation could change a transducer’s capacity, while a neural input mutation could change its range, but possibly not until the transducer receives further neural input signals.

And the last bullet here points out that we can also show non-mutation events in a similar way.

Let’s consider hypothetical accounts for two visual phenomena.

Slide 26
The first visual phenomenon is very familiar, red-green color blindness. These images are meant to illustrate how differently the same image might appear to a red-green color blind individual or “dichromat”, at left, and to a normally color sighted individual or “trichromat”, at right.

Researchers have proposed a cure for red-green color blindness.

Slide 27
Based in part on their experiments with monkeys and also on a color vision model, Mancuso et al. predict that human red-green color blindness could be cured by gene therapy that adds a missing photopigment or opsin to the retina, without changing the neural circuits that provide color vision.

This might sound like a great improvement to some people, but will it work?

Slide 28
This slide shows an optimistic scenario for color vision transducer circuits, but also helps us identify some things that might go wrong.

At far left in the diagram, some of our ancestors presumably had color vision transducers with low actual capacity, perhaps less than modern dichromatic primates have.

A Cyto-RTP mutation significantly increased capacity of color vision transducers in primates, enabling them to change between colors like a trichromat does.

Meanwhile, the eyes of someone who is red-green color blind provide neural input signals from which the transducer is only able to provide a range of color vision like that of dichromats.

Mancuso et al. injected genes that express an additional opsin, referred to as “L/M opsin”, into retinas of dichromatic monkeys, and the monkeys began to behave like trichromats. Mancuso et al. propose that similar gene therapy could be performed on a human subject to cure color blindness, enabling the subject to become a trichromat.

Note that this gene therapy would be very similar to a neural input mutation leading to the same L/M opsin. But note that the L/M opsin would be expressed after the transducer developed, rather than before, suggesting that the “cure” might not work, for reasons we could discuss later.

Setting aside for a moment whether insurance companies would pay for curing color blindness, we can use the optimistic scenario in this diagram to identify some reasons that gene therapy might not work. --For example, we don’t know for sure that the color vision transducer of a color blind adult has trichromatic capacity--it’s possible that dichromacy is due both to lack of L/M opsin and also to a dichromatic transducer; if this is the case, addition of L/M opsin will probably not immediately change the transducer’s capacity, and might never do so. --Another possible problem is that the gene therapy might not increase the transducer’s range, even though it might change the transducer’s neural input signals; for example, if the transducer is fully developed in a color blind adult, it might have reached a stable operating regime in which it is no longer capable of increasing its range. In contrast, if the L/M opsin were present in the individual’s retina throughout development, the individual‘s color vision transducer might have responded to neural input signals differently, by developing full trichromatic range.

Mancuso et al.’s monkey experiments are unfortunately not sufficient to rule out either of these possibilities--it is possible that monkey behavior changes unconsciously in response to neural input signals from L/M opsin, due to a blindsight-like visual pathway that changes behavior but does not produce trichromatic changes in a monkey’s conscious experience.

My sense, however, is that someone will try this type of gene therapy on a colorblind human subject sometime in the next few years; when the results are reported, we might learn something new about the transducers for human color vision.

Slide 29
The second visual phenomenon is actually a visuotactile phenomenon, but again one that is also a topic of current research.

As shown here, the word “Rose”, read visually, and its Braille equivalent, read tactilely, would have the same meaning to a person able to read English in both modes.

Slide 30
The insightful quotation here is from Prof. Gordon Legge of the Department of Psychology at the University of Minnesota, who reads both visually and in Braille, in part due to childhood damage to his visual system. The facts of Prof. Legge’s case are set forth in Cheung et al. (2009), and I am most grateful to him for his response to my questions about his experience, particularly about his feelings of understanding.

Slide 31
This slide shows another hypothetical scenario, but one similar in ways to Prof. Legge’s.

On the left, an individual has developed visual and tactile transducers with normal capacities and ranges and begins to develop a reading transducer through visual reading.

But then some sort of disease or trauma impairs the person’s vision, reducing the capacity and/or range of the vision transducer.

Faced with this loss, the person learned to read Braille, on the right. In this account, the person’s tactile transducer increased its range, causing the reading transducer’s range to increase, so the person’s reading vocabulary continued to grow.

A similar scenario might apply due to mutation.

Slide 32
And these points simply summarize the previous slide--Note the last item--vision impairment might be analogous to mutations.

Slide 33
So how much have we accomplished? We completed the first step of our strategy, and made a start on the second step, but have not yet started on the third step.

Slide 34
Here are several closing points.

The fourth item is a pointer to future work: It would be helpful to identify the mechanisms of cytoskeletal real-time plasticity and to also identify at least some of the mutations that affect it, the Cyto-RTP mutations.

If I still have a few seconds, I’ll show you a cursory brainstorm of where we might look.

Slide 35
Various cytoskeletal mechanisms might participate in RTP, separately or in combination, and here is a brainstormed list.

I feel we know much less about candidates for Cyto-RTP mutations--unlike the opsins for color vision and some other neural input mutations, we haven’t found key cytoskeletal proteins whose mutations correlate with changes in conscious experience.

I’m speculating here that the Cyto-RTP mutations might occur in DNA that affects rates of gene expression or trafficking of certain proteins in neurons.

Slide 36
So that’s the end of the conversation with the reviewer, at least for now. He (or she) seems cautiously optimistic, but still a little dubious--at least he (or she) is smiling!

Thank you for your attention.