← Brain Science with Ginger Campbell, MD: Neuroscience for Everyone
Brain Science with Ginger Campbell, MD: Neuroscience for Everyone
BS 147 Eve Marder's Life in Neuroscience- ad free version
00:43:29
Loading summary
Transcript1 lines
- [00:01]
A
This episode is a discussion of the book Lesson from the Eve Martyr's Work in Neuroscience by Charlotte Nosson. Unfortunately, I recorded most of this episode with a wrong pronunciation of Ms. Nosson's last name. We have tried to replace the mispronunciations with redubs, but I think there's going to probably be spots in here where the wrong pronouns pronunciation sneaks through. When that happens, I just want to express My apologies to Ms. Nosson. This is Brain Science, the podcast where we explore how recent discoveries in neuroscience are unraveling the mystery of how our brains make us human. I'm your host, Dr. Ginger Campbell, and this is episode 147. Today we're going to be talking about a book that I've been looking forward to for quite some time. It's called Lessons from the Eve Martyr's Work in Neuroscience by Charlotte Nosson. This book is an intellectual biography, which means Nossum takes us through Martyr's career through the lens of her ideas. Although this might seem like an unusual approach, it is quite fitting here because Martyr has made so many important contributions in her 40 plus years as a neuroscientist. Her career spans practically the entire history of neuroscience as a recognized field, and I'm sure I'm not the only one who feels that she deserves a Nobel Prize for one or more of her contributions. My goal today is to provide an overview of these contributions. If you're interested in Hearing directly from Dr. Marder, I encourage you to go back and listen to her interview. I'm changing that episode from premium back to free so that you'll be able to find it in whatever podcasting app you happen to be using. As always, you'll find complete show notes and episode transcripts@brainsciencepodcast.com and you can send me email@brainsciencepodcastmail.com Before I begin, I want to mention one thing about the author of Lessons from the Lobster. Charlotte is based in London and she has a background in neuroscience. This is essential to the approach she took in writing this biography. Her sources were Eve's notebooks and publications, interviews with Martyr's colleagues, and multiple interviews with Martyr herself. I think it's fairly unusual for a biographer to read all the work of a scientist, but this approach makes this book a rich source of information for anyone aspiring to a career in neuroscience. In the early years of Martyr's career, everyone kept handwritten lab journals. Reading these journals allowed NASSEM to follow some of Martyr's earliest intuitions forward to their fruition. Many years later. That's one reason I encourage students and working scientists to read this book, since time constraints will only allow me to hit the highlights. It's not an exaggeration to say that over her career Marder has made several paradigm shifting discoveries. These will be my focus today as I read I made a list of these key discoveries that Dr. Marder has made, so I'm going to review the list I made with you here. First, she discovered that acetylcholine is the main neurotransmitter in the neuromuscular junction in the somatogastric ganglion of the lobster. It had always been assumed to be glutamate. Second, she has established over and over again that it is dangerous to extrapolate results even between closely related species. Number three. She has realized that what may appear to be experimental error may actually be valuable clues. Fourth, she's had a key role in understanding the concept called neuromodulation. 5. She has established that the somatogastric ganglion is a central pattern generator and that there is no one size fits all model for this important phenomenon. Six Together with colleagues, she invented the dynamic clamp technique which has been used throughout neurophysiology. This is the sort of thing that Nobel Prizes are usually awarded for.7. She has established principles that show how neurons and circuits tune to target. Number eight. She has shown us the importance of variability and multiple solutions at the level of single neurons and in networks. No doubt this is an incomplete list since I've only read the book twice and I didn't have time to return to Martyr's groundbreaking papers, but hopefully what I have here is a manageable list. I'm not going to spend much time talking about the discovery that acetylcholine is the key neurotransmitter in the neuromuscular junction of the somatogastric ganglion of the lobster. Except to give you some context, this work was done while Marder was one of the first graduate students in neuroscience at the University of California in San Diego during the early 1970s. This was a time when only a few neurotransmitters had been discovered. In fact, Marder ambitiously set out to determine all the neurotransmitters in the lobster's somatogastric ganglion. Of course, we now know that there are probably hundreds of neurotransmitters, not to mention all the other so called neuromodulators. Nosson does a great job of describing the experimental challenges of those early days when electrodes were often made by hand. In fact, one reason the lobster was a popular animal was that it has fairly large neurons that could be consistently identified. Two excitatory neurotransmitters have been identified in the lobster, glutamate and acetylcholine. And because glutamate appeared to be the main neurotransmitter at the neuromuscular junction in some other crustaceans, it was assumed that this would also be true in the lobster. One challenge was that acetylcholine is very short lived. So what Marder did was show that neurons contained an enzyme necessary to produce acetylcholine. This was the sort of indirect proof that was the standard of the day. In fact, around the time she was doing this early work, another female neuroscientist, Jack Sue Caho, discovered that acetylcholine actually has three different receptors. This was the first clue that even a single neurotransmitter might have multiple effects. Marder submitted her findings to Nature, which is one of the most prestigious science journals in the world. And this was before she had even written her PhD dissertation. It was published in Nature in 1974. So I'm going to include this as one of the references in the show notes. Remarkably, in her conclusion, she was already pointing out that the implications this had with regards to assuming that results could be extrapolated between species. So in her very first publication, Marder challenged the status quo. And just as an aside, it turns out that in the lobster, both glutamate and acetylcholine act as neurotransmitters in neuromuscular junctions. It turns out the extrinsic muscles are cholinergic, which means they use acetylcholine and the intrinsic muscles use glutamate. So before we talk more about Marder's contributions to neuroscience, let's talk about the somatogastric ganglion, which is the structure that she has studied for over 40 years. A ganglion is simply a collection of neurons. In vertebrates, ganglia may be found both inside the brain, such as, for example, the basal ganglia, or they can be in the peripheral nervous system, such as near the spinal cord or other organs. However, in invertebrates, they're usually much more important, because invertebrates lack the larger brains seen in vertebrates. So the somatogastric ganglion has a function that relates to the unique way that lobsters and crabs digest their food. They essentially have teeth like structures inside their stomach so that the food is crushed and broken down after it is swallowed. The somatic gastric ganglion controls the muscles that make this happen. The somatic gastric ganglion of crustaceans like the lobster has become very popular because it has several important qualities. For one thing, it only has about 30 neurons, and these can all be readily identified. More importantly, at least from Martyr's point Of view, these 30 neurons represent a network that is small enough to study, but complex enough to reveal how networks of neurons really work. It's also important to note that all but about 6 of the neurons are motor neurons. This is what makes it possible to identify them reliably by which muscles they excite. Because I'm trying to focus on key principles, I will be neglecting several important elements of Nassem's book. Over the years, Marder has trained several generations of neuroscientists, and in the book, several key themes emerge. Marder's generosity as a mentor, her ability to see the big story buried in experimental data, and her role as a leader in her field. She unfailingly credits the creativity and insights of her students and colleagues. But I will not be naming most of these gifted scientists today. Just another reason for you to go ahead and read the book. Like all the great neuroscientists that I have interviewed over the years, Marder's career has been driven by the big questions. For example, she wanted to find all the neurotransmitters in the somatic gastric ganglion not for the sake of completeness, but because she thought it was essential to answering important questions about how neurons work. She wanted to know whether neurons could be treated as interchangeable units or whether there are factors that made each neuron unique. You should know the answer to that question by the end of this episode. To answer this question, she spent the early 80s exploring what came to be known as neuromodulation. Through a series of painstaking experiments, her lab established that different neurons could respond to the exact same stimulus in different ways. In a paper published in 1984, she wrote, we are beginning to define the mechanisms by which a small number of neurons with anatomically defined sets of synaptic connections can produce a wide variety of motor outputs. End quote. This was a big deal. Neuromodulation meant that the interaction between neurons could change without changing their physical connections, that is, synapses, and without changing the number or the strength of the synapses. Needless to say, this idea was resisted despite Martyr's evidence, especially by those who insisted that findings from invertebrates were not relevant to to vertebrates. Back in 1971, Marder had asked, what are the neurotransmitters in the somatogastric ganglion? Well, it turned out that about 30 neurons were only using two in the classic neuron to neuron synapse way. Glutamate, acetylcholine. But there were many other substances influencing the neurons responses, that is that they would prime or dampen their responses. Needless to say, the concept of neuromodulation brought an entirely new dimension to the study of neurons. Some examples of neuromodulators There are substance in the family of what's known as amines. This includes dopamine, histamine, serotonin, and one that is seen in crustaceans called octopamine. So you've probably heard of dopamine and serotonin as neurotransmitters. And it's an important point to note that many of these substances can either be a modulator or neurotransmitter, depending on where they are having their activity. Peptides are also very important neuromodulators and have been the focus of a lot of the work in Marder's lab. This work was very technically difficult, but it has been transformed over the years by the development of new tools like mass spectrography. One of Marder's students discovered over 200 novel peptides and they found over 20 in the somatic gastric ganglion. They also discovered that 20 to 30 members of a given peptide family could be found in a given crustacean species, and their effects might even be quite similar. Marder remains convinced that generating this kind of data is the key to a fuller understanding down the road. When I interviewed Dr. Marder, she talked about the fact the somatogastric ganglion is a central pattern generator and the fact that the ganglia is a network of interconnected neurons is one of its advantages as a model for study. It even has a presumed wiring diagram. Although Marder's work has led to some important changes in that wiring diagram, Marder has shown that the individual neurons in the somatic gastric ganglion are subject to what aimed to be called neuromodulation. The obvious implication is that this should also be true for the network as a whole. But at the time, the early 80s, people were focused on single neurons and even tended to focus on a single ion channel. Marder seemed to be the only one working on more than one neuron. Early on, she referred to it as a multifunctional Network martyr originated the term circuit neuromodulation, even though she did not coin the term neuromodulation. One of the features of the somatogastric ganglion that makes it attractive to neuroscience is the fact that it's called a central pattern generator, which means that it generates a rhythm on its own. In this case, the rhythm controls the muscles of the stomach. It appears to be an ideal central pattern generator because it will continue to fire even after it's removed from the lobster's body. Usually, the central pattern generating mechanisms starts in the interneurons, but in the somatic gastric ganglion, probably because it only has six interneurons, the motor neurons contribute on equal terms. This is important because it means the motor output of these neurons can be recorded, which is easier than trying to record from inner neurons. At this point, you should not be surprised when I tell you that Marder made discoveries here that again went against expectations. Expectations. The assumption was that everything could be explained by the wiring, and they were also sort of searching for the central pattern generator that would be the same for all animals with a similar motor pattern. However, by the early 80s, researchers that were working in a wide variety of animals had found that the neuronal circuits that produced similar motor patterns could have very different underlying structures, and that each circuit was different, peculiar, and unique to its species. One of the scientists working in this area was named Peter Gedding. He was also a pioneer of computational modeling, and he was working with sea slugs. And he came up with the similar idea to martyrs, which he called the polymorphic network. This was what she called the multifunctional network. Both of them came to the conclusion that the networks could be flexible, and they worked together to bring this idea to their colleagues. Marder saw this as a part of the bigger issue of neuromodulation. She showed that, based on observed phenomenon, central pattern generators were not explained solely by their wiring diagrams. And she proposed neuromodulation as the explanation. To put this into context, again, it's important to realize that this was the early 80s, and most of their contemporaries thought that the circuit properties were fixed and that all stable changes had to occur at the synapse. She published a paper in Trends in Neuroscience in 1984 examining the somatic gastric ganglion explicitly as a central pattern generator. This paper brought together neuromodulation, multifunctional circuits, and central pattern generators and brought her work to a wider audience. Unfortunately, Dr. Gedding's career was brought to a tragic end when he suffered a stroke in 1988. Dr. Marder continues to teach about his contributions in her major lectures. So here is a brief summary of Dr. Marder's work regarding central pattern generators. There are a great variety of pattern generating mechanisms, including some small canonical circuits that are found in numerous systems. But all central pattern generators are subject to multiple influences. Sensory feedback, hormonal influences, and other modulatory inputs. So there is no one size fits all description. The idiosyncrasies of each species probably reflect its unique evolutionary history. One of the things that became clear over the years was what Martyr calls a most surprising degree of flexibility at the circuit level. She thinks the reason for this is that living systems need to be both robust and flexible. So by the early 90s, Marder had led the way to new thinking about pattern generating circuits. Problem was, with all this variability came complexity and there was the need to analyze complex data. These were nonlinear circuits with nonlinear solutions. So now Marder had to take the lead in an entirely different direction. She began to work with computational models. Not only did this mean moving beyond pure neurophysiology, it meant collaborating with colleagues with entirely different skill sets. Marder showed how to do interdisciplinary before it became the buzzword it is today. This is what led to what is now known as the dynamic clamp method, which involved collaboration with Larry Abbott, a particle physicist at Brandeis. One reason that their collaboration was so amazingly fruitful was that Marder took the time to teach Abbott neuroscience prior to their work together. The early models, such as the celebrated Hopfield model, were indistinguishable from physics, which means that they virtually ignored the biology of how neurons really behave. Meanwhile, very few system neuroscientists at the time appreciated the need for for theoretical models. I don't have time to get into the fascinating story of how the models evolved over time, except to mention that the early models were focused on trying to model circuit oscillations, that is rhythms. One of the early models made some unexpected predictions that were then confirmed experimentally. Obviously, this was an important proof of their utility, but it also had far reaching implications for future experiments. The model suggested that the same network could have two different behaviors. Martyr noted that it would be common to assume that once you observed a behavior, that the behavior was always going to be at the same. On page 131, she is quoted as saying, quote, we were wrong. The danger of rushed generalizations was one really important lesson for me. The other was that laboratory experiments on preparations may only reveal part of the truth. Modeling the circuits led Abbott And Marder to the idea of hooking up real neurons to electronic circuits as a way of studying the dynamic behavior of the circuits. This is what led to the development of what's now known as the dynamic clamp technique. The basic idea of this technique is that two neurons are in separate dishes and they're connected by an electronic circuit that is used to model the behavior of other neurons. When they first submitted this for publication in 1992, they discovered that in 1990, researchers had connected isolated rabbit heart cells by means of an analog circuit, which they called an analog clamp. Even so, the dynamic clamp has become a standard tool for electrophysiologists around the world. So I want to talk a little bit about why this is such a cool idea and also to describe some of the technical advances that have made it more powerful. First, in my opinion, this basic idea is Nobel Prize worthy because it's been used to generate new knowledge by many other researchers. That's why I'm going to include this 1992 science paper in my references. But why would you want to do this to connect neurons or groups of neurons in this way? Well, first of all, it gives you a way to isolate them so that you can have one cell exposed to an entirely different environment from the other. And this could be also being exposed to different neuromodulators. Then the circuit, the electronic circuit, responds dynamically to the real time feedback from the neurons. And then the circuit itself can be manipulated to simulate a wide range of conditions, some of which might be difficult to do experimentally. Early on, the technique had two significant limitations. One was that it was slow compared to biological processes. And the other was that the effects of the currents injected by the electrode coming from the circuit were limited to a local area around the electrode. Now the speed problem has been resolved. As computers have gotten faster and more powerful, it's become possible to replace the analog circuit with computer slash digital simulations. The restrictions due to localization remain an issue. The goal was to build a computational model based on real world data. This led to many years of challenging, but fruitful research. I'm going to just give a few highlights of what was learned from this approach. Marder and everyone else had assumed that a neuron's membrane conductance was a fixed property. Conductance is a measure of how easily an electrical charge moves through the cell membrane. Those of you familiar with the basics of man made electrical circuits may also know that it is the inverse of resistance in an electrical circuit. We're talking about the flow of electrons, but in A living neuron, we're talking about specific ions such as potassium or calcium. Each ion has its own ion channel, so conductance has to be described separately for each ion. It was long assumed that a specific neuron would have a fixed conductance for each of the ions. In fact, identifying these conductances was considered essential to understanding a neuron's activity in a circuit. One of the experiments described in the book is one done by one of Marder's grad students on the so called LP neuron in the somatic gastric ganglion. The student found three outward and three inward currents. And they tried to build a model to explore which currents were important. This seemed logical because only a computer would be able to handle the large number of variables. But here's the kicker. They actually had to write this model in fortran, which many of you have probably never heard of. I think it disappeared pretty much in the 80s, but it was the original programming language used in science and involved putting your code onto punch cards, which you then gave to the operator who put them into this computer that was the size of a room. And then you got a printout. It was very tedious. Nelson's book actually includes a simplified equation for this model. She sort of jokes that every book is allowed to have one equation. They ran into some obstacles. One was that the model required data that they couldn't record. And another was, well, how do you choose which data to use? At any rate, even though the model was semi realistic, it was surprisingly hard to get to behave the way they expected. And she began to think that this is too hard, it just can't be the way cells are really working. Now remember, this is in the early days of using computational modeling to try to model biological phenomena. The people who were building the models were usually theorists who weren't collecting the data themselves. And they had to make do with whatever was in the literature. And they didn't have the background to make wise choices. For example, they didn't appreciate that there were differences between species. But eventually in Mars lab, they found a model that could work somewhat like a real neuron. And they found that in this model, in order for it to behave properly, the conductances needed to change. That was opposite from what everybody expected. The data confirmed it once they actually looked at the data. So basically what they had discovered was that for the LP's identity, it was a balance between the conductance, not a specific conductance. Individual conductances were not fixed. Why is this important? It's not just another example of martyr upending a Long standing dogma, one which she herself believed. It also shows what happens when you ask the wrong question. In this case, asking how the neuron's activity was shaped by the membrane currents was a wrong question. And if you ask the wrong question, then you can miss the clues in the data you already have. Marder has said, I have a very, very deep abiding respect for the actual data, whether it comes out of an animal or a computer model. And this is a theme we'll return to shortly. So let's consider the implications of this finding from the standpoint of neuroscience. It was the first step on what would become the far reaching examination of animal to animal variability. And it was a clue on the way to solving the question of how neurons can be both flexible and stable. Real life neurons are very robust or stable, but the early models were not. There seem to be countless variables that had to be fine tuned. And Martyr's instinct was that that just couldn't be what neurons really did. The key to creating better models seems to have hinged on the working relationship between Marder and Abbott. Marder told Nassim that the key to working with a talented theorist is that you have to be really clear about what you don't understand and you have to frame a coherent question for them to think about. At any rate, it was Abbott who finally realized that the activity of the neuron should be the controlled variable, not the activity of a particular ion channel. As I said, this is what opposite of everyone's expectation, including Martyrs. But this led them to models that not only allowed the conductances to change, but used these changes as a recording of the cell's past activity. They published this in Science in 1993, but it was years before it was really appreciated. At that point, her theory was kind of ahead of the data. Ironically, when Marder later submitted a paper to Science that had the experimental confirmation, it was initially turned down because it didn't contain anything new. So the key idea is that the neuron's own activity regulated its membrane properties. The key idea here is that the neuron's own activity regulates its membrane properties. This is the opposite from what had long been been assumed. So let's recap Marder's work up through the early 90s. Although she worked with the somatic gastric ganglion her entire career, early on, Marder suspected that the wiring diagram alone was not the whole story. She wanted to describe everything that was affecting the neurons and set out to determine which substances, neurotransmitters and other neuromodulators were involved. She also worked on describing the neuronal connections in more detail. But what made her work remarkable was that she always kept sight of the big picture, which was to ask, what are the bigger questions that we hope to answer? In this case, the bigger question is how is it that the nervous system maintains both stability and flexibility, or both robustness and plasticity? Most research had focused on what happens at the synapse. But by widening her focus, Marder was able to establish that although a given neuron's lifelong identity is determined by the kind of anion channels it expresses in its cell membrane, the number of these channels changes, and that's based on recent activity. So this is the source of their plasticity. Again, I emphasize that this is the opposite of long held assumptions. And even though it was originally predicted by a computational model, Marder backed it up with data. Marder was also interested in the principles of homeostasis as a mechanism for the stability. She brought these ideas together into a concept she called tuning to target, which basically means obtaining the output needed to to maintain function. Tuning to target is summarized on page 162 of the book. Basically, it means that the neuron's target is the activity that the circuit requires. Feedback allows the neuron to regulate its properties to maintain activity under changing circumstance. Each neuron's composition of ion channels is adjusted according to the composition of the whole population. And the rule for tuning always relies on feedback. So why do neurons need this mechanism? Well, it might help to remember that neurons generally last the lifetime of an organism, while ion channels may last only hours, days and sometimes weeks. And receptors last on the order of days to weeks. One reason I mention these time scales is to give you an appreciation for why these processes might be hard to detect experimentally. The average experiment is going to be done over a period of hours and so many of these processes just will not be easily detected. By the end of the 90s, Marder was focusing on variability and multiple solutions. As the models became more sophisticated, they predicted a three to five fold variation in conductance. And this is when they went back and looked at their older data and found that that was the exact range that they seemed to see. Nosson says Marder's paper, published in 1998 was a turning point because that's when she began to explicitly address the importance of variability, animal to animal, and moment to moment. I want to emphasize that Marder's team also did new experiments that showed this same three to five fold range of variability. It seems to be Highly reproducible because many other groups in other labs began to find this. It was like Marder said, it was there in the data if you looked. So two key ideas here. It seems like this range of three to five fold is something that's easy to maintain in the real world. And the other thing to remember is that the particular conductance measured is always a reflection of the recent activity of the neuron. So why didn't anyone notice this before everyone was busy averaging their results. Because this is the long standing practice. Martyr's work suggests that this practice should be discouraged or at least approached with caution. Now, as computers have become faster and more powerful, the models have become more and more sophisticated. And it's been now they can generate thousands of possibilities, things that can be sort of tested or not tested. There's one example on page 170 that really stood out for me. One of Martyr's colleagues created a model that generated about 2,000 neurons. And he was looking for the ones that generated one burst spikes. And that narrowed it down to 164 neurons. And then he took the averages of these conductances of these to generate a single average neuron. But this neuron actually produced three burst spikes. This means that the average had behavior that had not been seen in any of the neurons used to generate the average. The point is that averaging can result in something that doesn't respond to anything that was in the population under study. Kind of sobering. And the paper for this is a 2002 paper about failure of averaging, which you can find in the show notes. However, I think it's reasonable to say that the importance of this finding is still not widely appreciated. And even worse, there's not any way to predict when averaging will fail. Eve Marder was one of the first experimental neuroscientists to embrace the use of computational modeling. So let's take a moment to review the strengths and potential weaknesses of this approach. The strengths first, modeling allows you to deal with complex data sets and nonlinear processes. It can generate predictions that can inform future experiments. Sometimes it generates predictions that go against current assumptions, so that can stimulate paradigm shifts. You can run a large number of simulations which would be impractical to do via actual experiments. Probably the biggest problem is that these models are often generated by people that lack the knowledge of the real life phenomena being simulated. And choosing the inputs is fraught with potential errors. Common mistakes are making random choices from the literature and not appreciating interspecies variability. Ideally, there should be a close working relationship between an experimental neuroscientist and the person with the computational skills. That's why I think it would be reasonable if computational modeling was recognized as an important part of neuroscience, that is they train neuroscientists to do it. So I mentioned variability. What does variability mean when it comes to trying to figure out how neurons and networks of neurons work? For Marder, the idea of variability goes hand in hand with the idea of multiple solutions. Multiple solutions is a fancy way of saying there's more than one way to skin a cat. On page 172, Nassum reviews the paradigm shifts that Martyr has authored. I think it's worthwhile to go back through these again. 1. A specific type of neuron does not have canonical electrical properties that can be described once and for all and printed in a textbook. Instead, the conductances vary based on the neuron's recent activity number. 2 There is variation animal to animal as well as between neurons of the same type in the same animal. 3. The neuron's excitability is not the governing factor for its activity pattern. Instead, its recent activity regulates its excitability. 4. No exact balance of conductances is required to produce a particular activity. Many sets of conductances can lead to the same activity pattern. 5. The range of values of experimental data that should be treated as valid is likely to be wide because the variability between one animal and another is significant. Data that has previously been discarded as noise or error may actually be informative. In fact, Martyr tells her students not to delete the results of any experiment unless they have a clear and convincing reason. In conversation, Marder often comes back to the importance of studying networks of neurons. One of her most cited papers is a 2004 paper about the work her team did based on a three neuron network model. At that time they confirmed that the same principles of variability were seen in networks. Now this has been not surprisingly, challenged. More recently, Martyr's lab has been incorporating molecular techniques that allow them to measure the amount of messenger RNA present in neurons. The basic idea is that this is considered a reflection of gene expression. So they found a correlation between the gene expression and conductance. This confirms that the variability is not noise. Interestingly, this same three to five fold variance is again seen and as I said before, this seems to be a consistent range that people find in their data. Marder has worked to come to grips with the implication of this variability. It has changed the way experiments are performed in her lab, and it's not just about not discarding data that seems outside the expected range. They also try to do as much as possible with each preparation, they dissect. So instead of, say, doing 10 experiments in a row where you just measure one variable and then another 10 experiments where you measure something else, try to measure as many variables in each preparation. And as you might imagine, this greatly complicates experimental design. The book doesn't come up to the present because it's a book, but it does mention a few topics that are no doubt still relevant, including some ongoing challenges. One of these is the ongoing challenge to establish that these findings are relevant to vertebrates. And there's a big question that's unanswered, which is what happens in larger circuits? As I was reading this book, I found myself thinking about Michael Anderson's neural reuse hypothesis, which is based on the growing evidence that neurons in the human cortex can participate in multiple different networks. To me, this seems like a logical consequence of taking Martyr's work to a higher level. Reading Lessons from the Lobster Eve Martyr's Work in Neuroscience by Charlotte Nosson gave me a new appreciation for the immense contributions Dr. Eve Marder has made to neuroscience. I've already listed these contributions several times over the course of this discussion, so I want to close by mentioning some of the other reasons I think she is an incredible role model. First, I admire her original thinking and the way she can look beyond the data to find the story. She always gives others the credit they deserve, and she honors the scientific ideal of sharing knowledge and putting progress before competition. And she's an incredible mentor. All these priceless qualities shine through Nassim's book, which is one reason why I recommend Lessons from the Lobster to all students and working scientists. Earlier, I mentioned that I think Eve Martyr deserves a Nobel Prize. I'm sure she'd be pleased to share one with Larry Abbott for the dynamic clamp technique, but I want to emphasize that Nassim never mentions the Nobel Prize in this book. I don't want you to get the wrong impression. All the comments about the Nobel Prize are my personal opinion, based on years of observing the process from very far away. It's been a while since I did a non interview episode, so please feel free to share your feedback with me, positive or negative. And don't forget the next Brain Science Live will occur Thursday, August 2nd at 8pm Central Time on Facebook. We will be discussing Brain Science 143, which was about creativity with Elkanan Goldberg. If you want to submit a comment or a question for me to use during the broadcast, please send it to me by 8-1-2018. If we stay on schedule, this episode, which is 1:47, will be featured on December 6, 2018. This means if you're listening to this episode before December 1, 2018, you still have time to submit a question or comment for me to use on Facebook Live. My hope is that these episodes will contain more listener feedback as people learn the schedule, and I'm now keeping a list of the questions and comments for the other episodes aired this year so that hopefully the Facebook Live episodes will become more interesting as time goes on. So please do send me your feedback about this@brainsciencepodcastmail.com finally, last but not least, if you want to join me on my trip to Australia in May 2019, please email me as soon as possible. Right now. There's plenty of room, but There are only 16 slots and these are going to whoever signs up first. If you want to learn a little bit more before you write to me, the itinerary and prices are now available@brainsciencepodcast.com Australia2019. Thanks again for listening. I look forward to talking with you again very soon. Brain Science with Dr. Ginger Campbell is copyright 2018 to Virginia Campbell, MD. You can copy this show to share it with others, but for any other uses or derivatives, please contact me@brainsciencepodcastmail.com.