 Okay, thanks very much, and Dan's talk was a perfect introduction to my talk. For the last 10 years, I've been in programmatic mode in addition to my individual research. In 2003, I started EVOS, which teaches evolution across the curriculum. In 2006, I started to do community-based research from an evolutionary perspective, treating my city of Binghamton, New York as a field site for human-related research. And then in 2007, I helped to start the Evolution Institute, which formulates public policy from an evolutionary perspective on a variety of policy issues. And some of these projects, which are relevant to today's talk, is one with a collaboration with the National Evolutionary Synthesis Center, NESSINT, on evolutionary mismatch and what to do about it. So that's clearly very relevant to our topic today. We also have a project called ProSocial, which provides a framework for increasing the efficacy of groups of all sorts. And that raises an issue which, in addition to diet and exercise, another kind of mismatch has to do with our social environments. And I was very happy to see Dan's focus on exercise in addition to diet. And then also some of the talks in addition to my own are going to emphasize this theme of our social lives as something which can also be mismatched. And ProSocial is one of my favorite projects. And then finally, Evolution, this year of life, is an online magazine that approaches anything and everything from an evolutionary perspective. We've already covered some ancestral health issues. We look forward to covering more. And not only does this magazine serve as a medium of communication for the general public, but it also provides a intellectual forum at the professional level so that as we talk about these things at a professional level, we can do so in a way which is really accessible to people from all the full spectrum from novice to expert and across disciplines. I think we need this kind of forum at these levels. And so what I want to do today is I want to spend some more time building upon Dan's lecture, clarifying the concept of evolutionary mismatch. Ending up with the need to, stressing the need for both theory and empirical research. If we're not doing these, both of these, in a coordinated fashion, then we're not going to be making the progress that we need to. And then finally to describe our own effort to do this, the Evo's Lifestyle Project, which is an ambitious program, which we hope can be conducted by multiple investigators on the basis of whatever resources are at hand. And so I think that this is a new model for scientific research, which I hope to engage you all in. I mean, look at the number of people in this room and the interest in this topic. More science should be coming out the other end than is currently coming out the other end of this. And so we're trying to create a framework for doing that. Okay, so what is evolutionary mismatch building upon Dan's talk? It's a state of disequilibrium between organisms and their environment. This fascinates me. It is central to evolutionary theory. Whenever populations are not of equilibrium with their environments, then there will be mismatch. So there cannot be a more central topic to evolution. And yet there is a great need for clarification. And we especially need standards of evidence to be established. What's the evidential goal standard for documenting a case of mismatch so that we can avoid this problem of just so speculative storytelling, which is so often a criticism levied against this kind of work. And often with justification. And I think what we need to do today is similar to what John Endler and others did in the 1980s for the Catopic of Natural Selection. Of course, Natural Selection is the centerpiece of Darwin's theory. And so of course we've been studying Natural Selection all along. But when Endler published this classic book, Natural Selection in the Wild in 1986, it was a kind of a landmark for clarifying the concept. And especially what is required to demonstrate that Natural Selection has taken place in the wild. And so no matter how central the concept, there is still a need for a kind of a back to basics approach. Not only for the people from other disciplines who are encountering the concept for the first time, but even the professional evolutionists basically need to take stock, step back, and to examine this concept. And especially what are the standards of evidence for documenting. That's something counts as a case of mismatch. And so our project with Nessent is in their working group rubric. It involves four workshops over a period of two years. We have a large advisory group from which we draw our workshop participants. We're holding our third workshop this October. What this means is that there's still time to get involved in this project. And so I invite you, if you're interested in this, to contact me. And we can connect you so that you can be part of this larger advisory group. And then even perhaps some of you could attend one of the workshops, the third or fourth workshop. This will result in a special issue of probably of evolutionary applications. A basic tutorial has already been written, co-authored with Lisa Lloyd and Elliot Sober. They are both philosophers of biology. And to have philosophers involved in this, I think, is a very exciting contribution. Because the best philosophies of biology, such as these two, I mean, they're so good at rigorous thinking about problems in biology. And to have them involved, I think, is a real asset. And as I just said, back-to-basics approach is needed for professionals, in addition to those learning about mismatch for the first time. Now, this is not difficult, easy for us to wrap our heads around this. How do we define mismatch? It's a negative consequence that results from a trait that evolved in one environment being placed in another environment. In the typical example, the trait was an adaptation in the prior environment, but this is not required. So it's possible that the trait could be deleterious, neutral, a byproduct in the new environment, in the old environment, and acquires new harmful consequences in the altered environment. So when we think about the status of this trait in the ancestral environment, maybe an adaptation, maybe not. Nevertheless, either way, it has deleterious consequences in the altered environment. And so what is the evidence that's required to demonstrate that a case of evolutionary mismatch? First, a given trait, T, must be shown to be adaptive, neutral, or deleterious in relation to its ancestral environment. As with the case of natural selection, we first must demonstrate that this trait, what its status was in the ancestral environment, and then with mismatch, we have the additional requirement is that we have to demonstrate that this trait was maladaptive in a new environment. A dysfunctional will often be defined in terms of evolutionary fitness, but not always, and I think that this is important, especially in the case of health-related issues. Imagine that there's a disease that doesn't strike you until you're late in life, 67 years old. That's going to have very little effect on genetic fitness, but still something we want to cure because it has to do with human suffering. So when we think about mismatch, often we're going to be talking about genetic fitness, but we also need to be talking about other criteria for calling something dysfunctional. Now, sometimes cases of mismatch, these requirements for demonstrating that something is a mismatch are very easy. It's a kind of a no-brainer to identify a case of mismatch. One of my favorite examples involves aquatic insect species, which are often attracted to reflective surfaces. And so if you have a glass surface, for example, a solar panel, what you find is that they're a magnet for aquatic insects who not only fly to these, but even try to mate and to over-posit on these glass surfaces. You know, why would they do that? Well, it's pretty obvious that what these reflective surfaces are mimicking is water, and that what these insects have done is they've adapted to sense water through polarized surfaces, and then these man-made reflective surfaces are mimicking that. In fact, they hyper-mimic it. They're more attractive than water. And so that's what's creating this trait to do something which is so bad that it's fatal in the modern environment. So going through a little algorithm, we can say T, the trait, is an attraction to polarized surfaces. E1, the ancestral environment, is the surfaces of water that were present ancestrally and are still present. That's pretty easy. E2 is man-made polarized reflective surfaces. So this is a case of mismatch. Who would doubt it? This is not a just-so story. We have all the evidence we need to call this a case of mismatch. And we can do something about it because knowing the proximate mechanism, it turns out that by the simple remedy of taking non-reflective strips, then what the insects are typically searching for is not just bodies of water, but large bodies of water. And so if you take a large reflective surface and you divide it up into smaller surfaces with the use of non-reflective strips, then that becomes less attractive by an order of 10 to 26-fold. So mission accomplished. Not only have we identified a case of mismatch, but we've done something about it, at least for this example. So sometimes the required evidence is easy to obtain, and sometimes it's difficult. And in the cases that we're most interested in, it can be very difficult indeed. And so human health problems can be enormously complex, making it difficult even to identify the trait. I'm going to dwell on this theme a little more. Take a disease such as atherosclerosis. Is that a trait? If so, it's a very complex one, which it consists of many component traits. The salient aspects of the ancestral environment can be much more difficult to determine. It can be lost in time so much more than the reflective surface of water, which was present, both past and present. Many aspects of the ancestral environment is very difficult for us to infer. Not impossible, just hard. And then the salient aspects of the current environment can be much more difficult to determine than man-made reflective surfaces. So establishing a case of mismatch can be much more difficult. And then there are complications, which actually I did not consider until I started this project began, and we began to discuss this. This to me is one of the most interesting ones. Typically, when we think of mismatch, we think that the trait is stable. It's adaptive in E1, then the environment changes, and it becomes maladaptive in E2. Yet, we all know that developmentally, most traits evolve by virtue of a gene-by-environment interaction. There's an interaction between gene expression and in the environment in the development of most traits. And if that's the case, if the trait is a product of a gene-by-environment interaction, then when you change the environment, you might change the trait, in addition to just changing the fitness associated with that trait. And that's different. That's a different way of thinking about it. And back to atherosclerosis. We know from work that's been done that in populations subsisting on a pre-agricultural diet, the trait of atherosclerosis, it actually doesn't exist. There's nothing to measure. So when you change the environment, then you get the disease. You can measure it. It's genetically heritable, all that stuff. And yet, the trait did not exist in the ancestral environment. So these are very interesting new ways of thinking about mismatch. Again, I get to amplify what Dan said in his talk that one problem with the study of evolution, even among the professionals, is that it's very gene-centric. You say evolution, people think genes. But evolution is not about genes. Evolution is about heredity. And genes provide one mechanism of heredity. Epigenetics provides another mechanism. And so the question about epigenetics is timely. And then there's mechanisms that have to do with learning, social learning, which exists in many species, and symbolic thought, which is the basis of human cultural evolution that's nearly restricted to humans. And for every evolutionary process, no matter what the mechanism of inheritance, and no matter how fast or slow, there is an EEA. And so the concept of mismatch exists for every evolutionary process, no matter how fast or slow. And these all interact with each other. And so when we say ancestral, we should not think too narrowly about genetically ancestral. We need to be smart about thinking about evolution as something which goes beyond genetic evolution. And mismatch is something which applies to all evolutionary processes. And so all of this seems so very complex, and it is so very complex. I want to address this issue by actually quoting from our paper just a few paragraphs here. So the prospects for working out such a complex story might seem daunting. But consider the plight of a biomedical scientist trying to understand the causes of atherosclerosis without the help of evolutionary theory. They're faced with the same problem of understanding a complex interaction between a set of traits in the organism and a set of traits in the current environment, leading to pathological consequences. Is their plight more or less daunting than the plight of the evolutionist? We think that the problem faced by the evolutionist is more tractable because evolutionary theory offers an abundance of testable hypotheses that might not occur otherwise. This raises an important point about the just-so-story accusation that merits discussion in general terms before returning to the subject of mismatch. Just-so-story is just another term for untested hypothesis. The purpose of any theory is to generate hypotheses which always start out as untested. If hypotheses motivated by evolutionary theory were somehow less testable than other hypotheses, then they could be regarded as deficient, but there's no warrant for making this claim. Evolutionary theories for cundity as a generator of testable hypotheses should therefore be regarded as an asset rather than as a liability. It's not evolution that makes this stuff complex. It is complex. And it's evolution that makes it actually less complex by navigating this complex topic and generating many, many testable hypotheses. So the bottom line is that understanding cases of evolutionary mismatch impacting health, yes, it will be very complicated. Evolutionary theory is required to navigate the complexity. The best that any theory can do is to suggest alternative hypotheses. And unless there's an empirical research program for testing the hypotheses, they will remain just-so-stories. And so if we don't have good theory and a very strong and productive empirical research program, forget about it. We'll all just be speculating. And that will be all there is to it. And so if we look at current evidence for mismatch, especially in relation to diet, what we find is a lot of circumstantial evidence. And as far as we can tell, a very few direct tests of pre versus post-agricultural diets. So few that you can almost literally count them on one hand. Five is what I'm saying. So, and there's more challenges. Very few of us are gonna get an R01 grant, face it. And furthermore, even if we did the kind of tightly controlled studies have their own liabilities. Because in the first place, they're hard to implement. And when you do implement, you can't duplicate exactly the right conditions. And so they're not robust to changes. And so what we need is a new model of scientific research that can be conducted by multiple researchers in real-world settings with whatever the resources are at hand. Don't wait for the money, should be our model. And so this is the spirit in which we are trying to design our EVOS lifestyle project. I'm doing this with my graduate students, Sudhinder Gow, who's right there. So please memorize his face and talk to him and me during the workshop. So we're trying to create this framework for testing hypotheses about diet, exercise, and social organization in real-world settings. What we're trying to create is a standardized protocol that can be used by multiple investigators with background and outcome variables that could be measured at low cost. Actually, for all of the things we wanna measure, we wanna come up with low-cost measures that just about anybody can do. Plus, better measures that are high-cost for those that can do them. So why EVOS and not Paleo? Oh, I think there's a lot in a name and this is important to get through. So if by ancestral environment, if that meant the Paleolithic era, and if diet during this period was unambiguously healthy, then the term Paleo diet would be well-chosen. In other words, the ancestral diet in the ancestral environment would be good for us, the diet in the modern environment would be bad for us, and it would be just a matter of picking an ancestral Paleo diet. That's not that simple. In fact, we have all four combinations. What we eat in the ancestral and modern environment could be either good or bad. And so therefore, basically we're using evolutionary theory to assemble a good lifestyle and that might include elements of the ancestral environment or not. Therefore, we think that the EVOS diet and evolutionary studies diet is a more appropriate label than the Paleo diet. And the factors that we need to vary include not only diet and exercise, but also social organization. So, and this is all familiar to you, humans evolved in the context of small group, social interactions, and this leads to a very interesting question. If we're trying to assemble a healthy lifestyle, could we do this better as groups than as individuals? Could we use groups in order to accomplish an ancestral, a healthy diet? And the literature on this is also amazingly skimpy. Amazingly skimpy as to the question of individuals versus groups. And of course, all of these major categories require many, many comparisons within them. Many diets, many exercise regimes, many forms of social organization. So we have a huge parameter space to explore. How can we systematically explore this parameter space given the resources at hand? And so what we intend to do is a small number of planned comparisons, which can expand with the number of people doing it. And so we wanna pick initially two diets. So we'll have diet one versus diet two. Two social conditions, group versus individual. And then we wanna track as many variables as possible for correlation analysis, which we can then on the basis of the correlation analysis, then we can do new planned comparisons. And as I think you can see, this is gonna involve many, many repetitions. And that's why we need multiple investigators following a single protocol. And so the protocol needs to cover all of these points here, recruitment of participants. We need a randomized design, interventions, outcome measures. All of this decided hopefully by a consortium of researchers. So we've made quite a bit of progress developing the protocol. At this point we wanna assemble a consortium of researchers so we can agree upon the protocol and then begin to use it. For our purposes, we're gonna be recruiting participants from our university and our community. One great thing about the Binghamton Neighborhood Project is that we're in the city and in the community, in the neighborhood so that we can draw people from all walks of life, all ages. So that's a great asset. We're necessarily going to be opportunistic in our recruitment process. This again is important of just using, accepting the variation that's going to exist in real world settings. Cause if the difference is not robust enough to apply across a heterogeneous sample then this, maybe we shouldn't be so interested in it. Or if it only works for some and not others then we can use the correlation analysis to identify those interaction effects. DNA samples and genealogy is very important so we wanna get ancestry data, both self-report and a DNA sample for every subject. And so here are some of the details that we've arrived at ourselves and I wanna just kind of zip through these because the details are not so important at this point other than to say that we're close to developing our own protocol and it's a good time for us to consult with you about them. What's my time? I wanna do a time check at this point. Okay, I'm doing good. Okay, so obviously based on, we need to screen people. It needs to be above the age of 18. It needs to be exclusion criteria and so we need to basically have two conditions, two diets individual and group and so our expectations is, our preliminary expectations are that the best outcomes are going to be in a EVOS diet and our comparison diet is going to be USDA, it could be something else but currently we're choosing a USDA diet. And so here's the two diets. The USDA diet there, you can read down. There's many paleo diets. Here's the one that we've tentatively happened upon. Down here, it's important to stress that the purpose of this is not necessarily to lose weight. It's to adopt two different diets. Some people might wanna lose weight on this diet, some people might not and so that's something that we need to incorporate in this design. And to say a little bit about the psychological aspects of this, so when people have diet problems, it's that's because they have underlying psychological problems and the therapy for that is often to get them to step back and to realize that what they wanna do is not really to lose weight, is to do something else. And so this design in which the object is not necessarily to lose weight, accommodates that kind of variation which I think is important. And so individuals versus groups, for the individual treatment, the participants will interact directly with the investigators but there'll be no interactions among the participants. And for the group treatments, participants will form a club basically. So what this means is that if you were to become involved in this project, if you could assemble as few as five or 10 individuals, then those five or 10 individuals would be randomly allocated to one of these treatments, one of the diets and the individual versus the group treatment. In the group treatment, they would be introduced to each other and they would form a group and then that group would start to think about how they can work as a group, how they can cooperate, not only with respect to such things as morale boosting and so on but also that just the nuts and bolts, if we're gonna do this just logistically, what would make it easier for ourselves? And there's certain principles, design features that cause just about any group to function well as adaptive units, which is the subject of prosocial or other project. So that we can coach these groups and basically how to function well as cooperative groups at the same time and then we can apply that to their specific objective of just doing well in this experiment that they're formed the group to do. And eager to talk to you more about prosocial, a kind of a separate project which overlaps with this project in which our groups are groups that are adopting this diet. And so we wanna get background and outcome measures for a large number of categories that you see here. So we wanna have measures of all of these things that range from easy to measure that just about anybody can do to more sophisticated measures. And let me zoom in on a few of these range of measures, some low cost so that extensive resources are not required to participate. And so here's some of the biomedical measures that we have listed in our protocol so far. And enough is known so that we have expectations basically so that we can make predictions as to what the EVOS diet is likely to do. Again, I don't wanna concentrate on the details here only to say that enough is known so that we're not on a total fishing expedition there. We have some fairly concrete expectations that we can go on here. And here we go for anthropometric measurements. We have weight, BMI, waist to high ratio, bioelectrical impedance analysis, skin fold calipers. A lot of these are easy to do so that just about anybody could get these measurements. We're gonna have fitness assessment measures which are also quite easy to incorporate this protocol. And so again, we have expectations for all of these so that we can be not just in fishing expedition mode. Yes, we need to do a diet log and we need to select something which is there's lots of web-based stuff that's out there. We hope to get something that's free because we're trying to drive the cost down. Maybe we'll get some of the ones that have a fee associated. These are all decisions that we need to make. And then cognitive performance I think is very important that we wanna include measures of cognitive performance. So and again, there's a large number of options that we want to incorporate into our protocol. A lot of hypotheses that are based on circumstantial evidence that we now want to test directly. Ancestry is huge because there's evolution operates on all different time scales so we need to know ancestry. And also with a DNA sample, then we can go for specific candidate genes. We could be very sophisticated that if we build up a inventory of DNA samples. Easy to get the samples, easy to store them and then that can be the basis of analysis as we build up our capacity. Okay, so here's what I've tried to do in this talk. First, I've tried to clarify the concept of evolutionary mismatch. Second, to stress the need for both theory and empirical research. And then finally to discuss our effort to create a framework for having this balanced approach and especially a plan for an ambitious research program that can be conducted by multiple investigators on the basis of whatever resources are at hand. To repeat, there is a tremendous reservoir of interest and capacity in this audience. So many people are interested in this and yet it's not resulting in that much science coming out the other and it can if we coordinate. And so that's why I'm so happy to speak to you at this symposium on the first day and I've left quite a bit of time for questions. If you wanna come involved, please talk with Surtur with me and we can collaborate on the development of the protocol and trials can begin as early as this fall. We hope to begin trials as early as this fall. So thank you very much. It is impossible that there are no questions. Yes. I just wanna go up to the microphone. You had stated that your hypotheses were that the B would be greater than A but greater than D, greater than C. Do you have, have you formally stated those hypotheses and is there going to be prospective hypotheses set or is it all gonna be retrospective once you just have an idea of when the data is ready? Do you have specific end points for the trial and number of patients that are gonna be collected? All right, well with these trials we expect on the basis of the small literature that exists there are gonna be 12 week trials and then with follow up because the previous studies indicate that you can get an effect in 12 weeks. And we expect the pre-agricultural diet to exceed the USDA diet and we expect dieting and groups to be more successful than dieting as individual. We also have hypotheses about the specific improvements but we also expect those to change because we have this sort of intermediate level of knowledge. We don't know nothing and yet at the same time there's all sorts of surprises that we're going to have and so we feel that that's the best that we can do without being too specific. I guess what I was getting at is do you have the trial powered for statistical validity so that once the data is collected it can be used and maybe used as a registry for others if they want to conduct their own trials in the future, create a performance goal from the data that you've then collected? Yeah, right and so that's where we have the idea of kind of an accumulating the sample size so as these come in, we'll have the first group and it'll be randomly assigned so that there will be a randomization plan and as these groups accumulate then the sample size will build up. At some point we'll get firm conclusions and at that point we can go on to another set of comparisons, a more focused set of comparisons. Just as for instance, as Sudha and I were talking there's a question as to whether the difference is among grains so maybe it turns out that rice is not bad for you even though other grains are. In that case you could assemble at some point you can make that be the plan comparison but all of this requires just enough people doing it so that we can generate these sample sizes in order to make these comparisons and that can be done in an accumulative fashion. Great, thank you. The paleo diet is, I understand it the list of forbidden foods is based upon hunter-gatherers all over the world, all right? And they all seem to have these things in common. So I wonder if doing all this DNA analysis is kind of a expensive distraction from what you're trying to accomplish. Well, I would disagree and in the first place obtaining the DNA is not expensive at all. All you'd have to do is swish your mouth with scope mouthwash and you've got your sample and it can be in a repository and then we have that data. I think that ancestry is going to be hugely important. There's quite a lot of evidence for example that the reason that people of African heritage have more health problems than people of European heritage is because simply they have not had that history of genetic adaptation to agricultural diets as much. So I think that one it's extremely important to include ancestry and there's two ways to do this. One is just by self-report where did your ancestors come from? People know to a pretty good degree. And then the final degree of resolution is the DNA. And then finally for a real fine grain analysis this is gonna boil down to specific genes. And so it's so easy to get the DNA at the beginning and destroy it but why wouldn't we do that? And then as we build up capacity then we could actually use this very important form of information. What are your parameters for deciding what a group is? Are you going to use electronic groups or physical groups? Oh that's a great example, a great question. And I think that our preference would be for physical groups. Those groups could be assisted as that would be just fine. And then in cases where electronic groups is the only option then we should probably try that too and I think that we need to incorporate this degree of flexibility in our design. And then of course we need to track that so that we can do correlation analysis to see how well they succeed. But I think the preference would be in the first place for the groups to be as real groups as possible. But in the second place to give the groups a lot of discretion about how they organize themselves because it turns out one of the major design principles for groups functioning well is self-determination. And so you don't want to do top down so this is how your group must be. This is a group which needs to make its own decisions and to create its own rules. Therefore if we need to measure that and track that and that becomes kind of a cultural evolutionary process in its own right, we can see which groups worked well better than others and then we can take those what they did and we could import that so that we could actually manage the cultural evolutionary process. Okay, I predict you're gonna have some problems with retention in your standard American diet group unless you have some sort of weightless control. So at the end of 12 weeks that group can then participate in the 12 week dietary intervention. I think without, plus you'd also have the strength of a crossover. So I'd suggest that you consider a weightless control. Thank you very much and so that would either be, basically there's a waiting period before you go on the diet or there'd be a switch so that once you're on one diet you can go on the other diet. Exactly, yes that would be. A crossover I think would yield a wealth of information. Thank you and so please contact us so that you can get involved in our process. I am gonna come out and help you guys, yes. I have a somewhat similar question that the previous person at this microphone had regarding the physical versus electronic groups and as I look out across I have some of my mentors out here. I have some of my colleagues. They were all my online groups when I began. So I think most of the people in here and I'm speaking just for myself really thrived on the online presence of this society. So I would just second your statement for having flexibility because I think we're all somewhat product of online groupings and in this fast-paced age. It'd be kind of a practical for me to meet with 10 other people. Yeah, we're vastly expanding. The joy was last year's meeting a lot of these people that I met online like Aaron and so forth. So I just wanna throw that out there for you. Yeah, thank you very much. I mean, we could be much more systematic than we are and yet at the same time there will be inherent variability and so we simply have to manage that trade off. Yes, Mike, time is up. Oh, did I go to the right? One more, please. My question's about metrics. Would there be any way to look at tooth decay in the two different groups? I was involved with a Nessent meeting this spring on evolutionary dentistry. Right. And we looked backwards, but you could actually be looking forward. I don't know if you're running the tests that long. Oh, very interesting. So tooth decay would be something that I had not thought of but also something which might not take place in a 12-week period but certainly among the variables could well be included. Thanks. Yeah, okay. Yes, thank you very much. Thank you.