Evolutionary Psychology

It has been 27 years since the inception of the field of evolutionary psychology marked by the publication of Jerome Barkow, Leda Cosmides, and John Tooby’s The Adapted Mind. Although some might argue that the foundations of evolutionary psychology go much deeper with the publication of EO Wilson’s Sociobiology: The New Synthesis in 1975 (and yet others like Steven Pinker will go much, much deeper to George Williams’ Adaptation and Natural Selection in 1966), really the theoretical foundations of evolutionary psychology in its modern incarnation were laid out in The Adapted Mind. Since then, evolutionary psychology (or EP, for short) has been through its own series of trials and tribulations as it has fought to advance itself as a legitimate field of inquiry.

Unfortunately for evolutionary psychologists, the climb to establish itself has not been an easy one, and even still today many question whether EP has done well to not only fulfill its mission statement by staying true to its founding theoretical framework but also whether it’s accurately employing evolutionary principles in the same way that biologists do.

As it stands, I have some opinions about evolutionary psychology myself, albeit not very strong. As an evolutionary anthropologist, I have been working in the past year or so on at least two almost exclusively EP projects myself, all the while trying to pin down where it is that I stand on evolutionary psychology’s foundations. In February I appeared on Iona Italia’s Two For Tea Podcast with Will Buckner to discuss some of these issues and feel I didn’t bring much to the table when it came to EP aside from my suggestion that evolutionary psychology has gotten better with its development and the general agreement between all of us that at least some of the hypotheses traditionally posited by EP tend to be myopic and fail to properly take into account the environment of evolutionary adaptedness (the EEA, for short).

My blogpost today is in response to a recent pre-print on the evolution of left-handedness and is partially here as a service to myself to help me walk through and lay out some of my own general thoughts about EP and the controversies that the field has gone through.

What is EP?

The mission of evolutionary psychology seems straightforward: to apply the principles of evolutionary biology to our own human mental quirks. In a sense, ethologists, behavioral ecologists, and primatologists have been doing evolutionary psychology for a long time. Before evolutionary psychology, much of this work was conducted under the guises of the mission of sociobiology, a similar but separate movement which began with EO Wilson’s treatise on the subject in 1975.¹

Although many evolutionary psychologists see EP as the modern successor of sociobiology, the mission statements of these fields are somewhat different. Wilson’s definition of sociobiology sensu stricto was, “the systematic study of the biological basis of all social behavior.” While he included in this the study of the early social features and social organization of humankind (specifically taking into account genetics), the development of evolutionary psychology has diverged from this exclusive dedication to sociality. Modern evolutionary psychology focuses on the adaptive bases of almost all behaviors such as fear, consumption leading to obesity, and (almost ubiquitously) our sexuality.

So while one can study behavior using an evolutionary approach, modern EP is mostly based on the principles laid out in The Adapted Mind in 1992 and since carried forward by the husband-wife team of anthropologist John Tooby and psychologist Leda Cosmides based out of UC Santa Barbara.² This approach was laid out with very clear guiding principles that are still in use today:

The idea of modularism, known as the massive modularity hypothesis in EP, was developed by philosopher of mind Jerry Fodor who believed that certain parts of the mind were separate and non-interacting with each other and our broad-domain cognitive processes.

1. The brain is a physical system which functions as a computer with circuits designed to generate behavior appropriate to environmental circumstances.
2. These circuits were designed by natural selection to solve problems faced by humans in our evolutionary history.
3. Most problems that are easy to solve are actually hard to solve and require complex neural circuitry, a fact obfuscated by our consciousness.
4. Different neural circuits are specialized to solve different problems
5. Our modern skulls house a stone age mind³

Put simply, these principles laid out by Tooby & Cosmides dictate that the brain is composed of a system of modules, or areas, networks, and circuits of the brain dedicated to solving specific problems, and that these modules arose during our evolution in the Pleistocene in our environment of evolutionary adaptedness (EEA) and are still with us today. To many readers, it might appear that evolutionary psychology a la Tooby & Cosmides is making two separate unrelated claims: the first being modularity of mind and the second being adaptation in a paleolithic environment as the key for understanding how the mind works. Nonetheless, the importance of modularity for EP has been recapitulated by evolutionary psychology’s other founders including David Buss whose 1995 article, Evolutionary Psychology: A New Paradigm for Psychological Science, also served to highlight what he called domain-specific (rather than domain-general or general processing) processes as fundamental to evolutionary psychology.⁴

Theory: Massive Modularity – The Seven Sins of Evolutionary Psychology

The purpose of modules in EP can best be described with the jukebox analogy. The discs within the jukebox, containing tracks, are saved internally and later accessed (or evoked) when demanded in the modern environment.

Personally, I have struggled to understand why much of evolutionary psychology hinges on modularity as an essential feature of the field’s theoretical framework. In fact, in talking to many of my junior friends in the field of EP (i.e. those who were not doing EP in the early 90s), it seems to me that modularity seems to be a focus of fewer and fewer evolutionary psychologists and is really just a background justification for much of their work that some of them choose to pay lip service to. Although I’m not by any means sold by the massive modularity hypothesis (for example, shouldn’t it be easy for them to find the module responsible for men’s preference for big-breasted women with fMRI?), in terms of the health of the field, I think this drive away from mechanisms to hypothesis building is rather worrying. Although one can easily build testable and replicable evolutionary hypotheses for human behavior the same way we do in animal behavior, the justification that these arose at some point in the EEA is entirely dependent on the brain’s ability to save these behaviors tens of thousands of years ago and later retrieve them (this is more broadly recognized as evoked psychology).

Much of the contention between evolutionary psychology and the rest of the behavioral sciences has focused on an inability to locate modules for most of EP’s hypotheses (aside from the speech-processing areas of the brain such as Broca and Wernicke’s areas). Problematically, alternatives to most EP hypotheses, can be approached from the simple framework of game theory as is the case with animal behavior: most individuals are going to act in a way that is most advantageous to themselves, selfishness is mediated by a theory of self, and determining whether something is advantageous or not is really mediated by non-psychological, biological mechanisms.

One of the best critiques in this vein came from father-son team Jaak and Jules Panskepp in their article The Seven Sins of Evolutionary Psychology, published in 2000.⁵ Jaak, who had studied many hypotheses similar to EP hypotheses in rodents (for example, examining rats’ aversion to cat hair but not dog hair), argued extensively against the modularity hypothesis and EP’s dedication to hardwired, rather than plastic, explanations for behavior. His critique, which was that no one in EP had worked towards finding any of the modules they claimed existed, sought to highlight the role of general purpose systems and non-modular neurochemical explanations for behavior. For example, in examining the role of intrasexual competition Panskepp highlighted the counteracting roles of hormones such as vasopressin and oxytocin for regulating our behaviors in more general ways which serve to make one more or less competitive with and against other males. Although this seems like a rather baseline hypothesis, this is in strict contrast to David Buss’ explanation that a male sexual jealousy comes only when evoked by emotional infidelity, triggering a throwback to our last-Pleistocene ancestral environment.⁶

A canalized trait produces few deviations in and through development. Natural selection is constraining the trait and thus it’s more likely to be phylogenetically conserved – this is sometimes a standard assumption in EP. This is in contrast to an emphasis on plasticity. Variation in plastic traits can later canalize, especially during times of environmental stability when deviation is undesirable. It’s easier to see how the interplay between emotions and motivations via hormones can become canalized more easily than hypothesized modular brain pathways.

Rather than examining myopic evolutionary hypotheses, Panskepp advocated for an approach that took into account the intersection between motivations and their emotional systems. He argued that such a framework, while not as dependent on processual hypotheses as modular EP, could nonetheless constitute an evolutionary approach to behavior. I personally think that in defense of EP, one can argue that Panskepp’s argument is a conflation between the proximate (what you see in a trait) and ultimate (what the trait is really for) explanations for behavior, but I also think his argument is partially strengthened by the fact that both EP proponents and neuroscientists have few mechanistic explanations for how it is that a module is genetically assimilated (turned from a simple plastic trait into an inherited one) and further canalized (turned from simply inherited to always expressed in an organism during development). By focusing on the evolutionary pressures behind “epigenetically-motivated” and emotionally-driven phenomena, Panskepp believed we might come closer to a more accurate version of what parts of our psychology truly are evolved.

Practice: Adaptationism – Evolutionary Psychology, An Exchange

Most fundamental to evolutionary psychology and one of its primary critiques is the role that adaptationism plays in the formation of its hypotheses. By adaptationism, I am referring to a certain branch of anti-Darwinian evolutionary thinking which assumes that natural selection is the cause of all variation and sole explanation for all biological traits and structures in the world. The debate over adaptationism began in 1979 when Harvard paleontologist Stephen Jay Gould and geneticist Richard Lewontin published a seminal paper titled The spandrels of San Marco and the Panglossian paradigm.⁷

2. The most common and cliched criticism of Ev psych is that it tells "just-so stories," compelling sounding explanations that are just pure speculation. The criticism goes back to the eloquent but utterly unfair critic, Stephen Gould.

— Bo Winegard (@EPoe187) February 14, 2019

Within sociobiology and EP circles, this paper is easily the most controversial paper of all time, and to this day innumerable casual followers of the EP debate on the internet will tell you that this paper was written to discount natural selection as a process, rather than what it actually stood for. Simply put, the paper noted a tendency within all of evolutionary biology (including paleontology) to replace one hypothesis for the evolution of a trait invoking natural selection with another after the first had already failed. That is to say, rather than natural selection being invoked as an explanation for the evolution of a trait, it had somehow become an a priori requirement. Gould and Lewontin coined this tendency, “the adaptationist programme.” Thereafter, this critique would continuously (and sometimes unfairly) be employed against both sociobiology and evolutionary psychology.

Gould and Lewontin’s group at Harvard formed something of an anti-sociobiology clique from the 1970s onward, and as evolutionary psychology developed out of sociobiology, similar critiques by the same group were levied against EP. Most vocal of these critics was Gould, who as a paleontologist had revolutionized our view of evolution from the macro-evolutionary perspective. Gould’s main contention with EP was outlined in an exchange between him and cognitive psychologist/EP proponent Steven Pinker that took place in The New York Review of Books in 1997. Originally publishing his critique alone in an article titled “Evolution: The Pleasures of Pluralism,” Gould approached evolutionary psychology from a skeptic’s position on the basis that it was exclusively adaptationist.⁸

Humans are animals and the mind evolved; therefore, all curious people must support the quest for an evolutionary psychology. But the movement that has commandeered this name adopts a fatally restrictive view of the meaning and range of evolutionary explanation. “Evolutionary psychology” has, in short, fallen into the same ultra-Darwinian trap that ensnared Daniel Dennett and his confrères–for disciples of this new art confine evolutionary accounts to the workings of natural selection and consequent adaptation for personal reproductive success.

While Gould makes some now-disproven arguments such as the idea that we can never know what the EEA or the sociality of fossil hominis was like (we can, based on stable isotope evidence- mind you I’ve yet to read any EP papers referencing this), his ultimate critique was that EP’s a priori dedication to adaptationist explanations was ultimately misguided and would result in both poor reasoning and an abundance of Type I errors.⁹

The hole in the center of this shell is known as the umbilicus. Although it initially evolved as a structural by-product of the shell’s spiral shape, it was later co-opted in several species to help brood eggs. The hole is a spandrel, but its co-option is known as a functional shift.

Five months later, The New York Review of Books published a response to the article from Steven Pinker as well as Gould’s response to Pinker in another article titled “Evolutionary Psychology: An Exchange.”¹⁰ In it, Pinker made two primary arguments: first, that evolutionary psychologists are not adaptationists (according to him, they were, in fact, pluralists and accounted for all evolutionary processes, not just natural selection); and second, that they do not employ adaptationist explanations indiscriminately. Pinker states that while Gould had noted that many human mental traits, such as reading, had likely evolved as spandrels (whereby a trait is not the target of selection but is instead a byproduct of the evolution of a separate trait- for example, the presence of male nipples which serve no purpose but are nonetheless adaptive in females), the fact that natural selection had co-opted the trait nonetheless proves that natural selection was the process at play. As an example of this, Pinker highlights an example from Gould’s own field of malacology (the study of molluscs) whereby the center hole of a shell is sometimes later co-opted via natural selection for a separate purpose.

The problem for Pinker is that that evolutionary psychology, in principle, is inherently adaptationist. Laying out the principles of EP in 1997, Tooby & Cosmides admitted as much, stating:

The evolutionary process has two components: chance and natural selection. Natural selection is the only component of the evolutionary process that can introduce complex functional organization in to a species’ phenotype….In evolved systems, form follows function. The physical structure is there because it embodies a set of programs; the programs are there because they solved a particular problem in the past. This functional level of explanation is essential for understanding how natural selection designs organisms.³

Because of a strict rejection of domain-general processing and a dedication to modularity, the foundations of EP are based on the idea that human behavioral modules were akin to expensive organs and therefore are subject to the same levels of selection that any other biological systems are. In reality, almost any level of domain-general or cumulative processing would lead to the rejection of such a notion, writ large. In response to Pinker’s points regarding spandrels as the target of selection, Gould pointed out the adaptationist thinking inherent in Pinker’s argument. For Pinker, while EP is not adaptationist, adaptation is the primary mechanism at play in evolution. Gould asks of Pinker:

Which claim does Pinker want to make: that pluralism reigns in evolutionary psychology (and I characterized the field unfairly), or that adaptationism reigns as a synonym for “evolutionary reasoning” (and my warnings are sterile)? He can’t have them both.

While the specifics of this dispute can be semantically debated for days on end, the point Gould was making is that the current co-option of a trait is not representative of its evolutionary origins. While EP inherently seeks to identify the evolutionary history of certain mental traits we have, many may be spandrels which evolved alongside other traits. For example, as Mark Changizi argued in his book Harnessed, the evolution of reading, despite being adaptive in our current environment, likely has its origins in separate pattern recognition processes which evolved in our species.¹¹ That is, assuming modularism is true, there is no module for reading as it’s an emergent process culturally piggybacking separate software.

Behavior and Adaptation: Dual Inheritance Theory

Despite Gould’s objection on the basis of adaptation reigning supreme and in defense of EP’s approach, many of our behaviors are adaptive in some way (indeed, our entire life must be technically adaptive). The problem is where these adaptations come from and whether the source is the domain of evolutionary psychology or not. As I noted earlier, from the perspective of game theory, the modules in EP are not necessary for explaining adaptive behavior. As individuals, one can assume humans will act in their best interests. As members of a group, one can still assume humans will also act in their best interest. As I wrote for Areo Magazine last year, most of human cooperation lies in selfish but emergent properties coming from our own demands.

As a method for explaining human behavior as both adaptive and selfish but learned, anthropologist Robert Boyd and Peter Richerson (based on earlier work by Luca Cavalli-Sforza and Marcus Feldman) developed dual-inheritance theory (DIT, for short).^{12, 13} DIT, in essence, circumvents the dual fallacy that for a behavior to be adaptive it must be inherited and the corrolary that all inherited behaviors are expensive and therefore individually adaptive. Under DIT, learned behaviors operating under the principles of group selection and group coercion lead to the evolution of behaviorally adaptive traits, massive modularity aside.

Under DIT, learning and plasticity are emphasized for being expensive (the reason why cheaply produced organisms such as ants have complex behaviors yet rather small brains), rather than evoked psychology as a result of expensive individual modules. If anything, DIT is a compromise for the strictness that was inherent in EP and the idea that behavior is not adaptive. It requires neither modules nor genes for the spread of adaptive, beneficial behaviors in a population and opens the door even for maladaptive behavior to be explained.¹⁴ Under the EP paradigm, maladaptive behaviors are often seen as “harmful disfunctions,” whereby good behaviors taken out of their ancestral environments are suddenly maladaptive (this may still be true, as is the case with sugar tooths, but this is more necessitated by neurotransmitters than by modules).¹⁵

The advent of DIT has been well-received since its introduction in the 1980s, and in more recent years it has been applied (under the broader umbrella of cultural evolution) in the study of history, cross-cultural psychology, and memetics. Where DIT will end up going is yet to be seen, but thus far its applicability to biology and other non-psychological fields speaks well for its future.

Modern EP Myopicism

So aside from what other people have saide, what do I think about evolutionary psychology? As a strict theoretical approach, I’m conflicted. As an approach to the idea that humans will act in their best interest, that their minds (despite not having modules) are granularly shaped in some ways by evolutionary pressures, and that the full breadth of evolutionary explanations can in some way be used to explain behavioral universals, I am completely with evolutionary psychology. But unfortunately, this is not what modern EP is. Tooby & Cosmides in 1997 separated such an approach from their paradigm, stating:

It also separates evolutionary psychology from those approaches to human behavioral evolution in which it is assumed (usually implicitly) that “fitness-maximization” is a mentally (though not consciously) represented goal, and that the mind is composed of domain general mechanisms that can “figure out” what counts as fitness-maximizing behavior in any environment — even evolutionarily novel ones. Most EPs acknowledge the multipurpose flexibility of human thought and action, but believe this is caused by a cognitive achitecture that contains a large number of evolved “expert systems”.³

These approaches, despite being evolutionary are not evolutionary psychology sensu stricto, and I think it’s important for readers to recognize this. One can believe in evolutionary explanations for human psychology without being an evolutionary psychologist. My personal opinion is despite the critiques from biologists including Gould and Panskepp that EP has ironically failed to adapt themselves. Instead of countering adaptationism in the field, adaptationism has been doubled down upon. Instead of emphasizing plasticity, more myopic hypotheses have been wholeheartedly embraced.

Poor evolutionary psychology comes in the form of the work of Gordon Gallup from the 1990s, who believed that anti-depressants were present in sperm because of a correlation between self-reported unsafe sex and happiness in a sample of college students.¹⁶ It comes in the form of the fundamental EP papers that are failing to replicate in the 2010s. And finally, it comes in the form of novel and interesting, yet misguided papers that are being published today.

Published as a pre-print in bioRxiv

Just last month a paper was published purporting to find an adaptive link between fighting ability and left-handedness in humans.¹⁷ Looking at 13,800 professional boxers and MMA fighters, the authors found that left-handedness was associated with winning a fight at a rate higher than left-handedness is associated with people in the general population (that is, 17% of winners were left-handed, while 13% of the general population were left-handed). Using these results, the authors then concluded that their findings were, “consistent with the fighting hypothesis [for the evolution of left-handedness].”

So what’s wrong with this study? First is the assumption that the heritability of handedness is high or maintained by balancing selection (a genetic process)- heritability for handedness from twin-studies is only roughly 25% (the most recent figure from a population of 300,000 individuals found a heritability of only 2-3%.)¹⁸ Second is the assumption that highly technical boxing relates in any way to general fighting ability in the population. In what is a highly technical sport, training is even more important than what we find in the general population, so the influence of right-handed people training with right-handed people is even higher than in the general population. Third is the assumption that canalization for handedness would high enough to carry left-handedness in human populations through the Neolithic by the invention of weapons. By the time the Neolithic rolls around we’re using spears, swords, and axes on one another (not to mention the prevalence of boxing-esque fist fights in hunter-gatherer societies sounds bizarre), and with the population explosion associated with the advent of agriculture, cleansing of male phenotypes from massive warfare resulting in a significant male bottleneck 7,000 years ago, and highest incidences of violence in human history, actual selection would have to be accounted with the Neolithic²⁰, not in a caveman’s boxing ring. For such a combat-related and supposedly sexually-dimorphic phenotype, this would be the EEA If these results replicate for, say, amateur sabre fencing, I might be less skeptical. Finally, the paper (I believe fatally) fails to take into account the simple fact that chimpanzee males display similar measures of handedness to humans.¹⁹

A bonobo fishing for termites. Among chimpanzees, the prevalence of right and left-handedness varies between individuals is similar to humans¹⁹; some modern EP hypotheses fail to take into account our evolutionary similarities (and differences) we share with living primates.

While not a complete condemnation of evo-psych as a whole, one can scroll through the tweets associated with this paper and find the high praise it received despite unfortunately falling for the most elementary “correlation is not causation” trick in the book (which is a shame considering how amazing the data they collected is). Nor have I simply selected this paper to simply bully its authors- what we see here isn’t atypical for the most sensationalist and popularized articles in evo-psych, in some circles this is nearing the norm.

My own perspective on evo-psych is that it has a lot of work to do, but many (not all) authors are denying there is any work to be done. From a constructive critique the best practices one should mainting in approach evolutionary psychology hypotheses might involve:

1. Emphasizing strict, cross-cultural and developmentally or biologically-constrained patterns over broader ones
2. Moving to identify genetic mechanisms associated with these patterns
3. Identifying modules associated with these mechanisms (in this case, coupling with the ongoing connectome project could be fruitful) OR moving away from modules entirely and moving to a broader neurophysiological or hormonal-affect state models
4. Most importantly, emphasizing the room that spandrels might play in our current behaviors or how developmental scaffolds may build upon more baseline-level EP module-esque systems to build higher-order emergent properties (I think Tooby & Cosmides “computational theory of mind” partially covers this, but even so this seems to be less emphasized in EP than its other aspects)

When it comes to evo-psych, I am conflicted. I feel drawn to many of the hypotheses, but not the mechanisms. Sometimes I wonder if the emphasis on modules wasn’t an over-anticipation of what functional-MRI studies were about to bring to neuroscience. In places where I don’t agree with the hypotheses, I am often frustrated by the correlative (but not constrained or causative) nature of the data and am left wondering how to approach the field without being treated like an anti-reductionist skeptic. Personally, I don’t think any “death blow” has been dealt to EP, nor is there one coming, but in order for the field to survive I think it’s time we examine some of the baseline assumptions the field has made along the way.

Like my blog and want to support my writing? Please consider supporting me on Patreon.

References

¹Wilson, E.O., 1975. Sociobiology: the new synthesis (Belknap, Cambridge, MA).

²Barkow, J.H., Cosmides, L. and Tooby, J. eds., 1995. The adapted mind: Evolutionary psychology and the generation of culture. Oxford University Press, USA.

³Cosmides, L. and Tooby, J., 1997. Evolutionary psychology: A primer.

⁴Buss, D.M., 1995. Evolutionary psychology: A new paradigm for psychological science. Psychological inquiry, 6(1), pp.1-30.

⁵Panksepp, J. and Panksepp, J.B., 2000. The seven sins of evolutionary psychology. Evolution and cognition, 6(2), pp.108-131.

⁶Buss, D.M., Larsen, R.J., Westen, D. and Semmelroth, J., 1992. Sex differences in jealousy: Evolution, physiology, and psychology. Psychological science, 3(4), pp.251-256.

⁷Gould, S.J. and Lewontin, R.C., 1979. The spandrels of San Marco and the Panglossian paradigm: a critique of the adaptationist programme. Proceedings of the Royal Society of London. Series B. Biological Sciences, 205(1161), pp.581-598.

⁸Gould, S.J., 1997. Evolution: The pleasures of pluralism.

⁹Copeland, S.R., Sponheimer, M., de Ruiter, D.J., Lee-Thorp, J.A., Codron, D., le Roux, P.J., Grimes, V. and Richards, M.P., 2011. Strontium isotope evidence for landscape use by early hominins. Nature, 474(7349), p.76.

¹⁰Pinker, S., 1997. Evolutionary psychology: An exchange.

¹¹Changizi, M., 2011. Harnessed: How language and music mimicked nature and transformed ape to man. BenBella Books, Inc..

¹²Cavalli-Sforza, L.L. and Feldman, M.W., 1981. Cultural transmission and evolution: A quantitative approach (No. 16). Princeton University Press.

¹³Richerson, P.J. and Boyd, R., 2008. Not by genes alone: How culture transformed human evolution. University of Chicago press.

¹⁴Henrich, J., 2004. Demography and cultural evolution: how adaptive cultural processes can produce maladaptive losses—the Tasmanian case. American Antiquity, 69(2), pp.197-214.

¹⁵Wakefield, J.C., 1999. Evolutionary versus prototype analyses of the concept of disorder. Journal of abnormal psychology, 108(3), p.374.

¹⁶Gordon Jr, G., Burch, R.L. and Platek, S.M., 2002. Does semen have antidepressant properties?. Archives of Sexual Behavior, 31(3), pp.289-293.

¹⁷Richardson, T. and Gilman, R.T., 2019. Left-handedness is associated with greater fighting success in humans. bioRxiv, p.555912.

¹⁸Francks, C., 2019. The molecular genetics of hand preference revisited. Scientific reports, 9(1), pp.5986-5986.

¹⁹Hopkins, W.D., Reamer, L., Mareno, M.C. and Schapiro, S.J., 2015. Genetic basis in motor skill and hand preference for tool use in chimpanzees (Pan troglodytes). Proceedings of the Royal Society B: Biological Sciences, 282(1800), p.20141223.

²⁰Zeng, Tian Chen, Alan J. Aw, and Marcus W. Feldman. “Cultural hitchhiking and competition between patrilineal kin groups explain the post-Neolithic Y-chromosome bottleneck.” Nature communications 9, no. 1 (2018): 2077.