define runaway hypothesis

• Skip to primary navigation
• Skip to main content
• Skip to footer
• Image & Use Policy
• Translations

UC MUSEUM OF PALEONTOLOGY

Understanding Evolution

Your one-stop source for information on evolution

• ES en Español

When sexual selection runs away

The concept of runaway sexual selection illustrates one of the ways that sexual selection is hypothesized to work. Runaway sexual selection can help explain how traits that are appealing to members of the opposite sex, but pose a risk in terms of survival, can evolve – and how that preference could evolve in the first place.

The quandary of female choice

It makes sense for a female to choose a mate based on traits that help him survive. For example, a female bird would do well to choose a strong-looking, disease-free mate. That male likely carries “good” genes that allow him to resist disease and get sufficient food — and he will pass those genes on to his offspring.

However, there are many examples of females choosing mates based on less useful traits (e.g., song complexity) or even traits detrimental to survival (e.g., brightly colored plumage, as in the case of the peacock). These cases present evolutionary biologists with a bit of a puzzle. How did these preferences arise in the first place? If a female chooses a male with bright feathers, her sons will have bright feathers, which are likely to attract predators. A gene for choosing brightly colored males would seem to be disadvantageous. How do such genes spread through a population?

There are several possible answers to explain how these seemingly disadvantageous genes spread through the population, among them:

• Runaway selection Imagine a bird population in which females choose mates at random. Males with slightly longer tails fly a little more adeptly, avoid predation, and so, survive better than males with slightly shorter tails. In this situation, a gene for female choosiness (longer tail = sexier) will be favored, since — by choosing a long-tailed male — she will have sons with longer tails. This trait will spread through the population until most males have long tails and most females prefer long-tailed mates. So far so good.

However, once this has happened, the process may run out of control, until the male trait becomes so exaggerated that it is disadvantageous. In other words, female preference, instead of survival advantage, may begin to drive the evolution of ever-longer tails, until males are encumbered by showy plumage that no longer helps them avoid predation.

• Good genes Imagine another bird population in which females choose mates at random. Some males in the population have better genes for survival than others, but it is difficult to tell whether a male has good genes or not. In this scenario, long tails make it more difficult to survive — they are costly to produce and maintain. Because they are so costly, only males with good genes have the extra resources to produce them. In this situation, a long tail is an indicator of good genes. A gene for female choosiness (longer tail = sexier) will be favored, since — by choosing a long-tailed/good gene male — she will have sons with good genes. This trait will spread through the population until most females choose long-tailed mates and males that are able to produce long tails are favored.

If females choose males with “long and costly” tails, they are guaranteed to get good genes! If they choose males with “short and cheap” tails, they may get good or bad genes.

• Evo Examples

Learn more about sexual selection in context:

• Survival of the sneakiest , a comic strip with discussion questions.
• Evolution's dating and mating game , a news brief with discussion questions.
• Quick evolution leads to quiet crickets , a news brief with discussion questions.

Subscribe to our newsletter

• Teaching resource database
• Correcting misconceptions
• Conceptual framework and NGSS alignment
• Image and use policy
• Evo in the News
• The Tree Room
• Browse learning resources

• Search Menu
• Sign in through your institution
• Advance Articles
• Editor's Choice
• Author Guidelines
• Submission Site
• Open Access Options
• Self-Archiving Policy
• About Evolution
• About the Society for the Study of Evolution
• Editorial Board
• Advertising & Corporate Services

Article Contents

In fisher's words, correcting fisher's model, author contributions, acknowledgments, literature cited.

• < Previous

Fisher's lost model of runaway sexual selection

• Article contents
• Figures & tables
• Supplementary Data

Jonathan M. Henshaw, Adam G. Jones, Fisher's lost model of runaway sexual selection, Evolution , Volume 74, Issue 2, 1 February 2020, Pages 487–494, https://doi.org/10.1111/evo.13910

• Permissions Icon Permissions

The bizarre elaboration of sexually selected traits such as the peacock's tail was a puzzle to Charles Darwin and his 19th century followers. Ronald A. Fisher crafted an ingenious solution in the 1930s, positing that female preferences would become genetically correlated with preferred traits due to nonrandom mating. These genetic correlations would translate selection for preferred traits into selection for stronger preferences, leading to a self-reinforcing process of ever-elaborating traits and preferences. It is widely believed that Fisher provided only a verbal model of this “runaway” process. However, in correspondence with Charles Galton Darwin, Fisher also laid out a simple mathematical model that purportedly confirms his verbal prediction of runaway sexual selection. Unfortunately, Fisher's model contains inconsistencies that render his quantitative conclusions inaccurate. Here, we correct Fisher's model and show that it contains all the ingredients of a working runaway process. We derive quantitative predictions of his model using numerical techniques that were unavailable in Fisher's time. Depending on parameter values, mean traits and preferences may increase until genetic variance is depleted by selection, exaggerate exponentially while their variances remain stable, or both means and variances may increase super-exponentially. We thus present the earliest mathematical model of runaway sexual selection.

In The Descent of Man and Selection in Relation to Sex , Charles Darwin proposed that the elaborate ornaments of many species, borne most commonly by males, evolved due to preferences for such traits by the opposite sex (Darwin 1871 ). Famously, Darwin left alone the question of why such apparently extravagant preferences should evolve in the first place. Ronald A. Fisher provided a solution in his 1930 book The Genetical Theory of Natural Selection , expanding upon allusions in an earlier paper (Fisher 1915 ). If females have some modest initial preference for a particular trait, then this preference will become genetically correlated with the preferred trait. Sexual selection for males with larger trait values then indirectly favors genes for stronger preferences. This genetic association can lead to a self-reinforcing process, where both traits and preference become more extreme over evolutionary time.

It is widely believed that Fisher provided only a qualitative verbal model of the runaway process, leaving mathematical formalization to future generations of biologists (e.g., the models of O'Donald 1962 , 1980 ; Lande 1981 ; Kirkpatrick 1982 ). For instance, Karlin ( 1992 ) refers to Fisher's “qualitative scenario” and “verbal theory,” and says that “many (others) have tried to quantify Fisher's proposal.” O'Donald ( 1990 ) similarly writes that ‘polygenic models [of the runaway process]… were first studied by Lande in 1981’. Hoquet and Levandowsky ( 2015 ) remarked on the oddity that “Fisher, an early pioneer in the field of applied mathematical statistics, did not construct a mathematical model of the [runaway] process.” More recently, Prum ( 2017 ) repeats that “Fisher never presented an explicit mathematical model of his runaway process.”

Despite this widespread belief, Fisher did construct a mathematical model of the runaway process, although it was never published in his lifetime. Fisher's model was laid out in private correspondence with Charles Galton Darwin, a physicist who was also the grandson of his more famous eponymous ancestor. These letters, excerpts of which appear in Henry Bennett's variorum edition of The Genetical Theory of Natural Selection (Fisher 1999 ), were not included in earlier editions of the same book (Fisher 1930   1958 ). It is consequently no surprise that their impact has not been widely felt in the field. Indeed, apart from Bennett's ( 1999 ) own discussion and a passing mention by Edwards ( 2011 ), we were unable to find any mention at all of Fisher's model in the literature.

Fisher's model predates the next mathematical treatment of the runaway process (O'Donald 1962 ) by 30 years and the first quantitative genetic treatment (Lande 1981 ) by half a century. It is consequently of great interest what his model says and to what extent it anticipates future work. Here, we dissect Fisher's model in detail. We show that, due to a pair of mathematical inconsistencies, Fisher's quantitative predictions are incorrect. Nonetheless, his correspondence contains all the necessary conceptual and mathematical ingredients for a working model of the runaway process. We correct his model and derive its predictions using numerical techniques.

… Take x for cock beauty, and y for hen taste [NB: Fisher and Darwin discussed the runaway process via the example of mate choice in domestic fowl]. These will vary about some means x ¯ and y ¯ ⁠ ; of which x ¯ will not matter, for there is no natural zero for this measurement, but y ¯ will, for y = 0 would represent indifference, and y ¯ the average intensity of preference. We may suppose for convenience that x and y are genetic values so that their averages in the offspring are the averages for the two parents, and that for each the scale of measurement is so chosen that the mean values of ( x − x ¯ ) 2 and ( y − y ¯ ) 2 are both unity. They may be correlated to a degree which must be determined from the problem so we may put r for the average value of ( x − x ¯ ) ( y − y ¯ ) ⁠ .

We might suppose beauty to be measured objectively e.g. by the length of feathers in a ruff, but taste will have to be measured by actual performance. A hen with no taste would mate at random, i.e., on the average of a number of trials, the average x of the cock she mates with is x ¯ ⁠ . A selective hen will lose some opportunities for mating with ugly cocks and will score a higher average. On our scale of measurement I will say that her value is y if the average beauty of the cock she chooses is x ¯ + k y ⁠ . k is a datum depending on powers of discrimination, opportunities for choice, etc.

If a cock with specification x 1 , y 1 mates with a hen specified by x 2 , y 2 the offspring vary about the average x 1 + x 2 2 ⁠ , y 1 + y 2 2 ⁠ . The only hypothesis about heredity we need is that within this progeny x and y are uncorrelated; if this is true, then the mean product r in the progeny generation will be merely

1 4 ( x 1 − x ¯ + x 2 − x ¯ ) ( y 1 − y ¯ + y 2 − y ¯ ) F1

averaged over all matings. If this is the same as in the previous generation we can find r , for the average value of

( x 1 − x ¯ ) ( y 1 − y ¯ ) = r and ( x 2 − x ¯ ) ( y 2 − y ¯ ) = r F2

while for the rest

( x 1 − x ¯ ) ( y 2 − y ¯ ) = k y 2 ( y 2 − y ¯ ) = k and ( x 2 − x ¯ ) ( y 1 − y ¯ ) = r 2 k F3

as appears from averaging the kinds of cock which any particular hen x 2 , y 2 will mate with.

It appears then that

2 r = k ( 1 + r 2 ) or r = 1 − 1 − k 2 k F4

and a selection which raises the average of x by 1 2 k y ¯ in each generation must raise the average of y by 1 2 k r y ¯ i.e. y ¯ increases in geometrical progression, supposing k , and therefore r , to be constant. Of course in this I have ignored all checks, some of which may work slightly from the start, while others will certainly come in powerfully later.

Let me know if I have made any headway, as I found myself entirely dissatisfied with my inability to get the argument across, and I hope the point that x and y must be correlated may remove the difficulty you feel.

Sorry I left the ( x 2 − x ¯ ) ( y 1 − y ¯ ) evaluation obscure. The argument would go like this:

Hens selected for or the aggregate of hens having x 2 will have taste above the average by r ( x 2 − x ¯ ) ⁠ . They will therefore mate with cocks above the average in beauty by r k ( x 2 − x ¯ ) and therefore with cocks above the average in taste by r 2 k ( x 2 − x ¯ ) ⁠ . So the average of ( x 2 − x ¯ ) ( y 1 − y ¯ ) will be the average of r 2 k ( x 2 − x ¯ ) 2 = r 2 k ⁠ . The term does not matter, and I doubted its existence for a while, but it does belong.

I am sending this without answering the rest of your letter, so as to catch you with the point still in mind. Selous’ observations on the Ruff, where he has seen the hen passing with perfect self composure among the crowd of males, who await, but cannot hurry, her choice, provide a perfect ecological framework for this runaway type of selection. The hens choose the fashion of their sons’ ornaments.

The exponential element, which I agree is the kernel of the thing, arises from the rate of change in hen taste being proportional to the absolute average degree of taste ( δ y ∝ y ¯ ) ⁠ . The milk yield is, of course transmitted through the bull, but the intensity of selection in favour of higher milk yield is not determined by the average milk yield. Again the drone bee has large eyes probably only to see the queen during the nuptial flight, and this quality selected thus in the male is transmitted presumably both to his sons, and to his daughters’ sons, but the intensity of selection in no way depends on the actual average size of the eye, so there is no tendency to exponential increase.

As far as we know, this was the end of Fisher and Darwin's correspondence on runaway sexual selection, although they continued to exchange letters for many years (Fisher's correspondence with C. G. Darwin and many other individuals is available online in the University of Adelaide's R. A. Fisher Digital Archive at http://hdl.handle.net/2440/67635 ).

Fisher's model appears at first glance reasonable enough, but it contains two subtle inconsistencies. First, Fisher assumes that both traits x and preferences y have unit variance in the parental population. However, this normalization procedure is problematic when iterating the model over multiple generations. This is because re-normalizing female preferences each generation is a not a neutral “change of scale,” but rather transforms the preference distribution away from its evolved value, thereby altering the evolutionary trajectories of both traits and preferences. This inconsistency might be seen as fairly benign. After all, similar assumptions are made in most quantitative genetic models of Fisherian sexual selection (e.g., Lande 1981 ; Iwasa et al. 1991 ), which treat the genetic variances and covariances of traits and preferences as fixed parameters (for partial relaxations of this assumption, see Barton and Turelli 1991 ; Pomiankowski and Iwasa 1993 ). Nonetheless, faithfully accounting for the evolution of preference variation yields some interesting predictions, as we shall see.

More egregious is Fisher's assumption that the covariance between traits x 1 and unexpressed preferences y 1 among mating males equals their covariance in the parental population as a whole. This assumption is represented by the first part of equation ( (F2) ). In general, however, these two covariances differ. For simplicity, suppose that a female with preference y 2 always chooses partners with trait values of exactly   x ¯ + k y 2 (we relax this assumption in the Supporting Information). For ease of comparison, let us also retain Fisher's assumption that σ x 2 = σ y 2 = 1 ⁠ . In this case, the covariance between traits and preferences in the parental population is σ x y = r ⁠ , whereas the covariance among mating males is σ x 1 y 1 = k 2 r (for details, see Methods section). The intuition that these values must differ is most easily obtained when k = 0 ⁠ . In this case, all females choose males with trait values that are exactly average. There is consequently no variance in male trait values, and so the covariance σ x 1 y 1 = 0 ⁠ .

As a consequence of these two inconsistencies, the equilibrium condition derived in equation ( (F4) ) is incorrect. Below, we reconstruct and correct Fisher's model. Doing so requires three additional assumptions that are not explicit in the original model. First, we must specify a joint distribution of traits and preferences in the parental generation. In fact, Fisher's original argument does not hold for arbitrary joint distributions, but only those where individuals with trait values of x ¯ + Δ have preference values of y ¯ + Δ r on average, regardless of the value of Δ. We will assume that traits and preferences in the parental generation follow a bivariate normal distribution. Second, we require a more concrete specification of female choice. For simplicity, we assume that if a female has preference y 2 , her mates have trait values that are exactly k y 2 standard deviations above the mean (i.e., there is no variance around the average in Fisher's model). We relax this assumption in the Supporting Information, where we allow for errors in female mate choice. Third, we allow for the variance in traits and preferences to be renewed each generation by meiosis and mutation. Without such variational input, the variance in these characters would quickly be depleted in the absence of strongly disruptive or temporally variable selection.

Following Fisher, we distinguish among trait and preference values in three contexts: in the parental generation before mating occurs ( x and y ), among mating males ( x 1 and y 1 ) and among mating females ( x 2 and y 2 ). We assume that these values are entirely genetically determined (i.e., with no environmental contribution). We also suppose that ( x , y ) initially follows a bivariate normal distribution (which is implicit in Fisher's argument). Fisher assumes that x and y have unit variance in the parental generation, with the consequence that the covariance and the correlation between these traits coincide. In contrast, we normalize neither traits nor preferences. We write σ x 2 and σ y 2 for their variances, σ x y for their covariance, and r = σ x y σ x σ y for their correlation.

In particular, note that when σ x = σ y = 1 ⁠ , we have σ x 1 y 1 = k 2 r ⁠ , which differs from Fisher's value of r in equation (F2) (see above).

Given values for the parameters k , σ ξ x and σ ξ y ⁠ , and initial values for x ¯ ⁠ , y ¯ ⁠ , σ x ⁠ , σ y ⁠ , and r , we can iterate the above model numerically to derive the evolutionary trajectory of trait and preference means, variances, and correlations across generations. For some values of k , σ ξ x and σ ξ y there is a trajectory where the variances σ x 2 and σ y 2 and the covariance σ x y are stable across generations. We located such equilibria “equilibria” by setting the expressions in equation (12) equal to their values in the previous generation and solving numerically.

Our corrected model predicts the occurrence of three qualitatively different outcomes (Fig. (1) ; note that the scales on both the horizontal and vertical axes differ among panels):

Coevolution of mean trait values (blue) and preference values (yellow) under three different scenarios. Note that, due to the vast differences in trait evolution among these scenarios, the scales on both the horizontal and vertical axes differ among panels. (A) Classic runaway , where mean traits and preferences increase according to a geometric progression. Variances in traits and preferences and the correlation between them remain constant at a stable pseudo-equilibrium. Shown with σ y = 1 and σ ξ x 2 = σ ξ y 2 = 1 2 ⁠ . (B) Explosive runaway , where the means and variances of traits and preferences increase super-exponentially. Shown with σ y = 5 and σ ξ x 2 = σ ξ y 2 = 1 2 ⁠ . (C) Fizzle away , where variance in traits and preferences is depleted. Mean traits and preferences initially rise and then plateau. Show with σ y = 1 and σ ξ x 2 = σ ξ y 2 = 0 ⁠ . All panels are shown with k = 0.5 ⁠ , initial parameters x ¯ = 0 ⁠ , y ¯ = 0.1 ⁠ , r = 0 ⁠ , σ x = 1 ⁠ , and other parameters as noted above.

CLASSIC RUNAWAY

In this case, the mean values of traits and preferences increase without bound, while their variances and correlation approach a stable equilibrium. 2 This occurs when (1) the parameter k , the initial variance in preferences, and the variational input to preferences σ ξ y 2 are all not too large, (2) traits and preferences receive new variational input each generation (i.e., σ ξ x 2 , σ ξ y 2 > 0 ⁠ ), and (3) the initial mean preference is non-zero. At this pseudo-equilibrium, preferences increase geometrically by a fixed proportion of 1 2 k r σ y each generation, just as Fisher predicted. Both the correlation r and the proportional rate of increase at equilibrium are increasing functions of k (Fig. (2) ). Although the equilibrium is stable, it is not a global attractor.

Within-individual correlation r between trait and preference values (blue) and the proportional increase in mean preferences each generation (yellow) at equilibrium in the “classic runaway” scenario. The variances σ x 2 and σ y 2 in traits and preferences and the correlation r are stable across generations. Shown with variational inputs of σ ξ x 2 = σ ξ y 2 = 1 2 ⁠ . When k > 0.69 ⁠ , there is no stable equilibrium for these parameter values and explosive runaway occurs (see Fig. 2).

EXPLOSIVE RUNAWAY

The second case occurs when either the initial variance or the variational input of preferences is large. In this case, both the means and variances in traits and preferences increase super-exponentially, quickly reaching absurd values. It is notable that such explosive behavior can occur even if there is no new variational input: in this case, selection is so strong that extreme outliers in the original distributions are strongly favored, leading to a rapid increase in variance. Such outliers are always available to selection, because this quantitative genetic model implicitly assumes an infinite population size and an infinite number of loci of infinitesimal effect (Barton et al. 2017 ).

FIZZLE AWAY

Lastly, if there is no variational input (i.e., σ ξ x 2 = σ ξ y 2 = 0 ⁠ ) and the initial variance in preferences is low, then sexual selection “fizzles away.” Variation in both traits and preferences converges to zero, and the means of both traits and preferences plateau after an initial period of increase. If, alternatively, there is variational input to traits but not preferences ( ⁠ σ ξ x 2 > 0 and σ ξ y 2 = 0 ⁠ ), then preferences will plateau while traits increase indefinitely (data not shown).

Our main model assumes that mate choice is perfect, in the sense that females always choose males with trait values exactly matching their preferences. Suppose, on the other hand, that realized mate choice is noisy, such that the difference between a female's preference and her mate's trait value can be represented by a normally distributed “error.” Such noise introduces an additional source of variation in the trait values (and, by extension, preferences values) of mating males. This acts to maintain variation in both characters, and consequently has similar evolutionary implications to variational input via mutation (see Fig. S1 ).

We have corrected Fisher's unpublished model of 1932 to produce, posthumously, the first mathematical model of runaway sexual selection. Despite a pair of mathematical inconsistencies, Fisher's letters contain all of the necessary ingredients of a working runaway process. The corrected model is more complex than Fisher's analytic sketch, and our analysis of it relies partly on numerical techniques that were unavailable in Fisher's time. Nonetheless, the corrected model is very close to the conceptual spirit of both Fisher's sketch and the verbal model in The Genetical Theory of Natural Selection . Among modern models, the corrected model is closest to that of Karlin and Raper ( 1990 ), which, however, contains additional elements such as viability selection on male traits.

An obvious question is why Fisher never published a formal model of runaway sexual selection, and here we can only speculate. First, it is worth noting that the situation is hardly unique. Fisher often favored verbal over mathematical models in an attempt to reach a wider audience (Edwards 2011 ), and many verbal models in The Genetical Theory of Natural Selection were never formalized by Fisher himself. Perhaps Fisher thought that his verbal model was clear enough, and eschewed the dirty and detailed work of giving it specific form. On the other hand, there is a hint of frustration at the end of Fisher's letter, where he pronounces himself “entirely dissatisfied with my inability to get the argument across.” Maybe Fisher never derived a mathematical model that met his high standards. Notably, many later models of the runaway process (Lande 1981 ; Pomiankowski et al. 1991 ; Iwasa and Pomiankowski 1995 ; Day 2000 ; Kokko et al. 2015 ) and of sexual selection more generally (Grafen 1990 ; Iwasa et al. 1991 ; Tazzyman et al. 2014 ; Dhole et al. 2018 : reviewed in Kuijper et al. 2012 ) are characterized by considerable analytical sophistication and by numerical techniques that were unavailable or impractical in Fisher's pre-digital era.

In contrast to most later work, Fisher's model contains no fitness costs or evolutionary constraints that would curb the evolution of elaborate traits. Indeed, he “ignored all checks, some of which may work slightly from the start, while others will certainly come in powerfully later.” As a consequence, runaway evolution occurs very easily in this model. Indeed, a “classic” runaway is possible whenever initial female preference differs from zero on average and there is some variational input to both traits and preferences. Moreover, high initial variance in female preference can lead to an “explosive” runaway, where both the means and variances of traits and preferences increase super-exponentially.

It is amusing to imagine a peacock's tail the size of the universe, but in reality, of course, such elaboration will be dampened in multiple ways. First, there will be natural selection favoring smaller trait values (Lande 1981 ; Kirkpatrick 1982 ; Karlin and Raper 1990 ), which will likely increase in intensity as elaborate traits begin straining the limits of resource acquisition (Fromhage and Jennions 2016 ; Henshaw et al. 2019 ). Second, extreme female preferences will be selected against if they prevent females from finding a mate (de Jong and Sabelis 1991 ; Kokko and Mappes 2005 ; Priklopil et al. 2015 ; Dechaume-Moncharmont et al. 2016 ; Henshaw 2018 ). Third, strong selection for ever-elaborating traits will deplete genetic variance for those traits in finite populations (Borgia 1979 ; Kirkpatrick and Ryan 1991 ; Rowe and Houle 1996 ; Kotiaho et al. 2008 ). If traits already commandeer a large share of available resources, then presumably few mutations will arise that increase trait size while maintaining general viability. The combination of no-cost traits and unbounded genetic variability in Fisher's model enables the “explosive” runaway that would be impossible in a model with either realistic fitness trade-offs or a finite population or genetic structure.

Fisher's model was never intended to be realistic. Nonetheless, it clearly demonstrates the conceptual operation of the runaway process and would have provided an admirable formal basis for further work, had anyone known about it.

J.M.H. designed the model. J.M.H. and A.G.J. wrote the manuscript.

We are grateful to Henry Bennett for cataloguing and digitizing Fisher's correspondence, much of which is available online in the R. A. Fisher Digital Archive of The University of Adelaide. We also thank the Special Collections at The University of Adelaide for maintaining this outstanding resource and for permission to reproduce some of Fisher's correspondence in this article.

We are grateful to the Special Collections at The University of Adelaide Library for permission to reproduce these extracts from the R. A. Fisher Digital Archive (available online at http://hdl.handle.net/2440/67635 ). Equation numbers have been added for clarity.

For some parameter combinations, there is additionally an unstable equilibrium in which traits and preferences are very highly correlated (cf. Karlin and Raper 1990 )

Barton , N. H. , A. M. Etheridge , and A. Véber . 2017 . The infinitesimal model: definition, derivation, and implications . Theor. Pop. Biol.   118 : 50 – 73 .

Google Scholar

Barton , N. H. , and M. Turelli . 1991 . Natural and sexual selection on many loci . Genetics   127 : 229 – 255 .

Bennett , J. H.   1999 . Foreword. Pp. vi – xxii   in   The genetical theory of natural selection: a complete variorum edition . Oxford Univ. Press , Oxford, U.K .

Google Preview

Borgia , G.   1979 . Sexual selection and the evolution of mating systems. Pp. 19 – 80   in   M. S. Blum and N. A. Blum , eds. Sexual selection and reproductive competition in insects . Academic Press , New York, NY .

Darwin , C.   1871 . The descent of man, and selection in relation to sex . Princeton Univ. Press , Princeton, U.K.

Day , T.   2000 . Sexual selection and the evolution of costly female preferences: spatial effects . Evolution   54 : 715 – 730 .

de Jong , M. C. M. , and M. W. Sabelis . 1991 . Limits to runaway sexual selection: the wallflower paradox . J. Evol. Biol.   4 : 637 – 655 .

Dechaume-Moncharmont , F.-X. , T. Brom , and F. Cézilly . 2016 . Opportunity costs resulting from scramble competition within the choosy sex severely impair mate choosiness . Anim. Behav.   114 : 249 – 260 .

Dhole , S. , C. A. Stern , and M. R. Servedio . 2018 . Direct detection of male quality can facilitate the evolution of female choosiness and indicators of good genes: evolution across a continuum of indicator mechanisms . Evolution   72 : 770 – 784 .

Edwards , A. W. F.   2011 . Mathematizing Darwin . Behav. Ecol. Sociobiol.   65 : 421 – 430 .

Fisher , R. A.   1915 . The evolution of sexual preference . Eugen. Rev.   7 : 184 – 192 .

Fisher , R. A.   1930 . The genetical theory of natural selection . 1st ed.   Clarendon Press , Oxford, U.K .

Fisher , R. A.   1958 . The genetical theory of natural selection . 2nd ed.   Dover , New York, NY .

Fisher , R. A.   1999 . The genetical theory of natural selection: A complete variorum edition . Oxford Univ. Press , Oxford, U.K .

Fromhage , L. , and M. D. Jennions . 2016 . Coevolution of parental investment and sexually selected traits drives sex-role divergence . Nat. Commun.   7 :12517.

Grafen , A.   1990 . Sexual selection unhandicapped by the Fisher process . J. Theor. Biol.   144 : 473 – 516 .

Henshaw , J. M.   2018 . Finding the one: optimal choosiness under sequential mate choice . J. Evol. Biol.   31 : 1193 – 1203 .

Henshaw , J. M. , L. Fromhage , and A. G. Jones . 2019 . Sex roles and the evolution of parental care specialization . Proc. R. Soc. B   286 :20191312.

Hoquet , T. , and M. Levandowsky . 2015 . Utility vs beauty: Darwin, Wallace and the subsequent history of the debate on sexual selection. Pp. 19 – 44   in   T. Hoquet , ed. Current perspectives on sexual selection . Springer , Dordrecht, Germany .

Iwasa , Y. , and A. Pomiankowski . 1995 . Continual change in mate preferences . Nature   377 : 420 – 422 .

Iwasa , Y. , A. Pomiankowski , and S. Nee . 1991 . The evolution of costly mate preferences II. The “handicap” principle . Evolution   45 : 1431 – 1442 .

Karlin , S.   1992 . R. A. Fisher and evolutionary theory . Stat. Sci.   7 : 13 – 33 .

Karlin , S. , and J. Raper . 1990 . Evolution of sexual preferences in quantitative characters . Theor. Pop. Biol.   38 : 306 – 330 .

Kirkpatrick , M.   1982 . Sexual selection and the evolution of female choice . Evolution   36 : 1 – 12 .

Kirkpatrick , M. , and M. J. Ryan . 1991 . The evolution of mating preferences and the paradox of the lek . Nature   350 : 33 – 38 .

Kokko , H. , I. Booksmythe , and M. D. Jennions . 2015 . Mate-sampling costs and sexy sons . J. Evol. Biol.   28 : 259 – 266 .

Kokko , H. , and J. Mappes . 2005 . Sexual selection when fertilization is not guaranteed . Evolution   59 : 1876 – 1885 .

Kotiaho , J. S. , N. R. LeBas , M. Puurtinen , and J. L. Tomkins . 2008 . On the resolution of the lek paradox . Trends Ecol. Evol.   23 : 1 – 3 .

Kotz , S. , C. N. Balakrishnan , and N. L. Johnson . 2000 . Continuous multivariate distributions . 2nd ed.   John Wiley & Sons , New York, NY .

Kuijper , B. , I. Pen , and F. J. Weissing . 2012 . A guide to sexual selection theory . Ann. Rev. Ecol. Evol. Syst.   43 : 287 – 311 .

Lande , R.   1981 . Models of speciation by sexual selection on polygenic traits . Proc. Natl. Acad. Sci. USA   78 : 3721 – 3725 .

O'Donald , P.   1990 . Fisher's contributions to the theory of sexual selection as the basis of recent research . Theor. Pop. Biol.   38 : 285 – 300 .

O'Donald , P.   1980 . Genetic models of sexual selection . Cambridge Univ. Press , Cambridge, U.K .

O'Donald , P.   1962 . The theory of sexual selection . Heredity   17 : 541 – 552 .

Pomiankowski , A. , and Y. Iwasa . 1993 . Evolution of multiple sexual preferences by Fisher's runaway process of sexual selection . Proc. R. Soc. B   253 : 173 – 181 .

Pomiankowski , A. , Y. Iwasa , and S. Nee . 1991 . The evolution of costly mate preferences I. Fisher and biased mutation . Evolution   45 : 1422 – 1430 .

Priklopil , T. , É. Kisdi , and M. Gyllenberg . 2015 . Evolutionarily stable mating decisions for sequentially searching females and the stability of reproductive isolation by assortative mating . Evolution   69 : 1015 – 1026 .

Prum , R. O.   2017 . The evolution of beauty: how Darwin's forgotten theory of mate choice shapes the animal world—and us . Doubleday , New York, NY .

Rowe , L. , and D. Houle . 1996 . The lek paradox and the capture of genetic variance by condition dependent traits . Proc. R. Soc. B   263 : 1415 – 1421 .

Tazzyman , S. J. , Y. Iwasa , and A. Pomiankowski . 2014 . The handicap process favors exaggerated, rather than reduced, sexual ornaments . Evolution   68 : 2534 – 2549 .

Associate Editor: E. Kisdi

Handling Editor: M. Servedio

Author notes

[Corrections added on Jan 20, 2020 after first online publication: the second line of equation 12 is updated.]

Supplementary data

Figure S1 Coevolution of mean trait values (blue) and preference values (yellow) when females choose their mates with a normally distributed error with variance σ ε 2 = 0.5 .

Email alerts

Citing articles via.

• Recommend to Your Librarian
• Advertising and Corporate Services
• Journals Career Network

Affiliations

• Online ISSN 1558-5646
• Print ISSN 0014-3820
• Copyright © 2024 Society for the Study of Evolution
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

• school Campus Bookshelves
• menu_book Bookshelves
• perm_media Learning Objects
• login Login
• how_to_reg Request Instructor Account
• hub Instructor Commons

Margin Size

• Download Page (PDF)
• Download Full Book (PDF)
• Periodic Table
• Physics Constants
• Scientific Calculator
• Reference & Cite
• Tools expand_more
• Readability

selected template will load here

This action is not available.

4.11: Curbing runaway selection

• Last updated
• Save as PDF
• Page ID 4138

• Michael W. Klymkowsky and Melanie M. Cooper
• University of Colorado Boulder and Michigan State University

\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

\( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)

\( \newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\)

( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\)

\( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)

\( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\)

\( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\)

\( \newcommand{\Span}{\mathrm{span}}\)

\( \newcommand{\id}{\mathrm{id}}\)

\( \newcommand{\kernel}{\mathrm{null}\,}\)

\( \newcommand{\range}{\mathrm{range}\,}\)

\( \newcommand{\RealPart}{\mathrm{Re}}\)

\( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)

\( \newcommand{\Argument}{\mathrm{Arg}}\)

\( \newcommand{\norm}[1]{\| #1 \|}\)

\( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\AA}{\unicode[.8,0]{x212B}}\)

\( \newcommand{\vectorA}[1]{\vec{#1}}      % arrow\)

\( \newcommand{\vectorAt}[1]{\vec{\text{#1}}}      % arrow\)

\( \newcommand{\vectorB}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

\( \newcommand{\vectorC}[1]{\textbf{#1}} \)

\( \newcommand{\vectorD}[1]{\overrightarrow{#1}} \)

\( \newcommand{\vectorDt}[1]{\overrightarrow{\text{#1}}} \)

\( \newcommand{\vectE}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash{\mathbf {#1}}}} \)

Sexual selection can lead to what has been termed (but is not really) runaway selection. For example, the more prominent the peacock male's tail the more likely he will find a mate even though larger and larger tails may also have significant negative effects. All of which is to say that there will be both positive and negative selection for tail size, which will be influenced by the overall probability that a particular male mates successfully. Selection does not ever really run away, but settles down when the positive (in terms of sexual success) and negative (in turns of various costs) of a trait come to equal each other. Sufficient numbers of male peacocks emerge as reproductively successful even if many males are handicapped by their tails and fall prey to predators. For another example, consider the evolution of extremely large antlers associated with male dominance and mate accessibility, such as occurred in Megaloceros giganteous . These antlers could also act to inhibit the animal’s ability to move through heavily wooded areas. In a stable environment, the costs of generating antlers and benefits of effective sexual advertising would be expected to balance out; selection would produce an optimal solution. But if the environment changes, pre-existing behavior and phenotypes could act to limit an organism’s ability to adapt or to adapt fast enough to avoid extinction. In the end, as with all adaptations, there is a balance between the positive effects of a trait, which lead to increased reproductive success, and their negative effects, which can also influence survival. The optimal form of a trait may not be stable over time, particularly if the environment is changing.

Summary: Social and ecological interactions apply to all organisms, from bacteria to humans. They serve as a counter-balance to the common caricature of evolution as a ruthless and never ceasing competition between organisms. This hyper-competitive view, often known as the struggle for existence or Social Darwinism, was not supported by Darwin or by scientifically-established evolutionary mechanisms, but rather by a number of pundits who used it to justify various political (that is, inherently non-scientific) positions, particularly arguing against social programs that helped the poor (often characterized as the unfit) at the “expense” of the wealthy. Assuming that certain organisms were inherently less fit, and that they could be identified, this view of the world gave rise to Eugenics, the view that genetically inferior people should be killed, removed, or sterilized, before their "bad" traits overwhelmed a particular culture. Eugenics was a particularly influential ideology in the United States during the early part of the 20 th century and inspired the genocidal programs of the Nazis in Germany. What is particularly odd about this evolutionary perspective is that it is actually anti-evolutionary, since if the unfit really were unfit, they could not possibly take over a population. In addition, it completely ignores the deeply social (cooperative) aspect of the human species.

Contributors and Attributions

Michael W. Klymkowsky (University of Colorado Boulder) and Melanie M. Cooper (Michigan State University) with significant contributions by Emina Begovic & some editorial assistance of Rebecca Klymkowsky.

• Reference Manager
• Simple TEXT file

People also looked at

Review article, runaway social selection in human evolution.

• 1 Department of Biology, Simon Fraser University, Burnaby, BC, Canada
• 2 Department of Anthropology, Baylor University, Waco, TX, United States
• 3 Department of Biology, East Carolina University, Greenville, NC, United States

Darwin posited that social competition among conspecifics could be a powerful selective pressure. Alexander proposed a model of human evolution involving a runaway process of social competition based on Darwin’s insight. Here we briefly review Alexander’s logic, and then expand upon his model by elucidating six core arenas of social selection that involve runaway, positive-feedback processes, and that were likely involved in the evolution of the remarkable combination of adaptations in humans. We discuss how these ideas fit with the hypothesis that a key life history innovation that opened the door to runaway social selection, and cumulative culture, during hominin evolution was increased cooperation among individuals in small fission-fusion groups.

“There can be no doubt that a tribe including many members who, from possessing in a high degree the spirit of patriotism, fidelity, obedience, courage, and sympathy, were always ready to give aid to each other and to sacrifice themselves for the common good, would be victorious over most other tribes; and this would be natural selection”

Charles Darwin [1871: 166]

“Why are we all alone [in] … our tendency and ability to cooperate and compete in social groups of millions?”

Richard Alexander (1990: 1)

Introduction

In 1871–2 Charles Darwin tackled two major challenges to his theory of evolution: costly displays and weaponry used in courtship, and the descent of humans. He proposed that mating competition among conspecifics — sexual selection — was a potent evolutionary force that could explain the apparent enigmas of bright plumage, antlers, even the great horn of the Rhinoceros beetle. Sir Ronald Fisher (1930) furthered Darwin’s concept of sexual selection by positing that intraspecific competition could involve a process of “runaway” selection if choice was based on comparisons favoring relative extremes. Crook (1972) and West-Eberhard (1979 , 1983) expanded Darwin’s and Fisher’s concepts to include selection from all aspects of social interaction among conspecifics, termed “social selection.” Humphrey (1976) and later Dunbar (1993) and Tomasello (1999) suggested that intelligence could evolve in the context of social competition and cooperation among conspecifics: cleverness in the “social chess game.” Alexander (1990) ; (see also Alexander, 1974 , 1979 , 1989 ) integrated these ideas into a comprehensive model of “How humans evolved” involving a process of “runaway social selection” where hominins “increasingly became their own principal hostile force of nature,” with cooperation and coalitions posited as crucial and complementary aspects of hominin social environments ( Alexander, 1987 , 2006 ; Wrangham, 1999 ).

We use the term “runaway” selection here to refer broadly to selection that involves either: (1) arms races within species, whereby competition-based selection between individuals and groups leads to reciprocal, escalating trait expression and elaboration across evolutionary time, or (2) positive feedback between the selection pressures and evolutionary changes in one trait (e.g., a phenotype subject to choice), and selection pressures and evolutionary changes in a second trait (e.g., choice of the phenotype), such that evolutionary changes in both traits become mutually reinforcing across generations (e.g., Nesse, 2007 ; Nakamaru and Dieckmann, 2009 ; Piantadosi and Kidd, 2016 ; Bailey and Kölliker, 2019 ). As such, runaway selection can apply to a wide variety of sets of phenotypes, including aspects of social interactions, in addition to those involved in female choice and sexual selection (e.g., Bailey and Kölliker, 2019 ). Alexander focused on a process of “runaway social selection” involving arms-race competition among individuals for “social cleverness” (including language abilities, social skills, aptitudes for cultural information, coalition building, and multiple other types of intelligence) that became increasingly important in human evolution ( Alexander, 1990 ; Flinn and Alexander, 2007 ). There are other plausible models of human evolution involving social competition (e.g., Hrdy, 2009 ; van Schaik and Burkart, 2010 ; Wrangham, 2019 ); our main objective here is to describe, extend and expand upon Alexander’s runaway social selection model, which connects to both Darwin’s (1871 ) ideas about selection associated with competition among conspecifics and his model of human descent. As such, we consider runaway social selection involving both competitive arms races, and positive feedbacks driven by mutually reinforcing selection during choice-trait coevolution.

Darwin and the Descent of Humans

Darwin (1871) suggested an evolutionary scenario for humans involving a positive feedback loop between tool use and intelligence. Initially a “smart ape” began to use tools; this advantage led to further selection for intelligence and more sophisticated tools, and eventually to upright bipedal locomotion, precision hand control, reduced dentition, social cooperation, morality, and other human traits. Evidence from hominin paleontology and archeology over the past 150 years has not supported Darwin’s tool use model as he presented it. Fossils indicate that Australopithecines and perhaps even earlier hominins were habitual bipeds for >2 MY prior to significant changes in brain evolution ( McBrearty and Brooks, 2000 ; Antón et al., 2014 ; Almécija et al., 2021 ). Tool use also predates increases in cranial capacity by at least 1 MY ( McPherron et al., 2010 ), it is not restricted to hominins and it is not subserved by specialized neurobiological mechanisms ( Geary and Huffman, 2002 ; Geary, 2005 ; Sherwood et al., 2008 ; Bruner, 2021 ). Although technology is clearly a significant part of the human evolutionary story ( Osiurak and Reynaud, 2020 ), it apparently does not account for our extraordinary social mental aptitudes including such traits as empathy, language, mental time-travel, consciousness, and mind-reading ( Herrmann et al., 2007 ; Haber and Corriveau, 2020 ), or for the uniqueness of the hominin evolutionary trajectory.

Darwin recognized that culture — socially transmitted information and materials — was also a key selective pressure in human evolution. Indeed, he noted (1871, pp. 78–79) that “the formation of different languages and of distinct species, and the proofs that both have been developed through a gradual process, are curiously parallel.” Aptitudes for language, learning, and the sociality underpinned acquisition of information were increasingly important for hominin survival and reproduction, eventually resulting in the extraordinary Anthropocene niche that we inhabit today. Why humans are “the uniquely unique species” ( Alexander, 1990 ) who developed such extraordinary cognitive and cultural abilities remains an elusive evolutionary puzzle ( Tomasello, 1999 ; Henrich and McElreath, 2003 ; Laland and Seed, 2021 ). The problem is further complicated by inherent biases of humans trying to understand themselves ( Alexander, 1987 ; Varella, 2018 ).

Alexander and Runaway Social Selection

An important universal trait of mammals is maternal care of altricial (helpless) offspring. Many mammalian species, including most primates, also have varying levels of alloparental support and protection by relatives. Beyond these shared features, however, humans exhibit a suite of highly unusual traits ( Alexander, 1990 ; Chapais, 2009 ), many of which appear adaptively responsive to variable conditions. Humans are the only species characterized by the combination of stable breeding bonds; flexible and extensive alloparenting and considerable male parental effort within multi-male groups; lengthy childhoods; cryptic ovulation; extended bilateral, multi-generational, and affinal kin recognition; grandparenting; influence of relatives over mate choice; language; variable group composition and inter-group relationships; and a suite of other human-elaborated traits.

Alexander’s model of how hominins evolved this combination of traits is based on the concept that hominin evolution became an increasingly autonomous and self-reinforcing, runaway process. A key selective pressure on hominins was thus interactions with other hominins, particularly with regard to its selective effects on brain evolution ( Alexander, 1990 ; see also Flinn et al., 2005 ). Concomitant with the increased importance of competition and cooperation among conspecifics was an increase in “ecological dominance,” whereby predation and competition from other species became weaker and weaker selective forces on hominin phenotypes.

We describe a set of revisions and extensions to Alexander’s model, in the light of recent work on human social and sexual evolution and behavior. First, we discuss an explanation for the initial split in the evolutionary trajectories of hominins and Pan , and how it underpinned the divergence of the two lineages. Second, we operationalize Alexander’s model of runaway social evolution by explicitly describing the relevant arenas of social selection and social competition, what traits were selected for and how, and how different arenas and forms of social selection contributed to runaway effects.

Hominin Origins

Hominins and Panins last shared a common ancestor about 6.5–9 MYA ( Andrews, 2020 ; Almécija et al., 2021 ). From an orthograde last common ancestor, Hominins and Panins initially diverged into distinct niches, with associated changes in locomotion leading to upright bipedalism in hominins, and knuckle-walking in Pan , similar to Gorilla s. Arguably, the shift to bipedalism implies a more terrestrial niche, with different foraging opportunities and predation pressures ( Harcourt-Smith, 2007 ; Almécija et al., 2021 ). Early hominins likely developed a fluid sociality similar to that characteristic of contemporary hunter/gatherer societies, involving tolerance of, and interaction with, individuals from other small, local, low-density groups (in contrast to chimpanzees and gorillas; more similar in some but not all aspects to bonobos as discussed below), eventually encompassing flexible alliances and coalitions in their fission-fusion context ( Walker et al., 2011 ; Apicella et al., 2012 ; Macfarlan et al., 2014 ; Migliano et al., 2017 , 2020 ). Such alliances may have been beneficial, at least initially, in terms of cooperative foraging, food-sharing, and protection from predators ( Allen-Arave et al., 2008 ; Smith et al., 2016 ). This initial difference, possibly linked to the gradual shift to a more open and mobile, terrestrial niche, would have enabled a series of subsequent evolutionary changes: (1) a flexible, distinctive pattern of extended family relationships that supported longer periods of child development ( Washburn and Lancaster, 1968 ; Lovejoy, 1981 ; Hrdy, 2009 , 2014 ; Hawkes, 2020 ); (2) critical aspects of the fluid and complex coalitional sociality posited above ( Gavrilets et al., 2008 ; Chapais, 2009 , 2011 ; Hawkes et al., 2018 ); and (3) an environment in which cultural innovations were increasingly important for foraging, defense against predators, and success in cooperation and competition with conspecifics ( Hill et al., 2011 ; Lotem et al., 2017 ; Flinn, 2021 ; Garg et al., 2021 ).

Early hominins also diverged from Pan in which other sets of individuals were most important to them. For female hominins, relationships with mothers, sisters, daughters, aunts, and grandmothers were of increasing importance. But so too were fathers, mates, brothers, and sons. From the male hominin perspective, relationships with paternal relatives — fathers, brothers, and sons — were of increasing importance for cooperative defense and foraging. As posited above, female relatives — wives, mothers, sisters —, and children benefited from this crucial support from males.

Hence the conundrum, analogous to the “matrilineal puzzle” proposed by Richards (1950) ; (see also Irons, 1983 ; Macfarlan et al., 2014 ; Dyble et al., 2015 ) emerges. How can males and females be with kin who reside in different places? How to help both your sister and your wife? And how to effectively avoid inbreeding problems if male and female relatives — father-daughter, brother-sister — co-reside? The solution is found in most hunter-gatherer foraging-band societies: flexible, fluid camp residence and social networks. Individuals can choose to stay or visit with whomever is most useful to them at a particular time. With inter-camp group tolerance and cooperation, hominins got the best of both worlds; help from maternal and paternal kin, mates and affines. This pattern of “exploded fission-fusion” sociality ( Marlowe, 2004 ; Foley and Gamble, 2009 ; Macfarlan et al., 2014 ) stands in stark contrast to that of all other hominids.

A key consequence and benefit of this fluid interactive social system was an open door for cumulative culture and language. Socially transmitted information could move easily and rapidly across the hominin social landscape ( Hill et al., 2011 , 2014 ; Walker et al., 2011 ; Gowlett et al., 2012 ). A good idea (“meme”) would spread fast and far (go “viral”). “Good ideas” were not limited to tools, engineering, and technology, but include social tactics and strategies (e.g., Coward and Grove, 2011 ). And, as described below, such a complex matrix of self-selected interactions provides excellent opportunities for social selection to exert its runaway effects, by the various mechanisms, in the various contexts, that typify human cognition, behavior and culture.

Psychological Mechanisms of Social Selection

Selective arenas represent specific contexts within which social selection mediates variation in inclusive fitness among individuals. The primary psychological mechanisms of social selection, within these contexts, are cognitive and emotional. Thus, an individual will benefit the most in inclusive fitness under social selection if they can:

(1) Individually recognize all of the persons in their group;

(2) Discern the relationships of kinship, friendship, and sexuality among all interacting individuals in their group;

(3) Figure out, consciously or unconsciously, how each individual person who they interact with could best be manipulated, cooperated with, or competed with, in what way, to maximally increase their own lifetime inclusive fitness;

(4) Discern and infer, consciously or unconsciously, how any other individual would be expected to respond to these possible actions toward them, from being able to take their mental perspective regarding their strategies and abilities to maximize their own inclusive fitness.

Each other person in a group thus has some potential inclusive fitness value to a focal individual, that could be maximized by success in providing benefits, imposing costs, or taking control of behavior away. Ability to achieve this potential will be some function of asymmetries in information, physical and intellectual power, alliances, and leverage (control of a resource or service that cannot be taken by force; Strassmann and Queller, 2010 ; Watts, 2010 ; Bissonnette et al., 2015 ). An individual would also benefit tremendously from knowing their own abilities and best strategies for increasing inclusive fitness, in this complex multidimensional web of social interactions and their mental representations.

The cognitive challenges of being able to most-effectively maximize inclusive fitness via the four steps described above are open-ended and almost unimaginably complex, for any extended human group of reasonable size, such as 50 to 150. As a result, social selection and responses to the selection, in the context of evolving human social-cognitive-emotional abilities, can proceed virtually without limit, being constrained only by human neural computing power, manifest in brain size and modular specializations, and being driven by multiple forms of runaway social selection, as described in detail below.

The psychological mechanisms of social selection can be applied in a wide variety of specific contexts, or arenas, whereby pairs and larger sets of humans interact. These arenas can, in turn, delineate the different forms of runaway social selection that lead to accelerated human evolution for cognitive and cultural traits, and the remarkable suites of adaptations that result.

Arenas of Social Selection

Arenas of human social selection exemplify the different contexts of human interactions that contribute to runaway human social evolution under the broad umbrella of Alexander’s (1989 ; 1990 ) model. These arenas have been discussed before, but not integrated together, and they have not been considered in the framework of how humans evolved since their divergence from a shared ancestor with the genus Pan . The “runaway” component of runaway selection is especially important because it can help to account, via positive feedbacks, for the extraordinary rapidity of human cognitive and social evolution.

Arenas of social selection help to indicate the mechanisms of social interactions that can lead to the enhanced brains and more-complex social cognition and emotion that characterize humans. They represent pairs or larger sets of human interactions, ordered by sex and age and number and nature of groupings, that have been postulated to involve runaway effects. The key question in particular is how runaway social selection is expected to work in fluid populations of early humans, in terms of how increased social abilities can translate into higher inclusive fitness of individuals, and enhanced survival and proliferation of groups, and in terms of evolutionary dynamics across generations.

The First Arena: Arms Races

Most generally, runaway social selection effects within generations can be driven by social competition, social cooperation or social choice. The first arena of social selection described here is direct, symmetric and asymmetric arms-race competition within a group , whereby two or more individuals are engaged in some fitness-related conflict where the individual with better social skills (social “weaponry”) wins. Such competition represents a classic arms race, where the selective pressure is autocatalytic across generations, because the selective cause is persistent and self-reinforcing. Arms races are normally thought of in physical terms, where they become limited, across many generations, by the costs of armament and tradeoffs with other components of fitness. This limitation may apply to brain size, over the long term, due to the high costs of neural tissue. However, psychological arms races per se should be subject to no such constraints, since they are governed by “software” — neural organization — that can, in principle, complexify indefinitely, and involve cumulative learning and culture. In hominins, brains concomitantly evolved to become both larger and more “socialized” (specialized for social cognition), with material culture lagging behind ( Geary, 2005 ; Gowlett et al., 2012 ; Rilling, 2014 ), as expected under autocatalytic models driven by social selection.

Possible examples of psychological arms races would be enhanced abilities to read emotions and intentions in others, levels in metacognition and theory of mind (“I think that they think that I think,” etc.), ability to gain status (and recruit partners in reciprocity, as discussed below) through displays of cognitive abilities, and skills involved in strategically “out-thinking” adversaries in conflicts (e.g., Byrne and Whiten, 1988 ; Dunbar, 2014 ). The arsenal of social weapons would also include a broad range of cognitive and emotional phenotypes whose expression reciprocally selects on each other within and across generations, mainly in the general context of motivating other individuals to behave more, and more often, in the inclusive-fitness interests of the actor. A number of studies have examined coevolutionary arms races in the context of social selection and intelligence (e.g., McNally et al., 2012 ; dos Santos and West, 2018 ; Coen, 2019 ), and have supported the conclusion that such arms races can drive higher levels of intelligence, cooperation, and social complexity. Indeed, as noted by Darwin (1871 , p. 97) “natural selection, arising from the competition of tribe with tribe, would, under favorable conditions, have sufficed to raise man to his high position.”

Among males, psychological arms races should be most prominent where males are evenly matched physically, such that simple muscular dominance cannot determine competitive outcomes. Among females, it may commonly involve indirect forms of (non-physical) aggression, such as manipulations of social status and abilities in competition for allies (friends) and social support. Competition for useful allies is also likely to have been very important to males.

Mental arms races are the psychosocial equivalent of Darwin’s (1871 ) sexual selection by male-male competition. Such simple, one-on-one psychological arms races should ramify easily into one-on-multiple and multiple-on-multiple interactions, given the fluidity of human groups and kin-structured and reciprocity-structured organizations. Indeed, the high fluidity and organizational complexity of human groups are likely, in part, end products of such arms races as well. The “multiples” of these interactions are presumably allies of some sort, who can join psychological and physical forces to better increase their inclusive fitness at the expense of others. Once a competitive dyad expands, however, the dynamics necessarily change. Indeed, a pair of competing individuals may themselves be allies in some domains and adversaries in others. How they became so is a matter for the next selective arena.

The Second Arena: Partner Choice

The second arena of social selection is partner choice in the context of cooperative traits, which can lead to runaway choice-trait coevolution under a variety of conceptual and mathematical models (e.g., Nesse, 2007 , 2010 ; Debove et al., 2015 ). Here, “partner” refers to social partners, with whom one preferentially interacts, over a relatively long period, due to various benefits that accrue especially via mutualism and reciprocity, sometimes combined with kinship. In the same way that classical sexually selected mate choice is driven by attractive displays by one individual to others, social partner choice is driven by attractive social displays, usually involving demonstrations of prosocial traits such as honesty, reliability, cooperative cultural and religious beliefs, and generosity through social, informational and material assistance in times of relative need or potential for benefit ( Nesse, 2007 , 2010 ). There are some interesting parallels here between social selection and classical Fisherian runaway selection (typically involving choice of a few “top-ranked” individuals) and “complementary” mate choice (such as for immune system gene compatibility). Thus, for example, social selection may entail tradeoffs between choosing the “best” partner (i.e., individuals that are highly socially intelligent or skilled in some area), versus choosing individuals who will remain committed to, and focus on, the relationship, even when mutualistic reciprocal relationships demand consistent effort and attention, limiting the ability to engage effectively in many of them. The latter type of relationship should also be promoted by complementarity of different social (and other) abilities between members of a dyad, which increases the benefits accruing to both.

Social partners are, of course, conventionally regarded as “friends,” who in evolutionary terms represent allies who both gain inclusive fitness benefits, over the longer term, from their continued association. The partnerships can be of any dyadic combination of the two sexes, or can involve larger groups united through multiple partner choice events, merging into and overlapping with other such groups in complex social networks. Choice of partners in various contexts can also generate “markets” for partners, with complex dynamics that can enhance the competitive nature of the processes involved ( Barclay, 2016 ; Eisenbruch and Roney, 2017 ). Smith and Apicella (2020) describe how partner choice, for traits that include generosity and foraging ability, mediates campmate preferences among Hazda hunter-gatherers.

The long duration of the human lifespan makes social partnerships, in principle, highly beneficial to inclusive fitness, especially if they involve complementary abilities, knowledge or resources ( Nesse, 2007 , 2010 ). In humans, choices regarding memberships in coalitional groups, based on the traits of the group and its leader, should also be notably important ( Boyd and Richerson, 2009 ), and could themselves synergize with group against group arms races, as discussed in more detail below.

Partner choice, like arms races, can result in the runaway evolution of socially selected traits. By this process, the expression of the chosen trait, and the choice of the trait, come to be positively genetically associated with one another across generations ( Sachs et al., 2004 ), as they both increase rapidly in frequency. Cultural analogs of this process can also lead to culturally inherited patterns of association that do not require genetic change, although such changes can themselves impose selection for genetic change and gene-culture coevolution ( Richerson and Boyd, 2005 ; Lotem et al., 2017 ).

Nesse (2007 , 2010) described how runaway partner choice may have promoted a whole suite of uniquely human or elaborated-in-humans traits, including theory of mind, extreme forms of cooperation, capacities for morality, the importance of building and protecting one’s reputation, and self-domestication of the recent human species as a whole. Most generally, partner choice, and choice of leaders and groups, should select for finer and finer abilities to discriminate the socially salient qualities of other individuals and groups, in terms of if and how much interacting reciprocally with them, compared to alternatives, will result in gains to inclusive fitness. In humans (and possibly dolphins) the complexity of cooperation (and the intelligence required to negotiate alliances) increased dramatically in the context of triadic interactions among groups in nested hierarchies ( Connor, 2007 ; Gerber et al., 2021 ). The risk/reward ratio and the number of options (and potential outcomes) were probably critical with regard to selection for enhanced intelligence, and these increased dramatically with expansion in the number of levels in a nested hierarchy of interacting entities (individuals, groups, groups within groups, etc.) and associated potential alliances. As such, this arena of social selection should result in notably enhanced abilities to judge character, truthfulness, morality and social abilities, as well as the ability to display and communicate these sorts of traits, even if they sometimes conflict with one’s ability to maximize inclusive fitness by alternative, relatively selfish, and self-serving means.

The Third Arena: Mate Choice

The third arena of social selection, human mate choice, was, of course, originally formalized by Darwin (1871) . It represents a subset of partner choice that is sufficiently special and distinct to warrant its own domain. Classical Fisherian mate choice by sexual selection ( Fisher, 1930 ; Lande, 1981 ; Kirkpatrick, 1982 ) involves a process whereby choice by one sex (in animals, usually females) for one or more fitness-related traits in the other sex (usually males) results across generations in a positive genetic correlation between stronger choice for the trait (typically in females) and higher level of trait expression (typically in males) — a runaway process that stops only when the trait is so highly developed that it incurs strong costs in terms of some other component of fitness, typically survival. This dynamic appears responsible, at least in part, for the rapid evolution and high diversity of sexually selected, mate-choice-related traits among many non-human animal groups ( Arnold, 1983 ).

Human mating systems have diverged substantially from the Fisherian paradigm, in that (a) females, as well as males, exhibit forms of sexually selected “beauty,” that may be chosen by the opposite sex; (b) mate choice is commonly more or less joint and reciprocal, with both sexes engaging in choice of a partner based on some criteria (though often with social constraints on choice); and (c) mate choice engenders relatively long-term pair-bonding, with mutual contributions to the rearing of offspring (e.g., Miller, 2000 ; Buss and Schmitt, 2019 ; Geary, 2021 ).

For human mate choice, the main considerations in the second arena apply, specifically, to the situation where males and females each choose one individual of the other sex by some criteria. Individuals are thus under selection to display socially selected traits (e.g., intelligence, cleverness, humor, conversational ability, kindness, a variety of social skills), and to choose some overlapping constellation of such traits in others ( Etcoff, 1999 ; Miller, 2000 ). Choice of a good opposite-sex partner for mating, reproducing, providing, and parenting is probably a much more challenging task than making and maintaining a same-sex friend, and thereby should represent a stronger socially selective filter. Pair-bonded males and females are thus selected to be able to successfully navigate the psyche of their mate, in a much more intimate, cognitively complex, and fitness-salient way than for friends. Indeed, among animals, comparative analyses by Dunbar and Shultz (2007) have demonstrated that among carnivores, artiodactyls and bats, larger relative brain size is associated with pair-bonding, and that in primates it is linked with complex, enduring social relationships even more broadly (as well as with larger group size); they argue that these findings reflect “the cognitive demands of the behavioral coordination and synchrony that is necessary to maintain stable pair-bonded relationships.” Humans appear to represent an extreme of social selection and bonding effects on relative brain size and behavioral coordination, especially given the partially divergent optimal mating and parenting strategies of the two sexes, and the complex mixtures of confluence and conflicts of inclusive-fitness interest that ensue.

From an evolutionary perspective, mutual mate choice in humans becomes subject to runaway dynamics due to genetic correlations of socially selected pair-bond related traits with choice of these traits; for example, females choose kind, caring males, males with genes for these traits are selected for, and the genes for the choice and the traits become associated and rise in frequencies across generations. For this type of sexual-social selection, it remains unclear if there are evolutionary brakes on the process equivalent to those operative during natural selection by predation against, say, too-large of a train in peacocks. Possibilities for such “brakes” might be over-expression of choice (such that no or few individuals are deemed suitable for a mate), or expression of prosocial, altruistic, or parenting-related traits to such a degree that they became maladaptive in the context of maximizing inclusive fitness as a whole.

The Fourth Arena: Caregiver-Offspring Choice and Signaling

Runaway coevolution between social signals and their choice includes not just cooperation partners, and female-male pairs, but also offspring interacting with their caregivers, specifically mothers, alloparents, and fathers ( West-Eberhard, 2003 , p. 467; Hrdy, 2013 ). By this mechanism, offspring benefit from producing signals, such as high levels of subcutaneous fat, vigorous crying, smiling, eye contact, and other social interactions with caregivers (“other-regarding” in Hrdy’s term), that represent indicators of their phenotypic and genetic “quality” (inclusive fitness value) and that prompt increases in feeding and engagements that enhance social-emotional cognition and learning. Such signals are expected to be predominantly honest indicators of offspring value, but may include manipulative elements ( West-Eberhard, 2003 ), as in other models of signal-receiver interaction, that could reinforce increased discriminability of cues by receivers. Social selection and evolution should thus increase maternal, alloparental, and paternal sensitivities to offspring cues (to better reward higher-value offspring, invest less in lower-value ones, and tell honest from dishonest signals), and increase offspring aptitudes and success at solicitation. As for the other forms of signal-choice system, the result is genetic and/or cultural correlations and coevolution by a self-reinforcing runaway process ( West-Eberhard, 2003 ).

Runaway social selection between caregivers and offspring represents an integral component of the human life history evolving toward increased alloparental and paternal care, shorter interbirth intervals and higher reproductive rates, larger-brained offspring (which are more expensive to produce and rear), and neural precocity and plasticity combined with physical altriciality ( Alexander, 1990 ; Hrdy, 2009 ; Piantadosi and Kidd, 2016 ; Sherwood and Gómez-Robles, 2017 ). Neural precocity, in turn, forms part and parcel of social precocity and the evolution of greatly enhanced human social and emotional cognition, the acquisition of which is inherently developmental and centers around the elongated human childhood and adolescence ( Bogin, 1990 ; Flinn et al., 2011 ; Ponzi et al., 2020 ). This selective arena is especially important given that infant mortality has long represented such a substantial component of variation in fitness among humans, and that such mortality can be reduced in a variety of socially salient ways, including effective offspring solicitation, alloparental contributions to maternal nutrition and infant care, paternal augmentation of food supplies, and broader social support in the group for mothers who warrant or earn it. For example, in many human groups, there is evidence of strong positive associations of lower child mortality with higher cognitive abilities of the mothers (e.g., Sandiford et al., 1997 ; Piantadosi and Kidd, 2016 ).

The Fifth Arena: Cultural Traits and Social-Cultural Learning

A final arena of within-group social selection is culture, the human-created material and information-based aspects of the environment that underpin tools, customs, religion, arts, and beliefs ( Flinn and Alexander, 1982 ). All human phenotypes derive from interactions of genes with environments — especially cultural and social ones — but culture is special because it can be transmitted both vertically (like genes, or language as noted by Darwin, 1871 ) and horizontally (as memes), with horizontal transmission potentially proceeding at a very rapid pace. As such, human traits can evolve due to “gene-culture coevolution” (interactions of genetically based human phenotypes with cultural aspects of environments; Laland and Seed, 2021 ), commonly due to differential human adoption and perpetuation of different cultural phenotypes and culture acting as a causal agent for selection ( Whiten et al., 2017 ; Richerson et al., 2021 ).

Cultural change can proceed under a runaway process, whereby increases in cultural complexity and sophistication (the “traits”) generate environments that select for enhanced social-cultural learning and more-effective adoption of cultural behaviors (the “choices”), especially by young individuals, leading to runaway coevolution (e.g., Alexander, 1979 ; Flinn and Alexander, 2007 ; Boyd and Richerson, 2009 ; Rendell et al., 2011 ; Legare and Nielsen, 2015 ; Legare, 2017 ; Lotem et al., 2017 ; Muthukrishna et al., 2018 ; Markov and Markov, 2020 ). This process, coupled with “arms race” elements of cultural change, may have been especially effective in driving the recent and accelerating human cultural change that has so complexified human social environments. As such, and given the cumulative nature of human cultural change, this arena of social selection should be exerting stronger and stronger effects on human evolution as time proceeds, relative to the other four ( Birch and Heyes, 2021 ; cf. Wadley, 2021 ).

The Final Arena: Between-Group Competition

The five arenas of runaway social selection described above all operate within human groups, where groups may be delineated by various terms including families, bands, villages, tribes, or ethnic, linguistic, or cultural groups, of any sizes, that each has some conception of “us” in relation to “them.” Whereas social selection and competition occur extensively within such groups, their evolution should be constrained by demographic, ecological and anti-cooperative effects that weaken the group in the context of their competitive interactions with other groups ( Lahti and Weinstein, 2005 ). As such, a final, higher-level arena of social selection operates between groups, as described by Darwin (1871) in terms of competition between human “tribes,” and by Alexander specifically in terms of runaway imbalances of power. In particular, Alexander posits, following from Darwin’s (1871 ) views that conflict between tribes selects for within-tribe cooperation and morality, that human social evolution has been driven, in large part, by group against group competition that selects for enhanced within-group cooperation as a means to counter external threats ( Alexander, 1979 , 1987 , 1990 , 2006 ). Alexander’s balance/imbalance of power model represents a form of arms race, with both cognition and culture as weaponry, that selects for larger and larger, more and more cooperative groups, with better and better ways to compete, though larger group size may also exacerbate within-group variation and conflicts and dilute the benefits of winning. Such between-group conflicts would have originated in hominins on small scales (indeed, presumably reminiscent of the “warfare” of common chimpanzees; Mitani and Watts, 2001 ), as represented now in some extant human societies (e.g., Berndt, 1964 ; Chagnon, 1977 ; Macfarlan et al., 2014 ), but escalating as populations increase in size.

Perhaps the most telling evidence in support of Alexander’s model is the observation that human history is, in considerable part, the history of human warfare based on groups defined by culture, language, and ultimately, genes ( Bowles, 2009 ; Turchin et al., 2013 ; Bauer et al., 2016 ; Turchin, 2016 ). Warfare may, however, represent only the most extreme, obvious and effective form of between-group human competition, since humans compete, and cooperate to compete, in fluid, dynamic groups at all levels from families to nations, and based on biological kinship, ethnic markers of diffuse long-term ancestry, and cultural differences represented by kinship that can be mainly or purely psychological ( Jones, 2003 ). Groups may also form on the basis of complementary skill sets or interests. In this context, the fluidity of human groupings, with shifting of alliances across time and space as a universality rather than exception, may connect the early evolution of hominids, lost in prehistory, with the recent evolution of modern, historic humans — and all points in between.

The six arenas of social selection described here each generates, given any degree of heritability, social evolution of sets of psychological traits and abilities that have collectively “made us human” ( Table 1 ). What is striking about these sets of phenotypes is that they encompass a tremendous range of human-elaborated psychological and social traits, many of which are expected to be reinforcing across arenas (e.g., honesty and morality in partner choice and among-group arms races), transferrable across domains (e.g., finer-scale social discrimination in caregiver choice, partner choice, mate choice), or complementary (e.g., abilities to compromise with, lead, persuade, or control other individuals). The effects of these interacting arenas of social selection echo the emphasis of Laland and Seed (2021) on “dynamical feedbacks between mutually reinforcing aspects of cognition,” with human cognitive uniqueness arising from “trait interactions and feedbacks,” with the salient traits evolving squarely in the context of complex sociocultural landscapes (see also Dean et al., 2013 ; Whiten, 2018 ; Lombard and Högberg, 2021 ; Spikins et al., 2021 ). As such, runaway social selection and evolution appear to exhibit the breadth, power and scope to help explain, in principle, how modern humans evolved psychologically from chimp-human ancestors.

Table 1. Phenotypes and abilities postulated to be selected for, in six different arenas of human social selection and evolution.

Uncovering the selective pressures that gave rise to the first hominins, and to modern humans, has been a perpetual challenge ever since Darwin drafted the first clear hypotheses of human origins in 1871. From a broad perspective, it makes sense that the most exceptional human features, large social brains, complex cooperative and competitive interactions, and elaborate culture, should themselves reflect the selective pressures that guided their evolution. The logic of runaway social selection suggests that humans generated and became their own primary selective pressures, through diverse forms of arms races within and between groups, and through choice-trait coevolutionary-dynamic interactions involving allies, mates, and offspring with caregivers. By the hypotheses presented here, each of these arenas of social selection drove the evolution of different, interacting dimensions of human sociality and culture, that merged to create the humans inhabiting our world today ( Figure 1 ). This hypothesis is by no means incompatible with those based on other selective pressures postulated to be important in human evolution, such as alloparental care ( Hrdy, 2009 ; van Schaik and Burkart, 2010 ) and self-domestication ( Wrangham, 2019 ), but it stresses the importance of runaway social selection as a potential key factor in how and why modern humans evolved.

Figure 1. The six arenas (five within groups, one between groups), of human runaway social selection. See Table 1 for further details on the specific phenotypes selected for in each arena.

Tracing the selective history of humans relies on one part evolutionary logic, one part ecology, one part psychology and neuroscience, one part anthropology, and all parts grounded in phylogeny and evidence from archeological remains. An enduring part of the puzzle, the initial divergence of the eventual Homo and Pan lineages, is addressed here with the hypothesis that the divergent evolution of hominins was “kick-started,” in an ape with a small (i.e., 400 cc) brain living at relatively low densities, by ecological conditions that favored increased fluidity, connectivity, tolerance, and especially local cooperation within and between small social groups. At first, such cooperation need not be sophisticated or complex, and need not involve larger brains. But when conditions eventually arose that allowed the evolution of larger and more complex brains (e.g., cooking of foods, and use of more energy-dense foods), and the first manifestations of culture, such early humans would have been poised to enter a socio-ecological niche characterized by increased population densities, larger brain sizes, enhanced competition and cooperation, cumulative culture, and strategic social choices, that collectively encompassed the multiple mechanisms of runaway social selection described above.

Increases in understanding of Darwin’s (1871 ) “insensible grading” from an apelike form to humans requires clear and specific hypotheses that make testable predictions. Indeed, a primary criticism of the runaway social selection model is that it lacks concrete empirical support, in terms of the specific processes involved. The mechanisms that underpin the hypotheses described here, especially those relating to choice-trait coevolution among allies, mates, and offspring-caregiver interactions, can, however, be evaluated in extant human groups, and ecological benefits from relatively simple primate alliances can be evaluated in field populations using the most relevant taxa. In particular, empirical evaluation of the hypotheses described here regarding social selection in human evolution will require testing for evidence of the operation of each proposed process and link in the causal positive-feedback cycles, especially in small-scale human societies. The hypotheses would thus be falsified by robust findings that the within-generation processes that underpin runaway social selection, in any given arena, do not occur in human societies or, if they occur, do not impact upon variation in inclusive fitness. Neurology and neuroendocrinology may also provide salient evidence of mechanisms for social competition ( Dunbar and Shultz, 2007 ; Rilling, 2014 ; Shultz and Dunbar, 2014 ), and the evolutionary transitions of brain evolution ( Sherwood et al., 2008 ; Sherwood and Gómez-Robles, 2017 ; Stout and Hecht, 2017 ; Bruner, 2021 ). Although the challenges inherent in all analyses of the broad scope of human evolution are daunting, the intellectual rewards remain profound, in better comprehension of just what we as a species are, and how we came to be.

Author Contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work, and approved it for publication.

This work received funding from Baylor University Department of Anthropology and the Natural Science and Engineering Research Council of Canada, Discovery Grant 2019–04208.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Alexander, R. D. (1974). The evolution of social behavior. Ann. Rev. Ecol. Syst. 5, 325–383.

Google Scholar

Alexander, R. D. (1979). Darwinism and Human Affairs. Washington, DC: University of Washington Press.

Alexander, R. D. (1987). The Biology of Moral Systems. New Delhi: Aldine.

Alexander, R. D. (1989). “Evolution of the human psyche,” in The Human Revolution , eds P. Mellars and C. Stringer (Chicago, IL: University of Chicago Press), 455–513. doi: 10.1002/ajp.22357

PubMed Abstract | CrossRef Full Text | Google Scholar

Alexander, R. D. (1990). How Humans Evolved: Reflections on the Uniquely Unique Species. St, Ann Arbor, MI: University of Michigan.

Alexander, R. D. (2006). The challenge of human social behavior. Evol. Psychol. 4, 1–32.

Allen-Arave, W., Gurven, M., and Hill, K. (2008). Reciprocal altruism, rather than kin selection, maintains nepotistic food transfers on an Ache reservation. Evol. Hum. Behav. 29, 305–318.

Almécija, S., Hammond, A. S., Thompson, N. E., Pugh, K. D., Moyá-Solá, S., and Alba, D. M. (2021). Fossil apes and human evolution. Science 372:eabb4363. doi: 10.1126/science.abb4363

Andrews, P. (2020). Last common ancestor of apes and humans: morphology and environment. Folia Primatol. 91, 122–148. doi: 10.1159/000501557

Antón, S. C., Potts, R., and Aiello, L. C. (2014). Evolution of early Homo: an integrated biological perspective. Science 345:1236828. doi: 10.1126/science.1236828

Apicella, C. L., Marlowe, F. W., Fowler, J. H., and Christakis, N. A. (2012). Social networks and cooperation in hunter-gatherers. Nature 481, 497–501. doi: 10.1038/nature10736

Arnold, S. J. (1983). “Sexual selection: the interface of theory and empiricism,” in Mate Choice , ed. P. Bateson (Cambridge: Cambridge University Press), 67–108.

Bailey, N. W., and Kölliker, M. (2019). Social runaway: Fisherian elaboration (or reduction) of socially selected traits via indirect genetic effects. Evolution 73, 1549–1563. doi: 10.1111/evo.13791

Barclay, P. (2016). Biological markets and the effects of partner choice on cooperation and friendship. Curr. Opin. Psychol. 7, 33–38.

Bauer, M., Blattman, C., Chytilová, J., Henrich, J., Miguel, E., and Mitts, T. (2016). Can war foster cooperation? J. Econ. Perspect. 30, 249–274. doi: 10.1257/jep.30.3.249

CrossRef Full Text | Google Scholar

Berndt, R. M. (1964). Warfare in the New Guinea highlands. Am. Anthropol. 66, 183–203.

Birch, J., and Heyes, C. (2021). The cultural evolution of cultural evolution. Philos. Trans. R. Soc. Lond. B Biol. Sci. 376:20200051. doi: 10.1098/rstb.2020.0051

Bissonnette, A., Perry, S., Barrett, L., Mitani, J. C., Flinn, M., Gavrilets, S., et al. (2015). Coalitions in theory and reality: a review of pertinent variables and processes. Behaviour 152, 1–56. doi: 10.1163/1568539x-00003241

Bogin, B. (1990). The evolution of human childhood. Bioscience 40, 16–25.

Bowles, S. (2009). Did warfare among ancestral hunter-gatherers affect the evolution of human social behaviors? Science 324, 1293–1298. doi: 10.1126/science.1168112

Boyd, R., and Richerson, P. J. (2009). Culture and the evolution of human cooperation. Philos. Trans. R. Soc. Lond. B Biol. Sci. 364, 3281–3288. doi: 10.1098/rstb.2009.0134

Bruner, E. (2021). Evolving human brains: paleoneurology and the fate of middle pleistocene. J. Archaeol. Method Theory 28, 76–94.

Buss, D. M., and Schmitt, D. P. (2019). Mate preferences and their behavioral manifestations. Annu. Rev. Psychol . 70, 77–110. doi: 10.1146/annurev-psych-010418-103408

Byrne, R. W., and Whiten, A. (1988). Toward the next generation in data quality: a new survey of primate tactical deception. Behav. Brain Sci. 11, 267–273.

Chagnon, N. A. (1977). Yanomamö: The Fierce People. New York, NY: Holt, Rinehart and Winston.

Chapais, B. (2009). Primeval Kinship: How Pair-Bonding Gave Birth to Human Society. Cambridge, MA: Harvard University Press.

Chapais, B. (2011). The deep social structure of humankind. Science 331, 1276–1277. doi: 10.1126/science.1203281

Coen, E. (2019). The storytelling arms race: origin of human intelligence and the scientific mind. Heredity 123, 67–78. doi: 10.1038/s41437-019-0214-2

Connor, R. (2007). Dolphin social intelligence: complex alliance relationships in bottlenose dolphins and a consideration of selective environments for extreme brain size evolution in mammals. Philos. Trans. R. Soc. Lond. B Biol. Sci.: Biol. Sci. 362, 587–602. doi: 10.1098/rstb.2006.1997

Coward, F., and Grove, M. (2011). Special issue: innovation and the evolution of human behavior beyond the tools: social innovation and hominin evolution. PaleoAnthropology 111, 129. doi: 10.4207/PA.2011.ART46

Crook, J. (1972). ““Sexual selection, dimorphism, and social organization in the primates,”,” in Sexual Selection and the Descent of Man, 1871-1971 , ed. R. Trivers (New Delhi: Aldine Pub Co.), 231–281.

Darwin, C. (1871). The Descent of Man, and Selection in Relation to Sex. London: John Murray.

Dean, L. G., Kendal, R. L., Schapiro, S. J., Thierry, B., and Laland, K. N. (2013). Identification of the social and cognitive processes underlying human cumulative culture. Science 335, 1114–1118. doi: 10.1126/science.1213969

Debove, S., André, J. B., and Baumard, N. (2015). Partner choice creates fairness in humans. Proc. R. Soc. Lond. B Biol. Sci. 282:20150392. doi: 10.1098/rspb.2015.0392

dos Santos, M., and West, S. A. (2018). The coevolution of cooperation and cognition in humans. Proc. R. Soc. B 285:20180723. doi: 10.1098/rspb.2018.0723

Dunbar, R. I. M. (1993). Coevolution of neocortical size, group size and language in humans. Behav. Brain. Sci. 16, 681–735.

Dunbar, R. I. M. (2014). “Mind the gap: or why humans aren’t just great apes,” in Paper Presented Lucy to Language: The Benchmark Papers , (Oxford: Oxford University Press).

Dunbar, R. I., and Shultz, S. (2007). Evolution in the social brain. Science 317, 1344–1347.

Dyble, M., Salali, G. D., Chaudhary, N., Page, A., Smith, D., Thompson, J., et al. (2015). Sex equality can explain the unique social structure of hunter-gatherer bands. Science 348, 796–798. doi: 10.1126/science.aaa5139

Eisenbruch, A. B., and Roney, J. R. (2017). The skillful and the stingy: partner choice decisions and fairness intuitions suggest human adaptation for a biological market of cooperators. Evol. Psychol. Sci. 3, 364–378.

Etcoff, N. (1999). Survival of the Prettiest: The Science of Beauty. Anchor books: New York, NY.

Fisher, R. A. (1930). The Genetical Theory of Natural Selection. Oxford: Oxford University Press.

Flinn, M. V. (2021). The creative neurons. Front. Psychol. 12:765926. doi: 10.3389/fpsyg.2021.765926

Flinn, M. V., and Alexander, R. D. (1982). Culture theory: the developing synthesis from biology. Hum. Ecol. 10, 383–400.

Flinn, M. V., and Alexander, R. D. (2007). “Runaway social selection,” in The Evolution of Mind , eds S. W. Gangestad and J. A. Simpson (New York, NY: Guilford Press), 249–255.

Flinn, M. V., Geary, D. C., and Ward, C. V. (2005). Ecological dominance, social competition, and coalitionary arms races: why humans evolved extraordinary intelligence. Evol. Hum. Behav. 26, 10–46.

Flinn, M. V., Nepomnaschy, P., Muehlenbein, M. P., and Ponzi, D. (2011). Evolutionary functions of early social modulation of hypothalamic-pituitary-adrenal axis development in humans. Neurosci. Biobehav. Rev. 35, 1611–1629. doi: 10.1016/j.neubiorev.2011.01.005

Foley, R., and Gamble, C. (2009). The ecology of social transitions in human evolution. Phil. Trans. R. Soc. B 364, 3267–3279. doi: 10.1098/rstb.2009.0136

Garg, K., Padilla-Iglesias, C., Restrepo Ochoa, N., and Knight, V. B. (2021). Hunter-gatherer foraging networks promote information transmission. R. Soc. Open Sci. 8:211324. doi: 10.1098/rsos.211324

Gavrilets, S., Duenez-Guzman, E. A., and Vose, M. D. (2008). Dynamics of alliance formation and the egalitarian revolution. PLoS One 3:e3293. doi: 10.1371/journal.pone.0003293

Geary, D. C. (2005). The Origin of Mind. Washington, DC: American Psychological Association.

Geary, D. C. (2021). Male, Female: The Evolution of Human Sex Differences , third Edn. Washington, D.C: American Psychological Association.

Geary, D. C., and Huffman, K. J. (2002). Brain and cognitive evolution: forms of modularity and functions of mind. Psychol. Bull. 128, 667–698. doi: 10.1037/0033-2909.128.5.667

Gerber, L., Wittwer, S., Allen, S. J., Holmes, K. G., King, S. L., Sherwin, W. B., et al. (2021). Cooperative partner choice in multi-level male dolphin alliances. Sci. Rep. 11:6901. doi: 10.1038/s41598-021-85583-x

Gowlett, J., Gamble, C., and Dunbar, R. (2012). Human evolution and the archaeology of the social brain. Curr. Anthropol. 53, 693–722.

Haber, A., and Corriveau, K. (2020). Putting social cognitive mechanisms back into cumulative technological culture: social interactions serve as a mechanism for children’s early knowledge acquisition. Behav. Brain. Sci. 43:E166. doi: 10.1017/S0140525X20000084

Harcourt-Smith, W. E. (2007). “The origins of bipedal locomotion,” in Handbook of Paleoanthropology , eds W. Henke and I. Tattersall (Berlin: Springer).

Hawkes, K. (2020). The centrality of ancestral grandmothering in human evolution. Integr. Comp. Biol. 60, 765–781. doi: 10.1093/icb/icaa029

Hawkes, K., O’Connell, J. F., and Blurton Jones, N. (2018). Hunter-gatherers & human evolution: a very selective review. Am. J. Phys. Anthropol. 165, 777–800.

Henrich, J., and McElreath, R. (2003). The evolution of cultural evolution. Evol. Anthropol. 12, 123–135. doi: 10.1002/evan.10110

Herrmann, E., Call, J., Hernàndez-Lloreda, M. V., Hare, B., and Tomasello, M. (2007). Humans have evolved specialized skills of social cognition: the cultural intelligence hypothesis. Science 317, 1360–1366. doi: 10.1126/science.1146282

Hill, K. R., Walker, R. S., Božièeviæ, M., Eder, J., Headland, T., Hewlett, B., et al. (2011). Co-residence patterns in hunter-gatherer societies show unique human social structure. Science 331, 1286–1289. doi: 10.1126/science.1199071

Hill, K. R., Wood, B. M., Baggio, J., Hurtado, A. M., and Boyd, R. T. (2014). Hunter-gatherer inter-band interaction rates: implications for cumulative culture. PLoS One 9:e102806. doi: 10.1371/journal.pone.0102806

Hrdy, S. B. (2009). Mothers and Others: The Evolutionary Origins of Mutual Understanding. Cambridge, MA: Harvard University Press.

Hrdy, S. B. (2013). Evolution’s Empress: Darwinian Perspectives on the Nature of Women. Oxford: Oxford University Press.

Hrdy, S. B. (2014). “Development + social selection in the emergence of ‘emotionally modern’ humans,” in New Frontiers in Social Neuroscience, Research and Perspectives in Neurosciences , eds J. Decety and Y. Christen (Berlin: Springer International Publishing), 57–92. doi: 10.1007/978-3-319-02904-7_5

Humphrey, N. K. (1976). “The social function of intellect,” in Growing Points in Ethology , eds P. P. G. Bateson and R. A. Hinde (Cambridge: Cambridge University Press), 303–317.

Irons, W. G. (1983). “Human female reproductive strategies,” in Social Behavior of Female Vertebrates , ed. S. Wasser (Cambridge, MA: Academic Press), 169–213.

Jones, D. (2003). The generative psychology of kinship. Part 1. cognitive universals and evolutionary psychology. Evol. Hum. Behav. 24, 303–319.

Kirkpatrick, M. (1982). Sexual selection and the evolution of female choice. Evolution 36, 1–12.

Lahti, D. C., and Weinstein, B. S. (2005). The better angels of our nature: group stability and the evolution of moral tension. Evol. Hum. Behav. 26, 47–63.

Laland, K., and Seed, A. (2021). Understanding human cognitive uniqueness. Annu. Rev. Psychol. 72, 689–716. doi: 10.1146/annurev-psych-062220-051256

Lande, R. (1981). Models of speciation by sexual selection on polygenic traits. Proc. Natl. Acad. Sci. U S A. 78, 3721–3725. doi: 10.1073/pnas.78.6.3721

Legare, C. H. (2017). Cumulative cultural learning: development and diversity. Proc. Natl. Acad. Sci. U S A. 114, 7877–7883. doi: 10.1073/pnas.1620743114

Legare, C. H., and Nielsen, M. (2015). Imitation and innovation: the dual engines of cultural learning. Trends Cogn. Sci. 19, 688–699. doi: 10.1016/j.tics.2015.08.005

Lombard, M., and Högberg, A. (2021). Four-field co-evolutionary model for human cognition: variation in the middle stone age/middle palaeolithic. J. Archaeol. Method Theory 28, 142–177.

Lotem, A., Halpernb, J. Y., Edelmanc, S., and Kolodnyd, O. (2017). The evolution of cognitive mechanisms in response to cultural innovations. PNAS 114, 7915–7922. doi: 10.1073/pnas.1620742114

Lovejoy, C. O. (1981). The origin of man. Science 211, 341–350.

Macfarlan, S. J., Walker, R. S., Flinn, M. V., and Chagnon, N. A. (2014). Lethal coalitionary aggression and long-term alliances among Yanomamö men. Proc. Natl. Acad. Sci. U S A. 111, 16662–16669. doi: 10.1073/pnas.1418639111

Markov, A. V., and Markov, M. A. (2020). Runaway brain-culture coevolution as a reason for larger brains: exploring the “cultural drive” hypothesis by computer modeling. Ecol. Evol. 10, 6059–6077. doi: 10.1002/ece3.6350

Marlowe, F. (2004). Marital residence among foragers. Curr. Anthropol. 45, 277–284.

McBrearty, S., and Brooks, A. S. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behavior. J. Hum. Evol. 39, 453–563. doi: 10.1006/jhev.2000.0435

McNally, L., Brown, S. P., and Jackson, A. L. (2012). Cooperation and the evolution of intelligence. Proc. R. Soc. B 279, 3027–3034. doi: 10.1098/rspb.2012.0206

McPherron, S. P., Alemseged, Z., Marean, C. W., Wynn, J. G., Reed, D., Geraads, D., et al. (2010). Evidence for stone-tool-assisted consumption of animal tissues before 3.39 million years ago at Dikika, Ethiopia. Nature 466, 857–860. doi: 10.1038/nature09248

Migliano, A. B., Battiston, F., Viguier, S., Page, A. E., Dyble, M., Schlaepfer, et al. (2020). Hunter-gatherer multilevel sociality accelerates cumulative cultural evolution. Sci. Adv. 6:eaax5913. doi: 10.1126/sciadv.aax5913

Migliano, A. B., Page, A. E., Gómez-Gardeñes, J., Salali, G. D., Viguier, S., Dyble, M., et al. (2017). Characterization of hunter-gatherer networks and implications for cumulative culture. Nat. Hum. Behav. 1:0043.

Miller, G. F. (2000). The Mating Mind: How Sexual Choice Shaped the Evolution of Human Nature. New York, NY: Doubleday.

Mitani, J., and Watts, D. (2001). Boundary patrols and intergroup encounters in wild chimpanzees. Behaviour 138, 299–327. doi: 10.1007/BF02629601

Muthukrishna, M., Doebeli, M., Chudek, M., and Henrich, J. (2018). The cultural brain hypothesis: how culture drives brain expansion, sociality, and life history. PLoS Comput. Biol. 14:e1006504. doi: 10.1371/journal.pcbi.1006504

Nakamaru, M., and Dieckmann, U. (2009). Runaway selection for cooperation and strict-and-severe punishment. J. Theoret. Biol. 257, 1–8. doi: 10.1016/j.jtbi.2008.09.004

Nesse, R. M. (2007). Runaway social selection for displays of partner value and altruism. Biol. Theory 2, 143–155.

Nesse, R. M. (2010). “Social selection and the origins of culture,” in Evolution, Culture, and the Human Mind , eds M. Schaller, A. Norenzayan, S. J. Heine, T. Yamagishi, and T. Kameda (London: Psychology Press), 137–150.

Osiurak, F., and Reynaud, E. (2020). The elephant in the room: what matters cognitively in cumulative technological culture. Behav. Brain. Sci. 43:E156. doi: 10.1017/S0140525X19003236

Piantadosi, S. T., and Kidd, C. (2016). Extraordinary intelligence and the care of infants. Proc. Natl. Acad. Sci. U S A. 113, 6874–6879. doi: 10.1073/pnas.1506752113

Ponzi, D., Flinn, M. V., Muehlenbein, M. P., and Nepomnaschy, P. (2020). Hormones and human developmental plasticity. Mol. Cell. Endocrinol. 505:110721. doi: 10.1016/j.mce.2020.110721

Rendell, L., Fogarty, L., and Laland, K. N. (2011). Runaway cultural niche construction. Philos. Trans. R. Soc. Lond. B Biol. Sci. 366, 823–835. doi: 10.1098/rstb.2010.0256

Richards, A. I. (1950). “Some types of family structure amongst the central Bantu,” in African Systems of Kinship and Marriage , eds A. R. Radcliffe-Brown and D. Forde (Oxford: Oxford University Press), 207–251.

Richerson, P. J., and Boyd, R. (2005). Not by Genes Alone: How Culture Transformed Human Evolution. Chicago, IL: University of Chicago press.

Richerson, P. J., Gavrilets, S., and de Waal, F. B. (2021). Modern theories of human evolution foreshadowed by Darwin’s Descent of Man. Science 372:eaba3776. doi: 10.1126/science.aba3776

Rilling, J. K. (2014). Comparative primate neuroimaging: insights into human brain evolution. Trends Cogn. Sci. 18, 46–55. doi: 10.1016/j.tics.2013.09.013

Sachs, J. L., Mueller, U. G., Wilcox, T. P., and Bull, J. J. (2004). The evolution of cooperation. Q. Rev. Biol. 79, 135–160.

Sandiford, P., Cassel, J., Sanchez, G., and Coldham, C. (1997). Does intelligence account for the link between maternal literacy and child survival? Soc. Sci. Med. 45, 1231–1239. doi: 10.1016/s0277-9536(97)00042-7

Sherwood, C. C., and Gómez-Robles, A. (2017). Brain plasticity and human evolution. Ann. Rev. Anthropol. 46, 399–419.

Sherwood, C. C., Subiaul, F., and Zawidzki, T. W. (2008). A natural history of the human mind: tracing evolutionary changes in brain and cognition. J. Anat. 212, 426–454. doi: 10.1111/j.1469-7580.2008.00868.x

Shultz, S., and Dunbar, R. I. M. (2014). “The social brain hypothesis: an evolutionary perspective on the neurobiology of social behaviour,” in Lucy to Language: The Benchmark Papers , eds R. I. M. Dunbar, C. Gamble, and J. A. J. Gowlett (Oxford: Oxford University Press), 53–69. doi: 10.1017/S2045796021000135

Smith, D., Dyble, M., Thompson, J., Major, K., Page, A. E., Chaudhary, N., et al. (2016). Camp stability predicts patterns of hunter-gatherer cooperation. R. Soc. Open Sci. 3:160131. doi: 10.1098/rsos.160131

Smith, K. M., and Apicella, C. L. (2020). Partner choice in human evolution: the role of cooperation, foraging ability, and culture in Hadza campmate preferences. Evol. Hum. Behav. 41, 354–366.

Spikins, P., French, J. C., John-Wood, S., and Dytham, C. (2021). Theoretical and methodological approaches to ecological changes, social behaviour and human intergroup tolerance 300,000 to 30,000 BP. J. Archaeol. Method Theory 28, 53–75. doi: 10.1007/s10816-020-09503-5

Stout, D., and Hecht, E. E. (2017). Evolutionary neuroscience of cumulative culture. Proc. Natl. Acad. Sci. U S A. 114, 7861–7868. doi: 10.1073/pnas.1620738114

Strassmann, J. E., and Queller, D. C. (2010). The social organism: congresses, parties, and committees. Evolution 64, 605–616. doi: 10.1111/j.1558-5646.2009.00929.x

Tomasello, M. (1999). The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press.

Turchin, P. (2016). Ultrasociety. Chaplin, CT: Beresta Books.

Turchin, P., Currie, T. E., Turner, E. A. L., and Gavrilets, S. (2013). War, space, and the evolution of Old World complex societies. Proc. Natl. Acad. Sci. U S A. 110, 16384–16389.

van Schaik, C. P., and Burkart, J. M. (2010). “Mind the gap: cooperative breeding and the evolution of our unique features,” in Mind the Gap: Tracing the Origins of Human Universals , eds P. M. Kappeler and J. B. Silk (Berlin: Springer), 477–496.

Varella, M. A. C. (2018). The biology and evolution of the three psychological tendencies to anthropomorphize biology and evolution. Front. Psychol. 9:1839. doi: 10.3389/fpsyg.2018.01839

Wadley, L. (2021). What stimulated rapid, cumulative innovation after 100,000 years ago? J. Archaeol. Method Theory 28, 120–141. doi: 10.1007/s10816-020-09499-y

Walker, R. S., Hill, K., Flinn, M. V., and Ellsworth, R. (2011). Evolutionary history of hunter-gatherer marriage practices. PLoS One 6:e19066. journal.pone.0019066 doi: 10.1371/journal.pone.0019066

Washburn, S. L., and Lancaster, C. S. (1968). “The evolution of hunting,” in Man the Hunter , eds R. B. Lee and I. DeVore (Milton Park: Routledge), 293–303.

Watts, D. P. (2010). “Dominance, power, and politics in nonhuman and human primates,” in Mind the Gap , eds P. M. Kappeler and J. B. Silk (Berlin: Springer), 109–138. doi: 10.1371/journal.pone.0025117

West-Eberhard, M. J. (1979). Sexual selection, social competition, and evolution. Proc. Am. Philos. Soc. 123, 222–234.

West-Eberhard, M. J. (1983). Sexual selection, social competition, and speciation. Q. Rev. Biol. 58, 155–183.

West-Eberhard, M. J. (2003). Developmental Plasticity and Evolution. Oxford: Oxford University Press.

Whiten, A. (2018). “The evolution and ontogeny of ‘Deep Social Mind’ and the social brain,” in Minnesota Symposia in Child Development: Development of the Social Brain , eds J. Ellison and M. Serra (Hoboken, NJ: John Wiley & Sons, Inc).

Whiten, A., Ayala, F. J., Feldman, M. W., and Laland, K. N. (2017). The extension of biology through culture. Proc. Natl. Acad. Sci. U S A. 114, 7775–7781.

Wrangham, R. W. (1999). Evolution of coalitionary killing. Yearb. Phys. Anthropol. 42, 1–30. doi: 10.1002/(sici)1096-8644(1999)110:29+<1::aid-ajpa2>3.3.co;2-5

Wrangham, R. W. (2019). Hypotheses for the evolution of reduced reactive aggression in the context of human self-domestication. Front. Psychol . 10:1914. doi: 10.3389/fpsyg.2019.01914

Keywords : social selection, human evolution, cooperation, runaway processes, cumulative culture

Citation: Crespi BJ, Flinn MV and Summers K (2022) Runaway Social Selection in Human Evolution. Front. Ecol. Evol. 10:894506. doi: 10.3389/fevo.2022.894506

Received: 11 March 2022; Accepted: 13 May 2022; Published: 02 June 2022.

Reviewed by:

Copyright © 2022 Crespi, Flinn and Summers. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Mark V. Flinn, [email protected]

† These authors have contributed equally to this work and share first authorship

This article is part of the Research Topic

A 150 Years’ Celebration of Darwin’s Book on Human Evolution and Sexual Selection: Its Legacy and Future Prospects

Runaway Selection

• Living reference work entry
• First Online: 25 August 2017
• Cite this living reference work entry

• Laura M. Travers 3  

2925 Accesses

2 Citations

Fisherian selection

Runaway selection is a mechanism whereby a secondary sexual trait expressed in one sex becomes genetically correlated with a preference for the trait in the other sex. The genetic coupling of the trait and the preference leads to self-reinforcing loops of coevolution between the trait and preference for the trait. This process is known as runaway selection and can lead to accelerated evolution of exaggerated traits and preferences.

Introduction

One of the greatest challenges to Darwin’s theory of evolution by natural selection was the presence of exaggerated male traits (Darwin 1859 ). Such traits encompass a wide variety of elaborate visual, acoustic, chemical, and behavioral characteristics (e.g., tail of the peacock or courtship song in birds). These traits appear to contradict Darwin’s idea that selection acts on traits that increase survival. Elaborate ornaments and displays seem maladaptive to their bearer as they are costly to maintain...

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Institutional subscriptions

Andersson, M. B. (1994). Sexual selection . Princeton: Princeton University Press.

Google Scholar  

Arnqvist, G., & Nilsson, T. (2000). The evolution of polyandry: Multiple mating and female fitness in insects. Animal Behaviour, 60 (2), 145–164.

Article   PubMed   Google Scholar  

Bakker, T. C. M. (1993). Positive genetic correlation between female preference and preferred male ornament in sticklebacks. Nature, 363 , 1237–1266.

Article   Google Scholar  

Bakker, T. C. M. (1999). The study of intersexual selection using quantitative genetics. Behaviour, 136 (9), 1237–1266.

Bakker, T., & Pomiankowski, A. (1995). The genetic basis of female mate preferences. Journal of Evolutionary Biology, 8 (2), 129–171.

Curtsinger, J. W. (1991). Sperm competition and the evolution of multiple mating. American Naturalist, 138 , 93–102.

Darwin, C. (1859). On the origin of species by means of natural selection or the preservation of favoured races in the struggle for life . London: John Murray.

Book   Google Scholar  

Darwin, C. (1871). The descent of man, and selection in relation to sex . London: John Murray.

Fisher, R. A. (1930). The genetical theory of natural selection: A complete variorum edition . Oxford: Oxford University Press.

Folstad, I., & Karter, A. J. (1992). Parasites, bright males, and the immunocompetence handicap. The American Naturalist, 139 (3), 603–622.

Grafen, A. (1990). Sexual selection unhandicapped by the fisher process. Journal of Theoretical Biology, 144 (4), 473–516.

Greenfield, M. D., Alem, S., Limousin, D., & Bailey, N. W. (2014). The dilemma of Fisherian sexual selection: Mate choice for indirect benefits despite rarity and overall weakness of trait-preference genetic correlation. Evolution, 68 (12), 3524–3536.

Houde, A. E. (1994). Effect of artificial selection on male colour patterns on mating preference of female guppies. Proceedings of the Royal Society of London B: Biological Sciences, 256 (1346), 125–130.

Jennions, M. D., & Petrie, M. (2000). Why do females mate multiply? A review of the genetic benefits. Biological Reviews, 75 (1), 21–64.

Keller, L., & Reeve, H. K. (1995). Why do females mate with multiple males? The sexually selected sperm hypothesis. Advances in the Study of Behavior, 24 , 291–315.

Kirkpatrick, M. (1982). Sexual selection and the evolution of female choice. Evolution, 36 (1), 1–12.

Kirkpatrick, M., & Barton, N. H. (1997). The strength of indirect selection on female mating preferences. Proceedings of the National Academy of Sciences, 94 (4), 1282–1286.

Kirkpatrick, M., & Ryan, M. J. (1991). The evolution of mating preferences and the paradox of the lek. Nature, 350 (6313), 33–38.

Lande, R. (1981). Models of speciation by sexual selection on polygenic traits. Proceedings of the National Academy of Sciences, 78 (6), 3721–3725.

Lüpold, S., Manier, M. K., Puniamoorthy, N., Schoff, C., Starmer, W. T., Buckley Luepold, S. H., Belote, J. M., & Pitnick, S. (2016). How sexual selection can drive the evolution of costly sperm ornamentation. Nature , 533 (7604), 535–538.

Mead, L. S., & Arnold, S. J. (2004). Quantitative genetic models of sexual selection. Trends in Ecology & Evolution, 19 (5), 264–271.

Parker, G. A. (1970). Sperm competition and its evolutionary consequences in the insects. Biological Reviews, 45 (4), 525–567.

Prokop, Z. M., & Drobniak, S. M. (2016). Genetic variation in male attractiveness: It is time to see the forest for the trees. Evolution, 70 (4), 913–921.

Qvarnström, A., Brommer, J. E., & Gustafsson, L. (2006). Testing the genetics underlying the co-evolution of mate choice and ornament in the wild. Nature, 441 (7089), 84–86.

Ritchie, M. G., Saarikettu, M., & Hoikkala, A. (2005). Variation, but no covariance, in female preference functions and male song in a natural population of Drosophila montana . Animal Behaviour, 70 (4), 849–854.

Sharma, M. D., Wilson, A. J., & Hosken, D. J. (2016). Fisher’s sons’ effect in sexual selection: Absent, intermittent or just low experimental power? Journal of Evolutionary Biology, 29 (12), 2464–2470.

Simmons, L. W. (2001). Sperm competition and its evolutionary consequences in the insects . Princeton: Princeton University Press.

Simmons, L. W., & Kotiaho, J. S. (2007). Quantitative genetic correlation between trait and preference supports a sexually selected sperm process. Proceedings of the National Academy of Sciences, 104 (42), 16604.

Taylor, P. D., & Williams, G. C. (1982). The lek paradox is not resolved. Theoretical Population Biology, 22 (3), 392–409.

Taylor, M. L., Wedell, N., & Hosken, D. J. (2007). The heritability of attractiveness. Current Biology, 17 (22), 959–960.

Travers, L. M., Simmons, L. W., & Garcia-Gonzalez, F. (2016). Additive genetic variance in polyandry enables its evolution but polyandry is unlikely to evolve through sexy or good sperm processes. Journal of Evolutionary Biology, 29 (5), 918–928.

Zahavi, A. (1975). Mate selection—A selection for a handicap. Journal of Theoretical Biology, 53 (1), 205–214.

Download references

Author information

Authors and affiliations.

Centre for Ecology and Evolution, University of Exeter, Cornwall, UK

Laura M. Travers

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Laura M. Travers .

Editor information

Editors and affiliations.

Oakland University , Rochester, Michigan, USA

Jennifer Vonk

Department of Psychology, Oakland University Department of Psychology, Rochester, Michigan, USA

Todd Shackelford

Section Editor information

Averett University, Danville, VA, USA

David Hanbury

Rights and permissions

Reprints and permissions

Copyright information

© 2017 Springer International Publishing AG

About this entry

Cite this entry.

Travers, L.M. (2017). Runaway Selection. In: Vonk, J., Shackelford, T. (eds) Encyclopedia of Animal Cognition and Behavior. Springer, Cham. https://doi.org/10.1007/978-3-319-47829-6_430-1

Download citation

DOI : https://doi.org/10.1007/978-3-319-47829-6_430-1

Received : 01 August 2017

Accepted : 03 August 2017

Published : 25 August 2017

Publisher Name : Springer, Cham

Print ISBN : 978-3-319-47829-6

Online ISBN : 978-3-319-47829-6

eBook Packages : Springer Reference Behavioral Science and Psychology Reference Module Humanities and Social Sciences Reference Module Business, Economics and Social Sciences

• Publish with us

Policies and ethics

• Find a journal
• Track your research

• Subscriber Services
• For Authors
• Publications
• Archaeology
• Art & Architecture
• Bilingual dictionaries
• Classical studies
• Encyclopedias
• English Dictionaries and Thesauri
• Language reference
• Linguistics
• Media studies
• Medicine and health
• Names studies
• Performing arts
• Science and technology
• Social sciences
• Society and culture
• Overview Pages
• Subject Reference
• English Dictionaries
• Bilingual Dictionaries

Recently viewed (0)

• Save Search

Michael Allaby

• Find at OUP.com
• Google Preview
• Share This Facebook LinkedIn Twitter

More on this Topic

• View overview page for this topic

Related Content

In this work.

handicap principle

• natural selection

Related Overviews

sexual selection

View all related overviews »

• Publishing Information
• General Links for this Work
• Preface to the Third Edition
• Endangered animals
• The universal genetic code
• Geologic time-scale
• SI units (Système International d’Unités)
• Next Version

runaway hypothesis  

A hypothesis proposed by R. A. Fisher (1890–1962) in 1930 to explain the consequences of female selection of ...

Access to the complete content on Oxford Reference requires a subscription or purchase. Public users are able to search the site and view the abstracts and keywords for each book and chapter without a subscription.

Please subscribe or login to access full text content.

If you have purchased a print title that contains an access token, please see the token for information about how to register your code.

For questions on access or troubleshooting, please check our FAQs , and if you can''t find the answer there, please contact us .

• Oxford University Press

PRINTED FROM OXFORD REFERENCE (www.oxfordreference.com). (c) Copyright Oxford University Press, 2023. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single entry from a reference work in OR for personal use (for details see Privacy Policy and Legal Notice ).

date: 29 May 2024

• Cookie Policy
• Privacy Policy
• Legal Notice
• Accessibility
• [66.249.64.20|185.80.151.9]
• 185.80.151.9

Character limit 500 /500

RUNAWAY SELECTION

Hypothesis of female partner preference suggesting that particular attributes in males are sexually appealing to females, which select partners with such attributes and thus make certain any male offspring is likewise appealing to females, regardless of hereditary superiority. Compare with: good gene hypothesis.

Leave a Reply

Your email address will not be published. Required fields are marked *

Latest Posts

A Brief History of Brainwashing: The Science of Thought Control

A Deep Dive into the Social Psychology of Leadership

Counseling Approaches to Client Care: Theories to Apply in Practice

The Future Of Education: Can You Earn A Psychology Degree Online?

Insomnia & Mental Illness: What is the Correlation?

Stop Guessing: Here Are 3 Steps to Data-Driven Psychological Decisions

Getting Help with Grief: Understanding Therapy & How It Can Help

Exploring the Psychology of Risk and Reward

Understanding ADHD in Women: Symptoms, Treatment & Support

Meeting the Milestones: A Guide to Piaget's Child Developmental Stages

Counseling, Therapy, and Psychology: What Is The Difference?

The Psychology of Metaphysical Belief Systems

Popular psychology terms, medical model, hypermnesia, affirmation, brainwashing, backup reinforcer, message-learning approach, affiliative behavior, behavioral modeling, approach motivation.