Blute Blog

In an article “What is individual quality? An Evolutionary perspective” published in Trends in Ecology and Evolution 25(4) 207-204 in 2009 here, Wilson and Nussey reviewed the fact that the meaning of individual “quality” is commonly ambiguous in various biological literatures. Much of content of the article was an argument for better integration of (evolutionary) ecology with quantitative genetics. For better or worse, the vast majority of evolutionary (including behavioural) ecologists have opted for the “phenotypic gambit” as Alan Grafen famously called it. Presumably that is because they suspect that the genetic basis of the vast majority of the traits they are interested in, particularly behavioural ones, are not, and probably never will be known. That is an issue that cuts broadly across evolutionary ecology as does the issue of individual quality (e.g. as in offspring quantity versus quality), but the cuts intersect rather than coincide. Nevertheless, it is easy to agree with the article’s overall recommendation that “there is a need for authors to state more explicitly what they mean by individual quality in any given context”.

To reflect on this, consider some simple principles of evolutionary ecology blending foraging and life history theories – evolution under different densities, scales, in patchy, or uncertain environments (discussed slightly differently in Darwinian Sociocultural Evolution Chpt. 3).

1) Density (Figure 1): Assume spatio-temporal boundaries are fixed, but the amount of ecological resources varies. Ceteris paribus, low density relative to resources favours consuming/producing offspring individually while high density favours doing so by means of social interaction. As I noted in a recent review here, “after all, if most of the resources remain available in the ecological environment, it is towards it that one’s efforts should be directed; but if most have been absorbed into the population, then it is towards other population members that one’s efforts should be directed. The social interaction under high densities can be antagonistic (involving contact and aggression) or cooperative (based on economies of scale or specialization and exchange)” or a mixture of both. Given that sex is the most fundamental social relationship that exists among unrelated peers, it is not surprising that Graham Bell was able to show in 1982 in The Masterpiece of Nature that the strongest correlate of sex in animals is crowding. He showed that this is the case both intraspecifically (the easiest way to evoke sex in facultively sexual organisms is to crowd and starve them) and interspecifically (he characterized species-rich environments associated with sex like the tropics versus the poles, low versus high altitudes, and the ocean versus freshwater etc. as “biotically complex”).

FIGURE 1: DENSITY; BOUNDARIES FIXED, AMOUNT OF RESOURCES VARY

Low Density (Plentiful Resources) Versus High Density (Scarce Resources)

2) Scale (Figure 2): Assume the opposite i.e. the amount of resources are fixed but boundaries vary. Resources concentrated in time, space or niche favour fast, specialized consumption/production strategies; smeared out ones favour longer but slower, more generalized strategies.

FIGURE 2: SCALE; BOUNDARIES VARY, AMOUNT OF RESOURCES FIXED

Small Scale (Concentrated Resources) Versus Large Scale (Smeared Out Resources)

3) Patchiness: Assume both – boundaries and the amount of resources are fixed but resources may still be patchily distributed and individuals may be differently located with respect to them. Low local densities favour consumption/production, while high local densities favour 3M’s (maintenance/motility/mutability)/offspring 3M’s. For example, the most obvious way to devote resources to somatic maintenance is to digest more rather than eat more, thus deriving more breakdown products from each unit of resources consumed; and the most obvious way to devote resources to reproductive maintenance is to produce fewer, larger offspring, thus deriving more grand-offspring from each offspring produced. (On some of the complexities of the differences between 1) and 3) see footnote below.)

FIGURE 3: PATCHINESS; BOTH FIXED BUT PATCHILY DISTRIBUTED

4) Uncertainty: Simple uncertainty between any of these pairs of conditions favours bet hedging (doing both at random with some fixed probability); uncertainty with reliable cues favours adaptive phenotypic plasticity.

It seems natural to distinguish the alternative strategies in 3) informally as ones of “quantity” versus “quality”. (I have sometimes done so, usually resisting the temptation to put quality in scare quotes, but hopefully always in a context which makes the meaning explicit.). However, these three distinctions (maintenance, motility, mutability) are obviously not all substantively the same thing nor are they favoured under exactly the same conditions. What they do have in common is that they are all economic distinctions. They distinguish between spending on individual/demographic growth here/now versus investing in them for the future, whether in the current or in some other place/niche.

In the short run, selection acts in the short run but we also know that in the longer run it acts in the longer run. Given that the longer run is just the sum of a number of short runs, how is it possible for what is favoured in the two to be different? As long as it does not result in extinction, sacrificing quantity in the short run is possible if it is more than compensated for in some later short run. Hence quality in the short run is quantity in the longer run. What is favoured ultimately (assuming additive genetic variance) is what maximizes the net rate of individual/demographic growth across all temporal-spatial-niche regions occupied by the units subject to selection.

In some cases, quality may appear to be quantity right in the here and now, just less obviously so. Even if of the same size (volume/mass), different food prey/patches may differ in quantity even of energy content – an ounce of steak contains more calories than one of fruit and one of fruit more than one of foliage for example. Similarly among offspring – some of similar size might be constituted so as to be capable of producing more grand-offspring than others. Hence a fifth principle of evolutionary ecology seems needed: if the variance in relevant properties among resource items/offspring is low, it pays to consume/produce as much as possible; when it is high it pays to choose “quality”. On the other hand, this variance based principle may be viewed as just another way of talking about spending versus investing in a patchy environment. After all, how does one “choose” quality except by waiting (maintenance), moving (motility), or innovating (mutability)/provisioning offspring with these capabilities?

Going back to what selection maximizes, we know from complex life cycles and probably eusocial colonies that the units subject to selection can extend at least across more than one generation/conventional individual. However, how to identify these units in general remains a matter of much discussion and debate as anyone familiar with the modern literature on levels of selection, major transitions, etc., is aware.

A footnote on some complexities re the difference between 1) and 3):

One above is about absolute density while three is about local density under patchiness – but are they really different given that the upper left cells in 3) match the cells in 1) for example? Actually, the high density social strategies in 1) are liable to result in catastrophe – either because the population destroys itself through social conflict or because cooperation results in it hitting the carrying capacity of the environment under acceleration rather than deceleration, and subsequently crashing. However, if niche construction obtained i.e. if the ecological environment not only structured the population but the population also constructed the ecological environment, the outcome could be different. With niche construction included, low densities favour ecological strategies which construct high densities depleting ecological resources but giving social ones an opportunity to recover. Similarly, high densities favour social strategies which construct low densities depleting social resources but giving ecological ones an opportunity to recover. Were such a scenario to prevail, the likely outcome sooner or later would be negative frequency-dependence. Population members would evolve to anticipate ecological consequences and evolve instead to respond directly to frequencies in the population so that ecological strategies would be favoured if social ones were common and social ones favoured if ecological ones were common resulting in a stable equal allocation of effort in the population to the two alternatives.

In any event, by contrast, in 3) local density changes/differences are exogenous, imposed by ecological forces external to the interactions between the population and resource(s) at issue – for example, by seasonality. As a consequence, they can be responded to by the appropriate alternative – whether consuming/producing at low densities or 3M’s/offspring 3M’s at high. This still raises the question of whether, even given that at high local densities resources are expected to be plentiful elsewhere/in the future/in some other niche, could social strategies not evolve anyway? I think that is possible and the combination of the high density socially-oriented strategies from 1) and the ecologically-oriented ones from 3), whether simultaneous or sequential, may be what results in complex life cycles although that is another big topic which may be addressed in a future post.

(Revised, November 29, 2011.)