The differential biology reader

I am moving the blog to a new address at http://reader.differentialist.info. If you are subscribed to the RSS feed, it will begin redirecting automatically.

Because I had my own symposium yesterday on animal personality, I missed lots of great looking talks.

I am a sucker for methodology and pretty pictures so I quite enjoyed Niels Waller talk on fungible weights in regression. Waller used geometric principles to demonstrate that, while ordinary least-squares will also give the best fit to plain linear data, there are actually an infinite series of alternative solutions to a regression equation that give almost as good a fit. The paper inclues R code.

Other than that there is a lot of bashing about on the general factor of personality. There's a symposium this afternoon devoted to the idea that a single, underlying factor explains most of the variance in all personality traits. A lot of the argument is methodological. One side says that it is just an artifact of factor analyzing noise (to put it mildly) while the other side claims that of course you won't find it in personality inventories that have been designed to produce orthogonal factors. For me, the solution is to consider why (and when) we might expect a species to evolve a single underlying factor.

Waller. Fungible Weights in Multiple Regression. Psychometrika (2008) vol. 73 (4) pp. 691-703

I love when papers explore a really basic conceptual issue you'd never considered before that just makes so much sense:

In heterogeneous environments, sex-biased dispersal could lead to environmental adaptive parental effects, with offspring selected to perform in the same way as the parent dispersing least, because this parent is more likely to be locally adapted. We investigate this hypothesis by simulating varying levels of sex-biased dispersal in a patchy environment. The relative advantage of a strategy involving pure maternal (or paternal) inheritance is then compared with a strategy involving classical biparental inheritance in plants and in animals. We find that the advantage of the uniparental strategy over the biparental strategy is maximal when dispersal is more strongly sex-biased and when dispersal distances of the least mobile sex are much lower than the size of the environmental patches. In plants, only maternal effects can be selected for, in contrast to animals where the evolution of either paternal or maternal effects can be favoured. Moreover, the conditions for environmental adaptive maternal effects to be selected for are more easily fulfilled in plants than in animals.

This might have implications for parent–offspring resemblance in personality. Daughters in species with male-dispersal should be more like their mothers.

Anyway, any paper that can manage the axis labels "pollen/male dispersal" and "seed/female dispersal" is pretty cool in my book.

Emmanuelle Revardel, Alain Franc, and Remy J Petit. Sex-biased dispersal promotes adaptive parental effects BMC Evolutionary Biology 2010, 10:217 doi:10.1186/1471-2148-10-217

photo cc-by Jan Charles Linus Ekenstam

A coauthor had an great idea that we're going to try. While Creative Commons is an ideal for all aspects of scientific communication, there are still many, many good journals that either don't have open access policies or have high fees for making your work available this way. However, it is less common that a scholar will have the need to reproduce their work in ways that lie outwith fair use (such as distributing personal or classroom copies of an article). Out of all the content that makes up an article, the parts that do tend to find their way into other publications are figures and tables. What usually needs to happen is that authors has to seek permission from the publisher to reproduce the figure, even if said author was the one who originally drew it!

A CC license would allow us to bypass the publisher when incorporating figures and tables into future work and will let others reuse and build upon our figures as well. Effectively, we'll be asking the publisher to seek permission from us for the original use of the figure, a figure we'll provide to them under a Creative Commons Attribution license.

My hope is they won't have a problem with this. Let's call this a pragmatic baby step toward full open access.

Photo cc-by agentkevinski

I don't really know if the personality dimensions we record are "out there" or are primarily functions of perceivers. One of the things I am trying to do is see what happens when we take current trait theories seriously when thinking about how human personality came to be. The trick though, is not to think about our immediate ancestors but to look at our evolutionary cousins to see how else we might have turned out.

Nonhuman primates exhibit remarkably similar (although not exactly the same) trait dimensions as humans. Chimpanzees, one of our two closest living relatives, are primarily characterized by a personality trait called Dominance that binds up a lot of what appears in humans as agreeableness and neuroticism. Dominant chimps are not necessarily high in rank, but rather are savvy navigators of their social world (think Yereon). Orangutans, unlike chimpanzees and humans, don't have a distinct conscientiousness dimension but instead exhibit a blend of openness and conscientiousness called intellect.

If personality is about what we perceive, this at first seems pretty damaging to observer ratings of personality. Anthropomorphism is an old charge in ethology that is still ready to pop out at every corner. But the data just don't back up to this explanation, since it fails to explain the abundant interrater reliability and heritability of these traits (for example, a parent and child orangutan will have very similar personalities even though they live in different zoos and are observed by different sets of raters).

Perhaps the more likely implication is that we are picking with these up the latent traits that are most relevant to the common social sphere that primates inhabit. Examining the ecology of the species would indicate what might be missed from this picture, such as the personality of noctural prosimians or monkeys with highly seasonal mating behavior.

Srivastava. The Five-Factor Model Describes the Structure of Social Perceptions. In press, Psychological Inquiry
King and Figueredo. The Five-Factor Model plus Dominance in Chimpanzee Personality. J Res Pers (1997) vol. 31 pp. 257-271
Weiss et al. Personality and subjective well-being in orangutans (Pongo pygmaeus and Pongo abelii). J Pers Soc Psychol (2006) vol. 90 (3) pp. 501-11

The German photographer August Sander (1876-1964) pioneered the typological method with his seminal body of work, People of the Twentieth Century, a vast series of portraits, each of which is classified according to the profession and role of Sander’s individual subjects, in an aim to ‘provide a true psychology of our time.’

Combining an analytical discipline and ordering of his images, with a rigorous and objective approach to the photographs themselves, Sander produced each of his portraits from the same neutral distance, with the sequences dependent on the accumulation of mass for the similarities to emerge between each photograph. (Wayne Ford)

The central idea is that disorders are networks that consist of symptoms and causal relations between them. In a nutshell, what binds, say, the set of depression symptoms, is that they are thus connected through a dense set of strong causal relations. With regard to comorbidity, such a network approach presents a radically different conceptualization of comorbidity, in terms of direct relations between the symptoms of multiple disorders.

In contrast to existing perspectives, it is inappropriate to say that the symptoms measure the disorder in question.

There are lots of algorithms and metrics to throw at the problem once you start thinking of it this way.

Cramer et al. Comorbidity: A network perspective. Behavioral and Brain Sciences (2010), 33: 137-150 doi:10.1017/S0140525X09991567

photo cc-by clineclines

a rigid refusal to 'anthropomorphize' may have its scientific disadvantages. Obviously, the true objection to anthropomorphism is not to discovering a similarity of mechanism in human and animal behavior, but to inventing similarities that do not exist. A complete rejection of all concepts derived from experience with man would leave a vacuum in animal psychology, for the closer we come to man in the phylogenetic scale the more evident it is that some quite complex modes of human behavior occur in animals and one cannot help recognizing that the overt pattern is the same in each case. A discussion of jealousy in the earthworm is obvious nonsense, but not in primates.

 

Hebb. Emotion in man and animal: an analysis of the intuitive processes of recognition. Psychol Rev (1946) vol. 53 (2) pp. 88-106

Studies based on the Gallup-Healthways Well-Being Index of the US are starting to trickle out. Researchers have used this data to replicated a well-known finding: global measures of well-being are U-shaped, peaking when you are young and old, with the inflection point at around age 54. Hedonic measures of negative affect (such as stress and worry), on the other hand, consistently decrease as we age.

(download)

Click here to download:

Dont_worry_get_older.zip (199 KB)

When I first studied evolution as a sophmore in college the final exam included an open-ended question about whether humans were still evolving. I cannot remember what I wrote—that was before I really understood how to think about evolution—but my answer was probably all muddled.

Recent genetic evidence has gotten us used to the idea that humans have undergone strong and rabid selection regimes over the past 10,000 years. But there is still much to adapt to:

Infectious diseases have the potential to act as strong forces for genetic selection on the populations they affect. We find that an HIV infection similar to that currently affecting sub-Saharan Africa could not yet have caused more than a 3 per cent decrease in the proportion of individuals who progress quickly to disease. Such an infection is unlikely to cause major genetic change until 400 years have passed since HIV emergence. However, in very severely affected populations, there is a chance of observing such major genetic changes after another 50 years.

Despite the uncertainty in future climate-change impacts, it is often assumed that humans would be able to adapt to any possible warming. Here we argue that heat stress imposes a robust upper limit to such adaptation. Peak heat stress, quantified by the wet-bulb temperature TW, is surprisingly similar across diverse climates today. TW never exceeds 31 °C. Any exceedence of 35 °C for extended periods should induce hyperthermia in humans and other mammals, as dissipation of metabolic heat becomes impossible. While this never happens now, it would begin to occur with global-mean warming of about 7 °C, calling the habitability of some regions into question.

Cromer et al. How fast could HIV change gene frequencies in the human population? doi: 10.1098/rspb.2009.2073 Proc. R. Soc. B 7 July 2010 vol. 277 no. 1690 1981-1989

Sherwood & Huber. An adaptability limit to climate change due to heat stress. doi: 10.1073/pnas.0913352107 PNAS May 25, 2010 vol. 107 no. 21 9552-9555

Graphic by Karl Gerstner

Two perspectives are out this week on where animal personality has been and where it is going. Both show that the field of animal personality is gaining enough critical mass that review articles on a single taxon (primates) are backed by enough literature to start drawing broad conclusions and that there are enough workers in the field to make it worthwhile to advance conceptual shifts.

Freeman and Gosling come from a psychological slant to tackle primates and they synthesize work on all the different personality factors quite nicely:

Our analyses of the literature indicate that some domains (e.g., sex, age, rearing conditions) are more evenly represented in the literature than are others (e.g., species, research location). Studies examining personality structure (e.g., with factor analysis) have identified personality dimensions that can be divided into 14 broad categories, with Sociability, Confidence/Aggression, and Fearfulness receiving the most research attention. Analyses of the findings pertaining to inter-rater agreement, internal consistency, test–retest reliability, generally support not only the reliability of primate personality ratings scales but also point to the need for more psychometric studies and greater consistency in how the analyses are reported. Overall, the validity data present a somewhat mixed picture, suggesting that high levels of validity are attainable, but by no means guaranteed.

Recent studies of animal personality have focused on its proximate causation and its ecological and evolutionary significance, but have mostly ignored questions about its development, although an understanding of the latter is highly relevant to these other questions. One possible reason for this neglect is confusion about many of the concepts and terms that are necessary to study the development of animal personality. Here, we provide a framework for studying personality development that focuses on the properties of animal personality, and considers how and why these properties may change over time.

Stamps and Groothuis. The development of animal personality: relevance, concepts and perspectives. Biological Reviews (2010) vol. 85 (2) pp. 301-325 doi:10.1111/j.1469-185X.2009.00103.x

graphic via kevindooley