Patterns in the dusk

Via Bruce Schneier, it’s impressive the non-existent patterns that our minds will find in the environment:

During the war, rumors began to filter back to Britain about a German “engine-stopping ray.”2 The site of the supposed misadventure was invariably near a television tower. Jones relates,

As usually reported, the phenomenon consisted of a tourist driving his car on one of the roads in the vicinity, and the engine suddenly ceasing to operate. A German Air Force sentry would then appear from the side of the road and tell him that it was no use his trying to get the car going again for the time being. The sentry would, however, return and tell him when he would be able to do so. The sentry appeared in due course, and the engine started.3

The linked article suggests that in times of stress, people are more likely to find false patterns; the stress and fear are undermining our ability to think clearly.  The ‘stress hormone’ that people typically think of is cortisol, but I can’t find any papers linking cortisol to finding false patterns – or apophenia as I’ve learned it is called.  That’s probably because it’s a hard thing to search for; ‘cortisol false patterns’ will just result in articles about patterns of cortisol!  So if anyone has something knowledgable to say, please pass that knowledge on!

Feeding behaviors: the ultimate paper

I’ve been busy as a bee lately, so I have only just gotten time to start reading papers again.  I saw this guy about feeding behavior and, wow, that is what I call a real paper.  I was thinking of writing it up but I saw that someone else got to it first, and I don’t think that’s a discussion of the paper that I’ll be able to beat.  So go read it!

Testosterone and social aggression

Testosterone will probably always be linked in people’s minds with aggressive behaviors, but its role in behavior is a source of controversy.  Why it rises when it does – and whether it causes aggression or merely responds to it – is not clear, although recent studies that directly inject testosterone into an animal has begun to clarify things a little.  A little.  But we still don’t know whether increases in testosterone cause increases in aggression.

Birds are very territorial animals that have evolved signals to prevent the need from actually fighting over territory (too much).  Some of them have pretty fun ways to show off, too; the male red grouse grows a comb over its eyes almost like a rooster, with larger combs being a way to show off dominance.  Don’t want to mess with that guy, he’s got a really large red thing on his head!  In many animals, dominance and aggressive behaviors go hand-in-hand.  If your fellow red grouse is going to beat you silly over and over again, you’ll probably let him do what he wants.  But if aggression and testosterone are somehow related, and if aggression and dominance are somehow related, does testosterone control dominance?  We can start to answer this question by looking at red grouse comb size.

It turns out that testosterone kind of controls comb size.  When Vergara and Martinez-Padilla injected birds with testosterone, over time they grew larger combs, just as one would presuppose.  However, if you go ahead and stick a needle in the other birds – the ones that didn’t get the testosterone injection – they also had an increase in testosterone level.  In fact, they had the same level of testosterone in their bloodstream as those with the injection.  But these birds had smaller combs!  The lack of relationship between testosterone level and comb size (ie, dominance) shows that it doesn’t control displays of dominance.  Perhaps it’s a measure of one individual’s testosterone versus the population’s?  Or perhaps the birds who had increased testosterone started a feedback cycle in the population where everybody needed more testosterone, but only those birds with the initial increase got the rewards from it?

So testosterone levels per se don’t mediate (historical) dominance.  It’s also not at all clear whether testosterone is needed for aggression.  Take the example of another bird, the dark-eyed junco.  Dark-eyed juncos will defend their territory by squawking at unwanted intruders.  Some researchers like to go out and mess with these poor juncos by placing a little toy bird equipped with a speaker in the junco territory.  The result is a lot of swooping in, harassment and yelling at the decoys, in a futile attempt to make them go away.  And if you measure junco testosterone levels before and after introducing the fake bird, you’ll see the levels increase.  Now, if you go and put out just a speaker box – but not a fake bird – the local birds will still come out and and harass the speaker; they’ll approach close up and scream and throw a general fit.  But their testosterone levels don’t increase!  When Rosvall et al examined the baseline testosterone levels, though, it seems like these levels are (very weakly) correlated with how quickly the birds will quickly they’ll approach the speaker and how close they’ll get.  They can see this in a PCA analysis, but don’t mention the raw data so I don’t know if that’s significant; and it certainly isn’t for number of songs.  Just looking at the data makes me worry about outliers but who knows?  But at least for certain behaviors, testosterone surges aren’t associated with aggressive social behavior.

Aggressive behaviors clearly do not all require increases in testosterone, but perhaps it depends on the type of behavior.  In the Rosvall study, maybe the birds did not think they were in physical danger because they could never actually see the darn thing.  And maybe testosterone isn’t the right thing to measure at all!  Androgen and oestregon receptors regulate other genes after being activated by testosterone, and aromatase is a step required for testosterone to be useful.  If you measure the mRNA levels of these genes, they are significantly correlated with aggressive behaviors.  So the behavioral difference between individuals may not be the result of higher levels of the hormone itself, but levels of the receptor.  And since this is measured so quickly after the birds are observed to react to the false bird for six minutes (a mean of 4.5 minutes after the behavior), it seems unlikely that the transcript levels are changing because of the fake bird, but rather was there beforehand.

What role does testosterone play in aggression?  It doesn’t seem to be causing it, because aggressive behaviors happen without changes in testosterone.  It certainly doesn’t control dominance, because you can get the same increase in testosterone levels in birds with different levels of dominance displays.  But this doesn’t mean that testosterone is unrelated to aggression, because it could be receptor level that causes differences in behavior; one bird can do more with the testosterone it has than another.

Here’s my current working hypothesis: testosterone is more important as a way to prepare the body for the effects of aggression than for aggression itself.  Look at the studies above – the researchers only detected surges in testosterone level when it seemed like the bird saw another bird – when it thought it would need to be prepared to fight.  And since testosterone’s effect seems to be changes in gene transcription rather than direct depolarization of neural membranes, it is the role of the receptors to prepare the brain to respond to a different level of environmental aggression.  But hey, I’m new to the testosterone field so we’ll see what I think next week 😉


Rosvall KA, Bergeon Burns CM, Barske J, Goodson JL, Schlinger BA, Sengelaub DR, & Ketterson ED (2012). Neural sensitivity to sex steroids predicts individual differences in aggression: implications for behavioural evolution. Proceedings. Biological sciences / The Royal Society, 279 (1742), 3547-55 PMID: 22673360

Vergara P, & Martínez-Padilla J (2012). Social context decouples the relationship between a sexual ornament and testosterone levels in a male wild bird. Hormones and behavior PMID: 22841824

KA Rosvall, DG Reichard, SM Ferguson, DJ Whittaker, & ED Ketterson (2012). Robust behavioral effects of song playback in the absence of testosterone or
corticosterone release Hormones and Behavior DOI: 10.1016/j.yhbeh.2012.07.009

Photo from