Angry (Hungry) Birds

Note: This blog post is a joint Fatima Aboul-Seoud/Rob Kurzban production.

For some reason, I recently was interested in the relationship between food and aggression, and so I had a look at an old paper published in 2000 entitled “Effects of short-term hunger and competitive asymmetry on facultative aggression in nestling black guillemots Cepphus grylle.” In the research reported in this paper, the authors, Cook et al. were interested in something called the “food amount hypothesis.” Suppose that two siblings are being provisioned by parents. When times are good, inclusive fitness implies that the chicks will be, more or less, cooperative. When times are tough, however, the logic of parent/offspring conflict kicks in because the marginal benefit of food to me is now greater, so relationships between even full siblings might turn violent. As they put it, the prediction is that “dominant siblings become more aggressive during periods of food shortage, thereby obtaining a disproportionate share of total available parental resources.”

The happy couple.

The happy couple.

Black guillemots (pronounced: berdz) are a convenient model species because eggs are laid in pairs over a period of a few days. The authors set out to determine if the amount of food delivered was a proximate cue for sibling aggression. To do this, they recorded videos of 20 broods (5 in the control group and 15 in the experimental group). In both groups, they began observation before the eggs were laid and recorded when each egg was laid and hatched. The first chick to hatch was marked and called the A-chick, and the second chick to hatch was called the B-chick.

The experiment looked at what would happen if parents were prevented from feeding their chicks. The experimental manipulation, which is sort of awesome, is described this way: “[P]arental provisioning was prevented… by placing an adult scaring device—either balloons with painted eyes or a fiberglass great black-backed gull—near the entrance of the nest and in view of the returning adult.” So, for six hours out of a twelve hour observing cycle, parents were scared away from feeding their chicks.

Figure 1. Attacks by condition, over time.

Figure 1. Attacks by condition, over time.

In the control group, chicks were generally non-aggressive. (See Figure 1, from the original manuscript.)  In the experimental group, during period 1 (before the scaring device was introduced), rates of attacks between experimental groups and controls didn’t differ. Aggression rates were higher during period 2 (when the scaring device was present) and highest during period 3 (just after the device was removed).  The researchers also measured provisioning and, indicating the utility of aggression, after the deprivation phase of the study, A-chicks got a relatively larger fraction of feeds from parents relative to controls. The authors conclude (p. 285)”

A-chick aggression in black guillemot broods was elevated only after parental provisioning rates were experimentally reduced. When parental provisioning resumed, adults did not adjust their feeding rate in response to changes in their offsprings’ requirements, and the consequence of the intersibling aggression was a skew in the distribution of food received by siblings in favor of the dominant A-chick.

So, in summary, in these organisms, it appears that having less food causes increased aggression, even in the context of what would otherwise be a close, even cooperative, relationship. These results illustrate that aggression is deployed conditionally, in a principled way, depending on context.

Note that these results do not illustrate, at the risk of repeating myself, that chicks require parental provisioning in order to have the fuel necessary to muster the willpower not to harm their siblings. That would be an obviously silly way to understand these results. Right?

Anyway, part of the reason I looked at this paper was that I recently saw a talk by Johannes Haushofer, who ran studies investigating the effects of direct transfers of money to very poor households in Africa. Very generally, poor households endowed with more resources (compared to controls not so endowed) had (a little) better food security – because additional funds were used to buy food – and, in addition, domestic violence against women went down, as reported in a summary of the work. These results suggest (but of course can’t causally establish) that having less food leads to more aggression in the context what would otherwise be a close, even cooperative, relationship.

Oh, and, speaking of humans, and in completely unrelated other news, the Economist and other news outlets are reporting on a new paper in Proceedings of the National Academy of Sciences that shows a relationship between having less food and increased aggression in the context of what would otherwise be a close, cooperative relationship, the theme illustrated so well by the international “You’re Not You When You’re Hungry” Snickers ads, such as this one.


Cook, M. I., Monaghan, P., & Burns, M. D. (2000). Effects of short-term hunger and competitive asymmetry on facultative aggression in nestling black guillemots Cepphus grylle. Behavioral Ecology, 11(3), 282-287.

15. April 2014 by kurzbanepblog
Categories: Blog | 7 comments

Comments (7)

  1. Yeah, this seems pretty obvious. The same behavior appears to play a role in cross species cooperation as well. Really it appears that most animals will cooperate or at least behave nicely toward others as long as their basic needs are met, i.e. lions befriending goats, cats befriending mice and birds, etc. as has been observed to happen among captive animals from time to time.

    What’s interesting is the role of scarcity in cooperation. There appears to be some kind of bell curve.

    It seems that when resources are very scarce, then individuals of many species, including humans, become aggressive and anti-social. However, when there is low, but manageable scarcity then individuals are cooperative in ways to reduce scarcity. Yet, when scarcity is very low, or almost eliminated, then individuals become aggressive again.

    • It’s striking to me that humans often become aggressive or anti-social VERY shortly after they begin to feel hungry or thirsty or tired (as in, within minutes), and not just when resources are scarce in the long term. Certainly in the case of humans, it seems like a bad design to immediately turn on potential cooperators, allies, family members just because your body’s needs are suddenly not being perfectly met – wouldn’t it be better to recruit the help of these potential cooperators to find food, shelter etc. together rather than alienate them?

      I’ve observed these effects among my well-nourished friends and family. I can only imagine how much more severe the effects must be when energy isn’t as readily accessible. Does it make sense for individuals of a highly cooperative species to pursue an ‘every man for himself’ strategy as soon as one feels a twinge of hunger?

      • That’s an interesting question, Rachel. I wonder if our crappy modern diet has something to do with how common this problem seems to be. I’m not sure you’d see such wild blood sugar swings in people eating a more traditional diet, without all the super easily digestible carbs we eat today.

        Actually, I notice differences just within myself in how I respond to hunger when it’s clear my blood sugar has dropped like a stone (cortisol spike!) and when my diet has been better (more fat, protein, and fiber, less simple carbs). With the first, I get pretty touchy. With the second, I just feel slowly increasing hunger, with very little desire to kill nearby cooperators.

  2. Variability in the food source is the reason for laying eggs at different times. In good years, all the chicks survive; in bad years, at least one chick survives instead of all of them starving. Aggression by the biggest chick when it is hungry is a reasonable evolutionary behavior.

  3. Slightly off topic (though relevant to the PNAS paper you link to at the end): I notice that the editor for this paper is a frequent collaborator of the first author and is the graduate advisor of the second author. I’m genuinely curious about the “norms” in the field – do you (Rob, or any other readers) know if this is “normal,” or do editors typically decline handling such manuscripts in situations like this because of a conflict of interest (either real or only perceived by third parties)?

    Perhaps I’m misunderstanding something, though; I don’t see Roy Baumeister on the PNAS editorial board, so maybe the “edited by Roy Baumeister” note at the top does not refer to editorial decisions (i.e., selecting reviewers and making the final decision to accept versus reject the manuscript).

  4. Pingback: Angry (Hungry) Birds | Evolutionary Psychology ...

Skip to toolbar