Evolutionary Psychology Blog Archive

What Are Social Scientists Learning About Social Learning?

Two papers came out this past month on social learning, which seems to be an area in which there has been a slight recent uptick in attention. The first, by Morgan et al., is entitled “The evolutionary basis of human social Learning” and appeared in Proceedings of the Royal Society – B.  Their interest was in the strategies that people use to make decisions when they can do so using information available from the social world, in this case, what others have done when faced with the same choice that they themselves have to make. When do people use others’ choices to guide their own?

In the studies reported, subjects faced one of four tasks. For instance, in the “foraging task,” subjects had to forage for apples, picking one of two possible sites about which they had information; their task (presented by computer) was to pick the one that had the larger expected value of apples in it. Information about the right answer could come from one of two sources, either in the form of 1) a number of presentations of draws from each site’s distribution, or from 2) information about which of the two sites other people had chosen, with the former constituting the “social learning” choice. (You can think of this as choosing to go fishing in a pond based on how often you catch fish there, or basing your choice on which ponds other people choose to fish at.) Other tasks were a mental rotation task, a length-estimation task, and an auditory pitch discrimination task.

Their particular interest was in the factors that influenced when people used social information, such as how many other people chose a particular option, the degree of consensus in others’ observed choices, how hard the task was, and so on. So, for example, in, say, the foraging task, subjects might be provided information that five other subjects had chosen patch A over patch B, whereas three subjects had made the reverse choice. How does the information about others choices influence individuals’ choices?

Summarizing, the pattern of data suggest (roughly) the following: in these tasks, people are more likely to use social information when making a choice as the number of social models increases, as the consensus of the models they observe goes up, when individual learning is costly, when the models observed are doing particularly well, and when the subject’s own confidence in their answer goes down. These sorts of effects are predicted by various models that are likely to be familiar to readers of this blog.

(As a side note, to return to the Dennett quote from last time, my guess is that this work won’t be attacked as unscientific, post-hoc or storytelling because the ideas at stake here, having to do with social learning, don’t have the same political heat as issues other research programs have. “The evolutionary basis of human social learning” is a title that won’t raise hackles. But if one were to change the last two words – “conflict,” “aggression,” “sex differences” – well, that might change things…)

(A second side note, this time on terminology. A couple of posts ago, a commenter quoted Laland and Brown regarding their favored use of the terms “adaptation” and “adaptive,” with this passage:

“An adaptation is a character favored by natural selection for its effectiveness in a particular role. It is something that has evolved to fulfill that function, and it may or may not be adaptive in the current environment. Adaptive behavior is functional behaviour that increments reproductive success, and it may or may not be an adaptation.”

I mention this because this article — which includes Laland as an author –  uses the term “adaptively” and its variants in several places, including the abstract:

The number of demonstrators, consensus among demonstrators, confidence of subjects, task difficulty, number of sessions, cost of asocial learning, subject performance and demonstrator performance all influenced subjects’ use of social information, and did so adaptively.

My sense is that what they mean here is that people used social information in a way that made them more likely to choose the correct answer, and they render this as “adaptive” behavior.)

The second recent paper appeared in Evolution and Human Behavior, entitled “Prestige-biased cultural learning: Bystander’s differential attention to potential models influences children’s learning,” and investigated the possibility that young children use cues to prestige when making decisions about which model to imitate.

I thought the manipulation of prestige was particularly interesting. In the first study, 4-year-olds were shown a little clip in which two people looked either at a model to the right of them or to the left of them, the attention serving as the cue to which of the two models was the more prestigious. The two models then engaged in various behaviors – playing with different toys, for instance – and the young subjects were later asked to choose which of they toys they wanted to play with. The prediction was that children would choose to play with the one the “high prestige” model played with, which indeed turned out to be the case. This implies that four year olds monitor who others are attending to, using that information to shape their own subsequent choices.

The second study in this paper was slightly more complex. I won’t describe it in detail, but only mention that the most interesting result, to me, is the “cross domain” finding. If a child observes a model being attended to while exhibiting a food preference, the child follows the model’s food choice. However, the child is not more likely to show a preference for the toy that this individual played with. In other words, young children seem to infer that these models are domain-specific experts, which is pretty cool.

And completely unrelated… a recent paper in the Journal of Evolutionary Biology entitled, “Sick ants become unsociable” shows that ants infected with a fungus change their behavior, including spending most of their time outside the nest, possibly the result of an adaptation designed to reduce the communication of the pathogen to others. The authors conclude: “Our results provide evidence for the evolution of unsociability following pathogen infection in a social animal and suggest an important role of inclusive fitness in driving such evolution.” I just thought it was interesting, in part because anti-parasite adaptations are just really cool, and also I like these examples in biology journals in which data from behavioral experiments are considered (ho-hum, no big deal) relevant to an evolutionary claim.

References

Bos N., Lefèvre T., Jensen A. & D’Ettorre P. (2012) Sick ants become unsociable. Journal of Evolutionary Biology, 25, 342-351.

Morgan, T. J. H., Rendell, L. E., Ehn, M., Hoppitt, W., & Laland, K. N. (2012). The evolutionary basis of human social learning. Proceedings of the Royal Society – B, 279, 653-662.

Chudek, M., Heller, S., Birch, S. & Henrich, J. (2012) Prestige-Biased Cultural Learning: Bystander’s Differential Attention to Potential Models Influences Children’s Learning. Evolution and Human Behavior, 33, 46-56.

20. January 2012 by kurzbanepblog
Categories: Blog | 5 comments