Sociality and Cognitive Innovation -- Part 1

July 20, 2007

A short piece in Science magazine (May 15, 2009, p857) caught my eye . With the strange title of “6 heads are better than 2”, it told of a study on the relationship between group numbers and problem solving ability in house sparrows, the most common bird we see around our houses in the suburbs. The article attracted my attention because I had had the pleasure a couple of weeks ago of watching a house sparrow feed its two “toddlers” at my back yard bird feeder. The young sparrows were already capable of some flight, they were large birds, larger in fact than the parent, but apparently they had not yet learned how to feed themselves. They sat on the railing of my deck while the mamma bird (or pappa) took seeds from the feeder, hopped down to the railing, popped the seeds into an open chirping mouth, and returned for more. With two mouths two feed, this went on for some time, several days in a row.

The brief article reported that house sparrows (Passer domesticus), which are gregarious creatures, are able to better solve problems when in larger groups than when in small groups. But the fascinating aspect of this is that the birds were not working together. Rather, individuals in the group solved problems better if the group were larger.

I found myself caught up in trying to understand why.

Birds were captured and brought in from the wild to an experimental aviary where they had access to food from a feeder, presumably not unlike the feeder in my backyard. The birds were then given a problem -- how to obtain food from the feeder when access to the seeds was blocked by a transparent lid. Clever birds, they learned to dislodge the cap, opening the window to the food.

The important finding however was that birds in groups of six learned to remove the cap “10 times as quickly as smaller groups of two birds”. This indicated that individual birds in larger groups are “swifter at solving new problems than in smaller groups”. In addition, birds collected from urban environments were faster at problem solving than their rural counterparts.

One thing was immediately apparent; for house sparrows it is good to be in larger groups, because ALL of the birds in the group were able to access the food more readily once any ONE of the birds in the group could make it available. (And each would likely learn how to overcome that same problem if it arose again). Since birds in larger groups can access food faster, there is a clear evolutionary advantage to being in a larger group. For house sparrows, as for many other species of organism, it is good to be part of a larger group.

But could it also be that the real significance of this study is that it is good to be in larger groups because larger groups provides conditions for eliciting greater brain power among the individuals in the group? This possibility fascinated me. If so, that would be a remarkable finding. It would mean that what was important about being in a group is not so much that access to food improves, but that problem-solving intelligence improves in larger groups, and that socially induced intelligence provides for a variety of different conditions – food access being one – that make group dynamics a clear evolutionary winner. If so, it seemed to me, then this could speak to the early origins of the evolution of consciousness, and provide an understanding of why the phenomenon of consciousness, once it begins, could evolve so rapidly. How so? If the brain is constructed such that social interaction fosters the use of innate intelligence, and if active intelligence fosters the success of the social group, then one could also imagine that that problem solving could operate to better organize the group (eg foster co-operativity), which in turn would better elicit individual intelligence, etc. A positive feedback system could be constructed that could lead, inexorably, to the evolutionary development of the conscious mind.

Is that possible? Is it possible that simply being part of a larger group can somehow automatically elicit greater intelligent behavior in the individual members of the group? Is that why so many people enjoy living in cities? Does living in large groups provide greater fulfillment of our individual potential?

But I jump ahead of myself. This is all based on a big IF, and that is IF social interactions foster greater problem-solving ability in the members of the group, ie that native intelligence is evoked more when in larger groups. This is something that the brief article hinted at when it said that, “the pattern [of quicker problem-solving] was consistent across all individuals in the group”. If I understood that sentence correctly, then all of the birds in the larger group are above average, as Garrison Kiellor quips.

But I am not sure I understand that sentence correctly. What bothers me is that I can’t figure out how the investigators would know that. Once one of the birds solved the problem, all of the birds in the group could have learned that solution by observation. There seems to me to be no obvious way to know whether all of the birds in the larger groups were equally capable of more rapid problem-solving, or not.

So, we need to back up and consider some of the other possibilities for how birds in larger groups could be better at problem solving than birds in smaller groups.

One possibility can be ruled out. The article is not talking about cooperation. In this situation, the birds do not work together to remove the transparent caps from the feeder holes. Also, it can’t be based on experience. Experience could account for the greater success of urban birds compared to country birds, but would not explain why urban birds in smaller groups are not as good in solving problems as urban birds in larger groups. Besides, among the non-urban birds, the birds were set to task without accruing cap-opening experience.

Perhaps we are dealing with a statistical issue here. With more birds at work, it is likely that the larger group would get to the food faster. Or maybe, in a larger group it is more likely that one individual among the group was slightly more clever than average. The brief article in Science doesn’t mention these possibilities, and indeed, they do not seem consistent with the comment that, “the pattern was consistent across all individuals in the group”.

Further confusing things, (to my mind) the article concludes by saying, “Increased success at problem-solving in larger groups may reflect a wider diversity of experience and skill among the individuals in the group… “. But wait… Experience and skill are two different things. Experience depends upon history and learning. Skill can be learned or innate. Experience could explain the urban situation, but it cannot explain the group size differences. Skill, on the other hand, that is what I am curious about here.

“Success in problem solving in larger groups may reflect a wider diversity of … skill among the individuals in the group”, seems to suggest that larger groups succeed better because they are more likely to contain cleverer individuals among them, the statistical argument I mentioned above. That could be interesting, I suppose, because it would demonstrate the benefit of groups to less mentally well-endowed individuals. But it would be really important if the problem-solving skill observed in the larger group may reflect a new phenomenon, that being in a larger group can elicit better problem solving skills among the members of that group. Do you see the difference? The implications of the latter are far more significant, as I discussed above.

The next step is to read the original article, which appeared in the Proceedings of the National Academy of Sciences, in order to see what is really going on here. I will report next time on what I learned. I can tell you now, though, it does seem that, when it comes to social living and cognitive innovation, 6 is greater than 3x2.