An interesting paper I heard about on Bloggingheads argues that babies as young as 6 months can discern social interactions, and make judgment calls in favor of puppets that help, as opposed to those that harm. In the Bloggingheads interview, the interviewer takes a completely uncritical view of the study, accepting as a given that its results are true. But there are a number of problems with it:
1. These are preverbal babies, so the only way the researchers can know what they want is by some outward motion — in this case, which of two puppets the babies tries to grab. Of course, it’s essentially uncertain what this gesture means. Perhaps an average baby desires the object that he or she moves to grab, but maybe they want to throttle it, or throw it violently. We’ve all seen babies respond in many ways towards objects, but as you can see in this video of the experiment, grabbing the “helper” object was interpreted as preferring it. Even if there’s evidence to suggest that grabbing on average shows approval, that is hardly always the case.
2. This last point is especially important because the sample size was tiny: just 12 six-month old babies and 16 ten-month olds. Within that sample size the uniformity of choices were impressive: 14/16 and all 12 “chose” the helper. But again, because of the interpretation problems above, how are we to know what is determining these infants’ choices, and what their gestures even mean? It seems that, in a way, we’re learning more about Yale psychologists than we are about 6 month old infants.
And this is hardly a problem just for this study. Consider this fascinating TED talk by Rebecca Saxe. She shows a number of videos of how children at various ages understand problems of morality. But here again the evidence is sloppy. The children are at an age when they can speak and explain themselves, yes, but you can see them being steered in various directions by the researcher. There’s probably an effect this is showing — older children can understand motives and information of third-parties better than younger children — but it’s hard to see how big the effect is, and how stable the ages in which these skills arise.
This isn’t to say that we shouldn’t do any psychological tests, even with infants. But it’s probably better to focus on things that can be more easily discerned, and with much larger sample sizes. In general, we should be very careful about accepting studies like this at face value, even though it’s fun to test provocative theses.