New research on infant ‘intelligence’
Here’s a summary piece on recent research suggesting that infants have a host of previously unexpected abilities, including a surprisingly facility to distinguish languages by lipreading (!), discern among different monkey faces (that adult humans can’t distinguish), and recognize rhythms. I have a personal soft spot for infant research, in part because I have overly-romanticized images about how this research would actually take place. When I read the reports, they summon images of cute babies in front of video screens or stacks of toys or listening to music, startling when researchers change the stimulus, leading a bunch of adults to excitedly write down notes…
but enough of that. Anyway, the research again points to the possibility that humans enter the world with a substantial amount of structure to the way that they perceive, and that much of early development is actually the foreclosing of potential developmental pathways in perceptual ability. For example, infants seem to recognize more speech sounds than do older children, leading scientists to suggest that unused ability to differentiate these sounds eventually atrophies and disappears: use it or lose it. The process seems entirely consistent with what Gerard Edelman calls ‘neural Darwinism’ (no relation to illegitimate step-sibling, ‘social Darwinism’), a process of development in which an ‘over-wired’ neural system learns by eliminating unused connections.
Unfortunately, in anthropology, the people who have paid the closest attention to research findings like this have tended to use the research to argue for ‘innate’ ability, such as innate grammars or innate brain modules. The big story for me here though is not innateness, but the dynamics of development, working not from a ‘blank slate’ brain but instead from a promiscuously connected nervous system, one that is challenged to eliminate extraneous information and stimuli.
The presence of so much connectivity in the infant brain and nervous system does not serve the argument for ‘innate’ intelligence, like grammar, as much as proponents would like, in my opinion. In fact, it undermines one of their central arguments for suggesting innate, pre-programmed intelligence is intelligence: that, without it, the stimuli for producing something like grammar are too poor. That is, Chomsky’s approach to grammar is often argued for on the basis of poverty of stimulus; children learn grammar, and this allegedly can’t be explained by learning alone as the models they have to work from are too poor. Although language is a hard case (and I’m not entirely sure how I feel about it as I don’t do any original research on language), this approach often gets extended to other areas of intelligence or human activity, like motor learning and perception (which I do work on).
If infants’ brains are already hot-wired with an excess of connections, however, it may make sense that they respond to all sorts of stimuli, eventually getting control of their phenomenal worlds by learning what to ignore, what to pay attention to, and developing significant abilities to disregard irrelevant stimuli. The late Esther Thelen, along with colleagues like Linda Smith, did some brilliant work on how children learn to do basic movements, like grasping objects, which suggested something similar. Infant learning trajectories differed significantly; some had to learn to control movements, others to overcome inertia. ‘Learning’ to grasp involved developing different abilities depending on what the child’s pre-existing movement tendencies and control abilities were. Of course, what is ‘extraneous’ movement, like the question of what is ‘irrelevant’ in perception, language, or cognition, may vary among cultures, between individuals, or with relation to the training they undergo. And ‘unlearned’ ability may not completely disappear, but may go dormant and be subject to reinforcement and re-emergence later.
The display of promiscuous infant perceptions, however, may overcome one of the central arguments for innate intelligence: the poverty of stimulus argument. In fact, environmental factors don’t have to explain completely the associations that the brain and perceptual systems can make. Many of them exist already. Rather, environment is, at least in part, pruning and reinforcing certain neural connections and abilities, not creating them from scratch.
Although it’s far too early in the research to make these sorts of blanket statements, I do think that our account of infant enculturation will wind up being far more baroque and interesting than current models, so tilted as they are to simplistic nature v. nurture or ‘percentages’ of each models of how culture shapes cognition, perception, experience, or emotion.