By Emma Young
A newborn baby knows almost nothing about the world it comes into. To make sense of the onslaught of incoming sensory information, she or he must start to notice meaningful patterns and categorise them: that particular combination of visual data signifies a “face”, for example, while that noise is a “voice”. As the authors of a new paper in Developmental Science point out, “without this fundamental categorisation function, our nervous systems would be overwhelmed by the sheer diversity of our experience.”
It had been thought that infants form these categories using information from just one sense, whichever is the most relevant. Following this account, the category of “faces” results from an accumulation of visual information about what faces look like. However, an intriguing new study, involving four-month-old infants and their mothers’ smelly t-shirts, suggests that babies’ early acquisition of the faces category is a truly multi-sensory process.
Arnaud Leleu at the University of Burgundy, France, led the research involving 18 baby girls and boys aged four months. Before the testing started, the babies’ mothers wore the same T-shirt for three nights in a row, keeping it in a hermetically sealed bag at all other times. They and their infants were then taken into a quiet, odour-free lab space where the babies were shown a series of images. Some were of faces against natural backgrounds, while the majority featured animals, plants and manufactured objects. Each image was presented for only 167 milliseconds (less than a quarter of a second), meaning that perception had to happen at a glance.
Some of the time, the babies had their mother’s odorous T-shirt placed on their upper chest. (The T-shirt was folded so that the maximally-smelly underarm, breast and neck regions were closest to their nostrils.) Other times, they had a clean, unworn T-shirt placed near their noses. Throughout the experiment, the researchers used EEG (electroencephalography, which measures electrical activity via electrodes on the scalp) to monitor how the babies’ brains responded to the images.
The results were clear: the babies’ face-related brain activity in response to the images of faces was significantly greater when they could smell their mother’s well-used T-shirt at the same time, compared with when their mother’s odour was not present – suggesting that the odours strengthened their recognition of faces. The results provide strong support for the idea that multi-sensory, rather than single-sense, inputs drive our acquisition of categories, the researchers write.
Some earlier research findings lend support to this idea, in relation to faces, at least. Whenever you look at a face, neurons in a region of the visual cortex called the fusiform gyrus respond with a burst of activity (leading researchers to label this the “fusiform face area”). However, in 2009 a study showed that the smell of body odour, without any visual input, can also trigger activity in the fusiform gyrus. A region that had traditionally been thought to be responsive only to visual data is clearly receptive to other sensory signals when they indicate that a face is likely to be in the visual scene.
It makes sense that babies use odours to help them to spot faces, the researchers add. Odours are less fleeting than many visual stimuli, and of course body odours tend to co-occur with the appearance of a face.
The new work prompts many questions. Do babies rely more on smell than adults in their rapid recognition of faces? Does it matter to babies if the odour is their mother’s, or could it be someone else’s? (Only maternal odour was used in this study.) Which constituents of the body odour are most important for the effect? Do babies use smell to help them to form other “visual” categories? Only further research will tell.
The researchers also make this important point about the way that infant research is usually conducted: “Given that much evidence about visual categorization in infancy, and about virtually every neurocognitive process, has been obtained through testing infants seated on their parents’ lap, future studies should examine whether and how such parental sensory context, including body odour, mediates infants’ processing abilities.”