by Stella C. Gerdemann & Annie E. Wertz
Imagine picking and eating wild blueberries in a forest. After finishing with one patch, you walk a little further down the path and see another clump of bushes with blue-colored berries. But these bushes have differently shaped leaves. Do you eat the berries from these new bushes too? This could turn out to be a high stakes decision. The change in leaf shape is relatively subtle, but the fruits on the new bushes might be food or they might be fatal. In an article that recently appeared in Evolution and Human Behavior, we tested whether cognitive mechanisms to correctly categorize plants––and thereby distinguish between potentially safe and dangerous plants––may be present in the first years of life. To do this, we compared 18-month-olds’ use of leaf shape cues to categorize plants to their use of similar shape cues to categorize human-made artifacts.
Aside from the (hobby) botanist, or the person who might wonder whether they inadvertently bought kale instead of chard, categorizing plants correctly is not a problem many of us frequently face. In most modern environments, humans encounter plants in non-threatening contexts in which the hard work of distinguishing different kinds of plants has already been done for us. For example, we usually don’t need to know precisely what kind of plant we are dealing with in a grocery store to assess its safety. This, however, is not typical of the way plants were encountered throughout most of human evolution. In ancestral environments, humans foraged for plant resources in wild environments and had to correctly distinguish safe plants from dangerous ones to mitigate the costs of toxic plants. In fact, as some of our previous work published in EHB suggests, distinguishing between very similar looking plants is an essential component of social learning strategies for acquiring information about plants.
Another previous study had shown that 18-month-olds can distinguish between different plant types, but that study did not test whether or not infants use different features to categorize plants that they use to categorize other object types, such as artifacts. The previous study also did not address whether infants use cues that may be predictive of the type of plant they are dealing with, like the shape of the plants’ leaves, more than they would use non-predictive cues like pot color to categorize plants. Therefore, in our study, we assessed whether 18-month-olds preferentially rely on leaf shape cues to categorize plants, but not when categorizing artifacts.
To address our research question, we designed closely-matched plant and artifact objects that we presented to infants in our study. The plants were realistic-looking artificial plants with green leaves growing out of brightly-colored pots. The artifacts looked like feather dusters. They were made out of the same leaves as the plants, but instead of being green, the leafy parts of the feather duster artifacts were painted gold and silver and attached to brightly-colored wooden rods. For both types of objects, we pitted leaf shape cues against another visually-salient cue: the color of the pots (for the plants) or the rods (for the artifacts). If leaf shape cues play a special role in the early categorization of plants, then we expected infants to prioritize leaf shape over pot color when categorizing plants in our study. In contrast, we expected that, because these same leaf shape features would likely not be relevant for categorizing artifacts, infants would be less likely to use these cues to categorize our feather duster artifacts.
We tested these predictions by using a category matching task. In this task, we showed half of our 18-month-old infants a target plant and told them its name (“This is a Toma“). Then we showed them two new plants—one that shared the same leaf shape as the target plant, but has a different pot color, and one that shared the same pot color, but had a different leaf shape—and asked infants which one “goes with” the target plant. Infants responded by reaching for one of the two plants. The other half of our infants were given the exact same task using the artifacts.
Infants prioritized leaf shape cues when categorizing plants but did not prioritize those same shape cues when categorizing artifacts.
Our results supported our main hypothesis: infants prioritized leaf shape cues when categorizing plants but did not prioritize those same shape cues when categorizing artifacts. In a follow-up online study, we found that aspects of this categorization strategy persist into adulthood. Adults were less willing to categorize plants by cues other than leaf shape than they were for artifacts.
In sum, our study shows that infants rely on leaf shape cues to categorize plants, but not when categorizing feature-matched artifacts. These findings suggest that distinguishing between plants is indeed an important component of social learning strategies, and leave several open questions. For example, we still do not know whether early experiences influence infants’ emerging ability to use leaf shape cues to categorize plants. Our study did not show any evidence that infants’ early exposure to plants has a measurable influence on infants’ tendency to categorize plants by shared leaf shape cues. Nevertheless, there may be a role of experience that could be better assessed by using more fine-grained methods than the parental questionnaires that we used to measure infants’ experiences with plants. Cross-cultural studies of infants and young children could also shed light on the effects of growing up in environments that afford very different experiences with plants on these early emerging categorization abilities.
Correctly distinguishing between different kinds of plants may no longer be a regular feature of our everyday lives, but the cognitive systems that evolved to solve this problem seem to be very much a part of our modern-day minds.