Several of my studies ask how children learn words that refer to the domain of time, like “minute” and “yesterday.” Kids begin to say time words when they’re only 2 or 3 years old, but they make many humorous errors, and don’t seem to interpret time words in an adult-like way until several years later. So what do kids think these words mean, when they say them early on? Our findings suggest that preschoolers know a lot about time words: they know minutes are longer than seconds, and they know last year was before yesterday. But they don’t know how much longer minutes are, how how much farther away last year was. These patterns can teach us what kinds of information children use to learn early meanings and how time concepts are structured.
Space & time
Even though we don’t ever describe the past as being on our left side, English-speaking adults often think about time as a horizontal line, with earlier or past events on the left and later or future ones on the right. The “mental timeline” is reversed in speakers of languages that are read from right-to-left. I’m interested in when, how, and why children begin to conceive of time as a linear path with a particular direction. Are we predisposed to use a spatial framework to organize our experiences? Does this rely on experience with reading and writing and using spatial tools like calendars? How quickly do culture-specific space-time associations develop, and how is their development related to learning temporal language? My studies investigate how children interpret and use conventional timelines, and whether children use “mental timelines” spontaneously.
Languages vary widely in how they express abstract concepts like time. In English, for instance, we change verbs to indicate whether actions happened in the past, present, or future. However, some languages, like Chinese, don’t have a morphological tense system at all. Others, like Zulu, have multiple tenses within the past or future, indicating different distances from the present. So you’d use one form of the verb to talk about things that happened yesterday, and another to talk about things that happened longer ago. Language also vary in the types of time words they have. In Urdu, there is a single word that indicates “one day from now,” in either the past or future. Does this linguistic diversity affect how we think about time? Does it change how and when children acquire different types of time concepts? I’m conducting cross-cultural studies to find out!
When adults talk about time, we often use our hands to help communicate what we’re trying to say. For instance, we might gesture over our shoulder, or point over to the left, when we talk about the past in English. We usually don’t even notice we’re doing this, but if you pay attention to this when listening to someone else, you’ll see it. This is yet another way that adults use space to represent time. But what about kids, who are still learning to communicate about time in words? Do they also use space in this way? Could temporal gestures actually help them learn temporal language? My colleagues and I are cataloging the types of gestures children make while talking about time, and looking for connections between gesture patterns, temporal language comprehension, and kids’ ability to use spatial timelines.
Seeing a familiar object and naming it, or seeing a familiar word and reading it, seems effortless to adults. Explaining these abilities, and their developmental origins, however, is an enormous challenge. My early work asked how the visual system detects visual features and integrates them into a recognizable percept, like an object, word, or face. In work published in Nature Neuroscience, I synthesized the literature on a failure of object recognition called crowding, to propose a universal law characterizing the boundaries of visual space in which object recognition is possible. This work also produced key insights into word-recognition, allowing us to isolate three distinct cognitive mechanisms underlying reading, and to describe the perceptual limits on reading rate.