When time shifts the boundaries: Isolating the role of forgetting in children’s changing category representations

• In Event: It’s about time: Integrating the multiple timescales of attention and memory in early word learning.

Abstract

Learning categories allows children to structure and understand the world, affording inferences to new experiences. Most research on children’s categorization has examined how different types of new learning affect category representations. For example, category representations shift depending on the number or complexity of exemplars they see, or when they learn new perceptual, conceptual, or causal information about a category. However, at the same time that children are encoding new information about category members, their own internal mental processes are also at work; little is known about how these processes, such as forgetting, alone influence generalization . Thus, in the present study, we examined whether children’s category representations naturally shift over time, independent of new learning, and whether this shift could be due to forgetting.

In the first two experiments, participants (Exp1 N = 37, Mage = 4.02 years; Exp2 N = 32, Mage = 4.48 years) saw a novel object labeled by the experimenter (e.g., “This is called a wug!”) and then were shown five novel objects that had between one and five features changed from the exemplar. In Experiment 1, the presentation order of the objects was based on number of changed features (e.g., the first novel object had one feature changed, the second novel object had two features changed) and in Experiment 2, the objects were presented in a random order. For each object, the experimenter asked whether it was a member of the same category as the exemplar (e.g., “Is this a wug, or is this not a wug?”); children saw the objects either immediately after the exemplar or after a five-minute delay. In Experiment 3 (N = 34, Mage = 4.20 years), participants viewed the same exemplars used in Experiments 1 and 2. Then, either immediately or after a five-minute delay, children viewed seven individual object features and indicated whether each one had been part of the exemplar.

Children endorsed category membership at higher rates at immediate test than delayed test, suggesting that children’s category representations became narrower over time. In the third experiment, we investigated forgetting as a key mechanism underlying the narrowing found in Experiments 1 and 2. Indeed, children’s accuracy was lower after a delay, specifically in that they incorrectly rejected features that had been part of the exemplar. These results show that children forgot individual features in a way that is consistent with their category narrowing in Experiments 1 and 2.

The results of these experiments show that new learning does not need to take place in order for category representations to change: children made different judgments about objects’ category membership depending on whether they were asked immediately after viewing an exemplar, or after a delay. Children’s category boundaries likely became narrower due to their forgetting of individual features over time. This work provides further support for theoretical frameworks positing a critical role of forgetting in categorization, such as forgetting-as-abstraction theory, by suggesting that forgetting alone can narrow category boundaries.

Authors

• Christina Schonberg, University of Wisconsin - Madison
  Presenting Author

• Haley A Vlach, University of Wisconsin - Madison
  Non-Presenting Author