Acquisition of covariation information in elementary school children

Noseworthy, Jill E. (1995) Acquisition of covariation information in elementary school children. Masters thesis, Memorial University of Newfoundland.

[English] PDF (Migrated (PDF/A Conversion) from original format: (application/pdf)) - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (11MB) [English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. (Original Version)

Abstract

The study was designed to determine how children learn covariation information and whether increasing the number of irrelevant dimensions would facilitate implicit learning. Ninety-six fourth and fifth graders were trained on sets of three stimuli varying in size (large and small) and shape (curved and straight). Of the three stimuli, one represented the covariation between shape and size (e.g., large and curved) . Half of the participants were trained on sets with one irrelevant dimension (position of the stimulus on the computer screen) and half were trained on sets with two irrelevant dimensions (position and the stimulus pattern; open, filled, or striped). Following training, participants were exposed to a transfer task with novel stimuli, but the same covariation employed in training. Finally, participants were given a verbal awareness test requiring them to tell the experimenter how they solved the problem. This test resulted in three classifications: verbally aware (explicit learners), partially aware, and not verbally aware (implicit learners). -- All participants included in the analyses reached criterion during training, indicating that children can learn covariation information either explicitly or implicitly. As the complexity of the task increased, the learning rate for all participants decreased, particularly for the explicit learners who presumably relied on hypothesis testing. On transfer, explicit learners performed better than implicit learners. The implicit system was not particularly smart, perhaps due to a reliance on contextual cues acquired in associative learning. Partial learners performed like implicit learners on transfer when trained on one irrelevant dimension, and like explicit learners when trained on two irrelevant dimensions. -- From these results several assumptions were made about cognitive processes. First, both implicit and explicit pathways are activated in a learning task, with explicit learning rate falling off more steeply than the implicit learning rate as a function of increasing task difficulty. Second, an intersection occurs where both implicit and explicit learning are occurring at approximately the same rate; task difficulty at the point of intersection will vary between individuals. Therefore, an individual who usually learns implicitly has an intercept at a low level of task difficulty and learns difficult problems implicitly. An individual who usually learns explicitly has an intercept at a high level of task difficulty and learns easier problems explicitly. The third assumption was that partial learners acquire information at approximately the same rate implicitly and explicitly; in other words, each partial learner is at the point of intersection. When an individual is a partial learner on an easy task, their implicit and explicit learning curves are presumed to resemble those of individuals who usually learn implicitly. When an individual is a partial learner on a difficult task, their implicit and explicit learning curves are presumed to resemble those of individuals who usually learn explicitly. The final assumption was that learners will show a preference for accessing either implicit or explicit information based on how they usually solve similar problems.

Actions (login required)

View Item