Poster #25 - Examining Children’s Visual Statistical Learning and Learning of Patterns in Arithmetic Principles

• In Event: Poster Session 14

Abstract

Arithmetic operations follow general principles that capture regularities and patterns across and within problems (e.g., a problem with the same addends, regardless of order, will have the same sum). Children who demonstrate knowledge of these patterns solve arithmetic problems more quickly and accurately than their peers who do not (Canobi et al., 1998). Thus, we hypothesized that children who are able to more quickly recognize these patterns and regularities across arithmetic problems may have an advantage, particularly at the early stages of arithmetic learning. Specifically, we predicted that individual differences in first graders’ capacity for statistical learning – the cognitive process related to extracting regularities in the environment – would be positively correlated to children’s arithmetic accuracy, particularly for subtraction as it is less practiced. Further, we hypothesized that statistical learning capacity is a unique construct and would not be correlated with different executive function processes (i.e., inhibition and cognitive flexibility). In this preliminary study, first graders (N = 29) completed two measures of executive control (a cognitive inhibition and a cognitive flexibility task), an arithmetic task, and a visual statistical learning task, in that order. Children’s cognitive inhibition and flexibility were measured with the Shape School (Espy, 1997). Statistical learning was assessed using a modified version of an established measure of visual statistical learning (Fiser & Aslin, 2002; Brady & Olivia, 2008; Arciuli & Simpson, 2011). The arithmetic task was modeled after a measure by Canobi et al., 1998. It included pairs of problems that were related via the commutativity (7+2+4 = and 2 + 7 + 4) and inverse (12 – 6 – 4 = and 12 – 6 – 2 =) principles. As expected, individual differences in statistical learning were not correlated to either differences in inhibitory control (r =0.004, p =0.985) or cognitive flexibility (r =0.165, p =0.403). Consistent with previous findings, both executive functions were positively related to first-grade students’ arithmetic performance: cognitive flexibility (r =0.655, p < 0.001) and inhibition (r =0.390, p =0.40). In addition, individual differences in visual statistical learning capacity were positively correlated with children’s subtraction but not addition accuracy, r =0.596, p =0.032 and r =0.354, p =0.260, respectively. Within subtraction, there was a positive correlation amongst the low performing group, suggesting the process is most implicated in the early stages of learning arithmetic operations. While individual differences in children’s statistical learning have been associated with reading, this is the first study we are aware of that points to its potential involvement in math learning. Future work will continue to explore the associations with larger samples. The potential implications for understanding the processes underlying math learning and the design of instructional materials will be discussed.

Authors

• Hyun Young Cho, Boston College
  Presenting Author

• Elida V Laski

• Marina Vasilyeva