The role of domain-general attention and domain-specific processing in working memory in algebraic performance: An experimental approach.

We investigated the role of working memory in symbolic and spatial algebra and related tasks across five experiments. Each experiment combined a processing task (expression evaluation, arithmetic, coordinate plane, geometry, or mental rotation) with verbal and spatial memory loads in a dual-task design. Spatial memory was compromised in the presence of more difficult processing tasks, and verbal memory was only compromised in the presence of algebraic tasks. The latter was related to the demands of retaining quantities associated with variables in verbal memory. We suggest that both verbal and spatial working memory retention engage domain-general attention, but that their maintenance mechanisms differ. Verbal memory has attention-based and rehearsal-based mechanisms, and thus sustaining verbal information over a short period is less attention-demanding than holding spatial information. We suggest that effects of a memory load on processing (e.g., x = 6) depend on whether use of maintenance strategies are possible for the specific memory load while carrying out processing. In all, our results indicate that algebraic tasks use domain-general attention and include verbal processing of algebraic variables (i.e., information conveyed in x, y). We discuss the implications for algebra learning and working memory theories. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Examining the mutual relations between language and mathematics: A meta-analysis.

This study presents a meta-analysis of the relation between language and mathematics. A moderate relation between language and mathematics was found in 344 studies with 393 independent samples and more than 360,000 participants, r = .42, 95% CI [.40, .44]. Moderation and partial correlation analyses revealed the following: (a) more complicated language and mathematics skills were associated with stronger relations between language and mathematics; after partialing out working memory and intelligence, rapid automatized naming showed the strongest relation to numerical knowledge; (b) the relation between language and mathematics was stronger among native language speakers than among second-language learners, but this difference was not found after partialing out working memory and intelligence; (c) working memory and intelligence together explained over 50% of the variance in the relation between language and mathematics and explained more variance in such relations involving complex mathematics skills; (d) language and mathematics predicted the development of one another even after controlling for initial performance. These findings suggest that we may use language as a medium to communicate, represent, and retrieve mathematics knowledge as well as to facilitate working memory and reasoning during mathematics performance and learning. With development, the use of language to retrieve mathematics knowledge may be more important for foundational mathematics skills, which in turn further strengthens linguistic thought processes for performing more advanced mathematics tasks. Such use of language may boost the mutual effects of cognition and mathematics across development. (PsycInfo Database Record (c) 2020 APA, all rights reserved).