Magnitude processing of written number words is influenced by task, rather than notation

Authors

Becky Wong, National Institute of Education, Nanyang Technological University, Singapore. Electronic address: nie15661@e.ntu.edu.sg.
Rebecca Bull, National Institute of Education, Nanyang Technological University, Singapore.
Daniel Ansari, Numerical Cognition Laboratory, Department of Psychology, Brain and Mind Institute, The University of Western Ontario, Canada; National Institute of Education, Nanyang Technological University, Singapore.Follow

Document Type

Article

Publication Date

11-1-2018

Journal

Acta psychologica

Volume

191

First Page

160

Last Page

170

URL with Digital Object Identifier

10.1016/j.actpsy.2018.09.010

Abstract

The extent to which task and notation influence the processing of numerical magnitude is under theoretical and empirical debate. To date, behavioural studies have yielded a mixed body of evidence. Using the case of written number words in English and Chinese, we re-examined this issue. Thirty-nine bilingual participants who showed a balanced profile of language dominance in English and Chinese completed three tasks of numerical processing (Magnitude Comparison, Numerical Matching, and Language Matching) with pure English, pure Chinese, and mixed notation number words. We conducted frequentist and Bayesian statistics on the data. Magnitude processing, as indexed by the numerical distance effect (NDE), was found to be dependent on task. Specifically, the NDE occurred in all notation conditions in the Magnitude Comparison Task and mixed notation trials in the Numerical Matching Task only. However, the data indicated that magnitude processing was independent of notation. Task and notation had an interactive influence on overall speed of processing, where participants responded to Chinese number words significantly faster than other notations for the Magnitude Comparison and Numerical Matching Tasks only. Finally, Bayesian analyses indicated that task and notation do not interact to affect magnitude processing. Specifically, the Bayes Factor and posterior model probabilities of the Bayesian ANOVA yielded strongest support for the model with three main effects (Task, Notation, Numerical Distance) and two two-way interactions (Task × Numerical Distance, Task × Notation). These findings highlight the critical role of task in numerical magnitude processing, provide support for a notation-independent account of magnitude processing, and suggest that linguistic/orthographic factors, combined with task, may interact to affect overall speed of processing.