Exploring the role of operands’ magnitude representation in multiplication solving: evidence from the size congruity effect

Estudillo, A. J. 1 & García-Orza, J. 2

1 University of Edinburgh
2 Universidad de Málaga

According to some models, operands’ magnitude representations play a central role on multiplication solving (e.g., McCloskey, 1992). On the contrary, other models suggest multiplications are retrieved using verbal representations, claiming that operands’ magnitude representations, in the best case, would play a role in the process of re-coding the presented problem into a more familiar representation before accessing the corresponding verbal form (i.e.: 2x9 can be re-coded as 9x2, that is a more familiar representation) (Dehaene, 1992). This study explores whether operands’ magnitude representations are activated in multiplication problems and whether they are involved in their resolution. To accomplish this we present two experiments that manipulate the size congruity effect. The physical and numerical magnitude of the operands within each problem could be congruent (the operand with bigger numerical magnitude appears in bigger size), incongruent (the operand with lower numerical magnitude appears in bigger size) or neutral (both operands appears in the same physical size). Problem-size and the order of the operands (big x low vs. low x big) were also manipulated. In Experiment 1, thirty eight undergraduates participated in a verification (e.g.: are the following problems correct? 2x3=6; 2x3=7). In Experiment 2, twenty one undergraduates students participated in a production task (e.g.: say the result of the following problems 2x3=; 4x3=). Results showed longer response times in the incongruent condition than in the congruent conditions in both experiments. Although main effects of problem-size were also found, no interactions were observed between size congruity and the rest of variables. It is concluded that operands’ magnitude representations are automatically activated even in the context of multiplication problems, however, as suggested by the lack of interaction effects, we argue that this activation is not related to the multiplication solving process.