The effect of number codes in the comparison task of two-digit numbers

Abstract

Three experiments were conducted to evaluate the influence of number format on the magnitude processing of two-digit numbers. The participants decided the larger of two presented numbers while the unit-decade compatibility was manipulated. A trial was compatible if both decades and units led to the same response (e.g., 24-67), while the trial was incompatible if decades and units led to different responses (e.g., 64-27). The compatibility effect depended on the format of presented numbers. The decade-unit compatibility facilitated the comparison of Arabic numbers (Experiment 1), but it hampered the performance when a verbal code was used, written code (Experiment 2) and spoken code (Experiment 3). The results are not consistent with a format-independent processing of number magnitude of two-digit numbers (e.g., McCloskey, 1992), on the contrary, this processing seems to depend on the superficial code in which they are presented (e.g., Campbell, 1994).