Abstract:

The spatial-numerical association of response codes (SNARC) effect is a phenomenon in which the leftward response is faster than the rightward response for smaller numbers, whereas for larger numbers, the rightward response is faster than the leftward response. Although the existence of the SNARC effect has been examined in many studies, most of these studies focused on the symbolic SNARC effect and neglected to explore the non-symbolic SNARC effect. Little is known about how symbolic and non-symbolic SNARC effects develop and whether there are differences in the cognitive mechanisms involved in these two effects. The present study aimed to simultaneously investigate the developmental characteristics and cognitive mechanisms of symbolic and non-symbolic SNARC effects to contribute to the understanding of number processing.

In Experiment 1, a large-sample cross-sectional method was used with four age groups to explore the developmental characteristics of symbolic and non-symbolic SNARC effects. Thirty-six 6- to 7-year-old children (19 boys, mean age: 6.42 ± 0.47 years), 59 7- to 8-year-old children (30 boys, mean age: 7.56 ± 0.42 years), 69 8- to 9-year-old children (32 boys, mean age: 8.40 ± 0.38 years) and 31 adults (15 males, mean age: 21.76 ± 1.46 years) performed the symbolic and non-symbolic parity judgement task. Experiment 2 was based on dual coding theory and the findings from Experiment 1. In this experiment, 137 children aged 8 to 9 years (70 boys, mean age: 8.43 ± 0.75 years), the key age at which symbolic and non-symbolic SNARC effects are observed, were selected as participants and followed longitudinally for six months to explore whether the two SNARC effects had similar cognitive mechanisms. Phonological ability, visuospatial ability, visual working memory and phonological working memory were measured at T1. At T2 (after 6 months), the participants' symbolic and non-symbolic SNARC effects were measured. The symbolic and non-symbolic SNARC effects at T1 were controlled.

The findings of this study were as follows. (1) The non-symbolic SNARC effect emerged in 6- to 7-year-old children [t (35) = -4.20, p< 0.001], while the symbolic SNARC effect emerged in 8- to 9-year-old children [t(62) = -4.53, p< 0.001]. Thus, the non-symbolic SNARC effect emerged earlier than the symbolic SNARC effect (see Table 1, Table 2, Figure 1 and Figure 2). (2) There were no significant age differences in the symbolic or non-symbolic SNARC effects. (3) For 8- to 9-year-old children (r = 0.13, p = 0.33) and adults (r = -0.03, p = 0.86) with both symbolic SNARC effects and non-symbolic SNARC effects, these two effects were not significantly correlated. (4) Phonological ability (β = -0.81, SE = 0.19, p < 0.001) and phonological working memory (β = 0.45, SE = 0.09, p < 0.001) at T1 significantly predicted the development of the symbolic SNARC effect at T2 but not the development of the non-symbolic SNARC effect at T2. Visuospatial ability (β = -0.63, SE = 0.10, p < 0.001) and visual working memory (β = 0.29, SE = 0.10, p = 0.002) at T1 significantly predicted the development of the non-symbolic SNARC effect at T2 but not the development of the symbolic SNARC effect (see Table 3 and Figure 3).

In conclusion, 8 to 9 years is the critical age at which symbolic and non-symbolic SNARC effects emerge simultaneously, and there is no significant difference in the size of the SNARC effects according to age. Furthermore, phonological ability and phonological working memory contribute to the symbolic SNARC effect, whereas visuospatial ability and visual working memory contribute to the non-symbolic SNARC effect. These findings suggest a difference in the cognitive mechanisms of these two SNARC effects. These findings support the hypothesis of the separation of symbolic and non-symbolic SNARC effects and extend dual coding theory.

Key words: symbolic SNARC effect, non-symbolic SNARC effect, cognitive mechanism, developmental characteristic