Verbal-spatial and visuospatial coding of number–space interactions.

Abstract

A tight correspondence has been postulated between the representations of number and space. The spatial numerical association of response codes (SNARC) effect, which reflects the observation that people respond faster with the left-hand side to small numbers and with the right-hand side to large numbers, is regarded as strong evidence for this correspondence. The dominant explanation of the SNARC effect is that it results from visuospatial coding of magnitude (e.g., the mental number line hypothesis). In a series of experiments, we demonstrated that this is only part of the story and that verbal-spatial coding influences processes and representations that have been believed to be purely visuospatial. Additionally, when both accounts were directly contrasted, verbal-spatial coding was observed in absence of visuospatial coding. Relations to other number-space interactions and implications for other tasks are discussed.

The development of the SNARC effect: Evidence for early verbal coding

“…Verbal processing was observed as the dominant mechanism in adults (Gevers et al, 2010). In line with the conclusions of Van Galen and Reitsma (2008), it could be expected that verbal processing is also the dominant mechanism in children, at least when magnitude information is relevant to the task.…”

“…A previous study showed that verbal coding dominates the SNARC effect in adults (Gevers et al 2010). The present study shows that the SNARC effect in 9 and 11 year old children is also more dominantly driven by verbal coding.…”

“…In sum, the results support the verbal account, and indicate that both children and adults rely on verbal coding rather than on spatial coding. Similar to Gevers et al (2010), the SNARC effect was also tested by means of regression analyses. In this method, the difference between correct RTs on the right-hand side and correct RTs on the left-hand side is entered in a regression analysis for each participant separately with magnitude as predictor (e.g., Fias, Brysbaert, Geypens, & d'Ydewalle, 1996;Lorch & Myers, 1990).…”

“…Hence, this account predicts a negative regression slope when word position is congruent but a positive regression slope when word position is incongruent. The regression plots are shown in Figures 3A and 3B (for 3 rd and 5 th graders, respectively) and in Figure 3C (for adults, taken from Gevers et al, 2010;Experiment 4 …”

Reversing the Manual Digit Bias in Two-Digit Number Comparison

“…This work provides empirical data in support of recent computational models of number representation Gevers, Verguts, Reynvoet, Caessens, & Fias, 2006) and showing that a purely embodied perspective is likely insufficient to account for the observed associations between symbolic number and space (Santens & Gevers, 2008;Gevers et al, 2010). Further, the present study adds to a growing body of work on the dynamics of cognitive processing, representing diverse topics such as stereotype formation (Freeman & Ambady, 2009), voice processing (Sulpizio et al, 2015), language comprehension (Spivey et al, 2005;Incera & McLellan, 2015), memory (Abney, McBride, Conte, & Vinson, 2014;Papesh & Goldinger, 2012), and face processing (Freeman & Ambady, 2011;Hehman, Carpinella, Johnson, Leitner, & Freeman, 2014).…”