Is the Discrete Semantic System Part of the More General Conceptual or Linguistic Network?

Abstract

The ubiquity of numbers in our culture confers especial importance on the cognitive mechanisms enabling numerical processing and comparison. When asked to identify the smaller/larger one of two Arabic numbers in comparison tasks, participants’ performance improves with increasing numerical distance between the stimuli [1]. This so-called symbolic distance effect can be attributed to the Discrete Semantic System (DSS), a network in which numbers vary in their strength of association with concepts such as “small” or “large” [2]. Similar distance effects also occur when letters or months are compared [3]. This poses the question whether these distance effects might rely on a more general conceptual or linguistic network of associations or, in other words, whether the DSS is part of this domain-general network. If the numeric and non-numeric distance effects rely on the same mechanism, their slopes may correlate. Hence, to address our research question, we subjected adult participants (n = 65) to three different comparison tasks, using Arabic numbers, letters of the alphabet, and months of the year as stimuli. Subsequently, we calculated the reliability-adjusted correlations of the distance effect slopes for participants’ error rate and reaction time across the three domains. In line with previous studies, significant distance effects were observed in all three domains for both error rate and reaction time. The distance effect slopes correlated moderately to strongly between numbers and months (error rate) and between letters and months (error rate and reaction time). Further, we found weak correlations between the error rate distance effect slopes for numbers and letters as well as for the reaction time distance effect slopes for numbers and months. These results partly support the idea that the DSS drives symbolic comparisons beyond the number domain, at least in ordered lists as closely associated with numbers as months. However, the assumption that similarly-appearing distance effects indicate a single mechanism might have been flawed as participants might combine different strategies. In our case, number and letter comparisons might rely on two distinct mechanisms, with month comparison partly drawing on both of them. Future studies should therefore aim to disentangle the properties of different ordered lists to elucidate how the human mind processes numeric and non-numeric symbolic comparison.

References

[1] R. S. Moyer and T. K. Landauer, “Time required for judgements of numerical inequality,” Nature, vol. 215, no. 5109, pp. 1519–1520, 1967.

[2] A. Krajcsi, G. Lengyel, and P. Kojouharova, “The source of the symbolic numerical distance and size effects,” Frontiers in Psychology, vol. 7, 2016.

[3] B. A. Fairbank Jr., "Experiments on the temporal aspects of number perception." Doctoral Thesis, The University of Arizona, USA, 1969.