Syntax-driven number reading: The identification of digits is dominated by the number’s syntactic structure

Abstract

Humans have a remarkable ability to process complex syntactic structures using hierarchical representations. Within the domain of numbers, syntactic-hierarchical processing appears to exist in linguistic tasks involving verbal numbers. Here, I asked whether syntactic-hierarchical processing exists even within the nonverbal mechanism that parses visually-presented digit strings. I asked literate adults to read aloud multi-digit strings that were presented serially, digit after digit. The inter-digit SOA was identical in all positions except one SOA, which was longer – either between the thousand and hundred digits, congruent with the syntactic principles of the decimal system and how it divides numbers to triplets, or between the hundred and decade digits (incongruent grouping). The participants were sensitive to this visual manipulation: their accuracy was higher in the congruent condition than in the incongruent one, indicating that the visual analyzer divides the digit string into triplets in accord with the decimal system standards. This congruency effect was found when the participants read the numbers as syntactically-structured number names (“twelve”), and critically, also when they read each number as a syntax-less series of digit names (“one, two”). Namely, the triplet-grouping is not a top-down effect from syntactic verbal-production processes, it is a property of the visual analyzer itself. These results indicate that the visual analyzer intrinsically handles digit strings with at least 2 levels of hierarchy – the triplets, and the digits within each triplet.

Reading numbers is hard, and the difficulty is a syntactic one: A descriptive analysis of number-reading patterns in readers with and without dysnumeria

“…3; χ 2 (4) = 22.3, p < .001). This suggests that the processing of numbers is sensitive to the triplet-based structure, in accordance with previous studies showing that both the visual (Dotan, 2022) and the verbal (Dotan & Brutmann, 2022) processing of numbers are triplet-based. The triplet-based structure may be a part of a more sophisticated structure of the number, as a multi-level hierarchy (McCloskey et al, 1986).…”

Top‐Down Number Reading: Language Affects the Visual Identification of Digit Strings

“…A second relevant aspect of the visual analyzer is that it groups digits into triplets (Dotan, 2022; Dotan & Friedmann, 2018). This too may be a top‐down verbal effect, originating in the triplet‐based structure of the verbal number system.…”

“…Another distinction is between lexical processes, which handle the identity of each digit or number word, and syntactic processes, which handle the relations among these lexical elements. This distinction was shown by neuropsychological (Blanken et al, 1997;Cohen et al, 1997;Cuetos & Miera, 1998;Delazer & Bartha, 2001;Dotan et al, 2014;Dotan & Friedmann, 2018;Friedmann et al, 2021;Lochy et al, 2004;Noël & Seron, 1993), behavioral (Dotan, Breslavskiy, et al, 2021;García-Orza et al, 2017, 2022Hung et al, 2015;Lozin & Pinhas, 2022;Zuber et al, 2009), developmental (Barrouillet et al, 2010;Cheung & Ansari, 2020;Göbel et al, 2014;Moeller et al, 2011;, and brain imaging studies (Benavides-Varela et al, 2016;Furumoto, 2006a). The visual-verbal separation and lexical-syntactic separation are orthogonal, i.e., there are lexical and syntactic processes both in the visual analyzer and in the verbal production system.…”

Top-down number reading: Language affects the visual identification of digit strings

“…A second relevant aspect of the visual analyzer is that it groups digits into triplets (Dotan, 2022;Dotan & Friedmann, 2018). This too may be a top-down verbal effect, originating in the triplet-based structure of the verbal number system.…”

“…More specifically, I examined the order in which the visual analyzer scans the digits. For numbers sufficiently long (3-4 digits and more), the visual analyzer scans the digits serially, from left to right, even for speakers of a language such as Hebrew, in which words are written from right to left (Dotan, 2022;Dotan et al, 2021;Friedmann et al, 2010). A non-linguistic interpretation of the left-to-right scanning order is that it reflects a default intrinsic behavior of the visual analyzer.…”