Is 7 larger than 5? Investigation of behavioural effects not accounted for by models of Arabic digit comparison
(2026)
Files
Lepoittevin_Thesis.pdf
Embargo Access until 2030-01-01 - Adobe PDF
- 3.3 MB
Details
- Authors
- Supervisors
- Abstract
- (en) How do we compare numerals? This question has animated the field of numerical cognition since the 1960s. Diverse models have been proposed, which are primarily debated in terms of their respective abilities to account for the well-known and extensively studied numerical size and distance effects. The present work adopts a different approach to foster theoretical progress. In line with earlier research traditions, it proposes an in-depth examination of behavioural phenomena that fall, precisely, beyond the scope of current theoretical frameworks, in order to assess whether and how they may shed light on the processes and representations involved in the comparison of symbolic numbers. A first study focused on the examination of a phenomenon, namely the “sub-base-five effect”, which has been recently reported and proposed to possibly reflect the involvement of hand-based representations of numbers — inherited from our use of fingers for counting — when processing numerical-magnitude information from symbolic numbers. However, an in-depth examination of this effect led us to question question the very existence of this phenomenon, showing that it was likely an artefact of the analysis method used by the authors at the time of its identification. Another study conducted has investigated the hypothesis that the ease of comparing symbolic numbers may be influenced by the strength of their semantic links, as posited by a specific account of numeral comparison. To do so, the influence of whether the two numbers to be compared belong to a common set of multiples was tested, both on performance and on a complementary measure of processing load, namely the psychosensory pupillary response. No such effect was identified. Importantly, during the course of these investigations, a previously unnoticed phenomenon was identified serendipitously. This phenomenon is characterised by a selective difficulty in comparing pairs of Arabic digits involving 7 as the numerically larger of the two numbers. A subsequent study has been dedicated to an extensive examination of this effect. Although we did not identify the origin of this phenomenon, our investigation allowed us to rule out a number of pre-semantic, symbol-related, hypotheses as unique explanation for it. We argue that it is important to identify the causes of this “7-effect,” as it may help shed new light on the processes and representations involved in the processing of symbolic number comparison. The identification of this “7-effect,” as well as the questioning of the sub-base-five effect, constitute the central empirical contributions of this thesis.
- Affiliations
Citations
Lepoittevin, S. (2026). Is 7 larger than 5? Investigation of behavioural effects not accounted for by models of Arabic digit comparison. https://hdl.handle.net/2078.5/277320