Spatialization and its link to working memory capacity

The SPoARC (Spatial Positional Association of Response Codes) is a phenomenon in which information that does not have inherent ordinal property (e.g., a list of fruits) is mentally spatialized from left to right in working memory. This phenomenon, driven by cultural habits such as the reading and writing direction, is thought to support working memory by facilitating order encoding (Abrahamse et al., 2014; Guida & Lavielle-Guida, 2014), although this hypothesis has never been directly tested. To address this issue, our objective was to show a relationship between spatialization and working memory capacity. In the present study, we investigated i) whether spatialization could vary as a function of working memory load and ii) which component of the cognitive system (verbal vs. visual) could modulate the effect. We hypothesized that spatialization should be sensitive to capacity limits because anchoring ordinal information spatially may alleviate difficulties to retain information. We also predicted that verbal abilities would support spatialization more than visual abilities because spatialization has been shown to be preferentially linked to semantics (Ginsburg et al., 2017).In Experiment 1, 139 participants performed a spatialization task with sequences of either two, three, four, or five items. Verbal working memory capacity was assessed using a forward and backward digit span task. Using generalized linear mixed models, results showed a significant spatialization effect for all sequence lengths except for sequences of 5 items. However, no direct relationship was found between individual verbal spans and spatialization. In Experiment 2, 96 participants performed the same task for sequences of three, four, and five items, but using a within-subject design. Several abilities were measured, including verbal, spatial, and visual working memory capacity, as well as visuospatial ability, mental imagery, and spatial orientation ability. The absence of spatialization persisted for sequences of five items, whereas a significant effect was found for sequences of three and four items. Notably, this significant spatialization was related to the participants' verbal span. Additionally, larger spatialization was associated with lower spatial orientation abilities.The working memory capacity limit hypothesis is thus supported by our findings that spatialization can be best detected for sequences of four items and that participants with lower verbal spans spatialize more. The more surprising results concerning the low spatial orientation skills are discussed in relation to Bottini and Doeller's (2020) review on spatial representation of conceptual knowledge.