Abstract
<p>Visual scenes are routinely transformed into symbolic representations that ultimately become verbal language. However, the cognitive processes underlying this transformation remain only partially understood, largely because visual perception has traditionally been investigated through verbal descriptions that are already constrained by the grammatical organization of language. The present exploratory study proposes an alternative methodological approach by introducing pictogram-based representations as an intermediate level of conceptualization between visual perception and verbal description.Twenty-four native speakers of Puerto Rican Spanish viewed three experimentally designed visual scenes while their eye movements were recorded using Gazepoint GP3 eye tracker. Immediately afterward, participants described each scene first using pictograms and subsequently through verbal language. Qualitative inspection of scan paths revealed substantial variability in visual exploration across participants. In contrast, pictogram selections exhibited considerably greater convergence, suggesting that symbolic representations were not simple reproductions of gaze trajectories. Rather than demonstrating the existence of a visual grammar, the present study proposes a methodological framework for investigating how visual scenes become conceptual and symbolic representations before they are fully organized by linguistic structures. The findings suggest that pictogram-based representations may constitute an informative intermediate level for studying systematic visual organization at the vision–language interface.</p>