Sensor-based Knowledge Graph Modeling of Cognitive Pathways in Brand Information Recognition

The evaluation of consumer visual attention often relies on static area-of-interest (AOI) metrics or self-reports, which fail to capture continuous cognitive pathways.To address this limitation, high-frequency eye-tracking sensors are integrated with knowledge graph modeling to analyze dynamic brand recognition.Ocular data from 60 participants were recorded using the SensoMotoric Instruments Remote Eye Directorate (250 Hz sampling rate, 0.03° precision) during simulated e-commerce browsing.Raw gaze coordinates and pupil metrics were compressed into AOI sequences and modeled as a directed weighted graph.Weighted degree centrality identified price information (a 485.6 s fixation), main product images (450.2 s), and the reviews of first-time consumers in the digital marketing environment (users, 398.1 s) as the primary anchors of attention.Graph extraction results showed a dominant overview-to-valuation pathway (brand logo → main image → product title → price).New users showed higher path complexity (mean = 1.85, standard deviation = 0.42) than experienced users (mean = 1.31, standard deviation = 0.35), whereas impulsive users concentrated on promotional cues.Pupil dilation correlated with path complexity (r = 0.58, p < 0.01), confirming its role as a cognitive load indicator.The framework developed in this study advances sensor-based cognitive modeling and supports neuromorphic vision sensing and human-computer interaction.Limitations include reliance on desktop-mounted sensors and static layouts; future work must extend to mobile and adaptive interfaces.