Engineering product design relies on interdisciplinary group decision-making, but its complexity is affected by multiple factors, and traditional behavioral studies lack quantitative analysis of cognitive processes. Eye-tracking technology can capture attention allocation in real time, but its application in engineering decision-making is still in its infancy, especially the lack of systematic research on the correlation between cognitive synchronization and decision-making performance at the group level.

Experimental design:

✦✦Participants: 72 interdisciplinary participants (with backgrounds in structural engineering, aesthetic design, and environmental engineering) from Shanghai Jiao Tong University and shipbuilding companies, divided into 24 groups (3 people per group).

✦✦Task Flow:

▪Individual decision-making phase: evaluation of 3 cruise ship cabin design options (A/B/C) based on single-discipline knowledge;

▪Group decision-making phase: interdisciplinary discussion and consensus, using a Multi-Attribute Decision-Making Task (MADM) combined with the Think-Aloud Protocol to capture cognitive processes.

eye movement measurement: The

✦✦Equipment: Eye-Logic eye-tracker (60 Hz sampling frequency), recording gaze duration, number of gazes, thermograms, and trajectory consistency, and area of interest (AOI) segmentation into interdisciplinary technical focus (HCN), single-disciplinary technical focus (SE/AD/EE), and drawing area (DWG).

data analysis: The

✦✦A PLS-SEM model was used to validate the direct/indirect effects between the variables, with the independent variables being group mean gaze duration (X₁), number of gazes (X₂), and trajectory similarity (X₃/X₄), and the dependent variables being quality of decision making (Y₁), consensus (Y₂), efficiency (Y₃), satisfaction (Y₄), acceptability (Y ₅).

Attention allocation characteristics:

✦✦The interdisciplinary technology focus (HCN) had the highest length (23.761 TP4T) and number (25.151 TP4T) of gaze, and the drawing area (DWG) had a high variance in gaze (CV=0.966), suggesting that the group was more focused and consistent in their attention to the textual information.

Decision Performance Influence Pathways:

✦✦Direct effect: group mean gaze duration (X₁) → number of gazes (X₂) → satisfaction (Y₄, β = 0.849, p

✦✦Indirect effect: gaze duration and frequency indirectly affect efficiency (Y₁) through decision quality (Y₃), β = 0.871, p

✦✦Negative effect: excessive focus on the drawing area (DWG) reduces decision-making efficiency (β = -0.456, p

Disciplinary differences:

✦✦Structural engineering participants focused on functional parameters (e.g., partition materials), aesthetic design focused on color palettes, and environmental engineering focused on comfort metrics, but interdisciplinary discussions enhanced consensus (weighted mean 4.953).

Core mechanisms: The

✦✦Visual attention allocation affects group performance through the pathway of "deep information processing → cognitive synchronization → decision consensus", and the duration and number of cross-disciplinary technical focal points are key predictors.

Practice Recommendations: The

✦✦Optimize decision-making materials: highlight interdisciplinary technical focus through bolding and color annotation to reduce the distraction of abstract information in drawings;

✦✦Intensive cognitive training: interdisciplinary terminology inter-training to improve the ability to allocate attention to a heterodisciplinary focus;

✦✦Process design: using a multi-round discussion mechanism to balance single-disciplinary perspectives with group consensus.

Limitations and Prospects:

✦✦The sample size is small and limited to static decision-making scenarios, and the effect of population neural synchronization on decision-making needs to be explored in the future in combination with dynamic tasks and hyperscanning techniques.