Abstract
INTRODUCTION: In patients with Normal Pressure Hydrocephalus (NPH), the aging process results in decreased efficiency of cerebrospinal fluid circulation, leading to ventricular enlargement. This enlargement compresses several brain structures, impairing functions such as visual perception, semantic memory processing, and phonological encoding contributing to language processing difficulties. This study examines real-time language processing in NPH patients and healthy elderly (HE) controls by comparing their eye movements during correct and incorrect responses in a lexical retrieval task (LRT). METHODS: A total of 26 subjects participated, comprising 14 patients diagnosed with NPH (aged 65 or older) and 12 individuals in the HE group. A lexical retrieval task was administered as their eye movements were recorded. A Mann-Whitney U test was performed to compare LRT performance and eye-tracking metric results across different groups, respectively. Also, correlation analysis was performed to examine the relationship between LRT scores and eye-tracking scores. A two-way mixed ANOVA was conducted to assess the significance of eye-tracking metrics depending on response type (correct/incorrect). Additionally, a qualitative and quantitative comparison of heatmaps and scanpaths was conducted to visualize eye-tracking data for correct and incorrect items. RESULTS: The NPH group exhibited significantly lower performance in lexical retrieval compared to the HE group, accompanied by more counts and longer durations in both fixation and saccade metrics. A negative correlation was noted between LRT scores and eye-tracking metric values, with correlation coefficients predominantly at 0.50 or higher. Analysis of eye movements during correct and incorrect responses uncovered significant group and within-group effects across all metrics, with more intergroup differences during incorrect responses. Qualitative differences in eye movements were more noticeable in images associated with incorrect items. DISCUSSION: This study highlights previously under-explored language deficits in NPH patients using real-time visual processing analysis, underscoring the importance of targeted language interventions for these populations.