Mild cognitive impairment represents a transitional phase between healthy ageing and dementia, including Alzheimer’s disease. Early detection is essential for timely clinical intervention. This study explores the viability of smooth pursuit eye movements (SPEM) as a non-invasive biomarker for cognitive impairment. A total of 115 participants—62 with cognitive impairment and 53 cognitively healthy controls—underwent comprehensive neuropsychological assessments followed by an eye-tracking task involving smooth pursuit of horizontally and vertically moving stimuli at three different speeds. Quantitative metrics such as tracking accuracy were extracted from the eye movement recordings. These features were used to train machine learning models to distinguish cognitively impaired individuals from controls. The best-performing model achieved an area under the ROC curve (AUC) of approximately 68 %, suggesting that SPEM-based assessment has potential as part of an ensemble of eye-tracking based screening methods for early cognitive decline. Of course, additional paradigms or task designs are required to enhance diagnostic performance.
BACKGROUND Recent studies suggest that eye movements during tasks reflect cognitive processes and that analysis of eye movements using eye-tracking devices can identify developmental impairments in young children. Maintaining engagement during eye-tracking assessments in young children is challenging and often results in data loss due to distractions. This leads to incomplete recordings and repeated measurements, which can be taxing for young children. Gamification of eye-tracking procedures for cognitive diagnosis might increase engagement and help mitigate these problems, but its effects should be studied and quantified. OBJECTIVE This study compares a standard eye-tracking test battery designed by us with a gamified cartoon version to evaluate the effectiveness of gamification in reducing data loss. The gamified test incorporated child-friendly visuals to provide context for the stimuli presented. The study has two objectives: (1) to compare the data quality between the two versions of the test and (2) to investigate whether, by applying a dynamic stopping criterion to both tests, the higher data quality of the gamified procedure allows earlier test termination. METHODS Data were collected in a cohort of 25 children born preterm aged 5 years. We measured data quality using a metric derived from robustness, which we defined as the lost data index (LDI), along with task completion rates and feedback from participants. Data analyses were performed as follows: (1) direct comparison of the LDI for the two tests and (2) demonstrating that, although the base gamified test is longer, applying a stopping criterion results in comparable durations. The stopping criterion was based on the number of tasks with an LDI value below a predefined threshold. RESULTS The gamified version demonstrated a significant reduction in average LDI compared with the standard version in the first (P<.001, Mann-Whitney U test) and second (P=.01, U test) quarters of the test. In addition, a lower rate of missing values, concentrated at the beginning of the tests, allowed the cartoon test to be stopped after fewer tasks. This, together with the longer tasks of the cartoon test, resulted in comparable test lengths for all thresholds measured by area under the curve (P=.50, U test) and at the chosen threshold of 0.2 LDI (P=.21, U test). Increased engagement was further supported by positive feedback, with 79% (11/14) of the participants who provided feedback preferring the gamified version. CONCLUSIONS These findings highlight the potential of serious games in eye-tracking–based cognitive assessments for 5-year-old children born preterm. Specifically, gamification might reduce missing values and increase participant engagement, leading to higher retention rates and more effective tests, without significantly lengthening testing procedures. CLINICALTRIAL
Background: Recent studies suggest that eye movements during tasks reflect cognitive processes and that analysis of eye movements using eye-tracking devices can identify developmental impairments in young children. Maintaining engagement during eye-tracking assessments in young children is challenging and often results in data loss due to distractions. This leads to incomplete recordings and repeated measurements, which can be taxing for young children. Gamification of eye-tracking procedures for cognitive diagnosis might increase engagement and help mitigate these problems, but its effects should be studied and quantified. Objective: This study compares a standard eye-tracking test battery designed by us with a gamified cartoon version to evaluate the effectiveness of gamification in reducing data loss. The gamified test incorporated child-friendly visuals to provide context for the stimuli presented. The study has two objectives: (1) to compare the data quality between the two versions of the test and (2) to investigate whether, by applying a dynamic stopping criterion to both tests, the higher data quality of the gamified procedure allows earlier test termination. Methods: Data were collected in a cohort of 25 children born preterm aged 5 years. We measured data quality using a metric derived from robustness, which we defined as the lost data index (LDI), along with task completion rates and feedback from participants. Data analyses were performed as follows: (1) direct comparison of the LDI for the two tests and (2) demonstrating that, although the base gamified test is longer, applying a stopping criterion results in comparable durations. The stopping criterion was based on the number of tasks with an LDI value below a predefined threshold. Results: The gamified version demonstrated a significant reduction in average LDI compared with the standard version in the first (P<.001, Mann-Whitney U test) and second (P=.01, U test) quarters of the test. In addition, a lower rate of missing values, concentrated at the beginning of the tests, allowed the cartoon test to be stopped after fewer tasks. This, together with the longer tasks of the cartoon test, resulted in comparable test lengths for all thresholds measured by area under the curve (P=.50, U test) and at the chosen threshold of 0.2 LDI (P=.21, U test). Increased engagement was further supported by positive feedback, with 79% (11/14) of the participants who provided feedback preferring the gamified version. Conclusions: These findings highlight the potential of serious games in eye-tracking–based cognitive assessments for 5-year-old children born preterm. Specifically, gamification might reduce missing values and increase participant engagement, leading to higher retention rates and more effective tests, without significantly lengthening testing procedures.