Exploring eye movements of Parkinson's disease patients performing the Judgement of line orientation test.

INTRODUCTION: Parkinson's disease (PD) patients are reported to score significantly lower on the Judgement of Line Orientation (JLO) test compared with controls. The traditional method of scoring JLO ignores potentially interesting information on the mechanism of errors made. AIM: The aim of the current study was to analyse the performance of PD patients on the JLO while monitoring eye movements. Employing eye tracking methods while PD participants attempt JLO items may prove valuable in further characterising error-patterns. METHODS: We recruited three groups, each comprising 16 participants: PD participants with normal cognition (PD-N), PD participants with mild cognitive impairment (PD-MCI) and matched controls. RESULTS: The mean correct response rates were high: 93% (+/-6) for controls, 88% (+/-12) for PD-N and 87% (+/-11) for PD-MCI; the difference did not reach statistical significance (p = 0.21). Participants made more errors as they progressed from easy to harder item (r = 0.7; p = 0.02). Using the Ska classification, error types QO1 and QO3 were by far and away the most common. The mean amplitudes of saccadic eye movements were 5.9 degrees (+/-0.9) for controls, 5.7 degrees (+/-1.1) for PD-N, and 5.5 degrees (+/-1.0) for PD-MCI. The differences among the three groups did not reach statistical significance (p = 0.64). As a whole, participant fixation patterns were similar throughout the JLO task. For the reference lines, most fixations were made on the distal ends. Fixations on the test lines, on the other hand, appeared to vary among trials, dependent on whether the response was correct or incorrect. CONCLUSIONS: There were few differences among the study groups in test performance-eye movement associations. However, we gained important insights into oculomotor behaviour during JLO test completion in both healthy controls and PD patients which could reflect the underlying disease state as we hypothesised.