The misclassification of blinks as saccades: Implications for investigations of eye movement dysfunction in schizophrenia

Abstract

It is important to have a simple, accurate method for recording eye movements. Of the two popular approaches commonly adopted, electro-oculography (EOG) and infrared oculography (IROG), IROG is often accepted as the more accurate, and it is the method that is currently used most frequently to examine eye movements in schizophrenia. This study investigated whether the misclassification of blinks as saccades affects saccade rates when the presence of a blink is determined using only IROG recordings of eye position. Both vertical electro-oculography (VEOG), which can be used to objectively identify blinks, and IROG were recorded while 17 schizophrenia patients and 19 healthy controls were presented with sinusoidal stimuli. Of the blinks identified with the VEOG for the total group of participants, a substantial number (37%) were misclassified as catch-up and anticipatory saccades when only the IROG was used. Furthermore, in the schizophrenia group, but not in the healthy control group, the use of the IROG led to a significant misclassification of blinks as anticipatory saccades. Therefore, when IROG alone is used to identify blinks, the misclassification of blinks as saccades is likely to introduce measurement error into estimates of saccade rates, particularly estimates of anticipatory saccade rates in schizophrenia patients.