The Role of Predictability in Saccadic Eye Responses in the Suppression Head Impulse Test of Horizontal Semicircular Canal Function.

Abstract

In the suppression head impulse paradigm (SHIMP) vHIT protocol, the participant is instructed to follow with his gaze a mobile target generated by a laser placed on the participant's head. Recent studies have reported that the refixation saccade latencies are in relation with the time evolution of the vestibular dysfunction in both (standard and SHIMP) procedures. We hypothesized that some central mechanisms like head impulse prediction could be one of the causes for the differences in the saccadic eye responses. A prospective cohort non-randomized study was designed. For the SHIMP protocol, recorded with the ICS Impulse ver. 4.0® (Otometrics A/S, Taastrup, Denmark) vHIT device, three different algorithms were performed: "predictable," "less predictable," and "unpredictable" depending on the target's predictability. A mathematical method was developed to analyze the SHIMP responses. The method was implemented as an additional tool to the MATLAB open source script for the extended analysis of the vHIT responses named HITCal. In cohort 1, 52 participants were included in "predictable" SHIMP protocol. In cohort 2, 60 patients were included for the "less predictable" and 35 patients for the "unpredictable" SHIMP protocol. The participants made more early saccades when instructed to perform the "predictable" paradigm compared with the "less predictable" paradigm (p < 0.001). The less predictable protocol did not reveal any significant difference when compared with the unpredictable protocol (p = 0.189). For the latency of the first saccade, there was statistical difference between the "unpredictable" and "predictable" protocols (p < 0.001) and between the "less predictable" and "predictable" protocols (p < 0.001). Finally, we did not find any relationship between the horizontal vestibulo-ocular reflex (hVOR) gain and the latency of the saccades. We developed a specific method to analyze and detect early SHIMP saccades. Our findings offer evidence regarding the influence of predictability on the latency of the SHIMP saccadic responses, suggesting that early saccades are probably caused by a conditioned response of the participant. The lack of relationship between the hVOR gain and the latency of the saccades suggests that the predictive behavior that caused the early eye saccades are independent of the vestibular function.