An Accurate Measure of Reaction Time can Provide Objective Metrics of Concussion

There have been numerous reports of neurological assessments of post-concussed athletes and many deploy some type of reaction time assessment. However, most of the assessment tools currently deployed rely on consumer-grade computer systems to collect this data. In a previous report, we used robotic testing to demonstrate the inaccuracies introduced by typical consumer-grade computer systems (Holden et al, 2020). In that report, we described the accuracy of a tactile based reaction time test (administered with the Brain Gauge) of approximately 0.3 msec and discussed the shortcoming of other methods for collecting reaction time.The consumer-grade systems introduced latencies as high as 400 msec and variabilities as high as 80 msec, which greatly exceeds the control values reported for reaction time (200-220msec) and the control values for reaction time variability (10-20 msec). In this report, we examined the reaction time and reaction time variability from 396 concussed individuals and found that there were significant differences in the reaction time metrics obtained from concussed and non-concussed individuals for 14-21 days post-concussion. A survey of the literature did not reveal comparable sensitivity in reaction time testing in concussion studies using alternative methods. This finding was consistent with the prediction put forth by Holden and colleagues with robotics testing of the consumer grade computer systems that are commonly utilized by researchers conducting reaction time testing on concussed individuals. The significant difference in fidelity between the methods commonly used by concussion researchers is attributed to the differences in accuracy of the measures deployed and/or the increases in biological fidelity introduced by tactile based reaction times over visually administered reaction time tests. Additionally, while most of the commonly used computerized testing assessment tools require a pre-injury baseline test to predict a neurological insult, the tactile based methods reported in this paper did not utilize any baselines for comparisons. The reaction time data reported was one test of a battery of tests administered to the population studied, and this is the first of a series of papers that will examine each of those tests independently.