The research-grade ADOS is a broadly used instrument that informs and steers 12 much of the science of Autism. Despite its broad use, little is known about the empirical 13 variability inherently present in the scores of the ADOS scale, or their appropriateness to 14 define change, to repeatedly use this test to characterize neurodevelopmental trajectories.
15Here we examine the empirical distributions of research-grade ADOS scores from 1,324 16 records in a cross-section of the population comprising participants with autism between 5-17 65 years of age. We find that these empirical distributions violate the theoretical 18 requirements of normality and homogeneous variance, essential for independence between 19 bias and sensitivity. Further, we assess a subset of 52 typical controls vs. those with autism 20 and find lack of proper elements to characterize neurodevelopmental trajectories in a coping 21 nervous system changing at non-uniform, non-linear rates. Lastly, longitudinally repeating 22 the assessments over 4 visits in a subset of the participants with autism for whom verbal 23 criteria kept the same appropriate ADOS modules over the timespan of the 4 visits, reveals 24 that switching the clinician, changes the cutoff scores, and consequently, influences the 25 diagnosis, despite maintaining fidelity in the same test's modules, room conditions and 26 tasks' fluidity per visit. Given the changes in probability distribution shape and dispersion 27of these ADOS scores, the lack of appropriate metric spaces, and the impact that these 28 elements have on sensitivity-bias co-dependencies, and on longitudinal tracking of autism, 29 we invite a discussion on the use of this test for scientific purposes. 30