*3.3. Aim 3: Assessments with Low Feasibility*

The NEPSY-II Affect Recognition was the only measure to fall below the feasibility threshold in this study. To better understand the subset of the population within DS that this measure would be appropriate for, sensitivity and specificity calculations were completed (Table 4). Less restrictive guidelines (i.e., ABIQ deviation ≥ 20 or 30) provided higher sensitivity, indicating that completers of the measure were correctly identified. As guidelines become more restrictive (i.e., ABIQ deviation ≥ 40 or 50), sensitivity decreased, and not all participants who could complete the task were identified using the more limiting benchmarks. More restrictive ABIQ also led to higher specificity, indicating that those who were *not* able to complete the measure were correctly identified when using those more restrictive benchmarks. Chronological ages examined (8 and 10 years) revealed

minimal differences between sensitivity and specificity probabilities. Figure 1 illustrates the chronological age and ABIQ deviation scores of both completers and non-completers for the NEPSY-II Affect Recognition in our sample. those more restrictive benchmarks. Chronological ages examined (8 and 10 years) revealed minimal differences between sensitivity and specificity probabilities. Figure 1 illustrates the chronological age and ABIQ deviation scores of both completers and noncompleters for the NEPSY-II Affect Recognition in our sample.

The NEPSY-II Affect Recognition was the only measure to fall below the feasibility threshold in this study. To better understand the subset of the population within DS that this measure would be appropriate for, sensitivity and specificity calculations were completed (Table 4). Less restrictive guidelines (i.e., ABIQ deviation ≥ 20 or 30) provided higher sensitivity, indicating that completers of the measure were correctly identified. As guidelines become more restrictive (i.e., ABIQ deviation ≥ 40 or 50), sensitivity decreased, and not all participants who could complete the task were identified using the more limiting benchmarks. More restrictive ABIQ also led to higher specificity, indicating that those who were *not* able to complete the measure were correctly identified when using

positively correlated with ABIQ deviation scores and EVT-3 standard scores; however, no association was found with PPVT-5 standard scores. The SRS-2 had modest correlations with ABIQ deviation scores in the expected direction, such that more social behavior challenges were associated with lower ABIQ. In most cases, there was no significant correlation between the SRS-2 and PPVT-5 or EVT-3 standard scores. The majority of the measures were not associated with chronological age, with the exception of the NEPSY-II Affect Recognition Total raw score, which was positively correlated with age (*r* = 0.32).


**Table 4.** Post hoc sensitivity and specificity for the measure below feasibility criteria. **Table 4.** Post hoc sensitivity and specificity for the measure below feasibility criteria.

ABIQ <sup>a</sup> = Stanford-Binet, Fifth Edition abbreviated battery IQ deviation score. ABIQ <sup>a</sup> = Stanford-Binet, Fifth Edition abbreviated battery IQ deviation score.

*Brain Sci.* **2021**, *11*, x FOR PEER REVIEW 8 of 13

*3.3. Aim 3: Assessments with Low Feasibility*
