(1106) Validity Argument for the Cognitive Load Assessment Scales in Simulation Germane Load Scale

Introduction: Resuscitation education requires investigation of instructional features that improve learning efficiency. Cognitive load (CL) theory offers a framework for simulation design to improve learning outcomes and efficiency, but measures of CL types (intrinsic [IL], extraneous [EL], and germane load [GL]) are lacking. In developing the Cognitive Load Assessment Scales in Simulation (CLAS-Sim), we found preliminary validity evidence of IL and EL scales, but the GL scale requires evaluation. Rapid cycle deliberate practice (RCDP), shown previously to improve learning outcomes, may encourage schemata formation (augment GL) through repetition. We used Kane’s Framework to evaluate the validity evidence for the CLAS-Sim GL scale with the goal of providing a tool to assess whether modifications to simulation design are needed to optimize learners’ CL. We hypothesize that subjects exposed to a RCDP format will report higher GL immediately post-exposure and perform better in follow up.

Methods: We randomized 101 pediatric residents to RCDP or control simulation formats. Over an academic year, subjects completed 2 exposure simulations within their assigned arm and a follow-up simulation in control format 2-4 months later. After each simulation, subjects completed the CLAS-Sim. Principal component analysis (PCA) reduced dimensionality on each CL-type scale; Cronbach’s alpha measured internal consistency reliability of items on each component for each case. Independent samples t-tests determined if CL type and follow-up performance varied by arm.

Results: Sixty-nine subjects had complete data. PCA yielded a 3-component solution: 3 IL, 3 EL, and 5 GL items. Alphas for GL were at least 0.85 for each case. GL was significantly higher in the RCDP arm after Exposure 1 and 2 (p < 0.001), but not after follow-up. In follow-up, performance parameters (time to pulse check, no flow fraction, and appropriate chest compression rate and depth) were better in RCDP subjects (each p≤0.001).

Conclusions: CLAS-Sim GL-scale validity was supported by the 3-component solution, high alphas for the GL scale across cases, higher GL after the 2 exposure simulations and better follow-up performance in RCDP arm. The CLAS-Sim may assist educators in optimizing simulation design to improve learning efficiency.