Applying cognitive load theory to medical education: construct and measurement challenges

describes several excellent strategies to optimize intrinsic load and minimize extraneous load. As noted by the authors, there is legitimate disagreement about the construct of germane cognitive load and how it relates to intrinsic load. However, we do not agree with Leppink and Van den Heuvel’s rejection of germane load as an independent construct. Four of six prior studies performed by Leppink [1, 7] produced three-factor solutions in factor analysis for cognitive load sub-types. Factor analysis from our own research (manuscripts in preparation) has also clearly shown a third factor when looking at cognitive load during two different clinical procedures, colonoscopy and patient handovers. Moreover, the inconsistent correlation of germane load with learning found in the studies to date may in part be due to assessing the learning outcome too soon after the intervention. The effect of germane load on learning may not be captured within hours of a learning task (the time frame used in Leppink’s prior studies). Thus, a third factor seems to exist. Reconceiving this third factor as ‘subjective judgment of learning’ seems premature at this point. Future studies utilizing CLT should focus on further specifying, characterizing, and refining the third factor and, in so doing, assess its correlation with learning at a later time horizon. We believe that there is a reasonable theoretical argument for the construct of germane load. Some understand intrinsic load as related to task performance and germane load as related to task learning. In this view, germane load encompasses the mental activities related to schema construction and automation. Others argue that the construct of intrinsic load should include schema acquisition and that germane load should be limited to additional activities that enhance learning such as the conscious application of learning strategies (e.g., compare and contrast) [8]. Still others prefer to conceive of intrinsic load as including all of the activiLeppink and Van den Heuvel’s [1] article explores cognitive load theory (CLT), a framework for learning that has recently received increased attention in medical education [2]. CLT builds upon a model of human memory developed by Atkinson and Shiffrin [3] that includes three primary sub-systems (sensory, working, and long-term memory). Unlike sensory and long-term memory, working memory is severely constrained—it can only hold a limited number of information elements at any given moment [4]. Because of this constraint, CLT identifies working memory as a ‘bottleneck’ for learning. With working memory as the rate limiting step, CLT focuses on the cognitive load that learning tasks impose on the working memory. Originally, CLT described two types of cognitive load: intrinsic load (essential to the learning task) and extraneous load (non-essential to the learning task and often induced by poor instructional design) [5]. Later, CLT researchers proposed a third type, germane load, imposed by the deliberate use of cognitive strategies to enhance schema formation and automation. Overall, instructional techniques developed by CLT aim to optimize learning by minimizing extraneous load, matching intrinsic load to the developmental stage of the learner, and promoting germane load [6]. Leppink and Van den Heuvel’s article