Cognitive Biases in Screening Processes: Search Strategies in Educational Technology Research - A Systematic Review on Learning with Virtual Reality

The aim of this article is to empirically investigate the advantages and disadvantages of different search strategies in synthesizing research papers that use Virtual Reality (VR) educational technology. The aim is to identify cognitive biases on the part of the reviewers through different concrete searches. Using two search strategies, the study identifies the extremes between an AND search that finds as few irrelevant studies as possible but overlooks relevant ones, and an OR search that searches as broadly as possible but picks up many irrelevant studies. The article aims to show how systematic searches in educational research should be designed to adequately address the typical challenges of systematic analyses (e.g., recall-precision problem, cognitive load).

The search strategies were developed based on a Google Scholar search for existing systematic reviews on VR. Here, the two search strategies differed only in terms of their linkage between a technological (VR) and a pedagogical search term. The two elements were linked with either an AND or an OR. The search items were screened in a two-person cross design and evaluated on different measures of precision and recall.

There was no evidence that the more comprehensive search (OR) is superior to the narrower search (AND), but slight evidence of cognitive biases in the screening or search process in the more comprehensive search (OR). These results should be further evaluated, investigated, and, above all, replicated in further studies.