Monkey‐see, Monkey‐do: An Eye‐tracking Study Assessing the Efficacy of Feed‐forward Training in Histology Visual Literacy Development

Abstract

Histology is a foundational subject in professional health curricula requiring competency in visual identification of cells and tissues on microscope slides. The emergence of virtual microscopy (VM), an app that simulates optical microscopy (OM) in a digital interface, has transformed histology curricula in health sciences education by increasing teaching efficiency and accessibility to students while reducing costs and space associated with OM labs. Over 70% of North American medical schools now report using VM solely or in combination with OMs in histology education. Despite the numerous advantages reported in the literature, no studies have shown VM's ability to enhance acquisition of visual literacy. One method shown to be effective in training novices in a virtual environment is a feed‐forward system used in aircraft inspection training. Novices who are shown expert's eye scan‐paths conducting aircraft inspection perform better on subsequent tasks than those who are not. The objective of this study was to assess whether feed‐forward training would yield higher efficiency in visual literacy development in histology. An expert's eye scan path was recorded using an eye‐tracker while completing white blood cell identification tasks on VM. A scan path includes an overlay of the expert's fixations and saccades (rapid eye movements between fixations). In an IRB‐exempt study, first‐year graduate students enrolled in a histology course, who have not yet encountered the content, were randomly assigned to two groups. The experimental group (n=11) viewed videos of the experts' scan‐path completing blood cell identification tasks on VM. The control group (n=9) viewed the same videos, but without the expert's scan‐path overlay, instead, the correct cells were highlighted at the end of each video. Both groups then completed a series of blood cell identification tasks. Linear mixed model regression analyses were used to analyze group differences for the following parameter averages: (1) time to identification; (2) total number of fixations; (3) total number of saccades; and (4) average length of saccade path. These parameters have been used previously to characterize eye movement profiles for experts and novices. On average, identification time was 1.4% lower [95% CI: 24.6% lower to 17.5% higher] and average saccade length was 17.0% lower [95% CI: 54.5% lower to 11.3% higher] in the experimental group compared to the control group. The number of saccades per slide was 13.8% lower [95% CI: 31.2% lower to 8% higher], and the number of fixations per slide was 12.2% lower [95% CI: 27.7% lower to 6.6% higher] in the experimental group compared to the control group. These results indicate the eye movement parameters between the two groups are noticeably different, with the experimental groups being more similar to expert parameters. This study provides preliminary data that feed‐forward training may be a viable method to teach pattern recognition skills and visual literacy development in histology. Further investigation of this feed‐forward training method could have implications on developing educational interventions that influence future histology curricula.Support or Funding InformationThis abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.