Nestor Matthews¹ (), Samantha Kozlowski¹, Megan Broderick¹; ¹Denison University

Introduction: The COVID pandemic highlighted the need for health care professionals to create easy-to-read data visualizations for the general public. For example, plots showing COVID mortality rates by age (under 65 vs 65+) and vaccine status (vaccinated vs unvaccinated) could save or cost lives, depending on how easily general audiences perceptually organize the data visualization. We therefore conducted the present perceptual learning study to investigate how different data visualizations (bar graphs vs line graphs) affect naive participants’ success in detecting visual patterns. Method: 416 naive Prolific participants viewed black and white 2x2 bar graphs or line graphs and classified each graph into either of two initially unknown categories. The categories corresponded to significant versus non-significant effects in one of three randomly assigned 2x2-target-factors. These included Factor A Main Effects (left / right mean-height differences), Factor B Main Effects (black / white mean-height differences) or Interactions (slope differences). Before collecting data, we preregistered the prediction that failures to perform above chance would occur significantly more frequently for Factor A Main Effects than for each of the other two effects [https://osf.io/qwsfc]. We tested this prediction separately for line graphs and bar graphs. Results. For line graphs, failures to perform above chance occurred significantly more frequently for Factor A Main Effects than for Factor B Main Effects (p = < .001, Cramér's V = 0.58) and for Interactions (p = < .001, Cramér's V = 0.66). For bar graphs, failures to perform above chance occurred significantly more frequently for Factor A Main Effects than for Interactions (p = < .001, Cramér's V = 0.44) but not for Factor B Main Effects (p = 0.340, Cramér's V = 0.13, n.s.). Conclusion: General audiences benefit from 2x2 line graphs that show the most important main effect plotted as Factor B (black / white mean-height differences).