Revealing Differences in Legibility Between Typefaces Using Psychophysical Techniques: Implications for Glance Time and Cognitive Processing

This study adapts classical vision science techniques to an investigation of the relative legibility of two different typefaces across two different polarity (color) conditions. Participants performed a simple yes/no lexical decision task, with task difficulty controlled by an adaptive staircase in four typeface/polarity conditions. Stimulus duration thresholds (minimum time on screen needed for accurate reading) were sensitive to differences between both typefaces and polarities, with a humanist style typeface showing a legibility advantage compared to a square grotesque style typeface, and positive polarity text (black on white) showing a legibility advantage compared to negative polarity text (white on black). Legibility thresholds were found to increase with age. As expected, reaction time measures were not sensitive to differences in typeface or polarity, but they did reveal cognitive processing differences between correct and incorrect responses, as well as differences in processing words and pseudowords. There was also some evidence that switching to a new typeface and/or polarity may incur a quantifiable “task switching cost”. This study is broadly consistent with MIT’s previous effort to assess the impact of typestyle on interface demand in a simulated driving environment. We believe that this simplified, resource efficient methodology of assessing legibility differences can be adapted to investigate a wide array of questions relevant to typographic and graphic design in automotive as well as other interfaces.