A Stroop test | Cognitive Load.

A Stroop test is a measure of selective attention. Understanding how and why someone attends to stimuli in the environment is interesting and important. It is also valuable to understand the impact on attention as task complexity increases. This is especially important when designing user interfaces.

We conducted a Stoop test (pilot-test) to examine cognitive load on participant eye movements, pupil dilation, and electrodermal activity. We collected eye movement and electrodermal activity (GSR-Galvic Skin Response) data.

NOTE: This was not a formal study and we only performed it to test our laboratory equipment. We asked three people to participate.

Task 1: We asked participants to read aloud a list of words representing colors (e.g., blue, green, red).  Words were presented in a white font colored and displayed on a black background. 

Task 2: We asked participants to read a list of words representing colors (e.g., blue, green, red).  Words were presented in a various font colors (e.g., the word red appeared in the color yellow) and displayed on a black background. Participants read the word. If the word “blue” was presented in the color “green”, the participant was to say “blue”.

Task 3: We asked participants to read a list of words representing colors (e.g., blue, green, red).  Words were presented in a various font colors (e.g., the word red appears in the color yellow) and displayed on a black background. Participants named the color of the word. If the word “blue” was presented in the color “green”, the participant was to say “green”.

Naming the color of the word can create interference effects as participants inadvertently try to read the word rather than name the color in which it is displayed – naming the color interferes with reading the word. The interference presents both Stimulus–Stimulus and Stimulus–Response incompatibility (Proctor & Vu, 2016) and makes the reading-naming task difficult; and it potentially increases cognitive load.

In testing our lab equipment, we were interested to see if the Stroop interference effects impacted eye tracking scans, pupil size, and electrodermal activity (GSR). We were especially interested in determining how well we could collect and represent these data.

Figure 1 show a heat-map of participant normal reading of words (white words displayed on black background).

Figure 2 shows heat-map of participants naming the color of the word.

Reading Patterns and Time: All participants read words from left to right. When they completed a row of words, their eyes traversed back to the left most word on the successive row, which is interesting because they could have read the words in any order. When naming the word color (interference effect), participants took more time and had more dispersed eye scans (as shown in Figure 2).

Figure 3 shows electrodermal activity for normal reading (BW-blue line) and naming the color of the word (Color – Red line).  There appeared to be increased electrodermal activity when naming the word color (interference effect, Color – Red line) compared to reading the word. 

Pupil size: Pupil size increases with task demands, and pupillometry has been shown to be a stable measure of Stroop interference (Laeng, Ørbo, Holmlund, & Miozzo, 2011).

Figure 4 shows the left pupil size for normal reading (BW-blue line) and naming the color of the word (Color – Red line).  Pupil size was larger when naming the color of the word (interference effect, Color – Red line). 

References:

Krejtz, K., Duchowski, A., Niedzielska, A., Biele, C., and Krejtz, I. (2018). Eye tracking cognitive load using pupil diameter and microsaccades with fixed gaze. PLoS ONE 13(9): e0203629. https://doi.org/10.1371/journal.pone.0203629

Laeng, B. Ørbo, M., Holmlund, T., and Miozzo, M. (2011). Pupillary Stroop effects. Cognitive Process,12: pp. 13–21.

Proctor, R. and Vu, K. (2016). Principles for Designing Interfaces Compatible With Human Information Processing. Intl. Journal of Human–Computer Interaction, 32: pp. 2–22.

Lajante,M., Droulers, O., Dondaine, T. and Amarantini, D. (2012).Opening the “Black Box” of Electrodermal Activity in Consumer Neuroscience Research. Journal of Neuroscience, Psychology, and Economic, 5(4),pp. 238–249.