School of Communications Professor Kevin John shared three biometric techniques that can improve the accuracy of self-reporting studies in 2018 Beckham Lecture
Everybody lies. It’s human nature. While most of these lies are defined as harmless or “little white lies,” they have the ability to skew data when self-reporting techniques are used.
At the Winter 2018 Beckham Lecture, the first since the passing of series founder Raymond E. Beckham, Professor Kevin John presented on the limits of self-reporting surveys used in the majority of communications research and how biometric technology can aid in improving the accuracy of results and data. John is a professor in the School of Communications and the director of BYU’s eye tracking lab.
John began by explaining how over — or under — reporting is part of normal human behavior, even if the results are promised to remain confidential. In a study examining how often people consciously recognize they are lying, most participants were surprised by the number of lies they told in a short conversation, most of these being “little white lies.”
This social desirability bias is influenced by social pressures, including gender norms. John used the idea of a young man and woman watching a scary movie as an example to demonstrate the idea. The man is likely to under-report his fear in order to appear more “masculine” while the woman may over-report her fear in order to appear more “feminine.” While this situation may be an innocent example, John provided examples of when under-reporting can be more dangerous, such as in health communications research.
In order to combat skewing due to the human nature of self-reporting, John described three psychophysiological tools that can add additional data to create more accurate results.
The first technique described was eye tracking, which uses pupil and corneal reflections to track where someone is looking. John displayed a photo with an overlay of a map that showed where participants in a study had looked first and the trail their eyes took. The uses of eye tracking include analyzing what elements of an advertising campaign work and where they may need improvement. For example, teenagers’ eyes may glance over a “Drink Responsibly” tag on an ad, illustrating how a different tactic may need to be utilized in order to achieve the desired results.
Beyond advertising, John shared how eye tracking has been used to train the average person to identify potentially cancerous moles more accurately. Despite these advantages, the technique is dependent on knowing the context of what the participants are looking at and is not able to track or analyze a person’s emotions.
Galvanic skin response (GSR) is another technique that can track changes in emotions by measuring changes in the electrical resistance of a participant’s skin. The technology can quantify the intensity of an emotional reaction but cannot distinguish the type of emotion. In order to do that, researchers need the last technique John demonstrated: facial recognition.
Facial recognition uses 33 established points on a person’s face and tracks deviations in expressions. The participant cannot speak while being tracked and the technology can misread embarrassment or shame for joy, but facial recognition can lead to new data to enhance self-reported data.
To demonstrate how facial recognition technology can be used, John showed a video of a domestic violence PSA alongside a video of a woman watching it and graphs that tracked her reactions. The graphs revealed her spikes in anger and frustration throughout but also revealed that her anger did not resolve by the end in the way the creators of the PSA expected and hoped. The creators were able to ask why she still felt anger at the end and evaluate how to better get their message across to get the desired response.
Throughout the lecture, John emphasized that none of these techniques are meant to replace self-reporting data collection. Instead, they are tools that can be used to supplement research and studies by anticipating where human error can alter results. John concluded his lecture by adding a word of caution about weighing the dangers of potentially sacrificing privacy for the sake of research, but also expressed his hope for what biometric research can do in the future.