Previous research in educational psychology has indicated a positive correlation between confusion and cognitive performance. This is attributed to the “deep learning” process triggered by confusion. This study explores the nature of this relationship by examining whether confusion causally impacts cognitive performance.
The study designed two types of scenarios: one involving a memory task requiring understanding and recall of written information, and another involving an attention task requiring focus and effort. The experimental group and the control group were tested separately, with the experimental group receiving tasks containing content with inconsistencies compared to previous and subsequent information, introducing “confusion.” Facial expressions were monitored using the Facial Action Coding System (FACS), which describes facial movements through 44 different Action Units (AUs). AU4, commonly observed during states of confusion, was used as the primary indicator of confusion.
The results showed that confusion triggered activation of AU4 and had a positive effect on performance in memory tasks and general situations. However, no impact of confusion was found in the attention task. This may be because the increased attention caused by confusion was too brief and fleeting to positively influence performance in subsequent tasks. To continuously benefit from confusion, it is necessary for participants in a highly attentive state to invest effort in reflection, reconsideration, and deeper learning processes. Additionally, the study confirmed that the detection software’s sensitivity to facial expressions is capable of capturing signs of confusion, as evidenced by the significantly higher AU4 values during the confusion phase compared to the control group.
This research suggests that confusion might lead to more profound and reflective information processing. Such insights could inform the design of training tools in educational settings and decision-support tools in work environments. Moreover, the facial expression associated with AU4 triggered by confusion can be detected by software, helping individuals recognize and leverage their confusion to their advantage, and enabling precise recording of the timing of confusion for future research and development.