Each school day promises new opportunities for students to develop intellectually and socially. But there’s also plenty of room for anxiety if students sense that things may not go well. The source of this anxiety could be a specific social psychological phenomenon like stereotype threat—the fear that your actions will confirm a negative stereotype (e.g. “girls are bad at math”)—or it could be a more general feeling of stress that emerges from doubts about how you’ll do on an upcoming test.
This anxiety often manifests itself in conscious thoughts (e.g. “What if I fail?”, “What if I’m not smart?”, etc.), but it can also be experienced on a more unconscious level. In these latter instances the physiological and cognitive symptoms ramp up into a general feeling of unpleasantness, but they stop short of allowing a person to realize the specific concerns that are making them feel bad.
But what if this unconscious anxiety could be prevented? What if it were possible to make people aware of growing anxiety as it was happening?
Measuring and communicating levels of anxiety in real time is no easy task, but new technological improvements in our ability to observe physiological responses are rapidly making it more feasible. A new study by a group of researchers from the University of the Free State in South Africa shows how these new tools could be used to improve academic performance.
Their study used a brain-computer interface (BCI) to measure anxiety and inform students when their anxiety was increasing. In practice, this meant students wore a helmet that measured electrical activity along the scalp (also known as electroencephalography, or EEG). A program built into the BCI translated the signals to determine whether anxiety had crossed a certain threshold.
Participants in the study played a total of eight levels in a math computer game, and their gameplay was divided into four different sessions in which they played a pair of similar levels. After completing the first level in each session, a message flashed across the screen that alerted students about the extent if their anxiety. The BCI then continued to measure anxiety as students played the second level.
Before examining how the BCI’s messages influenced performance, the researchers compared the BCI measures of anxiety to student responses on the Fennema-Sherman Mathematics Anxiety Scale (FSMAS), a more conventional tool used to measure anxiety. The researchers found that students categorized as high anxiety on the FSMAS have significantly higher anxiety as measured by the BCI. They concluded that the physiological arousal measured by the BCI did an adequate job of measuring math anxiety.
Next the researchers examined changes in post-message anxiety and participant performance on the actual math games. They found that in all four gameplay sessions there was a significant reduction in anxiety between the first and second levels. In addition, the decrease in anxiety that followed the BCI’s anxiety warnings had a significant and positive impact on performance.
The big weakness of the study is that there was no true control group—it’s possible that some or all of the decrease in anxiety was due to the students acclimating to the game.
Still, the study highlights how technology can be used to make students more aware of what’s happening in their own bodies during learning. If combined with proven anxiety fighting techniques—whether it’s a basic conscious effort to relax, or strategies like self-affirmation (PDF)—the effects of such feedback could be greatly enhanced.
Physiological feedback could also be useful for evaluating students. If a BCI device indicates that students were experiencing high levels of anxiety during a particular assessment, teachers can use that information in determining how to weigh the results. They might decide they didn’t get an accurate representation of student knowledge, and therefore should provide additional opportunities for students to demonstrate their proficiency.
The impact of real-time, non-content-driven feedback doesn’t have to be limited to situations where there is anxiety or physiological arousal. There are a wide range of psychological pitfalls that have the potential to be identified and mitigated by computerized monitoring. For example, if a student chooses to give up on a question, but a program determines the student could probably figure it out, the program could respond with a message about themalleability of intelligence (PDF) or how difficulty is often a sign thatsomething is important (PDF).
In the short term, we won’t see classrooms full of students wearing EEG helmets that measure every single bodily impulse. But even very simple physiological data has the potential to make students more aware of what they’re feeling, thereby short-circuiting the most destructive feelings before they begin. We’re still a long way away from cheap, easy, lawful, and privacy-protecting tools, but such innovations present one more outside-the-box way that technology can improve learning outcomes.