A/B Testing
Emotion and the Ability to Understand the Video
A/B Test
A/B Test
Table of contents
Introduction
Experimental Setting
Results
Conclusion & Evaluation
01
03
02
04
Past studies and research about the emotion
Data collection, the method of OLS, and data analysis in A/B test
Using OLS to get the results
Conclusion and potential issues based on our results
Introduction
01
Past studies and research about the emotion
Past studies
Red card
Emotions vs. Cognitive Abilities
EMOTIONS v.s. MEMORY
EMOTIONS v.s.
HUMAN ATTENTION
EMOTIONS v.s. HUMAN LOGICAL REASONING
EMOTIONS v.s. PROBLEM SOLVING
Positive Emotions v.s. Learning Outcomes
Positive emotions created through mood induction would decrease the perceived difficulty of learning materials.
In addition, motivation, satisfaction, and perception of the learning contents would all be boosted by positive emotions and enacted as the energy behind one’s learning
(Um et al., 2012)
Negative Emotions v.s. Learning Outcomes
Negative emotions are likely to reduce a person’s time spent on learning since they automatically activate avoidance attitudes which will cause hindering the processing of the learning.
(Kremer, Telma, et al, 2019)
Experimental Study Question
To study if there are causal effects of different emotions, for instance: anger, contempt, disgust, fear, happiness, neutrality, sadness, and surprise, on an individual’s learning outcomes (based on pre & post video quizzes)
Experimental Setting
02
Data collection, the method of OLS, and data analysis in A/B test
Data Collection (survey)
The pre-quiz section helps us to get the correct rate before watching the video
Pre-quiz
Video
Post-quiz
Info Collection
automatically collect different frames of micro-expressions and calculate their emotions (Emot)
The post-quiz section helps us to figure out the correct rate after watching the video
Participants’ self-awareness of emotions and their demographic info
OLS
(Ordinary Least Squares)
A clear and straightforward method for analyzing the relationship between variables in a linear regression model. It helps find the model's coefficients, the residuals, and various statistical measures such as the R-squared value and p-values of the coefficients.
A/B Testing
A/B Test
Emot by machine recognition of facial expressions in each frame (we calculated the average emotion frame of each participant to analyze)
Eight different emotions: happiness, sadness, neutral, anger, contempt, disgust, surprise, and fear
In our A/B test, we
Self-reported emotion from the participants (subjective)
A/B Test
A variety of factors can influence the relationship.
Overall, A/B testing is an essential tool for understanding how different variables impact the effectiveness of another variable by comparing options in the same context. It helps validate assumptions, make informed decisions, and provide guidance.
OLS
A/B Test
Results
03
Using OLS to get the results
Data Analysis
Red card
Using OLS
Only significant feature is the fear emotion, which is totally negatively correlated to the quiz result.
Once scared by the video, people can hardly remember the knowledge learned or think carefully and deeply about the problem
Conclusion & Evaluation
04
Conclusion and potential issues based on our results
Based on the introduction & Experiment
Negative emotions such as fear can have a negative impact on learning, and may reduce the time for people to learn.
Positive emotions such as happiness may not have a significant positive impact on learning, according to the current results. However, there may be flaws in the experimental design.
Evaluation
One issue with the current experimental design is that only a small number of participants provided video data, which may deviate from the real situation and affect the results. More data should be collected in the future to improve the analysis.
Another issue is the use of the difference in the correct rate between two quizzes to measure the degree of understanding of the video. Adjusting the number of questions to be the same in both quizzes may increase the credibility of the results.
The experimental workflow can be improved in several aspects.
Resources
Tyng, Chai M., et al. “The Influences of Emotion on Learning and Memory.” Frontiers, Frontiers, 10 Aug. 2017, https://www.frontiersin.org/articles/10.3389/fpsyg.2017.01454/full#B166.
Um, Eunjoon “Rachel,” et al. “Emotional Design in Multimedia Learning.” Journal of Educational Psychology, vol. 104, no. 2, 2012, pp. 485–498., https://doi.org/10.1037/a0026609.
Vuilleumier, Patrik. “How Brains Beware: Neural Mechanisms of Emotional Attention.” Trends in Cognitive Sciences, vol. 9, no. 12, 2005, pp. 585–594.,https://doi.org/10.1016/j.tics.2005.10.011.
Jung, Nadine, et al. “How Emotions Affect Logical Reasoning: Evidence from Experiments with Mood-Manipulated Participants, Spider Phobics, and People with Exam Anxiety.” Frontiers in Psychology, vol. 5, 2014, https://doi.org/10.3389/fpsyg.2014.00570.
Isen, Alice M., et al. “Positive Affect Facilitates Creative Problem Solving.” Journal of Personality and Social Psychology, vol. 52, no. 6, 1987, pp. 1122–1131., https://doi.org/10.1037/0022-3514.52.6.1122.
Robinson, Peter. “Abilities to Learn: Cognitive Abilities.” Encyclopedia of the Sciences of Learning, 2012, pp. 17–20., https://doi.org/10.1007/978-1-4419-1428-6_620.
Kremer, Telma, et al. “Influence of Negative Emotions on Residents’ Learning of Scientific Information: An Experimental Study.” Perspectives on Medical Education, vol. 8, no. 4, 2019, pp. 209–215., https://doi.org/10.1007/s40037-019-00525-8.