The Effect of Anthropomorphic Design Cues on Increasing Chatbot Empathy
Riswanda Al Farisi, Ridi Ferdiana, Teguh Bharata Adji
Abstract
Empathy is the process of understanding and reacting to other people emotional expressions that involve cognitive and affective aspects. Empathy in education is very important, because service experiences in student academic services that are emotional and generate student affective responses have an influence on student motivation and academic achievement. Recent advances in artificial intelligence (AI) allow computers to gain the ability to express empathy by analyzing and reacting to user emotional expressions via chatbot conversational agents. Chatbots with human-like cues such as their language style can also affect user emotions. In order for chatbots to have human-like cues, it is necessary to apply the concepts in Human Computer Interaction (HCI) to chatbots. This research will apply one of the concepts in HCI, called Anthropomorphic Design Cues (ADC). Anthropomorphic is an attribution of human characteristics and behavior that is applied to non-human entities such as animals, plants, or technological devices like chatbot. This study will apply two designs from ADC on chatbots, verbal design and non-verbal design. This study will compare the level of empathy between two chatbots with ADC and without ADC. The results showed that based on the results of statistical tests using the Mann Whitney U Test ADC had a significant p-value of 0.000<0.05 on chatbot empathy, so it can be concluded that ADC had a significantly positive effect on increasing chatbot empathy.