AUTHOR=Wen Jianlan , Piao Yuming TITLE=Human–Computer Interaction-Oriented African Literature and African Philosophy Appreciation JOURNAL=Frontiers in Psychology VOLUME=12 YEAR=2022 URL=https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.808414 DOI=10.3389/fpsyg.2021.808414 ISSN=1664-1078 ABSTRACT=

African literature has played a major role in changing and shaping perceptions about African people and their way of life for the longest time. Unlike western cultures that are associated with advanced forms of writing, African literature is oral in nature, meaning it has to be recited and even performed. Although Africa has an old tribal culture, African philosophy is a new and strange idea among us. Although the problem of “universality” of African philosophy actually refers to the question of whether Africa has heckling of philosophy in the Western sense, obviously, the philosophy bred by Africa’s native culture must be acknowledged. Therefore, the human–computer interaction-oriented (HCI-oriented) method is proposed to appreciate African literature and African philosophy. To begin with, a physical object of tablet-aid is designed, and a depth camera is used to track the user’s hand and tablet-aid and then map them to the virtual scene, respectively. Then, a tactile redirection method is proposed to meet the user’s requirement of tactile consistency in head-mounted display virtual reality environment. Finally, electroencephalogram (EEG) emotion recognition, based on multiscale convolution kernel convolutional neural networks, is proposed to appreciate the reflection of African philosophy in African literature. The experimental results show that the proposed method has a strong immersion and a good interactive experience in navigation, selection, and manipulation. The proposed HCI method is not only easy to use, but also improves the interaction efficiency and accuracy during appreciation. In addition, the simulation of EEG emotion recognition reveals that the accuracy of emotion classification in 33-channel is 90.63%, almost close to the accuracy of the whole channel, and the proposed algorithm outperforms three baselines with respect to classification accuracy.