Discourse Comprehension: A Question Answering Framework to Represent Sentence Connections

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 1220 - 1250 December 7-11, 2022 ©2022 Association for Computational Linguistics Affective Idiosyncratic Responses to Music Sky CH-Wang◦ Evan Li◦ Oliver Li◦ Smaranda Muresan◦• Zhou Yu◦ ◦Department of Computer Science, Columbia University •Data Science Institute, Columbia University skywang@cs.columbia.edu {el3078, al4143, smara, zy2461}@columbia.edu Abstract Affective responses to music are highly per- sonal. Despite consensus that idiosyncratic fac- tors play a key role in regulating how listeners emotionally respond to music, precisely mea- suring the marginal effects of these variables has proved challenging. To address this gap, we develop computational methods to measure affective responses to music from over 403M listener comments on a Chinese social music platform. Building on studies from music psy- chology in systematic and quasi-causal anal- yses, we test for musical, lyrical, contextual, demographic, and mental health effects that drive listener affective responses. Finally, moti- vated by the social phenomenon known as 网抑 云 (wˇang-yì-yún), we identify influencing fac- tors of platform user self-disclosures, the social support they receive, and notable differences in discloser user activity. 1 Introduction Music can evoke powerful emotions in listeners (Meyer, 1956). However, our emotional reactions to it are not universal—affective responses to mu- sic are highly personal. Just as you may wonder why your friend is sobbing to a song that you only feel ambivalent about, a listener’s emotional re- sponse to music not only varies with inherent au- dio or lyrical features (Hevner, 1935; Webster and Weir, 2005; Van der Zwaag et al., 2011), but also with other factors such as a listener’s demograph- ics, mental health conditions, and surrounding en- vironment (Krugman, 1943; Robazza et al., 1994; Gregory and Varney, 1996; Juslin and Västfjäll, 2008; Saarikallio et al., 2013; Garrido et al., 2018). As a result of this idiosyncrasy, it has been ex- tremely difficult to precisely measure the marginal effects of these variables on a listener’s affective response (Yang et al., 2007; Beveridge and Knox, 2018). This difficulty is further compounded espe- cially when examining how a collection of these factors influence individual affective reactions in combination (Gómez-Cañón et al., 2021). Music psychology has long focused on identi- fying the relationships between human affect and music, both in those that are perceived and those that are felt. Perceiving and feeling emotions in mu- sic, while highly related, are not identical (Hunter and Schellenberg, 2010). Examining the latter has proved challenging, as in addition to insufficient scale for finding significance, measuring felt emo- tions in participatory studies often interferes with the experience itself (Gabrielsson and Lindström, 2010). While recent computational studies have at- tempted citizen science approaches for annotation (Gutiérrez Páez et al., 2021), reliability remains an issue; annotator confusion persists between the con- cepts of perceived and induced emotions (Juslin, 2019). Our work expands this line of research by examining affective responses to music in a natural setting: an online social music platform. We test for differences in affective responses to music by computationally measuring expressed …

Content truncated to 500 words