Modeling Protagonist Emotions for Emotion-Aware Storytelling

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 5277–5294, November 16–20, 2020. c⃝2020 Association for Computational Linguistics 5277 Modeling Protagonist Emotions for Emotion-Aware Storytelling Faeze Brahman University of California Santa Cruz fbrahman@ucsc.edu Snigdha Chaturvedi University of North Carolina Chapel Hill snigdha@cs.unc.edu Abstract Emotions and their evolution play a central role in creating a captivating story. In this pa- per, we present the first study on modeling the emotional trajectory of the protagonist in neu- ral storytelling. We design methods that gener- ate stories that adhere to given story titles and desired emotion arcs for the protagonist. Our models include Emotion Supervision (Emo- Sup) and two Emotion-Reinforced (EmoRL) models. The EmoRL models use special re- wards designed to regularize the story gener- ation process through reinforcement learning. Our automatic and manual evaluations demon- strate that these models are significantly bet- ter at generating stories that follow the desired emotion arcs compared to baseline methods, without sacrificing story quality. 1 Introduction Stories are an integral part of human culture. They allow us to express emotions, share knowledge, and to shape our perspective of the world (McKee, 2003). Stories are made interesting through emo- tions that connect the characters, their motivations, goals, and achievements (Vonnegut, 1981). Cognitive scientists have pinpointed the central role of emotions in storytelling (Parkinson and Manstead, 1993; Hogan, 2011). Early automatic storytelling systems based on symbolic planning also showed that addressing character emotions for plot construction resulted in more diverse and interesting stories (Theune et al., 2004; Pérez y Pérez, 2007; Méndez et al., 2016). However, these studies were rule-based and limited to small-scale data. The advent of deep learning has shifted com- putational storytelling efforts towards neural meth- ods (Martin et al., 2018; Yao et al., 2019). However, despite the broad recognition of its importance, neu- ral story generation methods have not explored the modeling of emotional trajectory. Title (input): Raw burger Emotion arc (input): joy → anger → sadness Story (output): Tom went to a burger place with his friends. He ordered a burger. When he got it , he noticed that it was raw. Tom yelled at the waiter for it being raw. He was really disappointed. Figure 1: An example story generated by our model for a given title and emotion arc of the protagonist. Story segments are highlighted with the emotions the protag- onist (Tom) experiences. In this paper, we present the first study to take into account the emotional trajectory of the pro- tagonist in neural story generation. Research in cognitive science has shown that while compre- hending narratives, readers closely monitor the pro- tagonist’s emotional states (Komeda and Kusumi, 2006; Gernsbacher et al., 1992). However, emo- tions experienced by the protagonist might differ from the general emotions expressed in the story. For example, the general emotion of “My boss was very angry and decided to fire me.” is anger, but the narrator’s emotional reaction would be to feel upset. At any point in a story, we represent …

Content truncated to 500 words