Learning to Use Narrative Function Words for the Organization and Communication of Experience

Abstract

How do people learn to talk about the causal and temporal relations between events, and the motivation behind why people do what they do? The narrative practice hypothesis of Hutto and Gallagher holds that children are exposed to narratives that provide training for understanding and expressing reasons for why people behave as they do. In this context, we have recently developed a model of narrative processing where a structured model of the developing situation (the situation model) is built up from experienced events, and enriched by sentences in a narrative that describe event meanings. The main interest is to develop a proof of concept for how narrative can be used to structure, organize and describe experience. Narrative sentences describe events, and they also define temporal and causal relations between events. These relations are specified by a class of narrative function words, including “because, before, after, first, finally.” The current research develops a proof of concept that by observing how people describe social events, a developmental robotic system can begin to acquire early knowledge of how to explain the reasons for events. We collect data from naïve subjects who use narrative function words to describe simple scenes of human-robot interaction, and then employ algorithms for extracting the statistical structure of how narrative function words link events in the situation model. By using these statistical regularities, the robot can thus learn from human experience about how to properly employ in question-answering dialogues with the human, and in generating canonical narratives for new experiences. The behavior of the system is demonstrated over several behavioral interactions, and associated narrative interaction sessions, while a more formal extended evaluation and user study will be the subject of future research. Clearly this is far removed from the power of the full blown narrative practice capability, but it provides a first step in the development of an experimental infrastructure for the study of socially situated narrative practice in human-robot interaction.

Conviction Narrative Theory: A theory of choice under radical uncertainty

“…We recently developed a model and situated robotic system that combines sensorimotor and perceptual inputs and narrative inputs to construct a situation model. Through its own experience and human narration, the system uses narrative practice to build up narrative constructions or patterns that are associated with representations of the causal and temporal structure of physical and mental states in the situation model (Mealier et al, 2017;Pointeau, Mirliaz, Mealier, & Dominey, 2021). In a proof of concept, we demonstrated how the system could use existing narrative constructions to understand (and thus potentially make decisions as required in CNT) about newly experienced situations, inspired by the narrative practice hypothesis of Hutto and Gallagher, and the notion of narrative structure of reality and perception of Bruner, Ricoeur, Gergen and Gergen, and Nelson.…”

Narrative Event Segmentation in the Cortical Reservoir

“…Related models have used recurrent networks to integrate meaning over multiple word utterances in order to predict the N400 (Brouwer et al 2017, Rabovsky et al 2018, further supporting the role of recurrent connections in accumulating information over multiple words. We have also started to address how structured meaning in the narrative is extracted to build up a structured situation model (Mealier et al 2017, Pointeau et al 2021, so that the system can learn to answer questions about perceived events and narrative.…”

Narrative event segmentation in the cortical reservoir

“…That is, the elements contributing to the narrative meaning in the input are combined in the high dimensional representation, and a trained readout should be able to extract the desired meaning representations. Related research has started to address how structured meaning in the narrative is extracted to build up a structured situation model [61,62], so that the system can learn to answer questions about perceived events and narrative. Baldassano et al [8] note that the event segmentation they observe in high areas including angular gyrus and posterior medial cortex exhibit properties associated with situation model representations including "long event timescales, event boundaries closely related to human annotations, generalization across modalities, hippocampal response at event boundaries, reactivation during free recall, and anticipatory coding for familiar narratives" (p. 717).…”