Grounding Emotional Descriptions to Electrovibration Haptic Signals

TL;DR

A computational pipeline was developed using natural language processing techniques, such as GPT-3.5 Turbo and word embedding methods, to extract sensory and emotional keywords and group them into semantic clusters (i.e., concepts), which linked the keyword clusters to haptic signal features using correlation analysis.

Abstract

Designing and displaying haptic signals with sensory and emotional attributes can improve the user experience in various applications. Free-form user language provides rich sensory and emotional information for haptic design (e.g., ``This signal feels smooth and exciting''), but little work exists on linking user descriptions to haptic signals (i.e., language grounding). To address this gap, we conducted a study where 12 users described the feel of 32 signals perceived on a surface haptics (i.e., electrovibration) display. We developed a computational pipeline using natural language processing (NLP) techniques, such as GPT-3.5 Turbo and word embedding methods, to extract sensory and emotional keywords and group them into semantic clusters (i.e., concepts). We linked the keyword clusters to haptic signal features (e.g., pulse count) using correlation analysis. The proposed pipeline demonstrates the viability of a computational approach to analyzing haptic experiences. We discuss our future plans for creating a predictive model of haptic experience.

Figures (4)

• Figure 1: The electrovibration haptic device in our data collection user study and an example of free-form user description.
• Figure 2: The overview of our computational pipeline with three main phases: (1) Keyword extraction identifies sensory and emotional keywords from the user descriptions and groups the keywords into positive and negative sentiments, (2) Keyword clustering clusters the keywords using four word embedding methods, and (3) Correlation analysis extracts statistical signal features and calculates the correlations between signal features and concept clusters.
• Figure 3: T-SNE visualization of the ConceptNet Numberbatch word embeddings and clusters of positive (left) and negative (right) keywords in our dataset. Circles showcase example clusters to denote coherence in the results.
• Figure 4: The correlations between signal features and the clusters for positive (left) and negative (right) keywords.