Predicting Affective States from Screen Text Sentiment

TL;DR

This study explores how analyzing screen text viewed on smartphones can predict users' emotional states, demonstrating that multi-shot prompting with large language models significantly improves affect prediction accuracy.

Contribution

It introduces a novel approach using large language models with multi-shot prompting to analyze screen text for affect prediction, outperforming traditional methods.

Findings

Multi-shot prompting outperforms linear regression and zero-shot methods.

Contextual information enhances affective state prediction.

Text and sentiment data are valuable for wellbeing assessments.

Abstract

The proliferation of mobile sensing technologies has enabled the study of various physiological and behavioural phenomena through unobtrusive data collection from smartphone sensors. This approach offers real-time insights into individuals' physical and mental states, creating opportunities for personalised treatment and interventions. However, the potential of analysing the textual content viewed on smartphones to predict affective states remains underexplored. To better understand how the screen text that users are exposed to and interact with can influence their affects, we investigated a subset of data obtained from a digital phenotyping study of Australian university students conducted in 2023. We employed linear regression, zero-shot, and multi-shot prompting using a large language model (LLM) to analyse relationships between screen text and affective states. Our findings indicate that multi-shot prompting substantially outperforms both linear regression and zero-shot prompting, highlighting the importance of context in affect prediction. We discuss the value of incorporating textual and sentiment data for improving affect prediction, providing a basis for future advancements in understanding smartphone use and wellbeing.