Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

From Delays to Densities: Exploring Data Uncertainty through Speech, Text, and Visualization

Chase Stokes, Chelsea Sanker, Bridget Cogley, Vidya Setlur

TL;DR

This work addresses how to communicate data uncertainty by comparing unimodal representations across speech, text, and visualization under a common decision task. It systematically maps a design space of concrete versus fuzzy uncertainty signals, using pilot studies to select stimuli and two crowdsourced experiments to evaluate effects on crossover decisions, rationality, confidence, and trust. The findings show that visualization and text improve rational decision-making, while speech engenders higher trust though potentially riskier choices, and that precision manipulations often fail to produce robust effects. The results inform future multimodal uncertainty communication and emphasize educational and accessibility considerations for conveying uncertain information in real-world contexts.

Abstract

Understanding and communicating data uncertainty is crucial for making informed decisions in sectors like finance and healthcare. Previous work has explored how to express uncertainty in various modes. For example, uncertainty can be expressed visually with quantile dot plots or linguistically with hedge words and prosody. Our research aims to systematically explore how variations within each mode contribute to communicating uncertainty to the user; this allows us to better understand each mode's affordances and limitations. We completed an exploration of the uncertainty design space based on pilot studies and ran two crowdsourced experiments examining how speech, text, and visualization modes and variants within them impact decision-making with uncertain data. Visualization and text were most effective for rational decision-making, though text resulted in lower confidence. Speech garnered the highest trust despite sometimes leading to risky decisions. Results from these studies indicate meaningful trade-offs among modes of information and encourage exploration of multimodal data representations.

From Delays to Densities: Exploring Data Uncertainty through Speech, Text, and Visualization

TL;DR

This work addresses how to communicate data uncertainty by comparing unimodal representations across speech, text, and visualization under a common decision task. It systematically maps a design space of concrete versus fuzzy uncertainty signals, using pilot studies to select stimuli and two crowdsourced experiments to evaluate effects on crossover decisions, rationality, confidence, and trust. The findings show that visualization and text improve rational decision-making, while speech engenders higher trust though potentially riskier choices, and that precision manipulations often fail to produce robust effects. The results inform future multimodal uncertainty communication and emphasize educational and accessibility considerations for conveying uncertain information in real-world contexts.

Abstract

Understanding and communicating data uncertainty is crucial for making informed decisions in sectors like finance and healthcare. Previous work has explored how to express uncertainty in various modes. For example, uncertainty can be expressed visually with quantile dot plots or linguistically with hedge words and prosody. Our research aims to systematically explore how variations within each mode contribute to communicating uncertainty to the user; this allows us to better understand each mode's affordances and limitations. We completed an exploration of the uncertainty design space based on pilot studies and ran two crowdsourced experiments examining how speech, text, and visualization modes and variants within them impact decision-making with uncertain data. Visualization and text were most effective for rational decision-making, though text resulted in lower confidence. Speech garnered the highest trust despite sometimes leading to risky decisions. Results from these studies indicate meaningful trade-offs among modes of information and encourage exploration of multimodal data representations.
Paper Structure (36 sections, 3 figures, 4 tables)

This paper contains 36 sections, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Communicating Uncertainty in Data Visualization
  4. Communicating Uncertainty in Text
  5. Communicating Uncertainty in Speech
  6. Decision-making and Reasoning with Uncertain Information
  7. Research Goals
  8. Design Space of Visualization, Text, and Speech Stimuli
  9. Visualization Stimuli
  10. Text Stimuli
  11. Speech Stimuli
  12. Discussion
  13. Experiment Overview
  14. Participants
  15. Task Design
  16. ...and 21 more sections

Figures (3)

  • Figure 1: Proportion of decision types for each condition in Experiment 1. Overall, decisions were mostly rational. Speech was the least rational mode, with a greater proportion of risky decisions.
  • Figure 2: Experiment 1 confidence ratings ranged from 50 to 100. Confidence was lower for text than for visualization forecasts.
  • Figure 3: Experiment 1 trust ratings. Participants trusted speech the most in comparison to both text and visualization.