Quantifying Low-Level Motor Effects of Emotion: Validating Fitts’ Law in Affective Contexts

Paper Title

Quantifying Low-Level Motor Effects of Emotion: Validating Fitts’ Law in Affective Contexts

Publication Info

• Topic area: Emotional modulation of motor performance in human-computer interaction.
• Keywords: Emotion, motor control, Fitts' Law, human-computer interaction, valence, arousal, task difficulty, throughput, affective computing, adaptive systems.

Background and Problem

• Problem / challenge: Prior research on emotion in human-computer interaction (HCI) has focused on high-level cognitive effects, neglecting how emotional states influence fine-grained motor behaviors like pointing and selection.
• Significance: Understanding these effects is critical for designing adaptive, emotion-aware interfaces that optimize user performance in real-world settings.
• Motivation and related work: While Fitts’ Law is a well-established framework for quantifying motor performance, its application in affective contexts remains unexplored. Existing research highlights the impact of emotion on attention and decision-making but lacks systematic analysis of its influence on low-level motor metrics like movement time, reaction time, error rate, and throughput.

Solution

• Proposed approach: The study investigates how emotional states, defined by valence and arousal, modulate motor performance in Fitts’ Law tasks under varying levels of difficulty.
• Novelty:
  1. Empirical demonstration of emotion’s impact on core motor behaviors in digital interaction.
  2. Integration of emotional dimensions into Fitts’ Law modeling to improve predictive accuracy.
  3. Context-sensitive analysis of emotional effects across low- and high-difficulty tasks.
• Procedure and key techniques:
  • Participants performed Fitts’ Law tasks under five emotional conditions (positive/negative valence, high/low arousal, neutral).
  • Emotional states were induced using validated image stimuli and measured using self-reports.
  • Task difficulty was varied using the Index of Difficulty (ID), with separate studies for low-ID (ballistic) and high-ID (corrective) tasks.
  • Performance metrics (movement time, reaction time, error rate, throughput) were analyzed using linear mixed-effects models.

Results

• Concrete findings:
  • Positive valence consistently improved throughput and reduced movement time across tasks.
  • High arousal facilitated performance in low-difficulty tasks but impaired it in high-difficulty tasks by slowing movements to prioritize accuracy.
  • Emotional effects were context-dependent, with optimal configurations shifting based on task demands.
• Advantage over baselines: Emotion-specific Fitts’ Law models achieved higher predictive accuracy (R² up to 0.910) compared to pooled models (R² = 0.764–0.577).
• Experiments / evaluation:
  • Study 1 (low ID): 22 participants completed tasks with IDs of 2.32–4.09.
  • Study 2 (high ID): 25 participants completed tasks with IDs of 5.04–7.01.
  • Emotional states were validated using self-reports and NASA-TLX workload ratings.
• Limitations and future work:
  • Static image-based emotion induction may limit ecological validity.
  • Fatigue effects from extensive trials and cursor occlusion may influence results.
  • Future studies should explore immersive emotion induction (e.g., VR), real-time emotion tracking, and individual differences in emotional reactivity.

Summary

This study demonstrates that emotional states systematically influence motor performance in Fitts’ Law tasks, with effects varying by task difficulty. Positive valence improved efficiency across conditions, while high arousal facilitated ballistic movements but impaired precision-dependent tasks. Emotion-specific modeling significantly enhanced predictive accuracy, highlighting the importance of incorporating emotional dimensions into motor performance frameworks. These findings provide actionable insights for designing adaptive, emotion-aware systems in domains requiring precision and low error tolerance, such as aviation and surgical robotics. Future work should address ecological validity and individual differences to further refine emotion-sensitive interaction models.