Manual Task Performance Under Distraction and Biomechanical Constraints
Overview
This project tested how distraction, screw-turn direction, and hand dominance affect performance on a constrained manual task.
Data and Experimental Design
Data came from 57 Cal Poly students. Each participant completed four trials covering dominant/nondominant hand and flexion/extension direction. Distraction (mental math vs. background noise) was assigned between subjects.
Outcome: time to fully unscrew a bolt (seconds).
Total observations: 228.
Design: mixed 2×2×2 factorial with repeated measures on hand and direction.
Methodology
We modeled log completion time with mixed-effects ANOVA. Hand, direction, and distraction were fixed effects; participant was a random effect to account for repeated measures. The initial model included full interactions, then was simplified by removing non-informative terms while retaining the hand-by-direction interaction.
Key Findings
Direction was the largest effect: extension was slower than flexion.
Nondominant-hand trials were also slower.
There was moderate hand-by-direction interaction evidence, suggesting the nondominant-hand penalty was smaller in extension than flexion.
Distraction showed no clear main effect or interaction.
Power Analysis and Design Implications
The study was powered for large biomechanical effects, but underpowered for small effects (like distraction). Power analysis suggested many more participants would be needed to detect subtle time differences. A better follow-up design is to make distraction a within-subject factor to improve sensitivity without increasing sample size as much.
Tools
R (mixed-effects modeling, ANOVA, power analysis).