How robust are measures of online control to offline mechanisms?

  • John de Grosbois University of Toronto
  • Luc Tremblay University of Toronto

Abstract

The importance of online visual feedback for the accurate completion of reaching movements is widely accepted. In contrast, the quantification of online feedback utilization can still be challenged. For example, many contemporary measures of online control are calculated across multiple trials, which brings about potential confounds via the contribution of offline trial-to-trial changes in performance. The current study sought to contrast the robustness of common measures of online control and a novel frequency-based measure (de Grosbois & Tremblay, 2015) to offline-induced changes in reaching trajectories. The main experimental conditions included reaches performed in no-vision conditions with: a) optical prisms that shifted the perceived location of the target and environment and b) terminal feedback [TF] via a 1-s window of vision after each movement. Also, control full-vision conditions were performed with and without TF. The measures of online control included two between-trial measures (i.e., variable error [VE] and normalized correlations of position at 75 % of movement time relative to the end-position [Z2]) and two within trial measures (i.e., the time-after-peak-velocity [TAPV] and a novel frequency analysis [pPower]). The results indicated that VE and TAPV were sensitive to the prism manipulation (i.e., not robust to offline control mechanism), whereas Z2 and pPower were relatively immune to the offline changes caused by the prisms. Supplementary analyses comparing the two full-vision conditions (i.e., with and without TF) indicated that the pPower measure was the most robust to offline influences.

Acknowledgments: Natural Sciences and Engineering Research Council of Canada (NSERC), Canada Foundation for Innovation (CFI), Ontario Research Fund (ORF)