Trial-to-trial dynamics and learning in a generalized, redundant reaching task

Abstract

Dingwell JB, Smallwood RF, Cusumano JP. Trial-to-trial dynamics and learning in a generalized, redundant reaching task. J Neurophysiol 109: 225-237, 2013. First published October 10, 2012 doi:10.1152/jn.00951.2011.-If humans exploit task redundancies as a general strategy, they should do so even if the redundancy is decoupled from the physical implementation of the task itself. Here, we derived a family of goal functions that explicitly defined infinite possible redundancies between distance (D) and time (T) for unidirectional reaching. All [T, D] combinations satisfying any specific goal function defined a goal-equivalent manifold (GEM). We tested how humans learned two such functions, D/T ϭ c (constant speed) and D·T ϭ c, that were very different but could both be achieved by neurophysiologically and biomechanically similar reaching movements. Subjects were never explicitly shown either relationship, but only instructed to minimize their errors. Subjects exhibited significant learning and consolidation of learning for both tasks. Initial error magnitudes were higher, but learning rates were faster, for the D·T task than for the D/T task. Learning the D/T task first facilitated subsequent learning of the D·T task. Conversely, learning the D·T task first interfered with subsequent learning of the D/T task. Analyses of trial-to-trial dynamics demonstrated that subjects actively corrected deviations perpendicular to each GEM faster than deviations along each GEM to the same degree for both tasks, despite exhibiting significantly greater variance ratios for the D/T task. Variance measures alone failed to capture critical features of trial-to-trial control. Humans actively exploited these abstract task redundancies, even though they did not have to. They did not use readily available alternative strategies that could have achieved the same performance.

An Uncontrolled Manifold Analysis of Arm Joint Variability in Virtual Planar Position and Orientation Telemanipulation

“…, Instead of being considered as a burden they proposed the role of motor redundancy in achieving high performance while minimizing the control effort [23]. Several methods were proposed to quantify the level of redundancy exploitation by evaluating the variability distribution in geometrically defined subspaces [24]- [26]. The Tolerance Noise Covariation (TNC) principle describes how humans cope with the neuromotor noise and decompose variability into exploitation of task tolerance, stochastic noise, and covariation between variables.…”

“…Thanks to the task dependent definition of the TIM, the TNC has been used to estimate how the three components of variance change during the learning process [24]. Another performance analysis of redundancy exploitation is represented by the Goal Equivalent Manifold (GEM) [25], [26]. The GEM is presented as the set of possible task solution strategies that arise from the theoretical definition of a goal function.…”

Effects of local and widespread muscle fatigue on movement timing

“…In fatiguing tasks, people exploit redundancy by altering activation patterns of individual muscles (Madeleine and Farina 2008; Holtermann et al 2010) or altering inter-joint and inter-muscular coordination (Gorelick et al 2003; Côté et al 2008). Identifying and exploiting redundant movement task solutions may be a broadly used motor control strategy (Cusumano and Dingwell 2013; Dingwell et al 2013). Thus, individuals alter their coordination strategies to maintain the same task outcomes when fatigued (Côté et al 2002; Selen et al 2007; Gates and Dingwell 2008; Fuller et al 2009; Fuller et al 2011).…”

“…The Goal Equivalent Manifold (GEM) approach provides an analysis that maps the observed dynamics of task performance, at the level of the body variables (e.g. position, velocity), onto a goal space that is defined independent of the subject’s actual performance (Cusumano and Cesari 2006; Gates and Dingwell 2008; Cusumano and Dingwell 2013; Dingwell et al 2013). The GEM is the set of all possible body state solutions that exactly meet the task goal.…”

“…In contrast, the UCM analysis decomposes variability about a manifold defined by the average of the observed experimental data, and hence assumes this average recorded trajectory reflects the desired movement goal. This can sometimes be problematic, however, as this average changes between subjects and conditions (Dingwell et al 2013). Additionally, for a given GEM, we can assess how changes in the body parameters affect the goal error during each movement repetition.…”