Movement-goal relevant object shape properties act as poor but viable cues for the attribution of motor errors to external objects

Shanaathanan Modchalingam; Maria N Ayala; Denise Y P Henriques

doi:10.1371/journal.pone.0300020

. 2024 Mar 28;19(3):e0300020. doi: 10.1371/journal.pone.0300020

Movement-goal relevant object shape properties act as poor but viable cues for the attribution of motor errors to external objects

Shanaathanan Modchalingam ^1,^2,^*,^#, Maria N Ayala ^1,^3,^#, Denise Y P Henriques ^1,²

Editor: Shenbing Kuang⁴

PMCID: PMC10977729 PMID: 38547216

Abstract

When a context change is detected during motor learning, motor memories—internal models for executing movements within some context—may be created or existing motor memories may be activated and modified. Assigning credit to plausible causes of errors can allow for fast retrieval and activation of a motor memory, or a combination of motor memories, when the presence of such causes is detected. Features of the movement-context intrinsic to the movement dynamics, such as posture of the end effector, are often effective cues for detecting context change whereas features extrinsic to the movement dynamics, such as the colour of an object being moved, are often not. These extrinsic cues are typically not relevant to the motor task at hand and can be safely ignored by the motor system. We conducted two experiments testing if extrinsic but movement-goal relevant object-shape cues during an object-transport task can act as viable contextual cues for error assignment to the object, and the creation of new, object-shape-associated motor memories. In the first experiment we find that despite the object-shape cues, errors are primarily attributed to the hand transporting the object. In a second experiment, we find participants can execute differing movements cued by the object shape in a dual adaptation task, but the extent of adaptation is small, suggesting that movement-goal relevant object-shape properties are poor but viable cues for creating context specific motor memories.

Introduction

Contextual information about the environment, the objects we interact with, and our own body is crucial for reliable motor control. Different environments may require different muscle activations to perform similar motor tasks, and changes to our end effectors, such as injuries to a hand or modifications to a tool, may require different movement kinematics. It is well established that the motor system can select from multiple previously-learned internal models, termed motor memories, cued by internal or external contexts to execute movements [1–6]. Along with features of the movement itself [1, 7], contextual cues can aid in reliably selecting the correct internal model [8–11].

When the predicted consequences of a movement does not match the actual outcome, the difference, known as the sensory prediction error, can be used to update the internal models used to make the prediction [12–14]. Alternatively, if the sensory prediction error is sufficiently large, existing models that better predict the observed outcome given the intended movement may be activated. In the absence of any existing internal models that predict the observed outcome, new internal models may be created [1, 7].

The decision on whether to update an existing internal model or to create a new one can be informed by several factors. Features of the movement itself, as compared to predicted consequences of internal models, can be used to select an internal model for a given context [1]. Attributing errors to specific causes using available contextual cues can also allow for quick and reliable selection of internal models when those causes are detected [4, 5]. During tool use for example, errors can be attributed to external sources, such as the environment or the object being interacted with, and internal sources, such as the arm or hand used in the interaction. If the attribution of errors is accurate, models specific to the currently detected context can be accessed and updated. Additionally, accurate attribution of errors can allow for the switching of internal models before a movement needs to be initiated, and thus an error needs to be committed and perceived, allowing fast switching between multiple internal models and simultaneous adaptation to multiple contexts. In this study, we explore the types of cues that can lead to reliable attribution of errors to external sources and in turn facilitate fast switching of motor plans.

Cues that are “intrinsic” to a movement, i.e., cues that directly affect movement dynamics, such as movement direction, type, sequence, or posture, reliably facilitate fast context switching [9, 11, 15–19]. Findings on cues “extrinsic” to movement dynamics on the other hand, such as background colour and object identity are mixed [17, 19–23]. In general, extrinsic cues are often not sufficient for accurate error attribution during motor learning [20, 24]. Recent findings revealed that some extrinsic cues, such as colour [8], the action-effects of the effector [25], and visual location of the movement [25] can elicit dual adaptation if explicit adaptation processes (e.g., strategy formation) are engaged during adaptation, explaining some of the mixed findings and suggesting a critical role of cognitive strategy in dual adaptation [8]. Furthermore, extrinsic cues that inform of the dynamics of the interaction-point of tools [26, 27] can facilitate distinct motor memory formation. In these studies, the property of the tool that may cue the presence of a perturbation is task relevant. For example, in a study by McGarity-Shipley and colleagues [26], the location of the control point of a tool–the point with which the object interacts with the world–both cued the presence of opposing perturbations to movements and was integral to completing the task. We test if task relevancy itself can be enough of a cue to facilitate multiple motor memory formation.

In this study, we increased relevancy of an extrinsic cue (the shape of an object) to task goals by having participants perform an object transport task in an immersive virtual-reality environment where they move a 3D object to one of two receptacles. The shape of the object determined the receptacle it must be transported to. To isolate the object-shape cue, we applied a visuomotor rotation (VMR) to movements only when transporting objects (VMR direction mapped to object shape) while ensuring that movement dynamics, object dynamics and interaction dynamics were the same between differently shaped objects. In one experiment, we tested whether such an extrinsic cue was sufficient to allow for associated motor errors to be attributed to the object itself, rather than the hand. In a second experiment, we tested whether further increasing task relevancy by forcing attention to the object-shape in a dual adaptation task allowed for the formation of distinct motor memories for differently shaped objects. We found that object-shape cues did not serve as sufficient contexts to elicit object-shape specific adaptation when adapting to a single visuomotor rotation, and were poor but viable cues for object-shape specific adaptation when adapting to multiple shape specific perturbations.

Methods

Participants

Sixty-two participants (47 Female, 19.87 ± 3.15 years of age, mean ± SD) participated in the study in one of two experiments: “Single Learning” (N = 32, 24 Female, 20.03 ± 3.21 years of age) and “Dual Learning” (N = 30, 23 Female, 19.69 ± 3.14 years of age). All participants were right-handed, had normal or corrected-to-normal vision, and were naïve to visuomotor learning experiments. Participation was voluntary and all participants provided written, informed consent. All data for this study was collected at the Sensorimotor Control Lab at York University between October 16, 2019, and March 03, 2020. The procedures used in this this study were approved by York University’s Human Participant Review Committee and all experiments were performed in accordance with institutional and international guidelines.

Apparatus

Participants sat on a height-adjustable chair facing a table placed at waist level. Participants donned a head-mounted display system (HMD: Oculus Rift Consumer Version 1; resolution 1080 by 1200 for each eye; refresh rate 90 Hz) and held an Oculus Touch controller in their right hand. Participants in the Single Learning experiments also held an Oculus Touch controller in their left hand. All visual feedback was provided via the HMD in an immersive virtual-reality experiment space developed in Unity 3D. We used the Unity Experiment Framework to handle trial and block schedules within the experiments [28]. Three Oculus Constellation sensors tracked the positions of the HMD and Oculus Touch controllers. To begin the experiments, participants placed their left hand on a magnetized docking apparatus (7.5 cm in diameter) attached to the table edge at body-midline. This apparatus acted as the starting position for each trial in the experiments and aided in returning the right hand to the starting position in the absence of visual feedback. The Oculus Touch controllers were equipped with magnets to aid attachment to the magnetized dock.

Object-transport task

All visual feedback during the task was provided in the immersive virtual reality (VR) environment. In all experiments, participants repeatedly performed the Object-transport task. Each trial of the task began when participants returned the Oculus Touch controller held in their right hand to the magnetized dock. The dock was located near the starting position for each trial, termed the “Home” position. At the start of each trial, participants were shown an avatar-like representation of their hand (Fig 1A and 1B). An object in the shape of a cube (2 cm x 2 cm x 2 cm) or a sphere (2 cm diameter) appeared 10 cm away from the Home position, and two receptacles (box and cylinder shaped, depth of 2 cm) appeared 15 cm or 12 cm away from the object for the Single and Dual Learning experiments respectively. The opening of the box receptacle was 4 cm x 4cm x 4 cm and the opening of the cylinder was 4 cm in diameter. Participants were required to transport the object within 1 cm of the center of the receptacle in any orientation to complete the task. The target of the transport movement during a given Object-transport task was the receptacle with an opening matching the shape of the object.

Fig 1 — A-B: Trial protocol during a Rotated Object-transport task in the Single (A) and Dual (B) Learning experiments. In both experiments, participants began a trial at the Home Position (H) and made a reaching movement toward an object. Each trial ended when participants transported the target to the corresponding receptacle. Solid lines represent visual movement direction and dashed lines represent real movement directions. C. The order of tasks during the Single Learning experiment. For each set of the Clamped Object-transport task, participants transported an object with the same or different shape as comparted to the object transported during the Rotated Object-transport tasks. D. The order of tasks during the Dual Learning experiment. All tasks in the Dual Learning experiment were performed using the right hand.

In the Single Learning experiment, participants transported objects of a single shape (counter-balanced between participants) during each block (Fig 1C). The trial-by-trial locations of the correct receptacles and the object’s spawn locations were pseudorandomized, appearing at 45, 90, or 135 degrees in polar coordinates relative to the Home position. In the Dual Learning experiment, participants transported objects of both shapes during each block. Here, the object-shapes appeared in trial-sets of 3, providing participants multiple opportunities to learn a shape-perturbation mapping before experiencing different perturbations to their movements. The trial-set-by-trial-set order of the object-shapes was pseudorandomized, ensuring that each object-shape appeared an equal number of times within a block. The trial-by-trial locations of the correct receptacles were pseudorandomized, appearing at 45, 90, or 135 degrees in polar coordinates centered on the object’s spawn location.

Each trial contained two steps: Reach and Transport. During the Reach step, participants moved directly towards the object and made a pinch gesture using the index finger and thumb to interact with it. Once the gesture was completed the Reach step ended and the Transport step began. To highlight the object identity, the hand representation disappeared, and participants transported the object to the corresponding receptacle. A trial ended when participants moved the object to the centre of the corresponding receptacle.

In each experiment, participants performed three variations of the Object-transport task: Aligned, Rotated, and Clamped. The Reach step, during which the visual representation of the hand was spatially aligned with the participant’s real hand, remained consistent across all variations. During the Transport step, the movement path of the object was aligned with the real movement path in the Aligned variation and a 30° visuomotor rotation (either clockwise or counterclockwise) was applied on the horizontal plane in the Rotated variation. In the Clamped variation, to prevent feedback-based corrections during a movement, the object’s movement path was constrained to travel directly to the appropriate receptacle during the Transport step, with the object’s distance from its starting position matched to the hand’s distance from the object’s initial location.

Experiment protocol

All participants began the experiment with their right-hand controller magnetically attached to the magnetized dock. They were instructed to make smooth, direct movements throughout the experiment. Although no strict time limits were imposed, participants were encouraged by the experimenter to move in a quick and uniform pace throughout the experiment and were warned if they moved too slowly. Before beginning the experiments, participants completed practice trials to familiarize themselves with the virtual environment and task. Following the practice trials, participants completed one of either the Single Learning or Dual Learning experiments. Both experiments involved an Aligned phase, used to measure individual baseline performances, and one or more Training phases (Fig 1). The Dual Learning experiment also included a short Test phase. In all phases, participants repeatedly performed the Object-transport task. Participants performed the Aligned Object-transport task during the Aligned phase of the experiments and the Rotated Object-transport task during the Training phase. Participants also performed Clamped Object-transport tasks in all phases.

In the Single Learning experiment, we investigated if adapting to a perturbation when transporting an object could be linked to the object’s shape if the shape was relevant to the task goal. Participants completed 3 phases in the experiment during a single session: The Aligned phase, and two Training phases (Fig 1C). To prevent fatigue, each phase of the experiment was preceded by a short break lasting a minimum of 30 seconds. During the Aligned phase, participants performed 60 trials of the Aligned Object-transport task using their dominant hand. Four 6-trial blocks of the Clamped Object-transport tasks were interspersed within the Aligned phase following each set of 15 Aligned trials. During the first Training phase, following 6 trials of the Aligned Object-transport task, participants adapted to a 30° visuomotor rotation (clockwise or counterclockwise, counter-balanced across participants) using a designated object-shape (either cube or sphere, counter-balanced across participants) for 180 trials of the Object-transport task. Following the first set of Rotated Object-transport tasks, we tested if adaptation transferred to using either a differently shaped object or a different hand using Clamped Object-transport tasks where participants transported differently shaped objects, used a different hand to transport the object, or both. If adaptation was specific to the object shape, it would persist when participants switched hands, but diminish when they encountered a different object shape. Participants first performed two 6-trial blocks of the Clamped Object-transport task, with a 15-trial block of the Rotated Object-transport task in between. Then, in the second Training Phase, participants first performed 24 trials of the Rotated Object-transport task using the designated object-shape, then performed six 6-trial blocks of the Clamped Object-transport task, with 15-trial blocks of the Rotated Object-transport task in between.

In the Dual Learning experiment, we investigated whether participants could concurrently adapt to two distinct visuomotor rotations at the same time when the perturbations were matched to specific object-shapes. Here, participants completed 4 phases in the experiment during a single session: the Aligned phase, followed by two Training phases and a Test phase. Each phase was again preceded by a short break lasting a minimum of 30 seconds. During the Aligned phase, participants performed 30 trials of the Aligned Object-transport task. During each trial, participants transported one of the two possible object-shapes (cube or sphere, order pseudorandomized using 3-trial sets to ensure each shape was transported an equal number of times). During both Training phases, participants adapted to two opposing 30° visuomotor rotations mapped to two distinct object-shapes (cube or sphere, object-shape-rotation mappings counterbalanced across participants, order pseudorandomized using 3-trial sets to ensure each shape was transported an equal number of times) for 180 trials. We tested whether participants dually adapted to both visuomotor perturbations during the Training phase and tested whether such adaptation relied on feedback-based corrections using Clamped Object-transport tasks. Following the 180-trial block of Rotated Object-transport tasks, participants performed 6 trials of the Clamped Object-transport task (object-shape order pseudo-randomized using 3-trial sets). Additionally, to ascertain if participants experienced implicit dual adaptation, we conducted a short Test phase using 12 trials of the Clamped Object-transport task (object-shape order pseudo-randomized using 3-trial sets). To determine if participants adapted implicitly, we used an altered version of a process dissociation procedure in which we instructed participants to refrain from using any strategies they may have developed during the training phase [29, 30].

Data analysis

All position measures were recorded using Unity 3D (2020.1.17, Unity Technologies, San Francisco, CA). Data was analyzed offline using R and was accessed for analysis from the start of data collection (October of 2019) to January of 2024. All data and analysis scripts, along with supplementary analysis notebooks, can be found in this project’s Open Science Framework repository (https://osf.io/twgc8/).

We assessed performance in the object-transport task by calculating the hand angle (HA) of a movement. We defined HA as the angular deviation from a straight-line movement to the target at 3 cm from the object’s starting position. HA is a measure of initial directional error, prior to effects of feedback-based corrections. We corrected for individual biases in transport behavior by subtracting individual baseline performances per Target location from the HA. Further, to compare across counterbalanced rotation directions, we normalized HA relative to the direction of the applied visuomotor rotation. We conducted all subsequent calculations and data analysis on the baseline corrected and normalized data.

To investigate performance changes over time, we compared mean HAs during the initial and final trial-sets of the first Training phase in each experiment, when participants were abruptly introduced to the visuomotor perturbations. The initial trial-set consisted of 3 trials: one reach to each of the 3 available targets. The final trial-set consisted of 9 trials.

To determine if participants in the Single Learning experiment adapted to the visuomotor rotation, we compared HA during the Reach and Transport step of the first and last trial-sets during the Training phase using separate paired sample t-tests. We then tested possible error attribution to either the object or the hand by comparing HA during the Clamped Object-transport task under various conditions. We performed a 2x2 repeated-measures analysis of variance (ANOVA) with the hand (same or different hand) and the object shape (same or different object shape) as within-subject factors. We used Tukey’s HSD to conduct pairwise analysis when post-hoc tests were necessary. For all statistical tests, we additionally computed Inclusion Bayes factors between models that include or do not include relevant effects [31].

To determine if participants in the Dual Learning experiment adapted to opposing rotations, we first determined if participants HAs changed over the course of the Training phase using paired sample t-tests. We separately compared the changes in HAs during the Reach and Transport steps. Additionally, to determine if participants performed differently without feedback-based corrections, we compared non-normalized HAs during the Transport step of Clamped Object-transport tasks in the Dual Learning experiment, where participants transported objects associated with positive and negative rotations using a paired sample t-test. To determine if object-shape related changes in HA were implicit, we compared non-normalized HAs during the Transport step of the Clamped Object-transport tasks when participants were instructed to not employ any strategies they may have learned to compensate for the visuomotor perturbations during the Training phase.

Finally, we tested whether the time taken to initiate and execute the Reach step, the time taken to initiate the Transport step, or the time taken to execute the Transport step changed during the Training phase of both experiments. For all three measures, we conducted separate 2x2 mixed ANOVAs with the experiment as a between-subject factor and the trial-set as a within-subject factor. When post-hoc tests were necessary, we used Tukey’s HSD to perform pairwise comparisons.

Results

We conducted two experiments to determine if goal-relevant visual properties of objects would suffice to enable error attribution to an object during Object-transport tasks. In the first experiment, participants transported an object with their right hand. A 30° visuomotor rotation was applied to their movement while transporting the object. To test for attribution of error, they were asked to transport a novel object, with either the trained hand or with their untrained, contralateral hand.

In the second experiment, to understand whether the goal-relevant extrinsic cues could facilitate dual adaptation, we mapped opposing visuomotor rotations (30° CCW or 30° CW) to distinct object shapes (sphere or cube). To test for dual adaptation, participants preformed Clamped Object-transport tasks where the direction of movement was clamped to head directly to the target. To test for possible implicit dual adaptation, participants also performed Clamped Object-transport tasks while disengaging any strategies they may have formed during the Training phase.

Experiment 1: Single learning

Visuomotor adaptation

We compared normalized hand angles (HA) during the initial and final trial-sets of the Training phase (Fig 2A–2C). We found no evidence of change in the HAs during the Reach step, where the location of the hand was aligned with the real hand, between the initial and final trial-sets of the Training phase (t₍₃₁₎ = -0.554, p = 0.583, BF₁₀ = 0.22). During the Transport step of the Object-transport task, participants adapted their movements to counter the visuomotor perturbation over the course of the Training phase (t₍₃₁₎ = 17.987, p < 0.001, BF₁₀ = 8.36x10¹⁴, Fig 2B). Over 180 trials (60 trial-sets), participants adapted to 79.0% of the rotation (23.7° hand angle during the final trial-set of the Training phase).

Error attribution

We first determined if retrieval of the adapted motor memory was cued by the object-shape. After adapting to a 30° visuomotor rotation, we tested HAs during the Transport step of Clamped Object-transport tasks. (Fig 2D: main effect of hand: F_{(1, 31)} = 174.588, p < 0.001, η² = 0.70, BF_incl = 3.10x10³⁶). We found moderate evidence that the object shape did not affect HA (F_{(1, 31)} = 0.030, p = 0.863, η² = 0.00007, BF_incl = 0.152).

After collapsing across object shape, we found moderate evidence that participants performed the same as during the Rotated Object-transport tasks when transporting objects with the trained hand (t₍₃₁₎ = -0.588, p = 0.560, BF₁₀ = 0.22). Participants also retained a small but significant portion (~19%) of the change in HA developed during the Transport step of the Object-transport task when performing with the untrained hand (t₍₃₁₎ = 4.7836, p < 0.001, BF₁₀ = 589.64).

Experiment 2: Dual learning

Visuomotor adaptation

Participants in the Dual Learning experiment completed a typical dual adaptation paradigm with object identity distinguishing the perturbation experienced on a given training trial. Overall, when collapsing across rotations, we found moderate evidence of no change in HA across the Training phase during the Reach step of the Object-transport task, where the location of the hand was aligned with the real hand (t₍₂₉₎ = 0.702, p = 0.488, BF₁₀ = 0.24). During the Transport step, participants were able to successfully counter 11.9% (3.59°) of the 30° visuomotor rotation by the final 3 trial-sets of the Training phase relative to their HAs during the Baseline phase (t₍₂₉₎ = 3.68, p < 0.001, BF₁₀ = 35.0). We failed to find any evidence for changes in HAs during the Transport step across the Training phase (t₍₂₉₎ = 1.90, p = 0.067, BF₁₀ = 0.95), suggesting learning was approaching saturation within the first three trials of experiencing a perturbation (Fig 3A–3C).

Fig 3 — Angular deviations from a straight-to-target movements (Hand Angle) during the Training phase in the Dual Learning experiment. A. Hand Angles over the first 180 trials in the Training phase. B. Hand Angles during the Reach step of the initial and final trial-sets during the Training phase. C. Hand Angles during the Transport step of the initial and final trial-sets during the Training phase. D. Non-normalized Hand Angles during the Transport step of Clamped Object-transport tasks while transporting the object associated with clockwise (CW) and counterclockwise (CCW) perturbations. E. Non-normalized Hand Angles during the Transport step of Clamped Object-transport tasks while transporting the object associated with clockwise (CW) and counterclockwise (CCW) perturbations when participants were instructed to exclude strategies that may have formed during the Training phase. Shaded areas represent 95% CIs.

For the Dual Learning experiment, we observed small changes in non-normalized HAs in the expected directions for each distinctive object during the Clamped Object-transport trials (Fig 3D). Participants’ HAs differed between the two object shapes (t₍₂₉₎ = -2.58, p = 0.0153, BF₁₀ = 3.14). The magnitudes of the changes when compared to baseline performance were small: 1.47° and 3. 56° when transporting the shape associated with the clockwise and counter-clockwise visuomotor rotation respectively (Fig 3C). These changes in non-normalized HA further decreased when participants were instructed not to employ a strategy, resulting in a lack of statistical reliability (Fig 3E: t₍₂₉₎ = -1.89, p = 0.069, BF₁₀ = 0.93). Given the large inter-subject variability and small magnitude of error, we do not make strong conclusions about implicit and explicit subprocesses underlying the observed dual adaptation.

Reaction and execution times

We also tested if the time taken to initiate and execute the Reach step, the time taken to initiate the Transport step, or the time taken to execute the Transport step changed during the Training phase of both experiments (Fig 4). Although there was an interaction between the trial-set and experiment in time taken to complete the Reach step (F_{(1, 60)} = 8.18, p = 0.006, η² = 0.062, BF_incl = 21.76), this was driven by higher step-completion times in only the initial trial-set of Dual Learning experiment when compared to the final trial-sets of both the Single Learning (Tukey’s HSD: t₍₆₀₎ = 3.62, p = 0.003), and Dual Learning (Tukey’s HSD: t₍₆₀₎ = 4.29, p < 0.001) experiments (Fig 4A). By the end of the Training phase, participants in both experiments took a similar amount of time to complete the Reach step (Tukey’s HSD: t₍₆₀₎ = -2.05, p = 0.182). This may be due to the confusing nature of abruptly experiencing opposing perturbations.

Both experiment (F_{(1, 60)} = 6.45, p = 0.014, η² = 0.060, BF_incl = 2.74) and trial-set (F_{(1, 60)} = 38.47, p < 0.001, η² = 0.205, BF_incl = 1.12x10⁶) affected reaction times during the Transport step across the Training phase (Fig 4B). Surprisingly, participants in the Dual Learning group were faster to start executing their transport movements by the end of training when compared to both the final trial-set of the Single Learning experiment (Tukey’s HSD: t₍₆₀₎ = -3.52, p = 0.005) and the initial trial-set of the Dual Learning experiment (Tukey’s HSD: t₍₆₀₎ = 5.08, p < 0.001). This may indicate low cognitive strategy use, leading to the small behavioural changes seen in the Rotated Object-transport task analysis.

Finally, the experiments differently affected execution time changes between initial and final trial-sets (Fig 4C: F_{(1, 60)} = 4.95, p = 0.030, η² = 0.033, BF_incl = 7.42). Despite similar execution times for the Transport step in both groups during the initial trial-set (Tukey’s HSD: t₍₆₀₎ = 0.543, p = 0.948), and participants in the Dual Learning group having to transport objects a shorter distance than those in the Single Learning group, execution times were higher in the Dual Learning group in the final trial-set (Tukey’s HSD: t₍₆₀₎ = 5.154, p < 0.001). This is likely due to additional correction times needed by participants making larger errors in the final trial-set of the Dual Learning experiment.

Discussion

The formation of new motor memories associated with detectable contexts enables quick switching between motor contexts by activating associated motor plans when a specific context is detected. Assigning errors from movements to plausible causes is crucial for associating motor memories with specific contexts, as these causes serve as contextual cues for future retrieval of motor commands. Properties of movement end effectors, such as human body parts or external objects, that are intrinsic to movement dynamics reliably aid in forming motor memories tied to specific contexts. The role of properties extrinsic to movement dynamics, like visual properties of the object of interaction, is less well understood. Our findings show that visual object-shape cues linked to a visuomotor perturbation are insufficient for creating distinct motor memories when adapting to a single visuomotor rotation, even when the object-shape was relevant to task goals. When the mapping between two distinct object-shapes and two distinct perturbations were repeatedly experienced in a dual-learning paradigm, participants’ performance differed between object-shapes, but expressed only partial adaptation.

In the Single Learning experiment, we directly tested if motor control changes to a visuomotor perturbation were linked to an object’s shape. Participants adapted to a visuomotor perturbation with a single object-shape, and we assessed if the adaptation persisted when the objectshape or the hand used was changed. In our experiment learning was entirely attributed to the hand and environment, not the object-shape. When using the same hand for differently shaped objects, the adaptation effects transferred fully to the new object. This result suggests that object shape was not a strong enough contextual cue to facilitate assignment of motor error. This is a common finding for cues extrinsic to movement dynamics [20, 24]. While some studies show that purely visual properties of a task, such as colour [22], or additional movement of a visual end-effector [26] can facilitate distinct motor memory formation, our findings indicate that movement-goal relevant object-shape properties alone are likely insufficient.

When using the non-adapted hand to move objects, both familiar and unfamiliar, ~19% of the total adaptation was retained. This component of learning is consistent with previous findings [32], and may reflect learning to perform the task itself or updates to a baseline internal model for performing object-transport movements [1].

When using the adapted hand, performance during clamped trials in the Single Learning experiment, where participants were not asked to exclude strategies, was similar to performance during late training. Although decay of adaptation is minimized in reaching tasks with clamped feedback, we did not expect complete retention from the rotated-feedback Object-transport tasks [33–36]. In our experiments, trials with clamped feedback and preceding trials with rotated feedback shared the same target locations requiring similar movement vectors to complete the task. In addition to model-based motor learning, model free mechanisms such as reinforcement learning to favour movement directions towards the learned target directions may have resulted in the apparent high retention and low decay of adaptation in our tasks with clamped feedback [37, 38].

Adaptation to errors attributed to objects has been observed in the past [32]. In a study testing the assignment of credit to errors during Object-transport tasks, Kong et al. [32] trained participants to transport familiar objects (cups). The researchers observed a reduction in measured adaptation when participants transported a different object while still using the trained hand. Since transporting cups is a common task, participants likely employed pre-existing internal models for this category of interactions that could be flexibly switched to given changes in object identity. Unlike previous studies, our study’s virtual reality environment, and the dynamics of interactions within that environment, was unfamiliar to most participants. Since participants transported objects with both the left and the right hand during the Aligned phase of the experiment, and since posture, perspective, and the used hand can lead to the formation of distinct motor memories and enable dual adaptation [9, 25, 39], new hand-specific internal models for object transport may have been created, leading to low transfer across hands. The additional haptic feedback from transported real-world objects could further facilitate the development of motor memories specific to the movements involved in transporting these objects. Cues intrinsic to movements such as the altered forces or postures required when transporting real-world objects can reliably facilitate object-specific adaptation [15, 40, 41]. In contrast, our virtual task required the same forces to move the hand with and without the objects. Indeed, participants expressed hand-specific adaptation despite participants not experiencing any perturbation during the Reach step of the Transport tasks. The two steps of the Transport task may have been poorly differentiated due to the absence of additional, non-visual cues.

In the Dual Learning experiment, participants could perform the object-transport movements differently when transporting objects with two distinct shapes matched to two distinct perturbations by the end of the Training phase. Using an altered process dissociation procedure, we attempted to qualify if these learning effects were due to the development of explicit strategies or implicit motor memories. We did not find conclusive evidence to suggest either subprocess of adaptation drove the overall learning effects. Studies exploring dual adaptation in conventional upper-limb motor learning tasks suggest a strong contribution of explicit processes [8, 42]. Although patterns of adaptation are similar in immersive virtual environments [39, 43–45], the effects of interactions within a new environments on the subprocesses of motor learning are not yet well explored. Early work suggests that adaptation in immersive environments may rely more heavily on explicit learning [43]. Further, when performing reaching tasks in immersive virtual environments, stable viewpoints while parallelly adapting to two perturbations, like in our study, may also lead to more explicit learning [39]. Furthermore, interacting in the virtual environment may have affected the outcome of the Dual Learning experiment due to the “newness” of the environment. Newly created internal models for new environmental contexts likely initially allow for high variability in predicted movements [1]. In immersive virtual environments where object interactions are unfamiliar to participants, new internal models of object transport may be flexible enough to encompass the variance of visual movements ascribable to opposing visuomotor perturbations. In future work, long-term exposure to virtual environments, or recruitment of participants with immersive-virtual-reality exposure, may mitigate such “newness” effects.

Considering people’s ability to switch between different tools and the strong reliance on the visual system in motor adaptation, we predicted that visual shape properties of objects relevant to the performed movement could viably cue activating context-dependent motor memories. However, in this study, we find they are not sufficient for motor memory formation and switching in object-transport tasks. It is possible that in our tasks, the object-shape, although relevant to the task of choosing the target of the reach, was not sufficiently relevant to the lower-level task of transporting an object. That is, despite our task requiring explicit attention to the object-shape to determine the movement goal, it was not relevant to the execution of the movement. Additionally, motor adaptation, and interference from opposing perturbations, can often occur independently of explicit attention [46]. Visual properties of objects that affect the execution of the movement may sufficiently cue motor memory formation and switching. Conversely, it is possible that shape properties are simply not sufficient and different properties of tools, such as visual endpoints or tactile/proprioceptive factors, are needed to create and switch to object-specific motor memories. Overall, our findings suggest that movement-goal relevant object-shape cues may not be sufficient to create new context-dependent internal models for motor control when adapting to a single object-specific perturbation, and may act as weak but viable cues when adapting to multiple object-specific perturbations.

Supporting information

S1 Checklist. PLOS ONE clinical studies checklist.

(DOCX)

pone.0300020.s001.docx^{(36.5KB, docx)}

Acknowledgments

We thank the members of the Sensorimotor Control Laboratory for all their help.

Data Availability

All data and analysis scripts can be found in this project’s Open Science Framework repository (https://osf.io/twgc8/).

Funding Statement

This work was supported by the Natural Science and Engineering Research Council of Canada Discovery Grant to DYPH and by the Natural Science and Engineering Research Council of Canada Post Graduate Scholarship - Doctoral to SM. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Oh Y, Schweighofer N. Minimizing precision-weighted sensory prediction errors via memory formation and switching in motor adaptation. J Neurosci. 2019;39: 9237–9250. doi: 10.1523/JNEUROSCI.3250-18.2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Wolpert DM, Kawato M. Multiple paired forward and inverse models for motor control. Neural Networks. 1998;11: 1317–1329. doi: 10.1016/s0893-6080(98)00066-5 [DOI] [PubMed] [Google Scholar]
3.Krakauer JW, Ghilardi MF, Ghez C. Independent learning of internal models for kinematic and dynamic control of reaching. Nat Neurosci. 1999. doi: 10.1038/14826 [DOI] [PubMed] [Google Scholar]
4.Kluzik JA, Diedrichsen J, Shadmehr R, Bastian AJ. Reach adaptation: What determines whether we learn an internal model of the tool or adapt the model of our arm? J Neurophysiol. 2008;100: 1455–1464. doi: 10.1152/jn.90334.2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Berniker M, Kording K. Estimating the sources of motor errors for adaptation and generalization. Nat Neurosci. 2008;11: 1454–1461. doi: 10.1038/nn.2229 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Cothros N, Wong JD, Gribble PL. Are there distinct neural representations of object and limb dynamics? Exp Brain Res. 2006;173: 689–697. doi: 10.1007/s00221-006-0411-0 [DOI] [PubMed] [Google Scholar]
7.Heald JB, Lengyel M, Wolpert DM. Contextual inference underlies the learning of sensorimotor repertoires. Nature. 2021;600: 489–493. doi: 10.1038/s41586-021-04129-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Forano M, Schween R, Taylor JA, Hegele M, Franklin DW. Direct and indirect cues can enable dual adaptation, but through different learning processes. J Neurophysiol. 2021;126: 1490–1506. doi: 10.1152/jn.00166.2021 [DOI] [PubMed] [Google Scholar]
9.Ayala MN, ‘t Hart BM, Henriques DYP. Concurrent adaptation to opposing visuomotor rotations by varying hand and body postures. Exp Brain Res. 2015;233: 3433–3445. doi: 10.1007/s00221-015-4411-9 [DOI] [PubMed] [Google Scholar]
10.Schween R, Taylor JA, Hegele M. Plan-based generalization shapes local implicit adaptation to opposing visuomotor transformations. J Neurophysiol. 2018;120: 2775–2787. doi: 10.1152/jn.00451.2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Howard IS, Wolpert DM, Franklin DW. The effect of contextual cues on the encoding of motor memories. J Neurophysiol. 2013;109: 2632–44. doi: 10.1152/jn.00773.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Mazzoni P, Krakauer JW. An implicit plan overrides an explicit strategy during visuomotor adaptation. J Neurosci. 2006;26: 3642–5. doi: 10.1523/JNEUROSCI.5317-05.2006 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Taylor JA, Ivry RB. The role of strategies in motor learning. Ann N Y Acad Sci. 2012;1251: 1–12. doi: 10.1111/j.1749-6632.2011.06430.x [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Wolpert DM, Flanagan JR. Motor prediction. Current biology: CB. 2001. doi: 10.1016/s0960-9822(01)00432-8 [DOI] [PubMed] [Google Scholar]
15.Ayala MN, Henriques D. Differential contributions of implicit and explicit learning mechanisms to various contextual cues in dual adaptation. 2020. doi: 10.31234/osf.io/an5wp [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Ayala MN, Henriques DYP. Context-dependent concurrent adaptation to static and moving targets. PLoS One. 2018. doi: 10.1371/journal.pone.0192476 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Baldeo R, Henriques DYP. Dual adaptation to opposing visuomotor rotations with similar hand movement trajectories. Exp Brain Res. 2013;227: 231–241. doi: 10.1007/s00221-013-3503-7 [DOI] [PubMed] [Google Scholar]
18.Howard IS, Ingram JN, Franklin DW, Wolpert DM. Gone in 0.6 seconds: the encoding of motor memories depends on recent sensorimotor states. J Neurosci. 2012;32: 12756–68. doi: 10.1523/JNEUROSCI.5909-11.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Woolley DG, Tresilian JR, Carson RG, Riek S. Dual adaptation to two opposing visuomotor rotations when each is associated with different regions of workspace. Exp Brain Res. 2007;179: 155–165. doi: 10.1007/s00221-006-0778-y [DOI] [PubMed] [Google Scholar]
20.Hinder MR, Woolley DG, Tresilian JR, Riek S, Carson RG. The efficacy of colour cues in facilitating adaptation to opposing visuomotor rotations. Exp Brain Res. 2008;191: 143–155. doi: 10.1007/s00221-008-1513-7 [DOI] [PubMed] [Google Scholar]
21.Gandolfo F, Mussa-Ivaldi FA, Bizzi E. Motor learning by field approximation. Proc Natl Acad Sci USA. 1996;93: 3843–3846. doi: 10.1073/pnas.93.9.3843 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Osu R, Hirai S, Yoshioka T, Kawato M. Random presentation enables subjects to adapt to two opposing forces on the hand. Nat Neurosci. 2004;7: 111–112. doi: 10.1038/nn1184 [DOI] [PubMed] [Google Scholar]
23.Hegele M, Heuer H. Implicit and explicit components of dual adaptation to visuomotor rotations. Conscious Cogn. 2010;19: 906–917. doi: 10.1016/j.concog.2010.05.005 [DOI] [PubMed] [Google Scholar]
24.Hinder MR, Tresilian JR, Riek S, Carson RG. The contribution of visual feedback to visuomotor adaptation: How much and when? Brain Res. 2008;1197: 123–134. doi: 10.1016/j.brainres.2007.12.067 [DOI] [PubMed] [Google Scholar]
25.Schween R, Langsdorf L, Taylor JA, Hegele M. How different effectors and action effects modulate the formation of separate motor memories. Sci Rep. 2019;9. doi: 10.1038/s41598-019-53543-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.McGarity-Shipley MR, Heald JB, Ingram JN, Gallivan JP, Wolpert DM, Flanagan JR. Motor memories in manipulation tasks are linked to contact goals between objects. J Neurophysiol. 2020;124: 994–1004. doi: 10.1152/jn.00252.2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Heald JB, Ingram JN, Flanagan JR, Wolpert DM. Multiple motor memories are learned to control different points on a tool. Nat Hum Behav. 2018;2: 300–311. doi: 10.1038/s41562-018-0324-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Brookes J, Warburton M, Alghadier M, Mon-Williams M, Mushtaq F. Studying human behavior with virtual reality: The Unity Experiment Framework. Behav Res Methods. 2020;52: 455–463. doi: 10.3758/s13428-019-01242-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Jacoby LL. A process dissociation framework: Separating automatic from intentional uses of memory. J Mem Lang. 1991;30: 513–541. doi: 10.1016/0749-596X(91)90025-F [DOI] [Google Scholar]
30.Werner S, Van Aken BC, Hulst T, Frens MA, Van Der Geest JN, Strüder HK, et al. Awareness of sensorimotor adaptation to visual rotations of different size. PLoS One. 2015;10. doi: 10.1371/journal.pone.0123321 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Clyde MA, Ghosh J, Littman ML. Bayesian adaptive sampling for variable selection and model averaging. J Comput Graph Stat. 2011;20: 80–101. doi: 10.1198/jcgs.2010.09049 [DOI] [Google Scholar]
32.Kong G, Zhou Z, Wang Q, Kording K, Wei K. Credit assignment between body and object probed by an object transportation task. Sci Rep. 2017;7. doi: 10.1038/s41598-017-13889-w [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Vaswani PA, Shmuelof L, Haith AM, Delnicki RJ, Huang VS, Mazzoni P, et al. Persistent residual errors in motor adaptation tasks: Reversion to baseline and exploratory escape. J Neurosci. 2015;35: 6969–6977. doi: 10.1523/JNEUROSCI.2656-14.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Krakauer JW, Hadjiosif AM, Xu J, Wong AL, Haith AM. Motor Learning. Compr Physiol. 2019;9: 613–663. doi: 10.1002/cphy.c170043 [DOI] [PubMed] [Google Scholar]
35.Kitago T, Ryan SL, Mazzoni P, Krakauer JW, Haith AM. Unlearning versus savings in visuomotor adaptation: comparing effects of washout, passage of time, and removal of errors on motor memory. Front Hum Neurosci. 2013;7: 1–7. doi: 10.3389/fnhum.2013.00307 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Brennan AE, Smith MA. The Decay of Motor Memories Is Independent of Context Change Detection. PLoS Comput Biol. 2015;11: 1–31. doi: 10.1371/journal.pcbi.1004278 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Diedrichsen J, White O, Newman D, Lally N. Use-dependent and error-based learning of motor behaviors. J Neurosci. 2010;30: 5159–5166. doi: 10.1523/JNEUROSCI.5406-09.2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Haith AM, Krakauer JW. Model-Based and Model-Free Mechanisms of Human Motor Learning. Adv Exp Med Biol. 2013;78. doi: 10.1007/978-1-4614-5465-6_1 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Verhulst A, Namikawa Y, Kasahara S. Parallel Adaptation: Switching between Two Virtual Bodies with Different Perspectives Enables Dual Motor Adaptation. IEEE International Symposium on Mixed and Augmented Reality (ISMAR). 2022. doi: 10.1109/ISMAR55827.2022.00031 [DOI]
40.Ghahramani Z, Wolpert DM. Modular decomposition in visuomotor learning. Nature. 1997;386. doi: 10.1038/386392a0 [DOI] [PubMed] [Google Scholar]
41.Seidler RD, Bloomberg JJ, Stelmach GE. Context-dependent arm pointing adaptation. Behav Brain Res. 2001;119. doi: 10.1016/s0166-4328(00)00347-8 [DOI] [PubMed] [Google Scholar]
42.Schween R, Langsdorf L, Taylor JA, Hegele M. Of hands, tools, and exploding dots: How different action states and effects separate visuomotor memories. bioRxiv. 2019. doi: 10.1101/548602 [DOI] [Google Scholar]
43.Anglin JM, Sugiyama T, Liew S-LL. Visuomotor adaptation in head-mounted virtual reality versus conventional training. Sci Rep. 2017. doi: 10.1038/srep45469 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Ballester BR, Oliva LS, Duff A, Verschure P. Accelerating motor adaptation by virtual reality based modulation of error memories. IEEE Int Conf Rehabil Robot. 2015;2015-Septe: 623–629. doi: 10.1109/ICORR.2015.7281270 [DOI] [Google Scholar]
45.Groen J, Werkhoven PJ. Visuomotor adaptation to virtual hand position in interactive virtual environments. Presence Teleoperators Virtual Environ. 1998;7: 429–446. doi: 10.1162/105474698565839 [DOI] [Google Scholar]
46.Kim HE, Morehead JR, Parvin DE, Moazzezi R, Ivry RB. Invariant errors reveal limitations in motor correction rather than constraints on error sensitivity. Commun Biol. 2018;1: 19. doi: 10.1038/s42003-018-0021-y [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0300020.r001

Decision Letter 0

Shenbing Kuang

27 Jul 2023

PONE-D-23-13216Task-relevant object shape properties act as poor but viable cues for the attribution of motor errors to external objectsPLOS ONE

Dear Dr. Modchalingam,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Your manuscript has been evaluated by three experts in the field. While reviewers consider your study interesting and well-conducted, they have also raised a few critical comments and concerns. In particular, there was a major concern regarding the appropriateness of the task design and the dependent variable used in this study.

Please submit your revised manuscript by Sep 10 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Shenbing Kuang, Ph.D.

Academic Editor

PLOS ONE

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating in your Funding Statement:

"This work was supported by the Natural Science and Engineering Research Council of Canada Discovery Grant to DYPH and by the Natural Science and Engineering Research Council of Canada Post Graduate Scholarship - Doctoral to SM."

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now. Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

3. We note that Figure 1 in your submission contain copyrighted images. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figure 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an ""Other"" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: No

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study examined an object transporting task using the right arm in an immersive virtual-reality environment and examined the following two questions:

1) Is the extrinsic cue (object shape in this study) sufficient to allow for associated motor errors to be attributed to the object itself rather than the hand?

2) Does increasing task relevancy (associating two object shapes with different directions of visuomotor perturbation) allow for forming distinct motor memories for different extrinsic cues?

To test the first question, the authors conducted the Single Learning experiment. They asked whether learning of a 30-degree visuomotor rotation with a single object can be transferred to a new object with a different shape. The results showed that the learning effect fully transferred to the new object, suggesting that object shape is not a sufficient contextual cue to promote motor error assignment. On the other hand, the adaptation effect did not transfer sufficiently to the untrained hand regardless of the congruence or incongruence of the object shape. This suggests that intrinsic cues to movements, like used hands, may be helpful for context switching during motor learning.

To test the second question, they conducted the Dual Learning experiment. Two objects with different shapes were associated with two opposite visuomotor rotations to test if they could simultaneously learn the two distinct perturbations. The results showed that the subjects could not fully learn the two rotations. A small learning effect was partially observed, but due to the explicit strategy. Based on these results, the authors concluded that task-relevant visual properties of objects are insufficient for motor memory switching.

In motor neuroscience, the question of what information provides the context cues for motor memory switching is of great importance and has been extensively debated. As the authors mention, intrinsic cues to movements are known to be possible contextual information for the simultaneous formation of different motor memories. On the other hand, concerning the extrinsic cue on which this study focuses, it has been considered that it does not allow memory switching, except in some studies (e.g., Osu et al., 2004).

This study is novel in at least two respects. First, their task design gives higher task relevance to the extrinsic cue (object shape). The goal receptacle for the spherical object was a cylinder, and for the cubic object, a box. It has yet to be sufficiently investigated if increasing the task relevancy of such extrinsic cues may facilitate memory switching. Second, while most previous studies have used more conventional laboratory environments with robotic manipulandum, the authors tested the question under more realistic conditions in an immersive virtual-reality environment. Motor research using HMDs has been increasing in recent years, but the effects of virtual reality environments still need to be thoroughly investigated. Thus, this experimental data has a novelty and will be a valuable accumulation of knowledge for the field.

The introduction is well-written. The discussion of the results also makes sense. On the other hand, there are several issues regarding the description of the methodology and the robustness of the analysis and results. Specifically, I would like to mention the following points.

Major concerns

1) Methods: Please add a more detailed description of the experimental protocols. I think there was no description about how many trials each phase consisted of in each experiment.

2) Results: The medians of the three receptacle positions in the first and last trial-set were used in the statistical analysis. I recommend using the average, as the magnitude of the errors may depend on the receptacle position.

3) Results: I would like to know how robust the results of Experiment-2 are. The change in the hand angle thorough the experiment looks very small. The authors used only the first and last trial-set (three trials for each trial-set) to be analyzed. Please analyze using the first and last three trial-sets, as many studies often take the average of about ten trials.

4) Results: For Fig3D, the significant change in hand angle from baseline was observed only for the CCW rotation. The authors considered this change may be due to explicit (Fig. 3E). However, if the behavioral change during the Dual Learning experiment is due to an explicit process, why is there no change in CW? Also, the explicit process is believed to allow for quick and large behavioral changes. In contrast, why is the observed behavioral change so small if the explicit process works? The authors need to discuss these two points.

5) Result of Fig3D: Related to the fourth point, the authors mention that, although it may result from explicit strategy, the participants could utilize the object shape cues to change their movements depending on the context for the opposing visuomotor rotations. However, considering that only the CCW rotation led to a significant movement change in Fig. 3E, it is possible to think that it exhibited the learning for one context and the failure to learn for the other rather than showing different adaptive changes for the two contexts. This possibility should also be discussed.

Minor concerns

6) Methods: The receptacles were located 45, 90, and 180 degrees from the object's spawn location. The 45 and 180-degree target locations are asymmetric with respect to the 90-degree target, and only the 180-degree location is far from the other two locations. Please explain why these three positions were chosen.

7) Results: Please report whether or not there was any difference in the results between the three receptacle positions.

8) Discussion: In the last sentence (lines 449-450): "our findings suggest that task-relevant object-shape cues may not be sufficient to create new context-dependent internal models for motor control but rather update existing models.” The logic is unclear to me. What evidence from your results supports this argument? Could you please further explain on this point?

Reviewer #2: This paper examines if and how object shape might be used consolidate context-dependent motor plans. In the experimental protocol, an association is made between object shape/target (e.g. cube with square receptacle) and the deviation of the visually perceived movement trajectory in a virtual environment. Performance was evaluated using the “heading angle”, a measure based on the angular deviation from a straight line to the target at 3cm along the 10cm movement trajectory. This heading angle was used to represent initial directional error, prior to corrections based on feedback. Clamped trials where visually perceived position of the end effector along the movement trajectory was fixed (based upon displacement from departure point) were used to evaluate the post effect.

Based upon the results, movement adaptation did not appear to be retained when participants used their opposite hand during the clamped trials. At the same time, movement adaptation appeared present when the object was changed whilst using the same hand. The interpretation is that participants attribute errors to the hand being used rather than the object which is transported. This finding is a little curious as it seems that the hand was never affected during the reach stage (per results line 277 and 309), only during the transportation stage when the hand interacts with the object. This peculiarity does not seem to be addressed in the discussion.

Overall, there are many pleasing aspects of this study. Generally speaking, I found the paper well-written and easy to follow. The inclusion of Bayes factors for gauging the likelihood of the null hypotheses was very helpful. Despite this, I have a non-trivial concern regarding the methodological approach. The main variable used here (heading angle) is based on the angular deviation from a straight line early in the movement trajectory. It appears that the underlying assumption is that movement trajectories for this task should be straight. However, I don’t feel that this holds for the experimental conditions examined. In effect, the use of a cube shaped object with a square shaped receptacle implies a significant task constraint, in that the object must be aligned to the form of the receptacle upon placement. As a consequence, the movement trajectory will likely differ to that which would be used for point to point reaching. A person could exploit different postural configurations through the upper limb (shoulder, elbow, wrist, finger) in order to remain close to a linear pathway, although even this will be constrained by the position of the target in the peripersonal space. It is more likely that they would choose a curvilinear path with comfortable joint configuration (as we tend to do when we take a mug by the handle).

It appears that the authors attempted to make some adjustments for this issue. They report that individual postures according to target location were controlled for by subtracting HA angles at baseline from HA angles in the training (experimental) phase (see line 223). Still, I remain unconvinced by this approach. I feel that the best solution would have been examining the 6-DOF position of the rectangular object along with the upper limb kinematics (in particular the wrist). Of course, I would of course be much less concerned if the deposit zone was more simply a larger 2D drop off zone, implying much less stringent spatial constraints. The specific receptacle presented to subjects is highly specific, adjusted in dimensions (including depth) to accommodate for the visually perceived object. As it stands, the results indeed show statistically significant differences in heading angle between the objects with different shapes (p=0.0153, BF10 = 3.14; line 328). With these elements in mind, my concern is that the study design and dependent variable used in this study are questionable. Given this, I would not recommend accepting this paper.

Reviewer #3: This paper presented a study in which the authors examined the extent to which object shape cues can be used to support context-specific motor adaptation to visuomotor rotations. Overall, the design is sound and the presentation is mostly fine. I have a few concerns about the clarity of the experimental protocol and how the results should be interpreted.

Major

1. The authors argue that the object shape in this experiment is ‘task-relevant’. However, I think this may require more discussion. In my opinion, the object shape is task-relevant in a same way as object color, which is only associative to visuomotor rotation. But in every-day life, object shape could be more ‘task-relevant’ as it could provide information about task dynamics (e.g., weight, weigh distribution, functional goal, etc.). Therefore, while I agree with the findings, I am skeptical about the generalizability of the results to a broader consideration of what ‘object shape’ could provide as contextual cues. There were a few sentences in the last paragraph about this limitation, but I think this point should be emphasized to avoid misunderstanding of the results. I would also recommend the authors reconsider the use of ‘task-relevant’ in the title.

2. What does ‘exclude strategies’ mean exactly and how did the subjects interpret it?

Minor

1. It was not clear what the number of trials was for the experiments (fig. 1). I assume Training 1 had 180 trials, but how about Training 2, Test, and Aligned? Also, how many clamp trials? Please clarify.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2024 Mar 28;19(3):e0300020. doi: 10.1371/journal.pone.0300020.r002

Author response to Decision Letter 0

26 Jan 2024

We thank the editor and reviewers for their comments and recommendations. We address each comment separately below.

In responses where we made changes to the manuscript, we indicated the relevant page and line numbers in the manuscript. For clarity, our responses are indicated in blue text in this document.

Thank you again for your time.

Shanaa Modchalingam,

Corresponding Author

General

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf2. Thank you for stating in your Funding Statement:

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

New funding statement:

This work was supported by the Natural Science and Engineering Research Council of Canada (NSERC) Discovery Grant for DYPH, by the NSERC Post Graduate Scholarship - Doctoral for SM and MNA, by the NSERC-CREATE-IRTG Brain in Action Doctoral Scholarship for SM and MNA, and by the Vision: Science to Application Doctoral Entrance Award for SM.

There was no additional external funding received for this study.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figure 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

Please upload the completed Content Permission Form or other proof of granted permissions as an ""Other"" file with your submission.

All images in Figure 1 were created by the corresponding author. All 3D renders (Figure 1 a and b) were created using Poser 11, and all plots and overlays were created using R and Inkscape respectively.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Partly

Reviewer #2: No

Reviewer #3: Yes

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

3. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

4. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

5. Review Comments to the Author

Reviewer 1

Reviewer #1: This study examined an object transporting task using the right arm in an immersive virtual-reality environment and examined the following two questions:

1) Is the extrinsic cue (object shape in this study) sufficient to allow for associated motor errors to be attributed to the object itself rather than the hand?

2) Does increasing task relevancy (associating two object shapes with different directions of visuomotor perturbation) allow for forming distinct motor memories for different extrinsic cues?

Major concerns

1) Methods: Please add a more detailed description of the experimental protocols. I think there was no description about how many trials each phase consisted of in each experiment.

Thank you, we have expanded the “Experiment protocol” section of the Methods substantially to add detail.

Page 9, Line 178.

We thank the reviewer for the suggestion. We originally used medians to be less sensitive to outliers as each trial set included a small number of trials (3). For the reasons suggested by the reviewer, we now use the mean. The choice between means and medians of the trial-sets does not impact any of our findings.

We tend to use small set of trial for the initial set (one trial for each unique target) because given that learning curve can rise quite quickly, averaging across a large set of trials to repeated targets can greatly inflate the size of early compensation. This would be the case if we averaged the first 9 trials.

However, there is no reason why we can’t use more trials for the Final set, given that learning should be quite stable by the end, so we use the last 3 trial-sets (9 trials) for the Final set. In this instance learning in the Dual-task is indeed unstable and small in the Dual Learning experiment. This change affected our conclusions about learning in the second experiment. We now do not find evidence of change in performance from the Initial to the Final set in the Training phase. Participants do however still reliably counter part of the rotation by the end of training (significantly different from Baseline performance: (t(29) = 3.68, p < 0.001, BF10 = 35.0)), suggesting a significant proportion of the observed learning occurs during the 3-trial Initial set.

We have changed the wording on page 16, line 309+ to account for the change.

We thank the reviewer for the suggestion. Given the size of the effect as mentioned by the reviewer, we no longer think this is strong enough evidence for an explicit process. There is evidence for dual adaptation in the Dual Learning experiment, but it is unclear whether there is a driving effect of either an implicit or explicit process.

We clarified this in the Results section: page 17, line 335.

As with the explicit strategies mentioned in our reply above, we don’t think the results are strong enough (the effect is small and variable) to argue that learning, as measured during the clamp trials, occurred for one rotation or another. We feel confident that participants could differentiate behaviour between the two perturbations.

We clarified this in the Results section: page 17, line 329.

Minor concerns

This was a mistake in the draft, targets were equidistant from each other at 45, 90, and 135 degrees relative to the object spawn location.

Fixed on page 8, line 148.

7) Results: Please report whether or not there was any difference in the results between the three receptacle positions.

We have added these calculations to a supplementary analysis notebook in the open-source repository.

There were no effects of target position on the visuomotor adaptation results during the Training phase of either experiment.

There was an effect of target position on the dual adaptation results during clamped trials (Figure 3d and e) but this effect did no interact with our findings (the effects of perturbation-direction). Hand angles were higher when transporting objects to the 45-degree target in all conditions.

The supplementary analysis can be found in the project repository (https://osf.io/twgc8/): “supplementary_analysis_nb.nb.html”.

We have removed the point about updating of existing models.

Reviewer 2:

We added this to the discussion on Page 21, Line 420+ when comparing our task to real-world object transport tasks. We think the lack of non-visual cues (e.g. tactile and weight cues) that occur in real-world object-transport tasks was lacking in our virtual environment. Since movement dynamics were the same when transporting objects, the transport step may have been interpreted as a pseudo-reach.

Thank you for this great point. Firstly, we ensured that participants did not need to orient the cube object to match the orientation of the receptacle when transporting to the target. This was made clear on the practice trials. We clarified this in the Methods section (page 7, lines 126).

Additionally, we conducted supplementary analyses to ensure participants do not differ in hand-position when initiating the Transport step of the task with different objects. The supplementary analysis can be found in the project repository (https://osf.io/twgc8/): “supplementary_analysis_nb.nb.html”.

It may be that participants used a slightly curvilinear path to move to certain targets, but the results from our Single Learning experiment suggest that participants do deviate the initial direction of the hand movement in the expected direction by 23.7 degrees. In order to achieve the target during these reaches, hand paths are likely slightly curved due to feedback-based corrections, but this is expected behaviour even in point-to-point reaching tasks.

Thank you for this suggestion. The receptacles openings were indeed larger than the objects being transported (2x the size of the objects) for this reason – twice the size of the transported object. Participants were not required to orient the objects to the receptacle and received practice trials to learn such nuances of the task. This was not clear in the original draft of the manuscript, and we have clarified.

Page 7, line 124.

Reviewer 3:

Major

We thank the reviewer for this feedback. We agree that task-relevant may not be the right terminology. The object shapes were relevant to the end-goal of the movements, whereas extrinsic cues like colour in past studies were not (that is, the movement task could be completed even while ignoring the colour cue). We have made changes throughout the Introduction and Discussion section of the manuscript to use the new “goal relevant” terminology.

We have also changed the title to accommodate the new terminology.

2. What does ‘exclude strategies’ mean exactly and how did the subjects interpret it?

We used an altered form of the Process Dissociation Procedure as used by Werner at al. (2015). We have clarified this in the Methods section (Page 11, Line 217+) and added relevant citations to direct the reader to more details.

Werner, S., Van Aken, B. C., Hulst, T., Frens, M. A., Van Der Geest, J. N., Strüder, H. K., & Donchin, O. (2015). Awareness of sensorimotor adaptation to visual rotations of different size. PLoS ONE, 10(4). https://doi.org/10.1371/journal.pone.0123321

Minor

Thank you for the suggestion. Due to this and Review 1’s comments, we have expanded the “Experiment protocol” section of the Methods substantially to add detail.

Page 9, Line 178.

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Attachment

Submitted filename: Response to Reviewers.docx

pone.0300020.s002.docx^{(36.2KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0300020.r003

Decision Letter 1

Shenbing Kuang

21 Feb 2024

Movement-goal relevant object shape properties act as poor but viable cues for the attribution of motor errors to external objects

PONE-D-23-13216R1

Dear Dr. Modchalingam,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Shenbing Kuang, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Partly

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #1: I thank the authors for the revised version of the manuscript. I found responses to previous concerns satisfactory. I have one last concern regarding Fig 3 for the Dual learning experiment.

In Fig 3A, as described in Methods, the authors used normalized HAs to align the direction of the HA change to two opposing visuomotor rotations. A change in the positive direction indicates that the HA changed to compensate for the rotation in each condition. In Fig 3C and D, I am wondering if the authors are using normalized HAs? If so, negative values in CCW indicate that learning performance got worse. This is relevant to the conclusion, so please be sure to provide an explanation for this in the text.

I would also like to see what the learning curve looks like using non-normalized HAs for each of the opposing rotations, and I would suggest that the authors present the figure in supplementary results.

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

**********

PLoS One. doi: 10.1371/journal.pone.0300020.r004

Acceptance letter

Shenbing Kuang

19 Mar 2024

PONE-D-23-13216R1

PLOS ONE

Dear Dr. Modchalingam,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Associate Professor Shenbing Kuang

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Checklist. PLOS ONE clinical studies checklist.

(DOCX)

pone.0300020.s001.docx^{(36.5KB, docx)}

Attachment

Submitted filename: Response to Reviewers.docx

pone.0300020.s002.docx^{(36.2KB, docx)}

Data Availability Statement

All data and analysis scripts can be found in this project’s Open Science Framework repository (https://osf.io/twgc8/).

[pone.0300020.ref001] 1.Oh Y, Schweighofer N. Minimizing precision-weighted sensory prediction errors via memory formation and switching in motor adaptation. J Neurosci. 2019;39: 9237–9250. doi: 10.1523/JNEUROSCI.3250-18.2019 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref002] 2.Wolpert DM, Kawato M. Multiple paired forward and inverse models for motor control. Neural Networks. 1998;11: 1317–1329. doi: 10.1016/s0893-6080(98)00066-5 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref003] 3.Krakauer JW, Ghilardi MF, Ghez C. Independent learning of internal models for kinematic and dynamic control of reaching. Nat Neurosci. 1999. doi: 10.1038/14826 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref004] 4.Kluzik JA, Diedrichsen J, Shadmehr R, Bastian AJ. Reach adaptation: What determines whether we learn an internal model of the tool or adapt the model of our arm? J Neurophysiol. 2008;100: 1455–1464. doi: 10.1152/jn.90334.2008 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref005] 5.Berniker M, Kording K. Estimating the sources of motor errors for adaptation and generalization. Nat Neurosci. 2008;11: 1454–1461. doi: 10.1038/nn.2229 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref006] 6.Cothros N, Wong JD, Gribble PL. Are there distinct neural representations of object and limb dynamics? Exp Brain Res. 2006;173: 689–697. doi: 10.1007/s00221-006-0411-0 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref007] 7.Heald JB, Lengyel M, Wolpert DM. Contextual inference underlies the learning of sensorimotor repertoires. Nature. 2021;600: 489–493. doi: 10.1038/s41586-021-04129-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref008] 8.Forano M, Schween R, Taylor JA, Hegele M, Franklin DW. Direct and indirect cues can enable dual adaptation, but through different learning processes. J Neurophysiol. 2021;126: 1490–1506. doi: 10.1152/jn.00166.2021 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref009] 9.Ayala MN, ‘t Hart BM, Henriques DYP. Concurrent adaptation to opposing visuomotor rotations by varying hand and body postures. Exp Brain Res. 2015;233: 3433–3445. doi: 10.1007/s00221-015-4411-9 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref010] 10.Schween R, Taylor JA, Hegele M. Plan-based generalization shapes local implicit adaptation to opposing visuomotor transformations. J Neurophysiol. 2018;120: 2775–2787. doi: 10.1152/jn.00451.2018 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref011] 11.Howard IS, Wolpert DM, Franklin DW. The effect of contextual cues on the encoding of motor memories. J Neurophysiol. 2013;109: 2632–44. doi: 10.1152/jn.00773.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref012] 12.Mazzoni P, Krakauer JW. An implicit plan overrides an explicit strategy during visuomotor adaptation. J Neurosci. 2006;26: 3642–5. doi: 10.1523/JNEUROSCI.5317-05.2006 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref013] 13.Taylor JA, Ivry RB. The role of strategies in motor learning. Ann N Y Acad Sci. 2012;1251: 1–12. doi: 10.1111/j.1749-6632.2011.06430.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref014] 14.Wolpert DM, Flanagan JR. Motor prediction. Current biology: CB. 2001. doi: 10.1016/s0960-9822(01)00432-8 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref015] 15.Ayala MN, Henriques D. Differential contributions of implicit and explicit learning mechanisms to various contextual cues in dual adaptation. 2020. doi: 10.31234/osf.io/an5wp [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref016] 16.Ayala MN, Henriques DYP. Context-dependent concurrent adaptation to static and moving targets. PLoS One. 2018. doi: 10.1371/journal.pone.0192476 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref017] 17.Baldeo R, Henriques DYP. Dual adaptation to opposing visuomotor rotations with similar hand movement trajectories. Exp Brain Res. 2013;227: 231–241. doi: 10.1007/s00221-013-3503-7 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref018] 18.Howard IS, Ingram JN, Franklin DW, Wolpert DM. Gone in 0.6 seconds: the encoding of motor memories depends on recent sensorimotor states. J Neurosci. 2012;32: 12756–68. doi: 10.1523/JNEUROSCI.5909-11.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref019] 19.Woolley DG, Tresilian JR, Carson RG, Riek S. Dual adaptation to two opposing visuomotor rotations when each is associated with different regions of workspace. Exp Brain Res. 2007;179: 155–165. doi: 10.1007/s00221-006-0778-y [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref020] 20.Hinder MR, Woolley DG, Tresilian JR, Riek S, Carson RG. The efficacy of colour cues in facilitating adaptation to opposing visuomotor rotations. Exp Brain Res. 2008;191: 143–155. doi: 10.1007/s00221-008-1513-7 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref021] 21.Gandolfo F, Mussa-Ivaldi FA, Bizzi E. Motor learning by field approximation. Proc Natl Acad Sci USA. 1996;93: 3843–3846. doi: 10.1073/pnas.93.9.3843 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref022] 22.Osu R, Hirai S, Yoshioka T, Kawato M. Random presentation enables subjects to adapt to two opposing forces on the hand. Nat Neurosci. 2004;7: 111–112. doi: 10.1038/nn1184 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref023] 23.Hegele M, Heuer H. Implicit and explicit components of dual adaptation to visuomotor rotations. Conscious Cogn. 2010;19: 906–917. doi: 10.1016/j.concog.2010.05.005 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref024] 24.Hinder MR, Tresilian JR, Riek S, Carson RG. The contribution of visual feedback to visuomotor adaptation: How much and when? Brain Res. 2008;1197: 123–134. doi: 10.1016/j.brainres.2007.12.067 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref025] 25.Schween R, Langsdorf L, Taylor JA, Hegele M. How different effectors and action effects modulate the formation of separate motor memories. Sci Rep. 2019;9. doi: 10.1038/s41598-019-53543-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref026] 26.McGarity-Shipley MR, Heald JB, Ingram JN, Gallivan JP, Wolpert DM, Flanagan JR. Motor memories in manipulation tasks are linked to contact goals between objects. J Neurophysiol. 2020;124: 994–1004. doi: 10.1152/jn.00252.2020 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref027] 27.Heald JB, Ingram JN, Flanagan JR, Wolpert DM. Multiple motor memories are learned to control different points on a tool. Nat Hum Behav. 2018;2: 300–311. doi: 10.1038/s41562-018-0324-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref028] 28.Brookes J, Warburton M, Alghadier M, Mon-Williams M, Mushtaq F. Studying human behavior with virtual reality: The Unity Experiment Framework. Behav Res Methods. 2020;52: 455–463. doi: 10.3758/s13428-019-01242-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref029] 29.Jacoby LL. A process dissociation framework: Separating automatic from intentional uses of memory. J Mem Lang. 1991;30: 513–541. doi: 10.1016/0749-596X(91)90025-F [DOI] [Google Scholar]

[pone.0300020.ref030] 30.Werner S, Van Aken BC, Hulst T, Frens MA, Van Der Geest JN, Strüder HK, et al. Awareness of sensorimotor adaptation to visual rotations of different size. PLoS One. 2015;10. doi: 10.1371/journal.pone.0123321 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref031] 31.Clyde MA, Ghosh J, Littman ML. Bayesian adaptive sampling for variable selection and model averaging. J Comput Graph Stat. 2011;20: 80–101. doi: 10.1198/jcgs.2010.09049 [DOI] [Google Scholar]

[pone.0300020.ref032] 32.Kong G, Zhou Z, Wang Q, Kording K, Wei K. Credit assignment between body and object probed by an object transportation task. Sci Rep. 2017;7. doi: 10.1038/s41598-017-13889-w [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref033] 33.Vaswani PA, Shmuelof L, Haith AM, Delnicki RJ, Huang VS, Mazzoni P, et al. Persistent residual errors in motor adaptation tasks: Reversion to baseline and exploratory escape. J Neurosci. 2015;35: 6969–6977. doi: 10.1523/JNEUROSCI.2656-14.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref034] 34.Krakauer JW, Hadjiosif AM, Xu J, Wong AL, Haith AM. Motor Learning. Compr Physiol. 2019;9: 613–663. doi: 10.1002/cphy.c170043 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref035] 35.Kitago T, Ryan SL, Mazzoni P, Krakauer JW, Haith AM. Unlearning versus savings in visuomotor adaptation: comparing effects of washout, passage of time, and removal of errors on motor memory. Front Hum Neurosci. 2013;7: 1–7. doi: 10.3389/fnhum.2013.00307 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref036] 36.Brennan AE, Smith MA. The Decay of Motor Memories Is Independent of Context Change Detection. PLoS Comput Biol. 2015;11: 1–31. doi: 10.1371/journal.pcbi.1004278 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref037] 37.Diedrichsen J, White O, Newman D, Lally N. Use-dependent and error-based learning of motor behaviors. J Neurosci. 2010;30: 5159–5166. doi: 10.1523/JNEUROSCI.5406-09.2010 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref038] 38.Haith AM, Krakauer JW. Model-Based and Model-Free Mechanisms of Human Motor Learning. Adv Exp Med Biol. 2013;78. doi: 10.1007/978-1-4614-5465-6_1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref039] 39.Verhulst A, Namikawa Y, Kasahara S. Parallel Adaptation: Switching between Two Virtual Bodies with Different Perspectives Enables Dual Motor Adaptation. IEEE International Symposium on Mixed and Augmented Reality (ISMAR). 2022. doi: 10.1109/ISMAR55827.2022.00031 [DOI]

[pone.0300020.ref040] 40.Ghahramani Z, Wolpert DM. Modular decomposition in visuomotor learning. Nature. 1997;386. doi: 10.1038/386392a0 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref041] 41.Seidler RD, Bloomberg JJ, Stelmach GE. Context-dependent arm pointing adaptation. Behav Brain Res. 2001;119. doi: 10.1016/s0166-4328(00)00347-8 [DOI] [PubMed] [Google Scholar]

[pone.0300020.ref042] 42.Schween R, Langsdorf L, Taylor JA, Hegele M. Of hands, tools, and exploding dots: How different action states and effects separate visuomotor memories. bioRxiv. 2019. doi: 10.1101/548602 [DOI] [Google Scholar]

[pone.0300020.ref043] 43.Anglin JM, Sugiyama T, Liew S-LL. Visuomotor adaptation in head-mounted virtual reality versus conventional training. Sci Rep. 2017. doi: 10.1038/srep45469 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300020.ref044] 44.Ballester BR, Oliva LS, Duff A, Verschure P. Accelerating motor adaptation by virtual reality based modulation of error memories. IEEE Int Conf Rehabil Robot. 2015;2015-Septe: 623–629. doi: 10.1109/ICORR.2015.7281270 [DOI] [Google Scholar]

[pone.0300020.ref045] 45.Groen J, Werkhoven PJ. Visuomotor adaptation to virtual hand position in interactive virtual environments. Presence Teleoperators Virtual Environ. 1998;7: 429–446. doi: 10.1162/105474698565839 [DOI] [Google Scholar]

[pone.0300020.ref046] 46.Kim HE, Morehead JR, Parvin DE, Moazzezi R, Ivry RB. Invariant errors reveal limitations in motor correction rather than constraints on error sensitivity. Commun Biol. 2018;1: 19. doi: 10.1038/s42003-018-0021-y [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Movement-goal relevant object shape properties act as poor but viable cues for the attribution of motor errors to external objects

Shanaathanan Modchalingam

Maria N Ayala

Denise Y P Henriques

Roles

Abstract

Introduction

Methods

Participants

Apparatus

Object-transport task

Fig 1. Experiment setup and procedure.

Experiment protocol

Data analysis

Results

Experiment 1: Single learning

Visuomotor adaptation

Fig 2. Hand angles: Single Learning experiment.

Error attribution

Experiment 2: Dual learning

Visuomotor adaptation

Fig 3. Hand angles: Dual Learning experiment.

Reaction and execution times

Fig 4. Reach and Transport step movement times and reaction times.

Discussion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Shenbing Kuang

Roles

Author response to Decision Letter 0

Decision Letter 1

Shenbing Kuang

Roles

Acceptance letter

Shenbing Kuang

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases