Luc Proteau

A Sensorimotor Basis for Motor Learning: Evidence Indicating Specificity of Practice

Our previous work (Proteau, Marteniuk, Girouard, & Dugas, 1987) was concerned with determining whether with relatively extensive practice on a movement aiming task, as the skill theoretically starts becoming open-loop, there would be evidence for a decreasing emphasis on visual feedback for motor control. We eliminated vision of the moving limb after moderate and extensive practice and found that the movement became more dependent on this feedback with greater amounts of practice. In the present study, we wished to test the hypothesis, developed from our previous work, that at the base of movement learning is a sensorimotor representation that consists of integrated information from central processes and sensory feedback derived from previous experiences on the movement task. A strong test of this hypothesis would be the prediction that for an aiming task, the addition of vision, after moderate and relatively extensive practice without vision, would lead to an increasingly large movement decrement, relative to appropriate controls. We found good support for this prediction. From these and previous results, and the idea of the sensorimotor representation underlying learning, we develop the idea that learning is specific to the conditions that prevail during skill acquisition. This has implications for the ideas of generalized motor program and schema theory.

Chapter 4 On The Specificity of Learning and the Role of Visual Information for Movement Control

Publisher Summary Hybrid control models propose that motor control is achieved by an interplay between central planning and processing of afferent information. This chapter reviews the major experimental evidence that has been used to suggest that learning a motor skill can be equated with either a reduction of the need for sensory information or a decrease in the importance of visual afference in the favor of kinesthetic feedback. It discusses the role played by visual information for movement control as a particular individuals expertise at the task increases. Several studies are presented in which the availability of visual information for the control of various types of movement is manipulated. The “ball catching” task is discussed, which describes that a normally available visual information is a major source of afference for movement control. A transfer paradigm is used to assess the effects of different sources of afference on movement learning and control.

On the type of information used to control and learn an aiming movement after moderate and extensive training

Abstract This experiment was conducted to see if, in an aiming task (MT = 550 msec), where subjects received moderate (200 trials) or extensive practice (2000 trials), performance would benefit from vision of the performing limb and the target to be reached when compared to a situation where only the target to be reached was visually available. As a second goal, a transfer paradigm was used to see to what extent learning was specific to the conditions under which practice occurred. The results indicated that performance was enhanced when subjects were permitted vision of the performing limb. Furthermore, the subjects who benefited from vision of the performing limb in the training period were not able, even after extensive training, to maintain performance in the transfer task (i.e., without vision of the performing limb). These results are consistent with the view that vision of the responding limb is particularly important in learning a perceptual-motor task. Moreover, practice does not decrease the importance of this information for guiding the movement as some of the past literature suggests might happen. The results are seen as supporting the notion that movement learning may involve the development of a complex sensorimotor reference mechanism that acts to control and, when necessary, modify the ongoing movement. Further, this would imply that movement learning is relatively specific to the conditions under which practice occurs.

Vision of the stylus in a manual aiming task: The effects of practice

Nearly a decade ago, Carlton (1981) showed that in a manual aiming task, vision of the ongoing stylus led to a better accuracy than when that source of information was not available. However, Elliott (1988) recently failed to replicate that finding and, rather, showed that being able to see an ongoing stylus did not result in a significant improvement of accuracy over a conventional no-vision condition. In the experiment to be reported here, we, on the one hand, replicated Elliotts results for the low level of practice condition (15 acquisition trials). On the other hand, the results obtained after moderate practice (150 acquisition trials) supported Carltons earlier conclusion. These results therefore indicate that after sufficient practice one is able to use effectively the information provided by the ongoing stylus to help control his/her movement. This conclusion was further substantiated by the results obtained in a transfer task. Collectively, the results gave added support to a specificity of learning hypothesis (Proteau, Marteniuk, Girouard, & Dugas, 1987; Proteau, Marteniuk, & Lévesque, 1990).

Practice does not diminish the role of visual information in on-line control of a precision walking task: Support for the specificity of practice hypothesis

It has been proposed that motor learning is specific to the sources of afferent information available during practice (Proteau, 1992). That hypothesis has obtained support from studies of aiming but not from studies of gross motor skills. The many procedural differences between the two sets of studies might have caused the conflicting results. In the present investigation of a precision walking task, the validity of the specificity of practice hypothesis was tested. Thirty-two participants were asked to walk for 20 m on a 2.5-cm-wide line. Subjects practiced the task for either 20 or 100 trials under normal visual conditions or while blindfolded. Following acquisition, all subjects performed the task for 20 additional trials while blindfolded and without knowledge of results. Practice improved the performance of blindfolded participants. Withdrawing vision in transfer resulted in a large and significant increase in error, however, a finding that supports the specificity of practice hypothesis.

Specificity of practice results from differences in movement planning strategies

Withdrawing visual feedback after practice of a manual aiming task results in a severe decrease in aiming accuracy. This decrease in accuracy is such that participants are often less accurate than controls who are beginning practice of the task without visual feedback. These results have been interpreted as evidence that motor learning is specific to the sources of afferent information optimizing performance, because it could be processed at the exclusion of other sources of afferent information. The goal of the present study was to test this hypothesis. To reach our goal we evaluated whether online visual feedback prevented kinesthetic information to be used for: (1) eliminating movement anisotropy resulting from difference in limb inertia when aiming in different directions and (2) creating an internal model of limb mechanics. Participants practiced a manual aiming task with or without visual feedback and with knowledge of results. After this acquisition phase, participants performed two transfer tests. The first transfer test was performed without visual feedback and/or knowledge of results. The second transfer test was similar to the first one but participants initiated their movements from a different starting base. The results showed strong specificity effects in that withdrawing visual feedback resulted in large pointing bias and variability. However, the results of the two transfer tests showed that the processing of visual feedback did not prevent the processing of kinesthetic information used to eliminate movement anisotropy or to create an internal model of limb mechanics. Rather, specificity of practice effects resulted from participants using the same motor plan in transfer as they did in acquisition even though they had no longer access to visual feedback to modulate their movement online. These results indicate that during acquisition participants adopted different movement planning strategies depending on the source of afferent information available.

Specificity of learning in a video-aiming task : Modifying the salience of dynamic visual cues

The authors investigated whether the salience of dynamic visual information in a video-aiming task mediates the specificity of practice. Thirty participants practiced video-aiming movements in a full-vision, a weak-vision, or a target-only condition before being transferred to the target-only condition without knowledge of results. The full- and weak-vision conditions resulted in less endpoint bias and variability in acquisition than did the target-only condition. Going from acquisition to transfer resulted in a large increase in endpoint variability for the full-vision group but not for the weak-vision or target-only groups. Kinematic analysis revealed that weak dynamic visual cues do not mask the processing of other sources of afferent information; unlike strong visual cues, weak visual cues help individuals calibrate less salient sources of afferent information, such as proprioception.

What causes specificity of practice in a manual aiming movement : Vision dominance or transformation errors?

Abstract The withdrawal of vision of the arm during a manual aiming task has been found to result in a large increase in aiming error, regardless of the amount of practice in normal vision before its withdrawal. In the present study, the authors investigated whether the increase in error reflects the domination of visual afferent information over the movement representation developed during practice to the detriment of other sources of afferent information or whether it reflects only transformation errors of the location of the target from an allocentric to an egocentric frame of reference. Participants (N = 40) performed aiming movements with their dominant or nondominant arm in a full-vision or target-only condition. The results of the present experiment supported both of those hypotheses. The data indicated that practice does not eliminate the need for visual information for optimizing movement accuracy and that learning is specific to the source or sources of afferent information more likely to ensure optimal accuracy during practice. In addition, the results indicated that movement planning in an allocentric frame of reference might require simultaneous vision of the arm and the target. Finally, practice in a target-only condition, with knowledge of results, was found to improve recoding of the target in an egocentric frame of reference.

Static visual information and the learning and control of a manual aiming movement

Abstract Our goal was to determine whether the use of static visual cues for motor control such as the position of the hand before (Prablanc, Echallier, Komilis and Jeannerod 1979a; Prablanc, Echallier, Jeannerod and Komilis 1979b) or after (Beaubaton and Hay 1986; Hay and Beaubaton 1986) movement execution changes as a function of practice with knowledge of results (KR). The results indicated that being able to see the surrounding environment prior to movement initiation is apparently of no use for either movement planning and/or control in a situation in which the individual has no simultaneous vision of both the hand and the target to be reached. Alternatively, the results of both experiments of the present study indicate that being able to see where ones hand lands on or around an aimed target is of importance for movement planning and/or control. The results suggest that this information provides an enriched form of knowledge of results. It is proposed that this enriched KR might show the negative effect of guidance sometimes associated with verbal KR (Salmoni, Schmidt and Walter 1984).

Straight ahead acts as a reference for visuomotor adaptation

One can adapt movement planning to compensate for a mismatch between vision and action. Previous research with prismatic lenses has shown this adaptation to be accompanied with a shift in the evaluation of one’s body midline, suggesting an important role of this reference for successful adaptation. This interpretation leads to the prediction that rotation adaptation could be more difficult to learn for some directions than others. Specifically, we hypothesized that targets seen to the right of the body midline but for which a rotation imposes a movement to its left would generate a conflict leading to a bias in movement planning. As expected, we observed different movement planning biases across movement directions. The same pattern of biases was observed in a second experiment in which the starting position was translated 15 cm to the right of the participants’ midline. This indicates that the “straight ahead” direction, not one’s midline, serves as an important reference for movement planning during rotation adaptation.