Conscious awareness of motor fluidity improves performance and decreases cognitive effort in sequence learning

https://doi.org/10.1016/j.concog.2021.103220Get rights and content

Highlights

  • Motor learning is improved when participants judged their movements fluidity.

  • This effect is called “conscious awareness” effect.

  • This effect could be mediated by cognitive effort.

  • Results revealed no mediation of effort for such movement consciousness.

Abstract

Motor skill learning is improved when participants are instructed to judge after each trial whether their performed movements have reached maximal fluidity. Consequently, the conscious awareness of this maximal fluidity can be classified as a genuine learning factor for motor sequences. However, it is unknown whether this effect of conscious awareness on motor learning could be mediated by the increased cognitive effort that may accompany such judgment making. The main aim of this study was to test this hypothesis in comparing two groups with, and without, the conscious awareness of the maximal fluidity. To assess the possible involvement of cognitive effort, we have recorded the pupillary dilation to the task, which is well-known to increase in proportion to cognitive effort. Results confirmed that conscious awareness indeed improved motor sequence learning of the trained sequence specifically. Pupil dilation was smaller during trained than during novel sequence performance, indicating that sequence learning decreased the cognitive cost of sequence execution. However, we found that in the group that had to judge on their maximal fluidity, pupil dilation during sequence production was smaller than in the control group, indicating that the motor improvement induced by the fluidity judgment does not involve additional cognitive effort. We discuss these results in the context of motor learning and cognitive effort theories.

Introduction

Motor skill learning is critical for survival. For example, efficient acquisition of a new evasive manoeuvre during a flight procedure can be decisive for a military pilot. For a building-worker, effective learning of a dockyard-tool can be critical for safety. Finally, a skilled-surgeon can make the difference between death and life in required situations. Consequently, the cognitive mechanisms at work during learning of many real-world motor skills represent a scientific challenge in our societies (see Harris et al., 2018, Schmidt and Lee, 2011 for reviews). These learning processes are accompanied with subjective, meta-learning percept of performance improvement, and more specifically, about the fluency of the motor skill. Such conscious awareness of the fluency of goal-based movements in turn reinforces future enactments (Boutin, Blandin, Massen, Heuer, & Badets, 2014). However, little is known about the specific mechanisms at work in this feedback effect of conscious feeling of fluency on learning.

The first empirical evidence on the effect of conscious awareness of action in motor learning comes from a study where participants were required to learn a motor sequence task with the instruction to detect, after each trial, the most maximal fluidity of the motor sequence (Boutin et al., 2014). Specifically, the task in this study was composed of a 12-element motor sequence on a keyboard, cued by on-screen visual stimuli, and participants were requested to press with the right-hand fingers the appropriate response keys as rapidly and accurately as possible. When such regular stimulation is repeated over time (after several trials), it is well known that participants encode the motor sequence by associative learning between keys presses, leading to optimized motor performance (Boutin et al., 2010, Jiménez et al., 2011, Miller, 1956, Solopchuk et al., 2016, see Sakai, Hikosaka, & Nakamura, 2004, for a review). The two main groups of this study were called the “awareness group”, and the “No-judgment control group”. For the awareness group, participants were required to judge after each trial whether their performed motor sequence had attained maximal fluidity; practically, if participants replied “yes” to this question, it meant that she/he believed that she/he was not able to improve motor fluidity further. For participants in the no-judgment group, no judgment was required after each trial, and the instruction emphasized only improving and learning the motor sequence. Results revealed a clear motor improvement for both groups at the end of the learning phase, and most importantly, that participants in the awareness group outperformed participants of the no-judgment group. For the authors, this is a first piece of evidence that the conscious awareness of the motor fluidity can be categorized as a factor that can influence motor sequence learning positively.

Accordingly to this finding, and as mentioned by Abrahamse, Ruitenberg, de Kleine, and Verwey (2013), motor sequence learning can be deeply improved by explicit sequence descriptions during the instruction. In the same vein, Jaynes, Schieber, and Mink (2016) have revealed that participants who became aware of their motor patterns during the learning phase performed the sequence with greater consistency with regard to their movement kinematics. Finally, Toner and Moran (2014) emphasized that expert sport performers and musicians can improve their performances throughout conscious cognitive activities in deploying for instance attention toward bodily movements.

However, altogether these results say little about the mechanisms at play in this motor learning improvement induced by conscious awareness. One hypothesis is that cognitive effort would be increased in the condition in which participants focus on reaching maximal fluidity of the movement, in turn leading to improved performance. Indeed, it is well known that some experimental factors which emphasize more cognitive effort can in turn improve motor learning (Schmidt & Lee, 2011). For example, it is accepted that practicing several motor tasks in a random practice format during the learning phase (e.g., the randomization of three motor tasks to-be learned), is more beneficial for a long-lasting learning than practicing the same tasks in a blocked format (e.g., with no or less randomization of these tasks). This beneficial effect of randomization for motor learning is known under the name “contextual interference effect” (Shea & Morgan, 1979). Accordingly, Li and Wright (2000) have found greater attentional demand for the motor planning of the task to-be learned during the learning phase of a random practice. To interpret these data, Wright and colleagues (2016, for a review) have recently suggested that encountering a contextual interference “in practice is “more effortful or demanding” due to the need to engage a broader set of cognitive operations needed to execute an action.” (p. 4). For these authors, such additional cognitive efforts can in turn improve the learning of the motor sequence tasks. Consequently, in following this theoretical argument of the cognitive effort effect on motor learning, we could speculate that an additional cognitive effort is at work when participants are instructed to attain the maximal fluidity of a motor sequence which in turn improves motor encoding.

While this cognitive-effort hypothesis seems well suited to interpret the effect of conscious awareness of action on motor learning, an alternative hypothesis would be that the improvement in sequence learning induced by fluidity judgment is mediated by other mechanisms than increased effort. Indeed, in order to assess the fluidity of their performance after each trial, participants have to mentally rehearse their action. Like an additional feedback on the movement, this additional rehearsal of performance in the fluidity judgment group could be responsible for the specific enhancement in their motor learning process (Aiken, Fairbrother, & Post, 2012). Moreover, Wulf and Lewthwaite (2016) showed that focusing attention on the enhancement of future performance outcomes is an important factor for motor learning. For these authors, “enhanced outcome is achieved with less effort, as indicated by reduced muscular activity, heart rate, oxygen consumption, and so forth. Thus, movement efficiency is enhanced as well.” (p. 1401). For example, it is well known that drawing attention on the environmental consequence of a performed action is more beneficial for motor learning than drawing attention on the different movements themselves to produce this action (see Lewthwaite & Wulf, 2017 for a review). Thus, in accordance with this “outcomes” hypothesis, it could be argued that the beneficial effect in drawing attention - at the end of a trial - on the maximal fluidity of an action can have the capacity to improve subsequently the learning of the motor sequence. Both the rehearsal and the action outcome hypotheses lead to predictions of lower effort associated with the fluidity judgment instructions.

The main aim of the present article is to disentangle these alternative hypotheses about the implication of cognitive effort in the effect of conscious awareness of action on motor learning. To that end, we have used the pupillary dilation paradigm, widely accepted as a physiological marker of cognitive effort (see van der Wel and van Steenbergen, 2018, Zénon, 2019, for reviews). In the domain of task learning, pupil size has been shown to decrease across trials (Foroughi, Sibley & Coyne, 2017), an effect interpreted as the signature of decreasing cognitive load induced by learning. In the present study, we recorded pupillary responses in both groups that learned a motor sequence task. Participants of the first “action awareness” group were instructed to execute the motor task (motor task A) as quickly and accurately as possible and to judge after each trial whether they reached their maximal motor sequence fluidity (Boutin et al., 2014). Participants of the second no-judgment “control” group were only instructed to perform the motor task as quickly and accurately as possible. For both groups, some test trials (motor task B) were inserted across practice blocks. These test trials served as a within-group control condition in order to assess the cognitive load when participants were facing an unpracticed, new motor task. Indeed, as suggested by Foroughi and colleagues (2017), we expected pupil dilation increases for both groups in the test trials.

More importantly, if the conscious awareness effect in motor learning implies increased cognitive effort (Wright et al., 2016), then the expected higher learning rate on motor task A for the awareness group should be associated with increased pupil dilation, in comparison to the control group. In contrast, with respect to both the mental rehearsal and effortless action-outcome hypotheses (Wulf & Lewthwaite, 2016), we expect inverse results during the practice of motor task A. Finally, from the perspective of motor performance, we also expected to replicate the pattern of behavioral results reported by Boutin et al. (2014), that is, a motor learning improvement for the awareness group in comparison to the control group. No difference between groups is expected for the new motor task B, in terms of task performance and pupil dilation.

Section snippets

Participants

30 students from the University of Bordeaux (age = 24.1 ± 3.8, 13 M) participated in this study. All were naive to the experimental procedure and task. All participants were right-handed and possessed normal or corrected-to-normal vision. Their informed consents to participate was obtained before the experiment, and the study was approved by the Ethical Review Board of the university. The inclusion criteria were the absence of any medical history related to the cognitive domain (e.g. dyslexia,

Results

The rank-sum test on the error rates did not highlight significant group differences in the occurrence of errors between the two groups (Z = 1.6, p = 0.11, d = 0.15). Specifically, the sum of errors of the control group had a mean of 5.8 and a median of 5 errors across all trials, and the awareness group’s sum of errors had a mean of 5.3 and a median of 4 errors across trials. This finding implies that any effect found between the RTs of groups could not be explained by diverging error rates.

Discussion

The main goal of this study was to tease apart two alternative hypotheses regarding the mechanisms at the origin of the effect of conscious awareness of action on motor learning. The cognitive effort hypothesis predicted that improved performance in the “awareness” condition would be associated with dilated pupil size (Wright et al., 2016). In contrast, the mental rehearsal and the action outcome hypotheses, assuming that participants would rehearse more their past actions and/or focus more

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (33)

  • A. Alamia et al.

    Pupil-Linked Arousal Responds to Unconscious Surprisal

    Journal of Neuroscience

    (2019 Jul 3)
  • A. Badets et al.

    The ideomotor recycling theory for tool use, language and foresight

    Experimental Brain Research

    (2017)
  • B. Eitam et al.

    Motivation from control

    Experimental Brain Research

    (2013)
  • C.K. Foroughi et al.

    Pupil size as a measure of within- task learning

    Psychophysiology

    (2017)
  • A.G. Greenwald

    Sensory feedback mechanisms in performance control: With special reference to the ideo-motor mechanism

    Psycho- logical Review

    (1970)
  • Harris, D. J., Vine, S. J., Wilson, M. R., McGrath, J. S., LeBel, M. E., & Buckingham, G. (2018). Action observation...
  • Cited by (1)

    1

    These authors have contributed equally for this work.

    View full text