Perceptual Learning Immediately Yields New Stable Motor Coordination

The reason 0° is easy while other relative phases are hard is that the requisite information is detected most readily at 0°. This test predicts that if a participant were to improve their ability to detect the requisite information then their movement stability would improve.

There were 12 participants. Half the participants were assigned to the Experimental Group and half to the Control. There were two types of experimental task: two alternative forced choice (2AFC) judgements, and coordinated rhythmic movement. There were then two types of session:

1. Assessment sessions consisted of both judgement trials (at both 90° and 180°) and movement trials (moving at 0°, 90°, and 180°). There was no feedback.

2. Training consisted only of judgement trials at 90° with feedback.

The Experimental group did three assessment sessions and up to 14 training sessions. The Control group did three sessions of movement trials, with no training or feedback.

Judgements (Perceptual Ability)

Participants had to identify a target phase of 90° in a pair of displays. Participants then trained with progressively harder perceptual discriminations around 90° with feedback.

Movement Stability (Baseline, Post Training and Retention)

Participants used a joystick to coordinate the movement of two dots on a screen at three relative phases. Participants were instructed to move so as to produce a mean relative phase of 0°, 180°, or 90°, three trials of each.

A repeated measures ANOVA was used to check the significance between the two groups. Significant differences were found in the experimental group which shows that perceptual training was successful. The movement task found that only participants in the Experimental group improved their movement stability across sessions and this improvement was restricted to 90°.

To conclude participants in this experiment improved their movement stability at 90° following training to improve their perceptual ability at 90°. Also improved perceptual discrimination of 90° led to improved performance in the movement task at 90° with no training. The improvement persisted until Retention without further exposure to either task.

The learning of 90° continuous relative phase with and without Lissajous feedback: External and internally generated bimanual coordination

The primary aim of this experiment was to find out if reducing the amount of lissajous feedback would allow participants to develop an internal representation of the coordinated pattern, enabling them to effectively perform bimanual coordination tasks without any feedback.

Participants were randomly assigned to one of nine groups that differed in terms of the percent of time they received Lissajous feedback (100%, 50% or 0%). For the 50% feedback group Lissajous feedback was presented in a fading schedule (25–25–20–20– 15–15–10–10–5–5 s) for each consecutive trial. The feedback was provided at the beginning of each trial and withdrawn according to the schedule above.

Apriori comparisons indicated that after 5 minutes of practice, participants who received 50% feedback had considerably higher RMSE values on the no-Lissajous test compared with participants who received 100% feedback, t(12)=4.63, pb.05, on the Lissajous test. After 20 minutes of practice, participants who received 50% feedback performed just as well as the group who had received 100% feedback in a test with no Lissajous feedback.

More practice while receiving 100% Lissajous feedback did not help participants to develop an internal representation of the task or improve their ability to perform when the Lissajous feedback was present. On practice trials while receiving 50% Lissajous feedback participants seek out and process other sources of information necessary to perform the task. This could result in participants acquiring the capability to detect and correct errors. The result is improved performance on tests without extrinsic feedback.

Perception & Action Lab: Perceptual Learning Immediately Yields New Stable Motor Coordination

Perception & Action Lab: Perceptual Learning Immediately Yields New Stable Motor Coordination

The purpose of this study is to determine whether perceptual training can significantly improve an unstable movement, which in this case is 90º.

12 participants (22-54 years old), half of which were assigned to an experimental group and the other to a controlled group. The experimental group participated in training sessions as well as assessment sessions, receiving feedback at the training sessions. The controlled group did three sessions of movement trials, with no feedback or training given. Experimental group took part in up to 14 training sessions (depending on how quickly plateau state of improvement occurred) and three assessment sessions. Each training set became increasingly harder; participants had a maximum of four repetitions to successfully complete each set otherwise they’d progress automatically to promote forced learning. Training did not involve any extensive practise regarding the movement task itself so learning is solely perception based.

Judgement data was analysed first to see whether the experimental group learnt as a result of the extensive training; which was followed by analysis of the movement data. A repeated measures ANOVA test was conducted to find out whether there were significant results regarding the two groups. Significant results were found in the experimental group, showing a significant improvement during training as well as improvements in movement assessment, which indicates a high correlation between the two. The repeated measures ANOVA test showed no significant results in the movement task (without perceptual training) concerning the controlled group.

This study concludes that participants improved their movement stability at 90 º due to improvements in perceptual training at 90 º.

Learning a co-ordinated Rhythmic movement with task- appropriate co-ordination feedback

Learning a co-ordinated Rhythmic movement with task- appropriate co-ordination feedback

One of the main assumptions made before this experiment was conducted was that when learning to produce a novel movement, people are unable to do it without training. This is explained to be because people are poor at discriminating the perceptual information required to co-ordinate and control the movement, which means people require additional (augmented) feedback to learn the novel task.

· Kelso (1995) stated that people can produce two stable co-ordination patterns without training 0° and 180°. In order to perform other i.e. 90° that pattern has to be learned. #

· Behaviour is said to be organised to the order parameter relative

· 0° and 180° are attractors in this space and in order to learn a new novel co-ordination movement a new attractor has to be created

Perception – action perspective

· This proposes that the phenomena of learning a co – ordinated rhythmic movement emerges from a task dynamic that includes perceptual information as a crucial element (Bingham 2001)

· Learning to produce a novel and unknown movement requires augmented feedback so that the learner knows what they are doing right or wrong

· Visual metronomes and Lissajous figures are used as a way of obtaining feedback; however these methods transform the task so that it no longer is a co-ordination task.

· Both methods show that relative direction is not defined in the feedback display and this alters the process of acquiring a novel co-ordination

· The objective of the task was to describe and test a new method for providing online augmented co-ordination feedback using a neutral colour cue

· Ten participants were split into 2 groups of 5. Group 1 received co-ordination feedback during training, group 2 received no feedback but an equal exposure to the task

· The study concluded that a new method for providing the augmented feedback necessary for learning novel co-ordinated rhythmic movements was developed

· An advantage of this new method is that it doesn’t alter or remove the visual information about the co-ordination task which the other methods do.

Learning a coordinated rhythmic movement with task appropriate coordination feedback

The aim of the study was to find out whether feedback aided individuals ability to correctly coordinate a specific rhythmic movement task. This task involved participants doing a coordinated rhythmic movement at 90˚ mean relative phase. Previous studies have suggested that individuals are able to produce 0˚ and 180˚ movement patterns without training; however 90˚ is a lot more difficult without the aid of feedback or training. Therefore the study measured whether the participants were to successfully learn a novel task both with feedback and without feedback.
10 participants were split into two groups of five individuals. Both groups had the same amount of time to complete the training under two conditions: group 1 (“Feedback”), received feedback during training, whereas group 2 (“Control”) received no feedback during their training. Both groups initially received a baseline assessment session where they viewed a demonstration of the relative movement of 90˚ on a computer screen which they then had 20 seconds to repeat the viewed movement, five times. A computer controlled dot and a metronome was used as a reference point for the individual to produce their movement.
Five later sessions involved the participants receiving training either using feedback for the “Feedback” group or no feedback for the “Control” group. Feedback included the same procedure as the bassline session but participants now had a green coloured dot which indicated that the individual was creating a coordinated movement of 90˚± mean relative phase. As training sessions progressed the error bandwidth decreased from 40˚ then 30˚, 20˚, 15˚ and finally 10˚±, therefore whilst the first training session triggered a green dot while the individual was moving between 50˚ and 130˚ the fifth training session triggered a green dot when the individual was moving between 80˚ and 100˚. The decrease in bandwidth would aim to promote an improvement in coordinated movement at 90˚ after each session. The “Control” group did the same amount of trials but received no such feedback.
The results found that participants who received feedback were significantly better at maintaining a 90˚ movement for longer periods of time than the control group, whereas the control group did not show any improvement whilst doing the same post-training session.
The results therefore show that feedback is vital to learning a new task as the control group were unable to complete the movement task at the same level as the feedback group even though they received the same amount of time to practice the movement as the feedback group.

Perceptual coupling in rhythmic movement coordination: stable perception leads to stable action

The aim of this study is to test a new technique that allows comparisons to be made between movement measures of between-trail (perceptual) and within-trail (movement) variability within the same perception-action task and person.

Experiment 1

Three groups of 8 students took part. At first participants produced 0° with three practice trails (0:0, 0:180 and 0:90). For the next three trials the cross-modal phase relation was set to 0° (0:0). This was followed by a block of six trials with the crossmodal phase relation set to 90° (0:90). The next six trials set the crossmodal relation to 180° (0:180), followed by three more 0:0 trials. Finally, there were two blocks of four trials in which the instructions were to produce either 180° or 90° visually.

First, they analyzed the MVLW data from the consistent conditions. Pairwise comparisons indicated the main effect of phase condition was due to within-trial stability at 0:0 being higher than the other two conditions, the main effect of frequency was due to stability being higher at 1 Hz. An average 90:90 and 180:180 were not different from each other.

Second, they analyzed the MVLW data from the 0° visual target conditions. Pairwise comparisons showed that the main effect of phase condition was due to within-trial stability at 0:0 being higher than the other two conditions and 0:90 being more stable than 0:180.

A clear relationship existed between within- and between trial stability. The matching scale of the two measures is evidence supporting the hypothesis that the differential movement stability is being caused by the differential perceptual stability.

The main new result is that the 0° visual target stabilized movements that were at a non-0° phase relationship to the dot being tracked.

Experiment 2

The five conditions from experiment 1 where replicated plus two new conditions – 90:0 and 180:0. Pairwise comparisons showed that the 0° visual target conditions were more stable than the non-0° visual target conditions. An ANOVA on the non-0° visual target conditions revealed they did not differ in their stability.

Learning a coordinated rhythmic movement with task appropriate coordination feedback

The aim of this study is to produce a novel coordinated rhythmic movement (90˚ mean relative phase) using coordination feedback during post training. This allows the participant to learn a novel task. This should not alter in any way the perceptual information. It is said that people cannot produce 90˚ novel movements stably without training. They can only produce two stable coordination patterns without training, 0˚ and 180˚. So progression and learning has to occur, in this case coordination feedback is used.

10 participants were split into two groups of five. Group 1 (‘Feedback’) received coordination feedback during training; Group 2 (‘No Feedback’) received no feedback but an equal exposure to the task.

There were two assessment sessions (Baseline and Post-training) and five training sessions. Participants viewed a demo of the target relative phase and then performed five trials. In the five training sessions participants performed ten 20 sec trials with a target mean relative phase of 90˚. There was a computer-controlled dot situated above the person-controlled dot. The participants then used a joystick to move the dot at a specific mean relative phase. For the Feedback group feedback was provided by the colour change of the person-controlled dot from white to green when the participant was moving at 90˚, ± an error bandwidth. In the first session the error bandwidth was set to 40˚. Any performance between 50˚ and 130˚ triggered the colour change. The bandwidth was decreased in each session to 30˚, 20˚, 15˚ and 10˚. This will drive learning as the participant improves after each bandwidth. A colour change was used as the coordinated feedback because it is said that colour has no affect on movement stability. The No Feedback group also did 50 trials, but with no feedback.

A repeated measures ANOVA was used. The results found that participants who received coordination feedback successfully and significantly improved their ability to maintain 90˚ coordination. The No Feedback group showed no improvement at any mean relative phase. Coordination feedback does not alter or remove the visual information (relative direction). Unlike visual metronomes and Lissajous figures which do alter the perceptual information.

People do not tend to suddenly acquire 90˚, as they are unable to move at 90˚ long enough to allow the required perceptual learning. The control group in this experiment received extensive practice at 90˚, but no help identifying when they were moving correctly, so failed to learn this coordination.

Perceptual Learning Immediately Yields New Stable Motor Coordination

This study is focused on the attainment and retention of coordinated rhythmic movements. Specifically, the study assesses the effect of perceptual learning in relation to the demonstration of the movement. Through previous studies, researchers have proved that movements at 0 º and 180 º are stable and can be spontaneously performed without much occurrence of phase variability. Movements at 90 º however, have exhibited an unstable result, concluding that this movement would have to be learned before attainment could be achieved.

The learning of a novel movement based on its stability is not simply attained through physical practise, relying solely on the limbs experience of the movement. Instead, perceptual consequences of the coordinated movement along with transformed feedback results in a non-0 º coordinated movement to be stabilised. Reasons as to why performance at 0 º is easily performed as opposed to other relative phases is due to the essential information being detected more promptly. Therefore, improvement in stability in novel coordination is the participants’ ability to detect essential information through his/her perceptual ability. This study’s prediction is that if the participant is able to improve his/her ability at detecting the essential information then an improvement in stability at 90 º will occur.

12 participants (22-54 years old), half of which were assigned to an experimental group and the other to a controlled group. There were two experimental tasks: two alternative forced choice judgments and coordinated rhythmic movement. Also included were assessment and training sessions. The assessment session included judgment trials and movement trials without feedback; and the training session consisted only of judgement trials with feedback. The controlled group did not participate in the training sessions and received no feedback. There were 21 different trial types regarding judgement trials- 10 different differences x 2 orders and a catch trial. Demonstration was given at the start of the assessment, without feedback throughout session. During the training session participants performed 12 blocks of “choose 90 º” with feedback; these were compared to four other phases. As sessions progressed, discrimination was made harder. If response was correct, they were told; if incorrect, they were given an example of 90 º. Performance during training determined whether participants progressed to a harder training session; however, after four repetitions, they were automatically progressed. During movement trials, force feedback feature was disabled and participants were seated comfortably controlling joystick without actually seeing it. To dots were displayed on the computer screen; top dot was under control of the computer and the bottom dot was controlled by the participant. The computer recorded joystick and computer controlled dots.

Repeated measures ANOVA of the judgement results revealed that participants learned as a result of the training they had received; showing that there was a significant difference for baseline vs. post training at 90 º, however, there was no significant difference when analysing 180 º. A repeated ANOVA was performed on the median proportion of time on task with the tolerance set at 20 º (movement task); two within subject factors were noted: phase levels and session levels. The study found that stability at 90 º improved for the experimental group after training was given. The experimental group showed a main effect of phase as well as an interaction between phase and session while the control group only showed an effect of phase.

This experiment shows that the perception-action couples aid the learning of a coordinated novel movement. This was determined through thorough testing on the experimental and control groups in which the control group exhibited no significant result without perceptual training. The study showed that once perceptual ability was improved, movement stability had been increased, which strongly supports the hypothesis that movement stability is a function of perceptual ability. Perception-action is a system in which there are informational and motor components which contribute towards the overall behaviour of the system.

Learning a coordinated rhythmic movement with task appropriate coordination feedback

The aim of this study is to develop and test a method of providing augmented feedback that could drive learning but not alter the informational content of the task. Ten participants split into two groups of five (one group received feedback and one group didn’t) took part in two assessment sessions and five training sessions. In the training sessions, they performed ten 20-s trials with a target mean relative phase of 90°, for a total of 50 trials. Feedback was provided to the feedback group by changing the colour of the person-controlled dot from white to green when the participant was moving at 90°, ± an error bandwidth. The error bandwidth faded; in the first session, it was set to 40° and was decreased in each session to 30°, 20°, 15°and 10° to drive learning.

A repeated measures ANOVA was carried out on the proportion time on task data (tolerance = 20°) with Session (2levels: Baseline, Post-training) and Phase (3 levels: 0°, 90°and 180°) as within subject factors and Group (2 levels: Feedback, No Feedback) as a between-subject variable. The group who received coordination feedback significantly improved their ability to maintain a 90° coordination. The No Feedback group showed no improvement at any mean relative phase.

This method demonstrated that it is effective at allowing people to acquire a novel coordination, and it does not change the overall perception action task dynamic. This method therefore will be useful in further studies examining the role of perceptual information or in studies that focus on the learning process.

Performing and learning 90° in older adults

Despite a wealth of knowledge regarding coordinated rhythmic movements in a healthy population of adolescents and younger adults this is not the case for their older counterparts. Following a review of the minimal literature the understanding of older adult populations appears inconclusive. From reviewing the main two papers similarities between the elderly and young populations are present. Firstly the elderly possess the ability to perform 0° and 180° at relative ease with little to no difference to the younger populations. With both populations able to learn the least stable phase of 90° when exposed to specific training.

However this is where the similarities cease and the differences begin. Despite both groups fully understanding the movement demands of a 90° phase the ability to produce this is diminished in older adults. Firstly older adults are more sensitive to different feedback types than younger adults. For example older adults did not improve when using augmented terminal feedback when the younger adults did. Furthermore, older adults show a greater improvement and retention of 90° when using concurrent augmented feedback when compared to other feedback methods used during training. Additionally learning 90° occurs at a slower rate in older adults, with the elderly initially drawn to perform at 180°; consequently leading to higher error rates during learning 90° in older adults. These error rates exist in variable movement frequencies, amplitudes and reduced accuracy with a 70° variation from the desired movement patterns. Therefore 90° cannot be considered stable but does become more so throughout training. However, the performance difference between the younger participants and the elderly becomes more pronounced with training.

So with poor performance by the elderly present in both papers when compared to the younger participants the question is, why? This may be explained through exploration of the feedback methods. Throughout training both terminal and concurrent augmented feedback was provided using a Lissajous display. In both papers subjects were tested at baseline and post training with differing perceptual information available. They were exposed to normal visual conditions, no vision of the hands and augmented feedback. The greatest performance in retention tests within older adults occurred under augmented concurrent feedback conditions. This was coupled with no significant difference between the no vision of the hands and normal vision conditions. Consequently this suggests that the perception of 90° or the requisite movement information is impeded in older adults. However, in slight contradiction to this, Swinnen et al. (1998) suggest that augmented visual feedback yields greater stability in retention under normal visual conditions. This may still suggest that older adults have improved their perceptual ability and are more able to gain the required requisite information due to training with augmented visual feedback.

In conclusion, by increasing the perceptual information available to older adults, novel coordinated rhythmic movements can be learnt. This may not result in movement stability on a par with younger populations, but yields improved performance when the augmented feedback is removed. However the research remains inconclusive and underwhelming in its amount and requires greater exploration within older adults.

References

Swinnen, S.P., S.M.P. Verschueren., H. Bogaerts., N. Dounskaia. (1998). Age-related deficits in motor learning and differences in feedback processing during the production of a bimanual coordination pattern. Cognitive neuropsycholocy, 15(5), 439-466.

Wishart, L.R., T.D. Lee., S.J. Cunningham., J.E. Murdoch. (2002). Age-related differences and the role of augmented visual feedback in learning a bimanual coordination pattern. Acta psychological, 110, 247-263.

Perception and coordinated rhythmic movements

When performing or learning skilled behaviour such as coordinated rhythmic movement’s perception and action go hand in hand. However to understand which of these is limiting performance the isolation of either perception or action is required. To achieve this Wilson, Snapp-Childs and Bingham (2010) used two- alternative forced choice judgements (2AFC) and coordinated rhythmic movement performance during their study.

Assessments were performed at baseline and post training. They involved judgement at 90° and 180° and movement at 0° 90° 180°. Further retention tests were performed a week after post training in which just the movement trials were performed. However to assess the role of perception no movement trials were performed during training. Therefore improved movement performance would be due to improved perception. The training consisted of judgement at 90° with up to 14 training sessions or to the point when performance plateaued. A control group did the assessment but no training.

Judgement assessments involved 2AFC at 90° and 180° with no feedback. This consisted of a pair of 2 moving dots with 1 of the pair moving at the target relative phase whilst the other was moving at the same or a different phase. Different phases were +/- 9°, 18°, 27°, 36°, 45°, producing 21 different trial types in randomised order. An additional 1 example assessment was given at the start of the trial

Judgement training was performed at 90°. There was 12 blocks of 2AFC 90° trials per training session. Each block compared 90° to 4 other phases 2 less than and 2 greater than 90°. Trial 1 and 4 were 90° +/- 40° reducing by 10° between each set. Trial 2 and 3 were 90° +/- 20° reducing by 5° between each set. Each of the 4 trials appeared in random order in each block. Feedback was given following judgements. When subjects achieved 85% correct they progressed to the next discrimination set. Maximum of 4 training sessions per set with a maximum 6 received on the last set.

References

Wilson, A. D., W. Snapp-Childs., & G. P. Bingham. (2010), Perceptual Learning Immediately Yields New Stable Motor Coordination. Journal of Experimental Psychology: Human Perception and Performance. 36(6), 1508-1514.