K211: Augmented Feedback + Observational Learning + Motor Imagery

Four reasons were presented as to why people may

choose to use physical guidance techniques. Which

of the following is NOT a reason?

a) To reduce risk of injury

b) To develop in the learner a 'sensory reference'

c) To increase variability

d) To encourage engagement/motivate

e) To allow attention to other task components

r

c) to increase variabilty

types of feedback

Inherent FB (intrinsic /response-produced sensory)

Augmented FB (extrinsic/ not naturally occurring)

Inherent FB (intrinsic / response-produced sensory)

sensory feedback internal to us

recall. closed loop control system. where would inherent feedback sources be

proprioceptive and exteroceptive feedback

Augmented FB (extrinsic/ not naturally occurring) is two types

Knowledge of results (KR)

knowledge of performance (KP)

where would augmented feedback be in closed loop system

measured outcome.

Augmented feedback definition:

movement-related information provided by extrinsic source

What is augmented FB?

1. adds to a performer’s intrinsic FB

2. provides info about the movement or movement outcome

3. Given by instructor, video or device

Four ways that Augmented Feedback functions

1. MOTIVATIONAL ROLE

• provides reward fx “good job”

2. ATTENTION-DIRECTING ROLE

• you can direct attention with FB

3. INFORMATIONAL ROLE

• provides error/accuracy info

4. DEPENDENCY ROLE

• guidance hypo: fb is used as a crutch

Knowledge of Results (KR) is? When is it important?

information on outcome (in relation to a goal)

It’s important when:

• Learner cannot assess their own success

• Helps improve next attempt by detecting error

Knowledge of Performance (KP) is

information about movement quality/form (kinematics).

• Concerned with quality of the mvmt pattern

• It doesn’t say anything about goal success

  • You can find out WHY the success worked

Common examples of KP

dartfish, ironpath

Video/motion feedback. What kind of FB is it and is it enough?

Video/motion FB is KP

Video FB is NOT enough, we want to supplement with additional info.

We want to direct people’s attention on where to look on the video

Bio or neuro-feedback

Electronic feedback of internal biological processes

e.g. EMG, HR, blood pressure

can KR and KP overlap?

Yes, they overlap when:

• situations where the goal is the movement e.g. diving

  • score that person is given is both KR and KP

• if goal is target HR, HR is both KR and KP

Problems with too much KP

1. Too much information (overload)

2. Directs attention internally

3. discourage ‘active learning’ via dependency

Direction of attention

Matters for performance. Can be internal or external. External (KR) is better than KP (internal)

• Two main types of augmented feedback

KR: related to an env goal

KP: how the mvmt itself was produced

• Often it's not enough to just show people what they did when

giving KP through video feedback for example, we often have

to

direct them to key information sources (error detection).

• Both KR and KP serve 4 functions:

Motivate, give info about error, direct attention, give guidance

• KP can inadvertently

1. direct focus of attention internally

2. Be overwhelming with info overload

3. cause reliance due to guidance

Feedback is similar to what, causing dependency..

physical guidance. Used as a crutch, stops active problem solving which causes issues in transfer when it’s not there

General term for how often/how much FB? What are the methods for reducing that?

Feedback frequency (every trial or some trials?)

-Methods for reducing frequency:
Bandwidth FB
Fading FB

Summary FB

Average FB

When to give FB?

During = concurrent

After = terminal

• Immediately or after a delay

2 ways to measure feedback frequency

1. Absolute Frequency:

• Total number of feedback presentations

2. Relative Frequency

• % of trials in which FB given

relative frequency formula

total number of feedback presentations / total number of movement attempts x 100

Is more frequent feedback better than less frequent feedback for learning?

Task: Using a joystick to track a repeating pattern

Study: Gun shooting with 100% vs 50% feedback with faded feedback (where you start with almost 100% FB but gradually drop to 50%)

Result: In retention without any FB, 50% group did better than 100% group

Guidance hypothesis

a failure to properly process intrinsic feedback

Coach giving info → don’t have to pay attn to internal FB → fail in transfer because coach not present there

The __________________ is critical for testing for

dependency

No-KR retention test

Guidance hypothesis can occur if FB is

provided too frequently

concurrent with performance

provided too quickly after practice attempt

What is one way to reduce amount of feedback?

Bandwidth FB

Bandwidth feedback (FB) is a

is a useful method to reduce feedback. Only give FB when errors fall outside ‘bandwidth of acceptability'

3 reasons why bandwidth feedback is more effective than feedback every trial

1. Motivation as less FB means they’re correct

2. Good for stability. Not trying to correct/adjust on every trial.

• *Correcting small errors from trial to trial can be harmful, especially as a novice. you don't have the motor proficiency to yet control those errors

3. Prevents dependency. Frequency of feedback decreased (faded schedule).

WHEN: FB given “during” trial is called? It’s similar to what?

Concurrent FB

Similar to physical guidance

Study that talks about WHEN of FB

Armstrong study

Task:

3 levels of feedback: Guidance, Concurrent and Terminal

Outcome

Only terminal FB (after-movement) aided learning

Concurrent and guidance fb were good at practice but horrible at transfer because crutch not there

When to give feedback post trial? why?

Feedback delay interval: If this interval is long then

After a feedback delay interval, NOT immediately

When feedback delay interval is longer, learning is enhanced

• This is because it teaches listening to intrinsic feedback which develops error detection capability

Longer pre-KR delay interval

encourages processing of intrinsic FB, enhancing error detection capabilities

If this is delay is long then learning is enhanced

Feedback delay interval such as PRE-KR delay interval

Feedback delay interval: If this interval is short then

not enough time to evaluate intrinsic FB

Evidence for delaying feedback interval.

Why was the outcome so?

Simulated batting task delaying feedback by ONLY 2 seconds lead to better performance!

Allows more time to process intrinsic fb and enhance error detection

Various techniques for alleviating effects of feedback and "Guidance"

Reduced Frequency of feedback

• Fading feedback - FB is gradually reduced

• Bandwidth feedback

• Summary feedback

Reduce when give feedback

• Terminal feedback vs. concurrent

• Increased KR delay

• Self-control of FB: Choose when to get FB

Not only is the content of feedback important, but so is the

timing.

Decreased frequency benefits are discussed in terms of the what hypo?

guidance hypothesis

...we fail to process our own intrinsic feedback of what we're doing (or what should be doing), when someone or something is doing this for us.

twos way to reduce guidance are

decreasing feedback over trials (i.e., reducing absolute and/or relative frequency)

OR

use a method to do that (e.g. bandwidth feedback - only giving feedback when outside of a predetermined band).

3 reasons were discussed for the potential benefits of bandwidth feedback

(motivation, stability, and prevention of guidance).

• In terms of when to give feedback it can be

during (concurrent)

or

after (terminal) a movement,

AND/OR

instantaneous or delayed (i.e., increased pre-KR interval).

Providing information at the end of a trial/after some delay

increases attention to other sources of naturally available, intrinsic sensory feedback - aids learning.

• There are many techniques for reducing guidance effects:

reducing, delaying, summarizing, fading & making decisions about when to receive feedback

refection 17A and B

What might be 2 inherent (intrinsic) feedback sources and 2 augmented (extrinsic) feedback sources in throwing a javelin, discus or shot-put?

From your answer to the question above about various types of feedback; how would you classify the sources of augmented/extrinsic feedback you listed with respect to KR and KP?

What is the main reason that would explain why not giving feedback on every trial is good for learning? What hypothesis has been proposed to explain this effect (and others) and what is thought to happen when feedback is given on every trial?

From the readings, what does it mean to say that feedback can have a DIRECT (vs indirect) motivational effect on learning? To answer this question, consider the study discussed in the book and how giving feedback for “successful” trials impacts learning (p255)

There are many possible answers. As to intrinsic, this could be how it felt/proprioceptive feel about the shoulder or elbow, maybe what it looked like (the ball or javelin in the air), what it sounded like as it hit the floor or how long it took to hear the sound of it hitting the ground etc.

Extrinsic feedback could be the actual outcome in cm as to where the object landed (you may see where it landed but wouldn't know the exact distance). Video feedback about technique or coach feedback would be other examples of extrinsic feedback.

KR = outcome feedback; so things such as where did it land (distance in cm)

KP = performance feedback, information about the quality/form of the movement, so things like video feedback or coach feedback about movement technique (e.g., elbow angle on release, flight path of the javelin).

Guidance hypothesis.

If feedback is given too much or too frequently, then the individual learner is not paying attention to their own intrinsic feedback sources and become dependent on the augmented feedback; such that when it's removed, performance suffers.

"Guidance hypothesis relates to a failure to properly process intrinsic feedback"

reflection 17B continued

From the readings, what does it mean to say that feedback can have a DIRECT (vs indirect) motivational effect on learning? To answer this question, consider the study discussed in the book and how giving feedback for "successful" trials impacts learning (p255)

Feedback can have a indirect motivational effect on learning, whereby the feedback may make you feel more motivated and this makes you practice for longer. If you practice for longer/more, then you're likely to show improved learning and as such, there was an indirect effect of feedback on learning (the direct effect was of practice amount). but, feedback has also been shown to directly impact learning (probably through reward/consolidation pathways in the brain). This direct impact of increased motivation on learning was shown in the research example where individuals only given feedback after successful trials, vs. less successful trials, showed improved retention .

Observational learning:

Observer watches a demonstration and adapts their perfomance afterwards. Changes their behavior

e.g.

watch videos or coach demos

if injured, can watch model

coach showing a serve

Motor imagery

Imagining ourselves doing an action, without actually moving

Official lol:

Cognitive simulation process that allows us to represent perceptual info in our minds in the absence of sensory input

What information could have been conveyed with a demonstration?

Cognitive-Spatial elements of the task

Perceptual-Motor elements of the task

Cognitive-Spatial elements of the task

• Goal / reference-of-correctness: knowledge of how the task should look, sound, and/or feel like

• Explicit strategies - observers will copy the strategy used by the model

Perceptual-Motor elements of the task

• Implicit (non-verbalizable) motor plans, potentially via

action simulation

Observation: explicit

more conscious of it, instructed to do it

Observation: implicit

non-conscious, non verbalized, process going on while we’re observing it

What is implicit mapping?

Implicit mapping through action simulation: Mapping observed actions onto our own motor plan

Evidence that we engage in 'simulation' when watching others

"Mirror Neuron System" or "Action observation network"

You as the observer are not doing the action, but the neurons are still firing as if you are. So we engage in sim when watching others

______ and _______ movements share a _____ neural code

Watching, executing, common

Study showing evidence of action sim in skilled performers

Task:

recruit expert dancers and martial artists to watch dance and martial arts

Outcome:

Skilled performers show evidence of action sim when observing motor skills they have a background in ie. neurons more active for familiar movements

The mirror neuron system is also involved during

observational practice of new skills

Evidence that mirror neuron system is involved during observational practice of NEW skills

Task:

Watch DDR demo with a human doing it or just with arrows

Outcome:

People learned best when they had someone to observe

Model skill level: Who should we watch?

Both

What do you learn from observing an expert?

• Reference of correctness

• More attention given to expert

• Could be demotivating

What do you learn from observing a novice?

• More problem-solving

  • Alerts to errors novices can make, helping avoid them

• More at par with stage of learner

• More variability in demo

Summary of features that can be picked up from watching others

Watching an expert and novice both good for learning, but thru diff mechanisms

Watching an expert:

More accurate template of goal/reference of correctness

Watching a novice:

Better parameterization from observed variability

Improvement in error detection and correction abilities from problem solving

What’s the best model: correct or learning?

Learning model with KR > Correct Model > Learning model

ie watching errors was GOOD as long as some extra info about those errors were given (Learning w/ KR)

Motor imagery

Cognitive process that allow us to represent perception into motor simulation without sensory input

Imagery modalities

Visual content imagery

Auditory

Olfactory

Kinesthetic/motor

We engage in __________ during imagery. Similar to?

"motor simulation"

Action simulation during observation

What components can you improve in motor imagery?

improvements with imagery practice. Such as:

1. Generation: bring the imagine to mind

2. Maintenance: be able to maintain that image

3. Manipulation: able to manipulate that image

4. Vividness: how clear/bright is that image

How do we measure motor imagery?

Questionnaire

Mental rotation

FMRI/TMS

Mental chronometry

Mental chronometry method

If action and imagery are “functionally equivalent”, they should have similar timing

ie. similar times between imagined vs physical action bc action and imagery are ‘functionally equivalent’

Functional equivalence:

imagined actions have been thought to be functionally equivalent to the overt action

Mental chronometry activity

Switching from the dominant to non-dominant hand should show similar MT increases in both physical task execution and motor imagery.

Because you 'have access' to your motor capabilities through motor simulation!

Aphantasia is a

Absence of ability to do visual imagery

• Can be acquired (injury) or congenital (from birth)

Motor imagery uses

simulative processes - idea that imagined & physically executed actions are 'functionally equivalent'

Motor imagery consists of

different processes, which can be inferred using a variety of measures such as

• Mental chronometry

• TMS/fMRI

• Questionnaires

• Mental rotation/laterality judgements

• For improved outcomes

ensure engagement

Alternate physical and mental practice

supplement imagery with observation

• Some people are unable to do motor imagery, termed

aphantasia, but it doesn't seem to have much impact on their general activities.