Kin 312 Augmented Feedback

Feedback Definition

a generic term that describes information a person receives about the performance of a skill during or after the performance

2 Categories of Feedback

1. Task Intrinsic Feedback (sensory)
2. Augmented Feedback (task-extrinsic)

Task Intrinsic Feedback

-performance related sensory information that is naturally available to the performer
o visual
o tactile
o auditory
o proprioceptive (naturally available)

Augmented Feedback: Definition

-information the person receives about a performance from sources external to themself
-the information adds to or enhances the task-intrinsic information

2 Types of Augmented Feedback

1. Knowledge of Results (KR): performance related information about the outcome of the performance

2. Knowledge of Performance (KP): performance related information about the movement characteristics that led to the outcome of the performance

How essential is augmented feedback for skill learning?

1) It can be essential for skill learning
2) It can hinder/slow skill learning

3 Situations Where Augmented Feedback Can be Essential for Skill Learning

1) Critical task-intrinsic feedback is not available (obstructed view)
ex: big team, no individual attention

2) Person is not capable of detecting the critical task-intrinsic feedback
ex: injury

3) Person is not capable of interpreting the critical task-intrinsic feedback
ex: lack of experience

3 Situations Where Augmented Feedback Can Hinder/Slow Skill Learning

1) Erroneous (incorrect) KR or KP
2) Concurrent presentation of KP
3) KR or KP presented too frequently

2 Issues with Content of Augmented Feedback

1) Errors v. correct aspects of performance
2) KR v. KP

Errors vs Correct Aspects of Performance

-Which type of information do we give the learner?
o Error-related information: action goal achievement role
o "Correct" : motivation role
o Combine both for needs of the learner

KR vs. KP

-both can provide valuable information

1.KR is especially beneficial to:
-confirm subjective assessment of task-intrinsic feedback
-determine performance outcome when task-intrinsic feedback is not available
-motivate the person to continue (ex: provide times for sprints; gives athlete a time they want to beat & therefore continue sprinting)
-establish a "discovery learning" practice strategy

2. KP is especially beneficial to:
-learn skills with specified movements--the outcome is assessed by movement characteristics (ex: gymnastics, diving)
-improve specific movements or correct specific movement errors
-learn to activate specific muscles
-facilitate learning when KR is redundant with task-intrinsic feedback (ex: "you missed the free throw" , "yeah no shit")

Types of KP

1) Descriptive KP: KP describes error
2) Prescriptive KP: KP describes error and prescribes how to correct it

-which one is better to facilitate skill learning?
o depends on the learner's stage of learning
-- initial stage of learning: prescriptive preferred
-- later stages of learning: descriptive preferred

Video Replay

the use of videotape replay should relate to the stage of learning

-initial stage: provide attention-directing or error-correction cues

-later stages: some assistance may be needed, but attention-directing cues may help

The Timing of Augmented Feedback: Three Issues

1. Concurrent and terminal augmented feedback
2. Time intervals related to terminal augmented feedback
3. Frequency concerns

Concurrent Augmented Feedback: Definition

augmented feedback available during the performance of a skill

Terminal Augmented Feedback: Definition

augmented feedback available after the performance of a skill

Research Based Conclusions: Concurrent Augmented Feedback

Concurrent augmented feedback can:
-facilitate the learning of skills in which critical task-intrinsic feedback is difficult to interpret
ex) learning to activate specific muscles
learning to produce a specific movement force

-Hinder or slow the learning of skills by leading to a dependency on the feedback
ex) performance poorer without the feedback than with it

Time Intervals Related to Terminal Augmented Feedback

-KR-Delay Interval: used to interpret task-intrinsic feedback
-Post-KR Interval: associate task-intrinsic with augmented feedback

Research Based Conclusions for the KR-Delay Interval and the Post-KR Interval

Interval Length: minimum length needed for augmented feedback to be effective; no evidence for maximum length

Activity during the interval:
\-most have no effect
\-some hinder learning (ex: similar learning processes required for both)
\-some beneficial for learning (KR-Delay Ex: subjective error estimation; Post-KR Ex: Problem solving activity)

Frequency of Presenting Augmented Feedback

-Traditional view: augmented feedback for every trial (100% frequency) is best for learning

-Current view: less than 100% frequency is best for learning
*based on research that has investigated several types of techniques that reduce augmented feedback frequency

2 Reasons Less Than 100% Frequency of Augmented Feedback is Better for Learning

1. Increases problem solving activity:
\-less frequency allows person to actively use their own problem solving strategies
\-100% frequency directs problem solving activity from the instructor's point of view

\
2. Guidance Hypothesis:
\-beneficial performance effects of receiving augmented feedback on every practice trial become negative learning effects
\-Why? : augmented feedback becomes a "crutch" on which the learner depends to perform the skill

4 Techniques that Reduce Augmented Feedback Frequency

1) Fading Technique
2) Performance Bandwidth Technique
3) Learner Requested Feedback Technique
4) Summary Technique

Fading Technique

-first described in experiment by Winstein & Schmidt (1990)
-systematically reduced KR frequency from every trial to 2/8 trials
-"fading" technique led to better learning than receiving KR every trial

Performance Bandwidth Technique

-leads to a less than 100% frequency
-results in better learning than KR/KP every trial

Learner Requested Feedback Technique

-KR/KP only given when learner requests it
-leads to a less than 100% frequency
-results in better learning than KR/KP every trial

Summary Technique

-KR/KP given for a set of trials (ex: every 5 trials)
-amount of KR/KP is same as 100% frequency, but given less frequently during a practice session
-results in better learning than KR/KP every trial