Week 6 Flashcards
learning
a relatively permanent change
adaptation
a transient change
2 types of test learning can be evaluated
- retention test
- transfer test
potential issues with learning a surgical skill in using a new method (virtual reality) and an old method (physical practice)
testing environment is familiar to one group but not the other
learning and performance variables
in retention and transfer tests we did not consider performance during acquisition at all
- changes in acquisition are NOT relatively permanent due to performance variables
performance variables definition
influence performance in transient ways
- the effect of the variable disappears when conditions are altered (intervention remvoed)
learning variables definition
influences performance in relatively permanent ways
- the effect of the variable stays when the conditions are altered
performance variables
feedback/tools that affect performance
learning variables
feedback and tools that lead to permanent changes
example of learning variables
- training wheels on a bike - when removed, the effect of the variable stays
- chair to hold postural support when skating
- feedback from a video tapes
well-practiced skills and assessing learning
for well-practiced skills it is hard to assess if additional practice will result in improvements
- ceiling effect
other ways of assessing learning
dual tasks to gauge is there is any improvements in learning
performance curves
indicated that skills can be learned using different patterns
overlearning
the process of having a person continue to practice after they have reached a performance ceiling
how would we measure if any more performance effects are happening during overlearning trials
- reach maximum score then keep going
- by computing a savings score
what can we do to assess the effect of overlearning?
calculate a savings score
savings score
how many trials the overlearning group saved in reaching their ceiling
what indicated overlearning?
more practice trials at ceiling
example of overlearning and ceiling effect
group A: 15 trials to reach ceiling
group B: 10 trials to reach ceiling
- therefore group B saved 5 trials relative to group A
other ways of assessing learning
- performance on a secondary task
- measuring indices of effort
- measuring response latency
- generalizability of learning
performance on a secondary task
gives us an idea of how much attention is needed to perform a task
- a well learned task requires less attention
measuring indices of effort
- physiological markers (learned task is to a greater extent)
- psychological markers
measuring response latency
speed of correct response or movement performance
generalizability of learning
varying the parameters of the task (linked to GMP)
the amount of learning
hard to quantify
- criterion that is used can affect the learning curve shapes
what is often used to quantify the rate of learning?
steepness of learning curves in acquisition period
2 problems with using the steepness of learning curves for quantifying rate of learning
- same-problems as the amount of learning
- inter-individual variability shows in practice
- can cloud results
when can you only assess learning?
in retention and transfer
- acquisition does not show learning
how should practice be distributed?
- massed practice
- distributed practice
massed practice
practicing with very little reset in between trials
distributed practice
practicing with longer rest periods in between
how is research conducted for practice distributions
using continuous tasks
- pursuit tracking task
short rest periods vs. long rest periods
short rest periods degrade practice relative to longer rest periods
confounding factor to short rest periods between trials?
fatigue = get tired quicker
fatigue
plays a role in tasks such as the pursuit-rotor task
- can affect performance
bourne and archer (1956)
zero seconds rest group performed worse in transfer
- groups with shorter rest periods learned to a lesser extent
ammons (1950)
used a pursuit rotor-task with 0s, 20s, 12 mins, and 24 hours for 36 trials
1. 20 min rest period then 36 transfer trials
2. 1-day transfer trials
two key findings from ammons (1950)
- only small differences remained after the last transfer trials
- the differences in performance re-emerged on the 24-hour transfer trial
- groups with MORE rest PERFORMED BETTER
baddeley and longman (1978)
looked at typing skills in postal workers and compared different distributions and looked at long-term retention
baddeley and longman (1978) study design
- each training session was an hour
- 1 session one per day (1 x 1)
- 2 sessions twice per day (2 x 2)
results of baddeley and longman (1978)
found that more distributed practice led to better retention when practice hours were help constant
- group that practiced the most = worst performance (same total amount but one distributed the most = better)
distributed vs massed practice application
- sports
- rehabilitation
- medical education
andersen et. al (2015)
explored “learning” in VR simulation of mastoidectomy
andersen et. al (2015) and training schedule
- distributed = Oct to Nov (2 procedures were separated by at least 3 days)
- massed = May 16, 17, 18 (3 days)
- all sessions completed in 3 days
(one group over 2 months, one group over 3 days)
results of andersen et. al (2015)
- both groups shows significant differences from session 1 to 12
- distributed performed better than massed at the end of session 12
- participants learned well due to distributed practice
how could we do better for andersen et. al (2015)
yes, didn’t do retention to transfer test
(need to include)
massed practice
reduced benefits per time training, but takes less total time
distributed practice
most learning per time training but takes more time
constant practice
one variation of a task per practice session
variable practice
many variations of a task within a practice session
what classification of skills does variable practice make sense for?
open skills
which practice produces more improvement in learning?
variable practice over constant practice
for what skills does variable practice appear to produce more learning than constant practice?
open and closed skills
(based on schmidts schema theory and the idea of motor programming
what else is variable practice better than constant practice for?
retention and transfer
shea and kohl, 1990
examined whether participants trained constant (specific) or variable practice conditions performed better in retention
- used a force production task where participants either produced a constant or variable force
shea and kohl, 1990 variable practice in retention
participants who performed with constant (specific) practice were worse than participants who trained with constant + variable practice
mckraken and stelmach, 1977
investigated the learning of an obstacle hit task
- participants had to move from a starting position to knock over a barrier in a goal time of 200 ms
- practiced either with constant practice or with variable practice
variable practice in transfer (mckraken and stelmach, 1977)
the group that practiced with variable practice actually performed better on the transfer task (when asked to hit new angles)
variability of practice
variable practice increased the generalizability for learning
- performance is better on delayed retention
- performance is better on transfer tests
(even if the test value was outside the range of values used in acquisition)
schema theory
with practice people develop rules or “schemas” about their own motor behaviour
characteristics of the schema theory
- participants learn schemas in practice context (contains relationship between environmental output and inputs)
- rule is maintained in memory and can be used to select a new set of parameters
- person is subsequently able to select parameters to produce a desired outcome
variable practice and schema theory
variable practice produces a schema for selecting the parameters of the GMP
(rule can be used for novel movement that relies on the same motor program)
characteristics of variable practice
- more powerful in children’s motor learning (Green et. al, 1995; Wulf, 1991)
- learning the absolute timing of a task (Wulf and Shea)
learning the absolute timing of a task
learning the relative timing of patterns within a task might be better with constant practice
does practicing the same thing over and over again really help you get better?
not really
