Lecture 3: Differential Reinforcement, Antecedent Control, Shaping and Chaining Flashcards
Three learning outcomes that occur during differential reinforcement
Response Differentiation:
* Differential Reinforcement of Talking
About Sports Over Problems.
* Joe talks about sports (R1) is followed
by step father’s attention (Sr+) which
increases the future rate of this
response (positive reinforcement).
* Joe talks about school problems (R2) is
not followed by step father’s attention
(Sr+) which decreased the future rate of
behaviour (extinction).
* The effect is that Joe learns that talking
about sports is reinforced and talking
about school problems is extinguished.
Joe learns response differenitation of
sports topics.
* That is, Joe learns what behaviours are
reinforced and others that are not.
Response Variation:
* Talking about Sports that Produce
Reinforcement
* Members of a response class (R1):
talking about baseball, football, hockey,
other sports topics etc. are all
reinforced by gaining his step father’s
attention (Sr+).
* Reinforcing all behaviours within a
response class gain reinforcement
increases the behavioural repitoire of
the child and promotes generalisation.
* For example, learning to eat soup with a
spoon, hands for pizza, and fork for
vegetables.
Response Narrowing:
* When only one specific behaviour
(rather than a response class) is
reinforced the child’s behavioural
repitorie may narrow as they only
engage in the behaviour that is
reinforced.
* For exmaple, only talk about hockey
because it is the only sports topic that is
reinforced.
Shaping via Successive Approximation
- To develop a response that differs
significantly from existing responses. - For example: Replace Reaching with
Use of a Manual Sign to Makes a
Request - Use shaping to teach students a new
skill. Rewarding successive
approximations until they can do it
independently without challenging
behaviours occurring.
How to Use Shaping
- Specify the target response (desired
behaviour) - Specify the positive reinforcer(s) to be
used. - Specify initial and intermediate
responses. - Reinforce the initial response each
time it occurs and withhold
reinforcement from other responses
until the initial response is performed
consistently. - Shift the criterion for reinforcement
from the initial response to an
intermediate response. - Reinforce the intermediate response
until it is performed consistently, then
shift the criterion for reinforcement
gradually to other intermediate
responses that are increasingly similar
to the target response. - Reinforce the target response when it
is performed.
Shaping behaviour
You need to know the child’s current level of functioning, reward it, and then increase the demands. Smaller steps take longer but guarantee skills is reached. Larger steps risk losing engagement of the child.
Chaining
- Shaping is similar to chaining.
Behaviour chaining are complex
behaviours that involve a series of steps
that are completed in a specific order to
achieve the task. - These are examples of complex
behaviours that include a behaviour
change:
o Linked sequences of responses leading
to a terminal outcome
o Making microwave popcorn
o Self-check out at Grocery store
o Using an ATM
o Dressing
o Preparing a sandwich
o Teeth brushing
Task Analysis
Task Analysis
* First step of behaviour changing is
conducting task analysis. Identifying the
sequence of teachable steps.
* Breaking multi-step/ complex tasks into
a sequence of smaller steps or actions.
How to write a Task Analysis
* From memory
* Record steps while performing task
* Record steps while watching others
perform task
* Use available Task
Sequencing Instruction
- There are three ways to teach
sequences of teachable steps in
behaviour chaining:
o Backward Chaining: instructor
completes the first steps and asks the
student to complete the last step to
proficiency. Then the demands are
increased to the last two, three etc.
until all steps are completed
independently. This enables students
who struggle to learn multi-step tasks
immediate reinforcement.
o Forward Chaining: instructor asks the
student to complete the first step and
the instructor completes the rest until
the sequence is completed and they
can be reinforced. Demands are
increased to first two steps, three,
fourth etc. until the master the chain by
themselves.
o Task Presentation: instructor asks
students to complete all steps in the
chain to gain reinforcement on all trials.
Errorless Learning
Errorless learning giving opportunity to respond and be reinforced whilst making it harder for them to make an error. What level of prompt is needed to get the child to respond appropriate behaviour to be reinforced. Fade out prompts to teach the child to do the behaviour themselves.
Prompts
- Context determines what level of
prompt is more or less intrusive (i.e.,
visual prompt is less intrusive than vocal
at a restaurant to tell someone they
have food in their teeth). - Hans the horse could not do maths
they just stamped their hoof until the
crowd cheered (kicked his hoof when
the crowd looked down and stopped
when they looked up). - Teachers can inadvertently (outer
directedness) cue the right response to
children rather than actually
comprehending the skill (i.e., look to the
teacher rather than the materials).
The Problem of the First Instance
- The problem of the first instance is how
do you reinforce a behaviour that
occurs at low rates or is not in the
child’s behavioural repertoire? You
prompt students to engage in target
behaviour in order to reinforce them. - Response Prompts
- Verbal
- Gesture
- Model
- Physical
- Prompt Fading
Delay
Hierarchies
Magnitude Fading
Stimulus Shaping
Changing of the physical dimensions of the stimulus over time.
