Module 5 – Part I. Contingencies, Self-Management, Rules and Instructions.
In this module, we will discuss some of the complex applications of behavior analysis.
Specifically, we will discuss instructions and rules, token economies, group
contingencies, contingency contracting, and self-management strategies.
We previously looked at how verbal instruction can be used as a prompt. In other
words, behavior can be controlled by a verbal stimulus. In fact, most teaching in the
school setting and daily life is through verbal instructions.
To effectively use verbal instructions and rules, you need to determine if the individual
behavior is indeed controlled by the verbal stimulus. For example, you can assess if the
learner responds to your instructions and follows the rules.
However, for a lot of the learners that you work with, they probably do not respond to
your instructions. In this case, teaching instructions and rules are critical. As you might
have guessed, teaching responding to instruction would involve a DTT procedure, in
which you would present the instruction, prompt the learner to respond to the
instruction, reinforce correct responses and use error correction for incorrect responses.
You will need to fade the prompts too.
As with all other skill-acquisition programs, you should start by simple instructions such
as “stand up,” “come here,” and “sit down.” Once the learner is able to respond to
simple basic instructions, you can move on and teach instructions that are more
complex. You can use skill assessments such as VB-MAPP and ABLLS, to determine
the instructions.
Initially, you will notice that the learner may take a long time to master one instruction,
but the trials or time to mastery will gradually decrease as they master more
instructions. This is called the learning-to-learn phenomenon.
You should further add rapid presentation of the learned instructions to train the learner
to respond fluently.
When you use instructions, make sure that you use simple and clear words that are in
the learner’s repertoire. Needless to say, if a practitioner uses some words that the
learner does not understand, it is likely the instruction won’t be followed. If instructions
are repeatedly not followed, your instructions will become s-deltas.
When teaching, present instructions at a rapid pace. If the learner responds correctly,
provide praise and reinforcer, and then immediately move on to the next instruction.
Past research has shown that rapid presentation of instructions decreases problem
behavior.
The effects of instructions during teaching can be enhanced with other strategies such
as modeling and different prompting procedures.
Direct Instruction model is one of the teaching models that relies on the rapid
presentation of instruction, frequent feedback, and prompts where necessary. Direct
Instruction is based on the philosophical principle that all children can be taught. It
places students at their skill levels, and students with a similar skill level form a group.
The programs are designed to ensure mastery through repeated instruction and
practices of smaller, easier instructional units before moving onto instructions that are
more complex and difficult. The instruction can also be modified to accommodate the
student’s rate of learning.
Project follow through was the most extensive educational experiment ever conducted.
It began in 1968, with more than 200,000 students in 178 communities. In 1977, the
evaluation showed clear and strong effects of Direct Instruction. It has produced higher
academic achievement, self-esteem, and self-confidence measures among the
students.
In the next slide, you will watch a series of Direct Instruction videos. You will have some
ASR questions on Direct Instruction after the videos.
Rules and instructions can also be used to manage problem behavior. If rules exert
control of behaviors, you can effectively increase appropriate behaviors such as on-task
and cooperation and decrease problem behaviors. To make rules effective, you should
involve individuals, such as your students, in the development of these rules. You can
ask them what behavior is appropriate, what goal they would want to achieve – and if
your rule states the reinforcer, what reinforcer they want to receive. Involving them in
the development of rules will increase their awareness and more likely result in them
following the rules.
When you construct rules, make sure these rules are stated positively. For example,
state the rule by telling the students what they should do, instead of what they should
not do. A rule that states, “keep your hands to yourself in class,” communicates your
expectations and the appropriate behavior, while a negatively stated rule, such as “don’t
hit your friends,” focuses on the problem behavior.
Relatedly, the rules should be concrete and specific. For example, when I visited a
center for children with autism a couple of years back, the instructor grouped a number
of children between 3 to 5 years old together and asked them to be respectful. It is
unlikely the children would follow the rules simply because they were not trained to
respond to conceptual terms. “Being respectful” is one of them. Instead, the instructor
should have communicated specific behaviors that the children can perform. For
example, the rule may be restated, “keep hands to yourself.” It is best to outline the
specific behaviors that you want the students to follow.
In addition, try to keep your rule as concise as possible because students are more
likely to remember and follow them.
Between classrooms and instructors, it may be beneficial to come up with common
rules to reduce potential confusion.
Last but not least, if the students do not have the skills to follow the rules. You should
teach the rules. I have discussed how to teach learners to follow instructions earlier.
Instruction on rule-following behavior is similar. You should provide a rule and ask the
students to follow it with prompts and repeated opportunities to practice rule-following.
Relatedly, make sure that you reinforce rule-following behavior. Specifically, when the
students follow a rule, deliver the reinforcer and praise the rule-following. Make sure
that you explicitly associate the reinforcer with rules. You can deliver a verbal statement
saying, “great job following the rule!” It is, of course, okay to praise appropriate
behavior, but it may only strengthen that specific behavior. Instead, emphasis on rulefollowing is necessary if you want to strengthen the overall rule-following behavior of
students.
Primary reinforcers such as edibles are often used as reinforcers during early
behavioral interventions for children with autism and other developmental disabilities.
Largely, this is because children with developmental disabilities do not have many
reinforcers, and edibles are most likely a safe option. However, the effects of edibles as
the reinforcers are affected by motivating operations such as satiation.
As such, generalized conditioned reinforcers should be established. Tokens are one of
these generalized conditioned reinforcers. In our daily life, money is a token that can be
exchanged for food, shelter, and various merchandise.
One of the benefits of the tokens is that they are not affected by motivating operations.
For example, just because you just ate, money will not suddenly lose value.
It is a result of money being exchanged for various reinforcers. Thus, when you use
token economies, make sure that you do not just pair the tokens with one reinforcer.
Otherwise, the tokens will only function as reinforcers when the learner is motivated for
the reinforcer that tokens can be exchanged for. One example is that food stamps and
vouchers are just exchanged for food – their value is correlated to the value of the food
item.
Additionally, tokens can bridge the temporal and spatial gaps between the behavior and
reinforcer. For example, after you have received money for the service that you
performed, you might not immediately exchange the money for goods. You could
exchange the money for goods at any time that you like. Similarly, you do not need to
exchange the money for goods in the same location where you have earned it.
As a result, token economies are widely used in home, school, and clinical settings.
Token economies have three components. First, you need to have target behaviors for
the tokens. Of course, you will also need to have the tokens – they could be any visual
stimuli such as points, marks, stars, or physical tokens. These tokens need to be
delivered for the target behavior. Last, you need a menu of backup reinforcers. The
reinforcer can be obtained by exchanging the earned tokens.
The level system is a type of token economies consists of a hierarchy of tiers or levels
across which students are moving up and sometimes down, contingent on meeting
specific performance criteria. A higher level is associated with more privileges but also
comes with a more stringent performance criterion.
For example, you can arrange less desired reinforcers in the earlier levels and more
desired reinforcers in the later levels. Students have to meet a behavior criterion to
move on to the next level. For example, you can use access to the internet as a
reinforcer. The earlier levels are associated with internet usage with staff supervision in
one location. Students can move up to the next level if they demonstrate 0 problem
behavior. The next level may be associated with the internet usage with staff
supervision but in different locations.
Alternatively, you can also let the students earn points for appropriate behavior. The
amount of points is associated with the level, and the levels are associated with different
menus of items. For example, in Level 1, you can have a menu of less desired
reinforcers, as selected using preference assessment. If a student earns between 1 –
30 points, they can only use the Level 1 menu. But in Level 2, you can have a menu of
more desirable reinforcers, and students need to earn more points to access the Level 2
menu. As the level progresses, the point requirement increases.
To design an effective token economy, you will need to identify the behaviors and the
goal first. Obviously, you need to select observable and measurable behavior. You need
to determine what constitutes task completion. You will begin with a smaller number of
behaviors that are easy for the learner to complete so that the learner can receive
tokens relatively easily, at least initially. Likewise, please select the behaviors for which
the learner has prerequisite skills. Otherwise, you will need to address the prerequisite
skills before targeting these behaviors. Last, token economies are an individualized
system; it should not be applied across the learners in one setting. Even if a number of
learners would use token economies, their systems should differ from each other to
meet individual needs.
When you select tokens, you can choose a number of different approaches. For
example, you can draw tallies on a piece of paper, punch holes in a card, deliver poker
chips or coupons. The rule of thumb is that the token should be safe and ageappropriate. For example, they should not be swallowed by young children, and
delivering a token with cartoon characters to a 20-year old may not be appropriate.
You should also make sure that the learner cannot make the tokens themselves. For
example, if you draw tallies on your datasheet, make sure that the learner does not
have access to your datasheet.
We had a child with autism on a token economy system. Whenever we tried to get
some materials for the next trial, he would grab a token and stick it onto his token board.
Well, clearly, the child found the tokens to be reinforcing, but this should not have
happened. So make sure that they cannot make their own tokens.
You should use tokens that are easily accessible to you and that are inexpensive so
that the replacement of tokens would be easy.
Last, past studies actually have found that using objects that the learner likes or
obsessed about can enhance the effects of tokens.
For example, we previously worked with a child with autism who likes balloons, and we
made balloon-shaped tokens. It worked very well with them
However, if the objects have a lot of details that may be distracting, such as movie
characters with details of these characters, you should refrain from using these tokens.
In token economies, you need to determine backup reinforcers. You should use
preference and reinforcer assessments to figure out. It is preferable that they are
naturally occurring activities, privileges, and events, although they could be tangible and
edibles in the beginning. For example, we used a token system to decrease the
aggression of a teenager with Down syndrome. We worked with his family to develop a
menu of backup reinforcers that he received at home, including going on a subway ride
– because he liked trains – and going out to have dinner. These activities were highly
reinforcing and were naturally occurring events.
Next, you should decide how many tokens the learner needs to exchange for a backup
reinforcer. In other words, what is the price of the backup reinforcer? When you start the
token economy, make sure that the learner has contact with the contingencies, for
example, you can set the initial ratio low. For young children with autism, you can
consider starting with a 1:1 ratio – a backup reinforcer for one token.
After the learner is familiar with the token system and the tokens have been established
as conditioned reinforcers, you should gradually increase the number of the target
behaviors. In this case, the learner will receive more tokens, and the cost of the backup
reinforcers should also increase.
Next, you should provide the menu of the backup items with costs for them on the
menu. For example, a menu may state “3 tokens for a potato chip.” It is also possible to
display all backup reinforcers on the table and with costs labeled next to them.
Likewise, you will need to determine how often tokens can be exchanged. Lower
functioning children may require more frequent exchanges, while higher functioning
children require fewer exchanges. Initially, you can include a higher frequency of token
exchanges, and the frequency is then gradually reduced. For example, initially, the
menu may only include tangibles and edibles that can be delivered within the same
session, so the learner can exchange the tokens at the end of every session. Later on,
you can add the after-school activities with higher costs associated with them. The
learner will then need to exchange the tokens before going home.
Now, what if the learner does not want the token or the backup reinforcer? In this case,
do not engage or argue with the learner. You can either respond neutrally by saying,
“okay, it is your decision.” Alternatively, you can ignore the learner and proceed to
deliver the tokens and reinforcers. If the learner throws the token away, do not respond
or replenish the token. Move on to the next task.
Likewise, if the reinforcer or reinforcer menu gets pushed away, or the learner does not
engage with it, take away the reinforcer or the menu. That said, you should always
involve the learner in determining the backup reinforcers.
A procedure that is often used with token economies is the response cost that we have
previously discussed. In response cost, a token is taken away contingent on the target
behavior. Bonus response cost may be easy to implement for the teachers in a larger
class to decrease behavior.
To choose whether you want to include the response-cost procedure, you can also
assess the learner preference. Some studies have found that some learners seemed to
prefer response cost over earning tokens.
If you choose to include the response-cost procedure, make sure that the students have
tokens that can be taken away.
Last, you should test your token economies for a few behaviors first. You will need to
assess if the learner has the prerequisite skills and determine if the criterion of task
completion should be adjusted.
As with all other procedures, token economies should be faded gradually. First, you
should pair social praise with the tokens so that praise can serve as the reinforcer when
the tokens are faded.
Second, as previously mentioned, you should increase the number of responses
required before exchanging for a backup reinforcer. Likewise, you should increase the
cost of the items.
Furthermore, you should consider reducing the time in which the tokens are used. You
can designate some time slots for the tokens and gradually reduce the duration.
Fourth, if you are using tangibles and edibles, you need to assess what the backup
reinforcers are in the natural settings, such as the school and home settings, and
gradually replace the tangibles with these naturally occurring reinforcers.
Last, you can fade the tokens physically. For example, when you start using the tokens,
you may be using poker chips. Gradually, you can start using tallies on a piece of paper
that you keep. Then you won’t show the learner when you draw a tally, and only
announce the number of tallies at the end. Last, you can remove the entire token
system and not announce the number of tallies.
Group contingencies are commonly used in school settings. In a group contingency, a
common consequence, such as a reward, is delivered on the behavior of one member
of the group, the behavior of a part of the group, or the behavior of everyone in the
group. As such, they are classified as dependent, independent, and interdependent
group contingencies.
Because you can use group contingencies to address the behaviors of multiple students
at the same time, they are economical and easier to implement than individual
contingencies. Because the common reward is placed on the members of the group, it
is possible to capitalize on peer support and monitoring. For example, if all students in a
group need to complete homework before having a recess, students may help each
other. Relatedly, this may foster positive social interaction and behavioral support within
the group.
In a study that we conducted, we used the group contingency, specifically, the
interdependent group contingency to improve the reciprocal conversation between three
pairs of children with autism. Specifically, we targeted three components of the
reciprocity: answer the peer’s question, elaborate on their answer, and ask a follow-up
question. Children we selected were all able to demonstrate these individual
components upon instructional demands from their instructors. When we pair them
together and ask them to chat with each other, children did not ask questions and rarely
elaborated their answers.
We noticed that as soon as the group contingency was in effect, children were able to
meet the conversational requirement and continued to carry on their conversation
without our assistance.
In addition, we further noticed that the intervention had effects on their social
preference. During baseline, all children preferred to be by themselves more often.
Upon intervention, they preferred to be with their peers more often.
As the study did not include a one-on-one format, group contingency may be more
economical as the instructor only needs an overarching contingency that applies to all
target children’s behaviors. The group contingency seems to further improve the social
relationship between the members in the group as the group receives reinforcer.
However, it should be pointed out that we specifically arranged rewards that require
children to interact with each other. Thus, the exact mechanism responsible for
preference change in our study was unclear.
In an independent group contingency, the contingency is stated to the entire group, but
group members will receive the rewards individually as they meet the criterion. An
example would be students who complete the homework will have recess. In this case,
while the contingency is stated to the entire group, rewards are contingent on individual
student’s performance.
In a dependent group contingency, the reward is delivered for the entire group,
contingent on the behaviors of specific members in the group. An example would be “if
Mary completes the homework, everyone can go have recess.” Past studies have found
that the dependent group contingency capitalizes on student support as the peers are
more likely to help specific students, such as Mary in the example, so that the group can
receive the reward. Of course, you should also note that if Mary is not able to complete
the task, the entire group won’t receive the reward. This may create negative social
attention. When arranging dependent group contingency, make sure that the target
behaviors of the specific students are those they can perform.
The interdependent group contingencies, as presented earlier, are the ones in which all
members in a group need to meet the criterion before they have access to the reward.
Because all students need to work towards a common goal, this procedure capitalizes
on peer pressure and group cohesiveness and support.
There are a couple of variations to the procedure. First, you can randomize the behavior
or the target students. If you have multiple target behaviors and a large class to teach, it
may be difficult to keep track of all students’ behaviors. Using randomization, at the
beginning of a session, you can randomly determine the target behavior for this session
and only observe the randomly selected behavior during the session. If you have a
larger class, you can also randomly select a number of students to observe during the
session and deliver the reward based on these students’ behavior. Of course, you
should not let the students know the target student and behaviors if randomization is
incorporated.
In the total-group-meets-criterion variation, you would determine a criterion for the group
to earn reinforcers. For example, the entire group needs to earn 50 points in a task
before a reward is delivered. In this case, as long as the group members earn 50 points,
no matter how individual students have contributed to the task, the entire group will
receive the reward.
In a group averaging interdependent group contingency, the successive criteria are
based on the average of the previous performance. In other words, you will have
increasing criteria throughout the intervention. For example, based on average group
performance in the baseline, you can determine the initial criterion. One week later, you
would calculate the average performance again and determine the next criterion.
The Good Behavior Game is an interdependent group contingency that aims to reduce
disruptive and problem behavior. In the Good Behavior Game, students are divided into
two to three teams. The teacher will let the students know for each disruptive behavior,
the teacher will place a mark.
By the end of the session, the team with the fewest marks will receive the reward. The
Good Behavior Game is one of the most researched interdependent group contingency
variations, and its effects have been well-evidenced. In addition to its rapid effects on
problem behavior, studies have reported the Good Behavior Game can be applied to
students across different ages and have great social validity.
The Good Student Game is similar to the Good Behavior Game, except that the
students may play as individuals. Additionally, the students will monitor themselves and
record their behaviors instead of the teacher. Reward delivery is dependent on either
individual or group performance. The teacher, in this case, does not provide feedback
for each rule-breaking behavior, but feedback is provided at intervals, such as every five
minutes.
When you use a group contingency, make sure that you include an effective reward. In
the past literature, rewards have been tangibles and social activities. It is advised that
you use generalized conditioned reinforcers as the rewards. In previous literature,
teachers presented a list of different rewards and let the students vote for their reward.
As discussed earlier, group contingency procedures can improve target behaviors and
other collateral behaviors such as peer cooperation and positive interactions between
peers. As such, you need to determine what behaviors to monitor. You should have a
measure of the primary behavior of interest, such as problem behavior or academic
behavior. In addition, you should have a measure of a secondary behavior such as
social interaction. Having two measures will allow you to examine how group
contingencies affect various behaviors of multiple students in the same group.
Before you start group contingency, you need to set the performance criterion. You can
use average, high, or low-performance levels of the groups.
Average group performance is the mean performance of the group, and the reward is
delivered if the mean performance is achieved. Likewise, a low-performance level
means rewards are delivered if low-level performance is achieved. You need to
consider your student characteristics before selecting a criterion. Students who have
difficulty meeting your criterion may perform the worst under high-performance
contingency.
As with all other behavioral tactics, group contingencies may be combined with other
procedures, such as differential reinforcement of low rates of behavior, differential
reinforcement of high rates of behavior, as well as prompting procedures. Combining
with other procedures may further enhance the effects of the group contingencies.
Depending on your goal, you need to select the most appropriate group contingency.
For example, if you have one student in the class whose behavior should be changed,
perhaps dependent group contingency may be more appropriate. If you would like to
change the behaviors of a group of students, the interdependent group contingency
should be used.
Last, you need to monitor the performance of the students who receive the group
contingency at both individual and group levels. Just because your group-level
performance has improved, it does not mean the behaviors of all individual students
have improved. In this case, additional procedures may need to be arranged.
A contingency contract also called a behavioral contract, is a document that specifies a
contingent relationship between a target behavior and a specific reward. It has been
used in school, home, and clinical settings to increase various appropriate behaviors
and decrease problem behaviors.
In a contingency contract, there are three basic components, the task, the reward, and
the task record. For an example of the document, please go to the Cooper book and
turn to page 673.
As you can see in the page, in the task column, you need to state four elements: who,
what, when, and how well. Who is the individual who will perform the task and receive
the reward. What is the behavior or the task that the individual needs to perform. When
is the time that the behavior or task needs to be performed by. How well includes the
specifics of the task. If it is a behavioral chain, you will need to state the key
components. You also need to state the criterion. For example, what is the accuracy or
fluency level?
The reward also includes four elements: who, what, when, and how much. Again, who
is the person who will control the reward. What is the reward. When is the time that the
reward will be delivered, and how much is the amount of the reward.
As you can tell, the contingency contract uses specific language to let the individual
know the exact response and the exact reward.
Of course, you will need signatures and dates.
The last component is the task record. The task record includes when the behavior was
correctly completed and when the reward was delivered. It provides the occasion for
both parties to review the contract. In addition, if the learner needs to have multiple
completed tasks, you can include a token economy, such as stars, in the task record.
For example, the learner needs to complete homework for five consecutive days to
receive the reward. The task record will provide the venue for the token to be placed on.
Even though in contingency contracting, the reward is provided after the learner has
completed the task. You should note that the reward is often delayed. For example, the
reward can be provided by the parents after school, when the learner completes the
task at school. In addition, learner behavior may not be shaped by reinforcement.
Rather, the learner’s behavior is evoked by the instruction or the rule in the contract.
Therefore, the behavior is governed by the rules, as stated in the contract.
To develop a contingency contract, first, you need to hold a meeting with everyone
involved, such as family members, teachers, or classmates, depending on the context.
At the beginning of the meeting, you should provide an overview of the contingency
contracting and make it very clear to both parties that the contract is a behaviorexchange process that requires inputs from the learner and the parents or teachers.
Then, you will have the learner list a number of possible tasks that they can do for
themselves and others. Likewise, you will ask the learner to list possible rewards that
they like to receive.
The last step is to write the document and sign the contract. Make sure that you include
who will perform the task, what the task is, when and how well the task should be done,
as well as who will control the reward, what is the reward, when and how much the
reward will be delivered. Both parties need to sign and date the contract.
For contingency contracting to work, the learner does not need to be a reader, but they
must have some verbal skills. That is, they need to be able to respond to the verbal
stimuli in the contract. If they do not respond to verbal instructions, you will need to
teach the following instructions and rules first.
Additionally, some learners may not be compliant. They may refuse to sign the contract.
You can simply ignore the noncompliance and shape any attempts that approximate
signing behavior. For example, pay attention to the learner’s curiosity towards the
contract or any questions regarding the contract.
You can use the learner’s sibling or a peer as a model to show how contingency
contract works.
Parents could also develop a contract with the learner with reversed roles. The parents
will complete the task, and the learner will control the reward to demonstrate the
contracting.
Similarly, the learner may use a self-contract. That is, the learner themselves would be
both parties. Of course, this should be monitored by parents or teachers. But as long as
the learner can determine their own criterion of tasks and rewards under the guidance
of parents and teachers, self-contract may be worth trying for a noncompliant learner.
When developing a contingency contract, make sure that your contract is fair. The
reward should reflect the difficulty and the amount of work in the task. For example, one
piece of chocolate for completing homework on time for five consecutive days would not
be fair.
The contract should include specifics with all necessary details to make sure the
contract is clear.
Both parties, especially the party that controls the reward, should follow through the
contract. If the learner has completed the task, the reward must be delivered in the
manner according to the contract.
You can further build in a response cost contingency in case the tasks are not
completed. For example, if you require the learner to wash the dishes for three days
before the reward can be delivered, you can remove a token or a star from the task
record as a response cost contingency if the task is not completed.
When the contract is written, make sure that you post it somewhere visible to the
learner so that it may prompt the learner. It may be helpful to review the contract once a
day, at least initially. That should prompt the learner to complete the task and be in
contact with the reward.
If either party is not happy with the contract, renegotiation should happen.
Last, as with other behavioral programs, the contract should be terminated when the
intervention goal has been achieved. Alternatively, if either party consistently fails to
follow through, then the contract should be terminated, and another procedure should
be used.
Self-management is the application of behavior change tactics to produce desired
improvement in behavior by the individual themselves. As the name suggests, selfmanagement skills promote independence and free learners from having to rely on
others.
It usually includes a goal and a procedure specified by the learner, self-implementation
of the procedure and self-monitoring of the progress, and self-evaluating of the
outcome.
To use self-management, you will need to sit down and work with your student to
develop a list of appropriate behaviors that you and your student want to see, and a list
of inappropriate behaviors that should be decreased. You will discuss with your student
how to define these behaviors using measurable and observable terms.
Then, either you or your student will choose the target behavior to be managed by the
student themselves. You will also need to determine an appropriate goal with your
student.
Additionally, you will provide a menu of possible rewards that the student can choose
from.
Then you will develop a recording form. In the recording form, consider including the
name of the student, the date, the target behavior and goal, a task record that is similar
to the contingency contract, self-feedback where the student can indicate if they have
met the goal, as well as whether they should receive the reward.
Sometimes, you will also include a self-management device. For example, if the
student’s goal is to attend all advising meetings, you can include a device such as a
mobile phone to schedule the event and deliver notifications.
You can also include a timer as the self-management device if a preset interval is
included in the student’s self-management goal.
Then, you will need to teach the student to recognize the target behavior, to use the
self-monitoring recording form, and the device if you use one. The teaching procedure is
dependent on the student. Some students only require verbal instructions and rules.
Others may require explicit instruction using prompt and prompt fading procedures. For
example, you can designate a session in which you will conduct the self-management
training. You can teach the student how to use the self-management device. Then, ask
the student to purposefully demonstrate the target behavior and have the student record
down the behavior. Praise and reinforcer are delivered for correct recording. Teaching
should be conducted until all prompts, and verbal instructions have been faded and that
the student can notice their target behavior and use the self-monitoring recording form
and the device. If you allow the student to self-administer the reinforcer for appropriate
behavior, you should include this component during teaching as well. However, make
sure that the student does not just deliver the reinforcer without demonstrating the
appropriate behavior. Monitoring is important in the early stages when establishing the
self-management repertoire.
Last, you can further have the student graph their progress to evaluate the progress,
and you can also review the progress with the student together.
You will have access to Koegel, Koegel, and Park’s self-management training manual
for individuals with disabilities in Canvas. The training manual provides a step-by-step
guide for teaching self-management. Please note that the manual was published in
1990 and has not been updated; you should include updated technology and other
effective teaching procedures in your procedure when necessary.
Module 5 – Part II. Data-Based Decisions
In the following presentation, we will take a look at data-based decisions.
Specifically, we will discuss the following topics: monitor client progress and treatment
integrity, as well as make data-based decision about the intervention effectiveness, the
need for revision and ongoing services.
Treatment integrity is the degree to which an intervention is implemented as planned.
For example, in a following-direction task, the protocol may include providing an
instruction, a prompt, reinforcement or error correction for the learner. Any deviation
from the listed step is considered a treatment drift.
Previous research has shown that variations in treatment components may affect the
skill-acquisition rate. Carroll et al. in 2016 found a more delayed reinforcer, such as 10
seconds as compared to 3-5 seconds, led to increased response variability and slower
skill acquisition for children with autism. Roxburgh and Carbone in 2012 found a longer
intertrial interval, 10 seconds as compared to 1 or 5 seconds, was associated with a
higher rate of problem behavior during treatment. Thus, when the implementation of a
program deviates from the plan, positive effects may not be achieved. Results from a
comparison study by Carroll et al. in 2013 showed that low treatment integrity, 33% as
compared to 67% and 100%, led to a lower rate of student acquisition of expressive
identification and functional play skills and even no skill acquisition at all for some
students. Unfortunately, treatment drift is very common in practice.
Because we will focus on monitoring treatment integrity, we will not take an in-depth
look at how to train the practitioners in this module. That said, this is the first step to
facilitate the accurate implementation of the task – if the practitioner has not learned
how to implement the treatment, treatment drift will occur. If you would like to know
more about the training protocol, please refer to Parson et al.’s article on behavioral skill
training protocol uploaded in Canvas.
Initial training is necessary but not sufficient, unfortunately. Even when the practitioners
are trained to mastery, treatment drifts are still common later on.
To monitor, you need to develop a treatment checklist that covers all essential steps
included in the intervention. Ross et al. presented the Teacher Performance Rate and
Accuracy Scale for discrete-trial arrangements. The checklist listed basic discrete trial
components, namely, Teacher Antecedent and Teacher Consequence, as the steps.
For more complicated discrete-trial arrangements, such as conditional discrimination
trials, you can modify this checklist to include more information, such as the
presentation of stimuli and the steps to evoke an observational response from the
learner.
When a practitioner is new and just completed their behavioral skill training, your
observation and monitoring of the program implementation should be frequent initially,
for example, observe daily for an entire session using the treatment checklist. Of
course, continuous observation is not sustainable. Once the practitioner starts to
demonstrate mastery in treatment implementation, you can start reducing the
observation duration, for example, from the whole session to 10 minutes per session.
You can also reduce the frequency of your observations. In research, we must collect
treatment integrity for at least 20% of the sessions, that is, one in every five sessions.
You can opt to reduce your observation frequency to once a week for an intervention
that is implemented daily. Some research has suggested that once every two weeks
would also allow the practitioner to maintain an acceptable level of treatment integrity.
After you have used the treatment checklist to observe the implementation, you need to
provide affirmative and constructive feedback. Affirmative feedback serves as verbal
praise for correctly performed components, while constructive feedback is the error
correction for incorrect components. Feedback has been well-researched and found to
improve teachers’ treatment integrity across settings and various procedures. It can also
be conducted delayed or immediately after the observation – both are effective. To
bridge the time and spatial gap in case the teacher does not remember the session, it
may be beneficial to video record the session and provide feedback while playing the
video.
We, as behavior analysts, rely on data to make decisions regarding any intervention.
There are different ways to collect data. You can collect continuous data, collect only
probe data before the daily intervention, or collect probe data after the daily intervention.
Regardless of the data collection, the data must be collected continuously on a daily
basis to monitor student progress.
To derive meaningful and accurate presentation of student’s progress, please make
sure that you collect at least three data points per phase. That means you will need to
collect at least three data points during baseline, intervention, and when you make any
changes to the intervention. Each change is a phase.
Three data points are required to demonstrate trend – is the student progressing or
regressing, variability – how variable or stable is the performance, and level – the
average student performance. One single data point may be an outlier and may not be
representative of the student’s current skill level. A single data point also does not
provide information regarding the trend or variability. Therefore, you need to ensure a
minimum of three data points.
There is, however, an exception to this rule. That is, when the client is demonstrating
severe problem behavior, one data point may be used to avoid potential harm.
However, you should note the limitation of the single data point in this case.
In clinical practice, we would analyze the student data using trend, variability, and level.
I will provide a brief description of each of them. If you have not already, you will learn
more in the experimental design course. A tutorial by Lane and Gast on how to conduct
visual analysis is also provided in Canvas.
A trend is the overall tendency of the data. In other words, is the response increasing or
decreasing? In the top figure, let’s assume A is the baseline, and B is the intervention.
As you see, the student performance decreased during the baseline and increased
during the intervention.
In the bottom figure, the level or the average performance, as demonstrated using a
straight solid line, is higher during intervention than during baseline. Lane and Gast
used a concept called stability envelope to determine variability. Their stability criterion
is 80% of the data points within 25% of the median in both directions. This envelope is
demonstrated using dashed lines.
To make decisions, you need to have your intervention goal. For example, in an identity
matching task, we can state the learner should independently match the same items
together for a minimum of 90% of trials across three sessions.
If the performance has met your intervention goal and minimal variability across the
required sessions, you can move onto the next phase of the instruction, such as
thinning the reinforcer and generalization training. You can also terminate the program if
the mastery criterion has been met.
If the performance is moving towards the goal, you should not make any changes, as it
seems that the intervention may be effective.
However, if the trend is relatively gradual and the improvement is not fast enough, or if
the data are variable, you can either wait for a couple of more sessions to see if
performance has been improved or stabilized. Otherwise, you can consider providing
additional intervention components to boost the effects. For example, using DRO to
decrease a behavior may be slow, you can add an NCR procedure to decrease the
problem behavior more quickly.
If the performance is not moving towards the goal, for example, the performance is still
at the baseline level. Or the performance had improved for a while, but the improvement
has stopped, your intervention may not be effective, and intervention should be revised
in this case.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
