Module ID# 52
Type: autism
Revised: 07/2007

Module Description - Instructional Outcomes - Instructional Events
Instructional Evaluation - Discussion List

Applied Behavioral Analysis

The increasing incidence of autism spectrum disorders has focused much attention on questioning which practices are evidence-based and effective. One of the strategies that has gained much popularity because of the evidence to support it is the use of Applied Behavioral Analysis (ABA). While ABA has been used with children on the autism spectrum for only about 30 years, ABA has been used with other populations and has more than 50 years of research.

Before we proceed, let’s clear up one confusion. The terms Discrete Trial Teaching and ABA are often used interchangeably. They are not the same. DTT is a strategy associated with ABA, but does not embrace all of the strategies associated with ABA. DTT involves teaching skills using multiple trials with a specific beginning and ending. So, for example, a child learning his/her colors would be asked (antecedent) to point to the green card. If the child correctly pointed (behavior), he/she would be rewarded (consequence). If not, then he/she would be corrected (consequence). There is a pause between trials and then the activity is repeated. Multiple trials are used. It is an effective way to teach rote skills such pointing to colors, specific items, or words. However, ABA is more than just DTT. For a description of Discrete Trial Teaching, visit the website at http://www.iidc.indiana.edu/irca/behavior/discretetrl.html.

Now let’s walk through the steps of ABA include defining the behavior, determining the function, identifying goals, teaching alternative behaviors, measuring outcomes, and programming for generalization. This module is not all inclusive of the strategies of ABA. Our interest is in providing a snap shot into this approach.

To facilitate this process, let’s use two case studies:

• Matthew

• Sara
  
  

Defining the Behavior

ABA is an effective technology that can address problematic behavior and enhance skill development. It is a systematic approach to teaching and addressing behavioral excesses and deficits. It is referred to as ABA, because the goal is for the approach to be A (applied and used for practical purposes), to focus on B (a behavior) and to engage in A (analysis or study of the individual parts to determined their relationship to the whole).

The first step is to operationally define the behavior you are focused on. Terms such as the child is angry or non-compliant are subjective. In other words, each person may view that behavior differently. However, phrases such as the student is spitting at others, or screaming are clear. All can agree when they observe these behaviors. For Sara, stating that she engages in inappropriate social behaviors is insufficient. Instead one would state that she stands too close to others, or it would be beneficial to make a list of the specific inappropriate statements that she makes to classmates. Based on the case study above, what are the behaviors exhibited by Matthew? Remember that they must be operationally defined.

Once you have operationally defined the behavior, it is helpful to gather more specific information about the behavior. You can determine this by looking at the rate (the number of occurrences), the duration (length of time the behavior occurs), the latency (length of time between episodes), topography (what the behavior looks like), force (intensity of the behavior), and locus (where the behavior occurs). You may want to take data on various dimensions of a behavior. For example, it may be helpful to know that the behavior intensifies in various settings. For Sara it would be helpful to know that on the playground she tends to make 20 bizarre statements to classmates and in the classroom only 2 bizarre statements are typically made. This means that we have to increase the support and training on the playground.

When conducting research, data is typically taken before we intervene with the behavior, during the intervention, and then after the intervention. In schools, it will be helpful to gather data on an ongoing basis to continually determine the impact of behavior programming. For example, if the focus is on decreasing spitting behavior, it may be helpful to develop a system for gathering this information daily to assist with program monitoring and implementation. Data collection can take many forms, including anecdotal records that report in a narrative format observations of the behavior. The website on teaching children with autism at http://www.polyxo.com/documents/#data has various forms that can be used in the data collection process.

Determining Function

Using an ABA approach, professionals and family members can use an ABC analysis to clarify details surrounding the behaviors. An ABC analysis can be one component of a complete Functional Behavioral Analysis. In the ABC analysis, A is for Antecedent, B is for Behavior, and C is for Consequence. So we may ask the following questions for each of these components:

A = Antecedent
What happens immediately before the behavior occurs?
Who is present?
What activity is occurring?

B = Behavior
Operationally describe the behavior. Does he hit, kick, strike out at others, or what is the behavior that we can all see and reach consensus on?

C = Consequences
What happens after the behavior?
How do other staff and students react?
Is he or she verbally reprimanded?

By charting out the ABCs of the behavior, those involved can begin to determine the events that may increase or decrease the likelihood of problematic behavior and to determine consequences that enhance or diminish behavior. For the purposes of this module, we will focus on teaching new behaviors based on data collected. Below is an example of an ABC analysis.

Antecedent	Behavior	Consequence
Teacher asks Matthew to stop pushing classmates.	Matthew continues.	Teacher again asks Matthew to stop pushing classmates.
Teacher asks Matthew to stop pushing on the classmates.	Matthew ignores teacher requests.	Teacher asks Matthew to stop pushing classmates…but this time louder.
Teacher asks Matthew to stop pushing classmates…but this time louder.	Matthew ignores and continues.	Teacher threatens Matthew with losing recess in the future.
Teacher threatens Matthew with losing recess in the future.	Matthew ignores.	Teacher screams louder and tells Matthew to stop immediately,
Teacher screams louder and tells Matthew to stop immediately,	Matthew ignores.	The bell rings signaling it is time to go in.  The teacher directs all to go into the building.
The bell rings signaling it is time to go in.  The teacher directs all to go into the building.	Matthew goes into the building.	The teacher does not mention the behavior to Matthew.

 

Looking at this ABC analysis, one can see that Matthew has learned that his teacher’s threats are empty threats. However, if he is a young man who loves attention…his behavior does get the teachers attention. Remember that students often do not care if this attention is negative or positive.

Identifying Goals

Once information has been gathered about the behavior, professionals and family members should collaborate on the selection of goals. Some ABA programs have very specific curriculum attached. However, remember that the technology of ABA can be used to teach anything. Goals that are chosen should follow four standards:

1. Goals should be written in objective and measurable terms. Stating that you want a child to behave is not measurable or objectively stated. However, stating that you want a child to follow two directions is measurable and observable. In Sara’s case you would state that you want her to brush her teeth without prompts or to greet two students each morning by saying, “Good Morning.”
2. Goals, if met, should improve the student’s quality of life.
3. Goals should be reasonable and attainable. For example, expecting Sara to play 100% of the time with another student when she currently only spends 2 minutes of the entire recess with others is unrealistic.
4. Goals should focus on helping students develop prosocial behavior, not merely suppress negative behavior. So, for example, rather then focusing on decreasing Matthew’s aggression on the playground, a goal for him should be focused on teaching him how to play with others. For example:
             During recess, Matthew will play a ball game with one other student for 5 minutes daily.
             To gain a students attention at recess, Matthew will use a hand sign or picture card twice daily.
   

Teaching New Behaviors

There is an array of technology associated with ABA that is used to teach new skills. Below just a few are highlighted:

Task analysis involves breaking down a task into the precise steps involved in learning the task. The size of each step will depend on the child’s ability to learn new tasks. One individual may require 15 steps to learn a task, while the next child will only need 7. For example, it may be helpful to create a task analysis to teach Sara the steps of brushing her teeth. Task analysis is used to teach skills through forward and backward chaining. An analysis for hand washing may include:

• Turn on Water
• Wet Hands
• Pick Up Soap
• Lather Hands
• Replace Soap
• Rub Hands Together
• Rinse Hands
• Turn Off Water
• Get Towel
• Dry Hands
• Put Towel Back

Visit Sandbox Learning at http://www.sandbox-learning.com/Default.asp?Page=33 for a form to use when designing a task analysis.

Chaining is the linking of component behavior into a more complex behavior. Chaining is especially useful for teaching self-help skills. So, in the example above, we would teach each step in sequence until all steps are achieved. In backward chaining, we would teach the last step first to give the student a sense of completion. Visit the website at http://www.polyxo.com/documents/#data  for a data collection form to use for documenting progress using task analysis and chaining.

Shaping refers to reinforcing successive approximations toward the desired behavior. In schools, for example, we would reinforce students for progressively improving their handwriting. We would not wait to reinforce for perfection, but would reinforce progressively more accurate efforts.

Prompting is cueing to guide the student in performing each step of the task. Prompts are the cues that indicate what should happen next. There are various types of prompts:

• Physical Prompts: Hand over hand prompting or physically manipulating the person’s body to do the task.
• Gestures: Prompts may be an additional visual hand signal of what the student is expected to do. For example, pointing toward the lights that you want them to turn on.
• Modeling: Prompts to simply show the learner what you want them to do. The student must be able to imitate in order to learn from a model prompt.
• Verbal: These prompts are most frequently used. Caution should be used to not overload the student with verbalizations. Keep verbal prompts specific and to the point. What exactly is the student expected to do? Care must be taken to not inadvertently cause students to become dependent on the verbal prompt. This could create a dependency that is very difficult to change. As soon as possible use visual prompts or gestures.
• Environmental: Prompts that are objects, like the carton of milk or box of crackers on the counter that are used to signal snack time.

The prompt should occur prior to the student reaching a point of frustration. The less frustration and difficulty the student has, the greater the rate of learning. The more times an activity is completed accurately, the greater the likelihood that the student will repeat it each time with fewer errors.

There are several factors involved in determining appropriate prompts and the sequence of prompts. Balancing these factors will differ with each child.

• The strategy for using prompts and the type of prompt used should be chosen based on the students’ skills and learning style. Prompts can be used in any order and in combination. It is dependent on the student and on observations of how the student learns. When a hierarchy of prompts has been determined, it is essential to follow through. If there is no response to the verbal prompt, or if the student begins to respond incorrectly, try a different and/or more intrusive prompt.
• For some, decrease assistance (most to least). Use the strongest prompt and move to a weaker one as the student masters the skill. If the student needs complete physical assistance to complete the step, the prompt would be used for a specified time span. If that does not produce success it is time to go to the next level of prompting. The idea is to ensure the student performs the skill correctly upfront.
• Make the prompt as weak as possible. The prompts should not distract the student from the task/activity that you ultimately want to have the student attend to. Be sure that the task/activity is clearly presented. You would not use physical guidance if a verbal redirect will accomplish the same thing. Using too strong of a prompt may delay the student’s ability to attend to the task/activity independently. The student can quickly become dependent upon the prompt itself, and be inattentive to the task/activity.
• Fade the prompt as quickly as possible. Whatever hierarchy of prompting is selected, a plan to fade the prompt should be developed as well. It is easy to continue prompting when it is no longer necessary. Another hazard is to not give the student an opportunity to try without the prompts. One strategy to avoid this is to build in a probe trial on a regular basis. A probe is giving the student the opportunity to complete the task without any prompting. Prompt dependency results from providing prompts when they are no longer necessary, or never planning how to fade your prompts.
• When the teaching program is developed, an essential decision is to determine how many attempts a student will make, at what level of prompting or how many errors before the task is stopped. There may be times when presenting the entire task, and prompting at the necessary level for each step is appropriate. The amount of time a student will require to process information is unique to the student. If too much time is given, the student will have the opportunity to make an error. However, if prompting is too quick, the student will not have the opportunity to become independent in completing a task. This requires ongoing judgment.

Time Delay can be used to determine if the student still requires a prompt. This simply requires the teacher to hesitate or wait a period of time before delivering the reinforcer.

Reinforcement is the response that occurs as a result of completing the task or certain portions of the task. Below are some guidelines for using reinforcement:

Reinforcers should be reinforcing. Care should be taken to choose reinforcements that are reinforcing to the student. If it is not reinforcing for the individual, it will not increase desired behavior. Reinforcers will vary depending on what your student finds reinforcing. If reinforcement is going to be effective, the student must be the one that finds it reinforcing. It is critical to continually monitor whether an item is reinforcing. The value of reinforcers can change rapidly. A variety of reinforcers should be used. By using a variety of reinforcers, there is less chance the individual will become satiated by the reinforcement. It will keep reinforcers fresh and more powerful.

Reinforcement should be contingent. Reinforcement should only be available when the target behavior occurs. If the student is rewarded without completing or if reinforcements are used at other times, the reinforcement loses its effectiveness. Try to select rewards that can be reserved for contingent rewards. Do not select a reinforcer if it cannot be withheld or if being withheld will create tremendous problems (e.g., foods).

Social reinforcement should be paired with primary reinforcement. Primary reinforcers (e.g., food, liquids, shelter, sleep) should be paired with social reinforcers, such as smiles and praise. The individual will associate them with the primary reinforcers and they will eventually become reinforcing as well. Development of social rewards allows the intermixing of social and primary rewards and ultimately should lead to fading primary rewards. Social reinforcers are the principle rewards used in everyday settings.

Use age-appropriate reinforcers. Age-appropriate reinforcers will increase peer acceptance and will be less evident and obvious. It will also help promote generalization when the child encounters the reinforcers in the environment.
Be Specific. The student should clearly understand what behavior is being rewarded. Verbal responses should state the action (e.g., Great, you turned on the water!). Do not simply say “great.”
Reinforcement should be faded over time. When a reinforcement plan is established, when and how to fade the reinforcement should be a part of the plan. The ultimate goal is for the student to complete the task, and for the reinforcement to be completion of the task and the natural satisfaction. Reinforcers that are directly related to the task will usually increase the student’s success with learning the task.

Reinforcement should be provided on a planned schedule. Reinforcement may be provided after completion of each step in a task, a predetermined number of steps, or at the end of the task. Varying the rate of reinforcement, or using intermittent reinforcement, is one of the most powerful tools for maintaining behavior. Continuous reinforcement is when reinforcement is provided for every attempt to complete the task or a portion of the task. Continuous reinforcement is powerful when teaching a new skill.

Again these are just a few examples of the technology associated with ABA. Strategies such as pivotal response training, incidental teaching, errorless teaching and others are also looked at within the ABA technology. Below are websites that describe these approaches.

Pivotal Response Training

How to Teach Pivotal Behaviors to Children with Autism: A Training Manual

USCD Autism Research

UCSB Koegel Autism Center 

Incidental Teaching

Project Spies for Parents

ICAN

Burkhart Project in Autism Education

Errorless Teaching

Patrick Mc Greevy

Documentation provides information about the progress that is being made. As mentioned earlier, initially a baseline needs to be established. This will indicate the skill/knowledge level of your student and guide the development of the plan. Careful observation of the student provides needed information in order to plan how to effectively teach. This documentation should provide insight into how detailed to make the instruction, and whether or not you need a task analysis.

A probe or other data should be gathered at specific intervals in the instructional plan. When assessing a student’s learning on a task, no prompting should be provided. Or if prompting is used, this should be documented in the data collection. This will help in assessing the effectiveness of the teaching plan and the rate of student learning. This documentation provides accountability, but more important, it assists in making ongoing instructional decisions.

The nature of the information needed will determine how frequently data gathering is required. What is written in the IEP will guide documentation. Information on the student’s rate of learning, as well as the frequency or duration of a behavior can be documented. Latency, or the period of time period between a prompt being given and when a student complies with the direction can be useful data to document.

After data has been gathered, it must be analyzed so that appropriate decisions regarding the success of the intervention can be determined. One of the simplest methods to analyze data is to put it into a chart. This can be done using a computer program like Excel. Data may also be manually charted on graphing paper. Once the data has been charted a visual inspection of the data can be made. Although this may seem rather crude, it shows changes that are significant. For instance when the data shows that the number of unprompted steps has gone from 1 to 7, it is likely that learning is occurring.

A visual inspection of the data can also show trends in student behavior. For instance, the data could show if a student is mastering more steps of the task analysis or using decreasing prompt levels. The data does not necessarily need to be taken each time the task is performed in order to be effective. The purpose of documentation is to provide sufficient information to allow decisions to be made regarding the student’s success with a task. That doesn’t mean data needs to be collected each time the task is completed. Timed math tests of addition may be done once a week, that is frequent enough to show change in student skills.

There are various forms that can be used in this process. See the following websites for examples of data collection forms:

ABA Educational Resources: http://www.abaresources.com/free.htm

Polyxo: http://www.polyxo.com/documents/#data 

Data gathering can answer questions such as:

• When is the student performing the task?
• Are there sections of the task that are improving?
• Are there steps that show no change?
• How often does data need to be taken and analyzed?
• What period of time should be given before making changes in instructional programming or in goals/objectives?

Generalization. Many students, and frequently those with autism, are not able to take a task that is learned in one context and use it in another environment.
In the beginning, teaching strategies should be very systematic. However, once a student is demonstrating proficiency on a task, instruction should be provided across settings. Students should be provided planned opportunities to demonstrate learned skills across instructors and using various types of equipment. For example, if a student is taught the color orange using a color card, then he/she should be taught about items in the room that are also orange.

For more information about the technology of applied behavioral analysis, read from the following resources:

Alberto, P.A. & Troutman, A.C. (1999). Applied behavior analysis for teachers (Fifth ed.). Upper Saddle River, NJ: Prentice Hall.

Donnellan, A.M. (1988). Progress without punishment: Effective approaches for learners with behavior problems. New York: Teacher’s College Press.

Leaf, R. & McEachin, J. (Eds.) (1999). A Work in Progress: Behavioral Management Strategies and a Curriculum for Intensive Behavioral Treatment of Autism. New York: Different Roads to Learning.

Schloss, P.J. & Smith, M.A. (1994). Applied behavior analysis in the classroom. Needham Heights, MA: Allyn and Bacon.

Wolery, M., Bailey, Jr., D.B., and Sugai, G.M. (1988). Effective teaching: Principles and procedures of applied behavior analysis with exceptional students. Needham, MA: Allyn and Bacon.