REGISTERED BEHAVIORAL TECHNICIAN (RBT®) Task List

The copyrighted RBT© task list is reproduced here with permission from the BACB©

Study guide materials are copyright of BAS

A. Measurement A­01 Prepare for data collection

Cumulative record of behaviors that have occurred over time (CUMULATIVE RECORD (GRAPHS) NEVER DECREASE). Flat line indicates period of no performance of the behavior, steep slope indicates more incidents of target behavior.

Say, for example, that you wanted to continuously graph the number of screams that a child emits during sessions. Let’s also say that the scream is well defined. A cumulative or continuous graph would look like this:

Notice that the graph never goes down. This is because each new behavior (scream) simply adds to the ones before it. In the first session, there was 1 scream. In the second session there were 2 more screams, for a total of 3, and so on. During session 4 & 5 there was no screaming behavior. This is why the graph is flat for these days. Cumulative graphs are a good visual way to see the rate of increase (the steepness of the upcurve) or the slowing/stopping of a behavior (relatively flat or flat line).

A­02 Implement continuous measurement procedures (e.g., frequency, duration).

Continuous measurement procedures involve all instances of a response class being detected. This will be contrasted with “discontinuous” measures later. For continuous measurement, simply remember that an observer watches and records every instance of the behavior, and that these measures are thought to be the most accurate.

This includes narrative event recording, ABC event recording and any direct observation (video recording would be included).

One very common type of event recording is ABC event recording. This stands for Antecedent, Behavior and Consequence. That is, the conditions before the behavior, the behavior itself and the consequence of that behavior. Below is an ABC chart.

Antecedent Behavior Consequence

With this simple chart, the observer simply writes down the target behavior in the middle column whenever it occurs. Immediately after this, write down what was happening before the behavior. Then, write down what happened as a consequence of that behavior. Antecedent: Who is present? What was the activity? Did someone interact with the client? Was it quiet or loud? Basically, what was happening? Behavior: This is the predetermined behavior that you have been tasked to look for. It could be “on­task” behavior, smiling, answering questions. It could also be kicking, screaming, tapping a pencil, getting out of one’s seat, etc. Consequence: The change in the environment as a result of the behavior. That is, what’s different now? Did a teacher give the student an evil eye? Did they talk to or otherwise engage the student? Did the client avoid the activity? Did the student stop the behavior or do it more?

Did anyone notice the behavior, or did the client do it without looking to anyone else? Was the behavior ignored? This “running record” continues through time as an event occurs, people behave and there are consequences. Then, it all starts again. Below is an example of a somewhat completed chart.

Antecedent Behavior Consequence

1) Lights flicker 2) Teacher prompts to get in seat and open book to p58 3) Teacher says, “No whining”

­[client] gasps ­ Says, “No, not math.” Whines loudly for 3 secs. ­Whines louder for 5 seconds

­Teacher says, “It’s fine” and lights stay on ­Teacher says “no whining” ­Teacher interrupts whining and says, “you need to step outside” Client does so.

We can’t tell too much from this limited sample. Perhaps the student avoids math by engaging in this behavior. If this is true, we would expect to see this pattern continue. The point of this example, however, is more to show how this record flows from before a behavior, recording the behavior and then to determining relevant consequences. After observing for a longer period of time, patterns may emerge. The major patterns to look for are (1) what antecedents are followed by the target behavior, and (2) is the consequence of behavior consistent in terms of avoiding something or gaining something.

Measure frequency (i.e., count).

Definition: A direct measure. Count.

Example: Counting the number of cartwheels, steps or words spoken. Think, “How many instances occurred?” or “How many times did that happen?”

This is a very common behavioral dimension to measure. It sounds easy to count how many times a behavior occurs, and it usually is. Any difficulty usually arises from a lack of clarity in the target behavior definition. Imagine that you are asked to count how many target behaviors occur within one hour, but are told to record “disruptions or off­task behavior.” You might record something totally different than someone else. If the target behavior were, instead, defined as “the hitting of a pencil onto another object with enough force to be audible within 25 feet”, you might get more accurate results.

Here’s a simple chart of frequency. You only need to know what behavior you are observing and to count the behaviors. In the chart below, you see that “licking other’s hands” occurred 8 times.

FREQUENCY CHART

Behavior: Licking other’s hands

You would be asked to use frequency only when you consistently observe for a set amount of time (e.g., 1 hour per day, each school day, 2 hours per week). Usually, a time interval is predetermined. As you’ll see in the next section, a total of behavior per time period is known as rate.

Measure rate (i.e., count per unit of time).

Definition: Ratio of count per observation time (often expressed as per minute, per session, etc.); a direct measure. This is really just frequency with the addition of a time period. In fact, Cooper, Heron and Heward (2007) use the terms frequency and rate interchangeably. The ratio is the division of frequency/time period, such as 10 pointing behaviors/hour.

Example: Counting the number of cartwheels per minute, steps per day or words spoken during a class period. Think, “Frequency within a defined amount of time.”

Utility:

A scenario might mention frequency over a time period ­ this is rate. Also, rate is best used for behaviors of short duration and with easily identified

behaviors. That is, observers with less training may still be able to record an accurate rate.

You may be asked which dimension of behavior is most appropriate in a given situation. Rate would is most appropriate when the post­intervention goal may be measured in behaviors per time period (e.g., requests per day or correct answers per class period).

Do not use rate when the behaviors occur almost continuously or for long periods of time in each occurrence.

Most charts of frequency are going to include a time period. Below is an example, based on the chart above.

RATE CHART

Behavior: Licking other’s hands

Hour 1 Hour 2 Hour 3 Hour 4

0 0

Now, we not only know that the behavior occurred 8 times, but that it occurred 8 times in 4 hours or twice per hour. In addition, one can see that the behavior occurred in an inconsistent pattern. Hours 1 & 3 include the occurrence of the behavior, while hours 2 & 4 did not. In collecting this data and working with your supervisor, you might determine that there is something different about the environmental conditions in hours 1 and 3. You may be asked to take more data to further investigate the reasons for the differences in behavior.

Application:

1. Whatever rate chart you may be given, it will still include the number of behaviors in each time period.

2. The total behaviors per time period can then be used to determine the overall rate per hour, day, week, etc.

3. Different intervals will likely be used to fit the situation, such as tracking the target behavior occurrence every 10 minutes, 1 hour or per week.

Below are some other examples of more complex rate charts that you may be asked to complete.

http://www.escambia.k12.fl.us/pbis/data/OPM/ACTIVITY%20FREQUENCY­RATE%20DATA%20SHEET.pdf

http://www.escambia.k12.fl.us/pbis/data/OPM/30%20MINUTE%20%20FREQUENCY­RATE%20DATA%20SHEET.pdf

http://www.escambia.k12.fl.us/pbis/data/OPM/DAILY%20FREQUENCY­RATE%20DATA%20SHEET.pdf

Measure duration.

Definition: Total extent of time that a behavior occurs. Direct; (continuous) measurement. Also known as temporal extent.

Examples: Total time of each tantrum, time it takes to chew each bite, or time spent scrubbing hands when washing.

Note: When measuring duration, to avoid confusion, define the behavior in simple terms but well enough to be observed and understood by other people. If you see an instance of a target behavior and it ends, you need to know how long to wait before timing a new occurrence. For instance, if someone clears their throat for 10 seconds, pauses for 1 second, clears for another 10 seconds, pauses for 2 seconds and clears again for 22 seconds; is this one behavior of a 45 second duration or three separate behaviors? The answer is in how you define the target behavior. If you say that any instance of throat­clearing separated by more than 5 seconds is a separate instance, then the above example would be a target behavior duration of 45 seconds. If, instead, you say that a pause of 1 second or more indicates a separate occurrence, then the above example would be three separate instances of the target behavior (10 seconds, 10 seconds and 22 seconds). If you followed this example, you might think that it seems a bit like splitting hairs.

The definition of a behavior seems like a small thing . This is true, and it still matters. Why it matters is not so much about the definition, but what it represents. You, your client, your supervisor, your client’s parents, insurance companies and anyone else involved, want to know if the behavior is improving or not. In order to show whether this is happening, we need to have a consistent and accurate view of what we mean when we say that the “behavior” is improving.

Here’s another example:

Take the example of “nose picking” behavior. If you are asked to take an hour and record the duration of each occurrence you might record:

Target behavior: Inserting any digit into the nasal cavity. Any occurrence is considered at least one second.

1st: 35 seconds

2nd: 26 second

3rd: 24 seconds

4th: 65 seconds

5th: 30 seconds

Total Duration: 180 seconds

Rate: 3 minutes per hour

If you were asked to report total duration, you would add all occurrences together and come up with (35+26+14+65+30) 180 seconds. Since you observed for an hour, this could also be reported as a rate of “nose picking” per hour (about 3 minutes per hour).

Utility:

Duration is a good measure to use for behaviors that last for a longer period of time. Useful for behaviors that are targeted for an increase or decrease in temporal extent. Use for behaviors that do not start and stop frequently ­ this would make timing the

behavior difficult.

Application:

1. Tracking sheets will include the target behavior being tracked 2. Duration charting sheets will include a place to indicate the total observation period (e.g.,

from 10:00 a.m. until 3 p.m.) 3. Duration charting sheets will include a space to track the duration of each behavior

occurrence and will often calculate a total duration for the hour, day, week, etc. 4. Tracking sheets may also ask for a calculation of the percentage of time that the

behavior occurred. For instance, if you observe for an hour and the target behavior occurred for a total of 15 minutes, that is 25% of the time that you observed.

Here are some examples of tracking sheets:

http://www.escambia.k12.fl.us/pbis/data/OPM/Duration%20Data%20­%20Version%201.pdf

http://www.positivelyautism.com/RecordingSheet_Duration.pdf

You will likely be given tracking sheets to use, but could even design your own. Again, the key is to note the target behavior, the observation period and the duration of each behavior. The rest is up to you and your supervisor.

Measure latency.

Definition: Time from onset of a stimulus to initiation of the response. Another way to think about this is: how long it takes for someone to respond to a cue, prompt, question, etc.

Example: A teacher calls out a student’s name. The student orients to the teacher 8 seconds later. The behavior/response occurred 8 seconds after the stimulus.

During a feeding program, you want to know how long it takes for the student to accept the spoon of food to touch his/her lips. From the presentation of the food, you would be calculating latency: the time from the presentation of the spoon with food on it to the time the client oriented toward the spoon and it touches the lip. Another example might be measuring the time from the spoon touching the lips to the opening of a client’s mouth.

Utility: You will see latency used when one wants to implement interventions to increase or decrease the time between a cue and the beginning of a behavior.

Application:

1. Tracking sheets will include the target behavior 2. These sheets will also need to include what the beginning of the target behavior is (for

example, the client looks toward you) 3. They will also include the “cue” that timing needs to start (e.g., a direction is given) 4. You will need to record the time that it takes from the cue to the beginning of the target

behavior

Here’s a nice example form: http://gulfportschools.org/cms/lib07/MS01910520/Centricity/Domain/80/BEHAVIOR%20TRACKING%20LATECNY%20FORM.pdf

Measure interresponse time (IRT).

Definition: The time between the end of one response and the beginning of another (same) response.

Example: A child hits his or her forehead with an open hand with only 2­10 seconds in between, and then hits his/her forehead again.

IRT is like duration in that it measures time, although IRT measures the time between behavior rather than the duration of the behavior itself.

Utility: Increasing the IRT means lowering the overall rate of behavior; decreasing IRT means increasing the rate of behavior. This concept may be useful for slowing or speeding up target behavior rates by manipulating IRT.

Example 1: If a student blurts out with a similar sound ­ that draws attention and disrupts other students ­ every few minutes, an intervention might focus on increasing the IRT. If successful, the student would have longer periods between blurting out. This has the effect of decreasing the overall rate of behavior.

Example 2: If a student blinks every two minutes, an intervention may want to increase the rate of this behavior. This would mean decreasing the time between blinks (IRT) to a more normal rate.

Application:

1. Tracking sheets will need to indicate the target behavior 2. You’ll need a definition of behavior that includes the end of the behavior and what’s

constitutes the beginning of the behavior. 3. Time from the end of one behavior until the beginning of the next. 4. You’ll have to wait until the target behavior occurs and ends, before you can begin

timing. 5. IRT will usually be expressed as a rate. That is, the sum of all the IRT data divided by

the number of behaviors recorded. For example, if IRT data show 2, 8, 3, 4, 6 & 7 seconds, this is a IRT rate of 5 seconds ((2+8+3+4+6+7)/5 observations) = 5 seconds per occurrence.

A­03 Implement discontinuous measurement procedures (e.g., partial & whole interval, momentary time sampling).

Discontinuous measurement­ Some instances of the behavior may not be detected. That is, you aren’t observing every occurrence of the behavior. This is practical, as you can imagine the difficulties in observing someone for all of their waking hours. Using discontinuous measurement simply means that you are estimating the actual occurrence of the behavior, rather than recording every instance. The three ways of recording discontinuous measurement are discussed below ­ Partial interval, whole interval and momentary time sampling (MTS).

Direct measurement­ Behavior is actually observed; it is the same behavior that is the focus of the data collection. Contrast this with indirect measures such as interviews or questionnaires.

Partial Interval­ Discontinuous, direct time sampling procedure where a behavior is recorded as occurring, if it occurs at least once during the predetermined interval. May overestimate occurrence of the target behavior. Findings are reported as a percentage or rate.

If you are asked to record partial interval data, you would watch to see if the target behavior occurred at any time during that interval. If it did, you mark that interval. If it did not, you don’t.

Whole interval­ Discontinuous, direct time sampling method. A behavior is recorded as occurring if it is observed throughout an entire predetermined interval. May underestimate occurrence of target behavior.

When asked to record whole interval data, you would watch to see if the target behavior occurred for the entire or whole interval. If it did, you would record the behavior as having occurred in that interval. If it did not occur for the entire interval, you would not.

Momentary time sampling­ Discontinuous, direct data sampling procedure when (at the end of a predetermined interval) a recording is made if the target behavior is occurring AT THAT MOMENT. May underestimate occurrence of behavior. Findings are reported as a percentage or rate.

Asked to record MTS data, you would watch to see if the target behavior occurred at the end of the interval. If the behavior is occurring at that moment, you would mark that interval. If it was not occurring at that moment, then you would not.

Example: Let’s say that a client blurts out movie one­liners during class such as, “You can’t handle the truth” or “I’ll be back”. You want to use a discontinuous measurement to record the behavior’s occurrence. One behavior occurrence lasts about 3­10 seconds. Let’s look at each measurement tool separately, using 15­second recording intervals:

Partial Interval ­ You would record the behavior as having occurred when the “one­liners” happen during any part of a fifteen second interval. You would probably record most behaviors, but would miss any time that a behavior occurs more than once during the interval. You would record a behavior twice if it started at the end of one interval and “spilled over” into the next interval. This is how partial­interval tends to overestimate occurrence.

Whole interval ­ You’d record the “one­liners” if they happened for an entire 15­second interval. If the behavior occurred for 14 of the 15 seconds, you would not record the behavior as having occurred. This is how whole­interval tends to underestimate.

Momentary Time Sampling (MTS) ­ Record “one liners” if they are occurring in the last moment of the 15­second interval. If the behavior is not occurring at this time, it is not recorded. MTS can over, or underestimate. In this case, it would probably underestimate the occurrence.

Take away: All of these discontinuous measurements may produce artifacts, in which you are being inaccurate in your measurement. Direct measurements, in contrast, capture all of the instances of the behavior. Which discontinuous measurement is used will be up to your supervisor, but you need to know which one is appropriate for the dimensions of the target behavior.

Another (visual) example:

Let’s suppose that the Xs and Os below indicate the occurrence (X) or non­occurrence (O) of our target behavior over a period of 2 minutes that you are observing. XXOXXXXOOOOOXOXOXXXXOOOXXXOXOOOOOOXXXXXXOXXXOOOXXXXX Now, let’s introduce 30­second intervals and put our data into these intervals:

XXOX XXXO OOOO

XOXO XXXX OOOX OOOO

OOXX XXXX OXXX OOOX XXXX

Let’s evaluate this data using all three discontinuous measurements… Partial Interval XXOX XXXO OOO

O XOXO XXXX OOOX OOO

O OOXX XXXX OXXX OOOX XXXX

Y Y N Y Y Y N Y Y Y Y Y

If the target behavior occurred at any time during the interval, we’ll mark a “Y” for yes. If it did not occur at any time during an interval, we’ll mark “N” for no. This results in a total of 10 occurrences over in 2 minutes, or 5 per minute. Whole Interval XXOX XXXO OOO

O XOXO XXXX OOOX OOO

O OOXX XXXX OXXX OOOX XXXX

N N N N Y N N N Y N N Y

If the target behavior occurred for the entire interval, we’ll mark “Y” for yes and “N” for no. This method reveals 3 occurrences over 2 minutes, or 1.5 per minute. Momentary Time Sampling XXOX XXXO OOO

O XOXO XXXX OOOX OOO

O OOXX XXXX OXXX OOOX XXXX

Y N N N Y Y N Y Y Y Y Y

If the target behavior was occurring at the end of the interval (look at the last letter of each interval) we mark “Y”. If not, we mark “N”. This reveals 8 occurrences in 2 minutes, or 4 per minute.

You can see from this visual presentation that given the same behavioral data, partial interval tends to overestimate (5 behaviors per minute), whole interval tends to underestimate (1.5 behaviors per minute) and MTS is usually between the two (4 behaviors per minute). More accuracy results from using smaller intervals. You’ll have data sheets to use from your supervisor/employer, but here are some examples: http://nyspbis.org/Regional%20Forum1314/Assessment%20Tools/Progress%20Monitoring%20Tools%202.pdf http://www.escambia.k12.fl.us/pbis/data/OPM/Momentary_Sample_blankrevised.pdf http://www.specialconnections.ku.edu/~specconn/page/assessment/ddm/pdf/Partial_Interval_blank.pdf A­04 Implement permanent product recording procedures.

Permanent products are a fancy way to say, “The things that people produce as a result of their behavior.” Examples might be a quiz, written essay, completed project, house built, dinner made or picture taken. These are useful because you can examine these items in order to determine if a task is complete. Also, you don’t need to directly observe the behavior that produced these products. Instead, you can infer what behavior did or didn’t occur from the permanent product. As you try the following examples, try to fill in any missing blanks:

Behavior Task Permanent Product

Spell words correctly with 90% accuracy Spelling test grade?

Assemble a picnic bench from instructions Picnic Bench completed?

Prompt ­ “Don’t get your clothes dirty!” State of clothes when coming home?

Prompt ­ “Clean your room by 5 p.m.”

Cat has food and clean water

Schedule a doctor’s appointment

File cabinet is organized alphabetically

A­05 Enter data and update graphs

All of the following rely on visual examination of graphically presented data. Our interpretations of progress, success, failure or stagnation are based on a visual examination of our collected information. The three ways to visually evaluate a graph are level, trend and variability.

Level ­ Look at the range between the lowest data point in a set and the highest. This range is the level of behavior occurrence. The data on the left of the dotted line shows a level between 9­12. On the right side of the dotted line, the level is between 0­4. There is a change in level across the dotted line, because the two data sets don’t “cross.” That is, all of the data on the left is at or above 9, all data on the right is at or below 4. If the two intersected, you might say that there was not a change in level.

Trend ­ Draw a line through the data points that is closest to matching each individual point. In a statistical analysis, this line would be defined by the “Pearson’s r coefficient”, but can be drawn, as shown. The line on the left is fairly stable, only slightly downward. The trend line on the right shows a pretty clear descending trend that is a change from the data on the left.

Variability ­ This is a somewhat subjective measure, at least when evaluated visually. It is simply how consistent the data points are to each other. A stable pattern is almost completely at the same level, looks like a horizontal line if you connect data points and lacks variability. When variability is high, it is difficult or impossible to draw the trend line described above. The interpretation of variable data is that no clear conclusion can be made. The exception is that variable data in a functional analysis might suggest an automatic reinforcement function. A variable set of data is displayed below.

Now that you know the ideas of level, trend and variability, you should be able to help interpret the graph with your supervisor.

Entering data into graphs and updating them will vary depending on who you work for. Many different programs are available to make charts, such as Google Sheets or Microsoft Excel. Many behavioral providers use software that charts behavior as you record it. It is the interpretation of this graphed information that informs decisions about what next steps need to be taken.

The image below is a screenshot of Google Sheets. The behavior of nail biting is being tracked. Pretend that the data that you took on this behavior, daily, is represented on the left. The first five numbers indicate the baseline condition (i.e., before any intervention) and the rest of the session behavior data are listed from the 6th session to the 15th. When you graph this data, you see the baseline on the left side of the black line. The intervention data is on the right side of the black line. In this case, the graph shows a clear reduction in nail biting. You might be asked to continue to enter data into a program like this.

B. Assessment B­01 Describe Behavior and environment in observable and measureable terms

Not just observable and measurable, but described well enough for another observer to independently be able to track the occurrence of a behavior with accuracy. Also, measurable in the dimension that best fits the occurrence of the behavior and the intended intervention goal.

If I said that I wanted to track unprompted vocalizations that are not directed toward another person, are not words and are uttered with closed lips, occurring nearly continuously during the school day; this might be a fairly accurate definition. Whereas “vocalizations” may not be accurate enough to capture what I intend to capture. Also, choosing a discontinuous measurement such as whole or partial interval recording might be appropriate. If we wanted the

vocalizations to decrease, we would want to choose partial interval recording rather than whole interval.

Take aways: The more precise the definition, the more accurate the recording of data is likely to be.

Avoid judgemental descriptions such as “aggressive, bad, mean, etc.” Consider the topography (the motor movements involved in producing the behavior) of

the behavior in the definition Consider including the function of the behavior (escape, sensory, attention or access) in

the definition, if known Consider using known antecedents in the definition

Example: Entangling fingers of the right hand in one’s own hair by twisting and then pulling the hand away from the scalp (topographical) when given verbal instruction to begin academic tasks (antecedent­based), which has sometimes been reinforced by escape from these tasks (functional).

Application:

1. Start with a clear definition of the target behavior 2. Ideally, observe the behavior with your supervisor so that you both have the same model

of the target behavior 3. Don’t change your data collection, even as the behavior changes, without consulting with

your supervisor. Changing what you mark as the target behavior occurrence, will impact the ability to interpret the data.

4. Stay consistent with the behavior definition, or change it with input from your supervisor.

B­02,03 &04 Conduct and assist with preference assessments

Choice measures are the formal evaluation of what items or activities act as reinforcers. The design of a reinforcer preference assessment varies with the client and situation, but usually includes presenting several items or activities to a client at one time and observing/recording the “choice” that a client makes. This is most often recorded as the duration of time that a client spends with a particular item/activity within the array, or the orienting/grabbing/touching of an item.

Example: A small room is cleared of tangible items (change in environmental conditions that may change motivating operations as well) except for 7 toys. A child is given the opportunity to play with any of the items by taking one. After choosing, the child is allowed one minute to play with the item. After one minute, the item is removed and the child is asked to choose from the remaining, “unchosen” toys left. This is repeated until the order of the child’s choices is recorded and all choices are exhausted. Another variation of this design would be to offer free access to all of the toys and note the duration that the child played with each particular toy.

Other ways to determine likely reinforcers are:

1. Surveys or questionnaires 2. Observation 3. Withholding items/activities versus allowing free access 4. Asking others that are knowledgeable of the client

Application:

1. You will likely be given surveys, questionnaires or other forms for clients or surrogates to complete. You will need to work with your supervisor to understand how to administer these.

2. You may be asked to observe a client in a “natural” setting such as home, work or school. It is common practice to note the items/activities that clients spend time engaging with, in order to understand what they seem to prefer. This leads to a list of potential items/activities that may be used as reinforcers. You will likely be asked to record the total time that a client spends with items/activities or simply record what free­time items/activities that a client engages with.

3. Another way to determine potential reinforcers is to withhold items/activities, instead of allowing clients to freely access them. You may be asked to withhold items and see if clients ask for them, or to require clients to perform other tasks in order to access the item/activity. An example is requiring a student to read the playground rules aloud before being able to get a rubber ball to play with.

4. An easy way to determine what a client might work for, is to ask parents, caregivers, teachers, supervisors, etc. The people that know the client may have good ideas of what will work as reinforcers.

Individualized behavioral assessment procedures.

Individualized means that each client receives an assessment of behavior that is tailored to their particular behaviors, settings, complexity and to thoughts about intervention and maintenance. Whereas observation for 30 minutes and descriptive assessment surveys may be sufficient to identify a behavior and its function in one instance, a more complete assessment may be necessary in another. Surveys, questionnaires, interviews, observation and experimental manipulation may be necessary for some, including reassessment of the hypothesized function if intervention does not change behavior in the required direction or magnitude.

Any specific assessment will need to be discussed with your supervisor. Your supervisor will need to design the assessment and train you in specific, individualized procedures.

Functional assessment procedures.

FBAs or functional behavior assessments are usually divided into three types; indirect, descriptive and functional analyses.

Indirect assessments are those that gather information without actual direct observation of the behavior in question. Think of interviews with parents, caregivers and teachers. This also includes completion of surveys or questionnaires. These methods relay some information about the behavior without recreating conditions or having the client react to the presence of an observer. However, these are generally considered to be insufficient to form an accurate functional hypothesis of behavior. They are most often used in combination with observation of the client and/or other assessments.

Descriptive measures involve the actual observation of the behavior and surrounding events. Use of ABC recording is quite common. This tool allows the observer to identify the immediately contiguous antecedents and consequences to the target behavior. Other common tools are the scatterplot, interval recording methods and momentary time sampling. These and others are summarized nicely on page 190 of Behavior Analysis for Lasting Change, 3rd edition (Mayer, Sulzer­Azaroff & Wallace, 2014). I’ll attempt my own summary here:

ABC event recording ­ Records the behavior (usually in the middle column of the paper), as well as what immediate events preceded it and what occurred as a consequence. This allows the behavior analyst to note, especially over time, what events are likely “triggers” (SDs) for the behavior and what may be maintaining consequences. This is the procedure that you will most likely be tasked to complete.

Antecedent manipulation: You can track the percentage of target behavior occurrence in the presence of a particular antecedent. You can also make likely antecedent “triggers” occur more frequently than they normally would and evaluate the client’s responses. These antecedent­based descriptive assessments include time sampling, interval recording, conditional probability, contingency space analysis and structured descriptive assessment. This procedure would most likely be done in tandem with your supervisor or by your supervisor alone, although this is certainly a procedure that you may receive training in and have experience doing.

Ecobehavioral assessment: This is intriguing, because this method attempts to not only identify antecedents and consequences that are contiguous to behavior (i.e. occur immediately before or after responses), but analyzes patterns of responses that may suggest how the client will react to the given situation. An example is a student that has a history of eloping (puts her head down, stops speaking, leaves and runs away) when some activities are presented or when others are denied. Knowing that this client tends to run away “when things get difficult”, might help predict behavior in new circumstances better than an analysis of the immediate antecedents or consequences. Again, you would most likely consult with your supervisor regarding this kind of assessment. It may be very beneficial to your client if you are able to identify these larger patterns and relay this information to your supervisor.

C. Skills Acquisition

C­01 Identify the essential components of a well written skills acquisition plan

A well­written skill acquisition plan includes: 1. A definition of the skill to be acquired 2. A baseline measurement of the skill or the behavior it is replacing 3. A clear goal that indicates skill acquisition 4. A detailed description of the procedures that will be used to implement the plan 5. Reactive strategies to employ when undesired behavior occurs or the skill is not

immediately acquired. 6. Data collection and display methods that are to be used 7. Frequent review of data and circumstances in order to make minor adjustments to the

plan 8. A plan for maintenance and termination of services

Your responsibility, line­by­line:

1. The definition will likely come from your supervisor or other source. Make sure that you review the definition and, ideally, observe with your supervisor. Reviewing the definition at each supervision session is also recommended. This is so that the definition of behavior is clearly understood by all and that you are accurately recording behavior.

2. You will be asked to record the behavior prior to intervention, in order to learn the current rate. This is also done to compare behavior before intervention to behavior during and/or after intervention. In most cases, the undesirable behavior is recorded. In many cases, the undesirable behavior and the new skill are recorded. When recording both, you should see a concurrent reduction in one behavior and an increase in the skill being taught.

3. Goals are extremely important. They must be carefully designed so that when reached, all parties know that intervention is complete, or new goals are developed. Remember that it is your client that ultimately needs to benefit from service, but that surrogates, supervisors, insurance companies, etc. also have a desire to know if treatment is effective.

4. A detailed description should be so clear that one could pick up the plan and implement it with complete fidelity. This is rarely the case, however. Even with the best­written plan, behaviors and circumstances evolve quickly and revisions to the original plan are almost certain. Your responsibility will be questioning your supervisor on specifics that may be misunderstood, vague or are missing from the plan. Once the plan is understood, your duty is to implement it with complete fidelity. Again, the plan may need changing. Rather than change the plan yourself, however, you should regularly meet with your supervisor to discuss developments.

5. The plan should discuss how you will punish or extinguish (basically, ignore) undesirable behaviors that the new skill is meant to replace. It should also outline how to react to rapid skill acquisition or acquisition that is not progressing. Other issues that arise will need to be discussed with your supervisor.

6. The sheets or program that you use to collect data needs to be part of the plan. How, when and to whom you report this data should be specified. Also, how the data will be displayed and communicated to clients or surrogates.

7. Frequent review is probably weekly. The minimum amount of supervision set by the BACB is 5% of the amount of time spent with clients and at least twice per month. At least weekly is recommended, because minor changes to a plan are generally necessary after each session, especially when beginning intervention.

8. A well­developed plan will need to think about what happens when intervention is no longer needed. What will remain in place and who will implement the plan once you are no longer working with the client? This generally is more vague at the beginning of an intervention and becomes more cemented as the goal(s) become closer to being reached. Once you see that a goal is close to being met, you should discuss termination with your supervisor.

C­02 Prepare for the session as required by the skill acquisition plan

This section is dependent on the skills acquisition plan. If the plan is relatively simple and easy to implement, especially if you have experience with similar plans, preparation may be simple. More complex plans may require some practice to implement. In general, it would be recommended to:

1. Read the plan 2. Act out the intervention to yourself 3. Identify any areas that require clarification 4. Review the plan with your supervisor and clarify those points that you already identified 5. Decide with your supervisor what, who and how the plan will be communicated to a

client or surrogate. 6. Consider a checklist or shortened procedural manual for yourself 7. Gather any materials necessary for the intervention, including tracking sheets 8. Implement the plan

C­03 unconditioned reinforcement

A stimulus change that increases the future frequency of a behavior, these are primary reinforcers, like food, water, shelter, sleep and sexual stimulation; things that are needed to survive and need no learning history. Other terms that are synonymous with unconditioned reinforcement: primary and unlearned.

Example: When I am really cold, I am going to seek out a coat, blanket or heat to find a way to become warm (warmth is the reinforcer). Likewise, when we are hungry or thirsty, we look for things in our environment to eat and drink.

conditioned reinforcement

A stimulus reinforces, or strengthens, set behaviors through its association with a primary reinforcer. Conditioned reinforcement is learned, as opposed to unconditioned reinforcers that require no learning history. Money is a conditioned reinforcer because it is paired with what it can get a person (food, shelter, clothing, etc.). Imagine someone who is unfamiliar with money or payment who wants food. If food is offered, but also some meaningless green paper along with it, eventually the green paper will take on some reinforcing properties. Also, requiring that this person hand you some of that green paper before you give them food (food contingent on providing money) will tend to make obtaining money a conditioned reinforcer. Conditioned reinforcement increases the future likelihood of a behavior.

unconditioned punishment

Unconditioned punishment is unlearned. The first time we touch a hot stove we instinctively remove our finger (reflexive action), this decreases one’s likelihood of touching the stove in the future (this becomes a learned punishment because of the history of consequences). Examples are pain, excessive cold or heat, too much light or lack of light, loud sounds, etc.

conditioned punishment

Conditioned punishment is learned punishment. The first time we touch a hot stove we instinctively remove our finger (reflexive action­­unconditioned), this decreases one’s likelihood of touching the stove in the future. If you touched a black wood stove and experienced pain, you’ll likely act differently around a black wood stove the next time that you encounter it. The pain needs no learning to be aversive. However, the black stove was not aversive until you touched it and experienced pain. In this way, you learned that something previously neutral (black wood stove) became a possible source of pain. Neutral things that are paired with aversives, or aversive consequences are avoided in the future.

Schedules of reinforcement and punishment

Simple schedules of reinforcement and punishment have two dimensions to them. The first is either fixed or variable. The second is either ratio or interval. That leaves these possibilities:

1. Fixed ratio 2. Fixed interval 3. Variable ratio 4. Variable interval

A fixed schedule has a set number of responses or time required before reinforcement is offered. A variable schedule has a set number of responses or time, but is an average. Ratio schedules require a number of responses before reinforcement is offered. Interval schedules require a period of time to elapse before reinforcement for a response is available.

Fixed Ratio (FR)­ delivery of reinforcement based on a fixed number of responses; produces a high steady rate of responding with a post reinforcement pause (a pause in responses immediately following reinforcement).

Example: You want a student to read more. You offer ten cents (and praise) for every ten words correctly read aloud. This is a FR10 schedule. That is, you require ten responses before reinforcement is given.

Fixed Interval (FI)­ reinforcement delivered for the first response following a passage of a set duration of time since the last response; results in increasing rate as the end of the interval approaches; post reinforcement pause.

Example: You want a student to read more. You offer ten cents (and praise) when your student is reading aloud for the entire time. The student may only earn ten cents per minute. This is a FI1 min. The student starts to read and the timer starts. At the end of one minute, reinforcement is given. If the student continues to read, reinforcement is available only after another minute has passed, and so on.

Variable Ratio (VR)­ variable number of responses required for reinforcement; produces steady, high rate of responding.

Example: You want a student to read more. You offer to pay them ten cents (and praise) for an average of every ten words correctly read aloud. Sometimes the student receives reinforcement after only 2 words. Sometimes the student receives reinforcement after 20 words. This is a VR10 schedule. Probably because of the uncertainty of when reinforcement occurs, variable schedules are not associated with a pause in responses immediately after reinforcement. That is, the student in this example would likely read at a consistent rate until stopping altogether.

Variable interval (VI)­ reinforcement for the first correct response following the elapse of a variable duration of time occurring in a random or unpredictable order; produces constant stable rates of responding.

Example: You want a student to read more. You offer ten cents (and praise) when your student is reading aloud for the entire time. You give reinforcement, on average, once per minute. Sometimes the student will receive reinforcement when reading for 30 seconds; sometimes after 2 minutes. This is a VI1min. schedule. The reinforcement availability is unknown to the student.

Take aways: Fixed schedules are associated with a post­reinforcement pause. Variable schedules are more resistant to extinction (and fixed schedules are more easily extinguished).

Application:

1. Count the number of responses and give reinforcement after the set amount for fixed ratio

2. Use a clock or timer to create a fixed interval 3. Variable ratio is the most difficult to implement, except if you have a random number

generator. These applications are freely available for iPhone and Android phones. For an average of 10, set the generator to a number between 1 and 20. Then, generate numbers for each ratio.

4. The same number generator can be used for variable intervals. Another idea is to take the entire session time (say 60 minutes) and divide it by the schedule (say 5 minute average). This results in 12 possible reinforcement periods per session, on average. You can vary these 12 opportunities, accordingly.

C­04 Implement discrete trial teaching procedures

A “discrete” trial is just what it sounds like. It is one opportunity, by itself, to produce a response. If someone asks you, “What is 6+4?” this is a discrete trial. You either answer it correctly or you don’t. Once you answer, there is no other opportunity for response nor for reinforcement until the next question is asked. Most DTT procedures are a bit more complex; we’ll get into that in a bit. First, let’s mention the free­operant arrangement, which is the opposite of the discrete trial. This is often defined as, “a behavior that results in minimal displacement of the individual in time and space and requires minimal time for completion.” What does that mean? It means that instead of one response and one opportunity for reinforcement, as in the DTT, a free operant procedure allows for basically unlimited responses and reinforcements.

As usual, an example may help to clarify. If I ask you what 6+4 is and you answer “ten”, I can say “good” to reinforce you. That is a discrete trial. If instead, I put a sheet with 100 math problems on it and say “go”, then each problem that you complete leaves you in a position to emit another response (solve another math problem) and receive reinforcement (e.g. “good”, “right”, “you got it”, etc.). I might, in this example praise you for the first two or three completions, move to praising every other one and then every ten, on average. The “free operant” arrangement has to do with setting up tasks so that many quick responses may be emitted, rather than just one at a time.

Since this section is about DTT, and you’ll likely be involved in applying DTT, let’s talk about the basic procedure. Although the specific instructions may vary from client to client, or with different supervisors, there are basically 5+1 parts to a discrete trial training procedure:

1. Presenting a stimulus that is meant to evoke a response 2. A predetermined time period in which to wait

a. This is the “+1”. Sometimes, especially at first, it is necessary to prompt or guide the client to a correct response and provide reinforcement.

3. Providing a reinforcer for a correct response 4. A procedure for correcting or extinguishing incorrect responses 5. An inter­trial interval between the end of one trial and the beginning of the next

Again, specific instructions will vary and you’ll need to discuss these with your supervisor, even if you’ve had experience in using discrete trial training. Here’s one example of the above steps:

1. A teacher turns over a card and lays it on the table, face up. It’s the 9 of diamonds. 2. A 3­second wait is part of this plan. The student looks at the card, but doesn’t respond

for 3 seconds. a. Because the student didn’t respond, a verbal cue is offered: “Ni.. b. The student slowly says “nine.” c. Another verbal cue is offered: “Nine of di…” d. The student says “Nine of diamonds.” This is correct

3. A smile and an enthusiastic, “that’s right” is offered. Every 5th correct response will also include a half of a pretzel stick.

4. The student responded correctly after prompting on this trial. If prompting was not given and the student said, “nine of reds”, then a correction might be given. This might be saying “It’s the nine of diamonds”, turning the card face down, turning the card face up and repeating the prompt. It might also be as simple as waiting, without comment for another 3 seconds.

5. Once the trial is ended, the child is allowed to eat the pretzel piece, comment and talk to the examiner for 20 seconds before a new trial begins.

Discrete trial training involves very specific procedures and is usually used for teaching very early skills. This method has a great deal of utility with early learners, but tends to limit response opportunities for learners who have a wider repertoire.

C­05 Implement naturalistic teaching procedures (e.g. incidental teaching)

This technique uses motivating operations to facilitate behavior. It is often associated with communication. What client’s seem to want, as evidenced by reaching for or otherwise attempting to access, is given contingent upon a desired behavior. The items/activities may simply be what someone asks for or wants in any given moment or a behavior analyst may introduce items/activities into the environment to try and elicit responses.

Example: A speech therapist sets toys around the play room and observes what a child tends to play with. The therapist then places these items out of the child’s reach or blocks access to them until she has taught the child to verbally request for the items.

Manipulating motivating operations: We speak of satiation and deprivation in motivating operations. This means, simply, that people who have not had access to something that they enjoy (deprivation) will work harder to get that something than someone who has had their fill. Someone will likely work harder for an edible reinforcer before lunch (deprivation) than immediately after eating (satiation). A child may want a plush toy if they see it on a shelf where they can’t reach it (deprivation). Someone who has been sitting and working for a long period (satiation) may be motivated to work for an active game. If you are tired of typing and want to go to bed (satiation), you may be motivated to finish your sentence and get to bed.

Helpful hints:

Look for states of relative deprivation Look for states of relative satiation Restrict access or provide access, depending on the state you are trying to evoke

Take away: This is also known as naturalistic teaching, natural teaching or incidental learning. Again, it uses what a person wants (motivating operations) to elicit behaviors ­ very often communication.

C­06 Implement task analyzed training procedures

Breaking down a complex skill or behavior chain into its individual steps. You, the behavior analyst, can:

1. Do the task yourself and break it into constituent parts, 2. Rely on an expert to tell you how they have task­analyzed a similar chain of behaviors,

or 3. You may observe someone demonstrating the skill and describe their actions as a series

of steps.

More difficult behaviors may be broken into more steps or even multiple chains of behavior. For simpler tasks, or for clients that are more able to quickly complete steps, a shorter task analysis may be appropriate.

For example, putting an IKEA cabinet together might take 98 steps (or more, it seems). One step might be to “screw the sides of the shelf together.” This direction might be attainable for some. For others, this particular direction might need to be task analyzed into, perhaps “hold the shelf as shown, pick up a screw that looks like this, insert the screw into this slot, pick up a phillips­head screwdriver (pictured), put the screwdriver end on the screw, turn clockwise.”

Here’s a link to a task analysis on making the perfect martini (man I love ABA): http://behavioradvisor.ipower.com/TaskAnalysisMartini.html

Use chaining to link individual isolated responses to create a complex behavior chain. Complex chains are broken down by steps (task analysis).

Chaining is used to teach independent living skills, or add behaviors to an existing repertoire. It can be combined with other behavior change procedures like prompting, instructing and reinforcement

Factors affecting performance of a behavior chain are: completeness of task analysis, length and complexity of the chain, schedule of reinforcement, stimulus variation and response variation

Forward chaining is beginning with successful completion of the first step in a chain (prompted or unprompted) to access reinforcement. After the first step is mastered, the next steps are introduced in successive order, always beginning with the first step. Step 1, steps 1­2, steps 1­2­3, 1­2­3­4, etc. The reinforcement for completion of one step is the introduction of the next step, and reinforcement is also given at the completion of the last step. Forward chaining would be best utilized when the initial steps are easier to complete, but the entire task is unlikely to be mastered without some repetition. Forward chaining may not be appropriate for tasks that need to be completed in the moment (everyone is hungry and you want to teach your son to make dinner, e.g.), as this method takes some patience and time.

Example: Getting to watch a television show means that a client needs to learn how to turn on the TV and get to the channel that their show is on. You task analyze this into:

1. Pick up the Samsung remote (not others) 2. Press power button 3. Press the cable button 4. Press “guide” button 5. Search for channel 256 by scrolling down 6. Press “select”

Your client is able to do the first and second tasks without further assistance ­ you provide praise and model pressing the cable button. Next, the client starts again and is able to complete steps 1­2­3­4. You again provide praise and model the next step. On the third trial, the client is able to complete all steps and gets to watch her show and hear your praise.

Backward chaining actually follows the steps of a chain forward, but with the behavior analyst or parent completing all except the last step. This method allows the entire chain to be observed and reinforcement to be delivered upon completion of the last step. After a client has mastered the last step, the behavior analyst completes all but the final two steps and so on.

Example: Using our TV example from above, you would slowly model each step for the client until the last one. On this last step, you prompt and guide the client to the “select” button and they press it. They hear your praise and get to watch their show. On the next trial (perhaps at a commercial) you show the steps again, but stop at step 5 and ask the client to complete steps 5 and 6. Give them as much help as they need, but also allow them time and an opportunity to complete the tasks. Training continues in this fashion, until the client is able to complete the entire chain.

Total task presentation uses prompting, as necessary, to complete all of the steps in a chain on every trial. This has the advantage, like backward chaining, of completing the chain. This is really a prompt fading procedure. As you guide/prompt a client through all steps, you gradually fade these prompts to allow the client to demonstrate the skill without them. Once the client has completed all steps without prompting, the task has been mastered.

Example: Again using the TV example, you would present all steps in each trial or attempt. Perhaps on the first trial, the client has trouble with steps 4 and 5. You assist only as much as necessary in order for them to continue and offer praise (and the show) only after all steps have been completed ­ with or without help. Each trial continues to allow the client to complete each response in the chain more and more independently until they are able to complete the entire chain.

Note: Visual task analyses and/or written instructions are often used in conjunction with chaining. These tools are also useful to help maintain behaviors after training has ended. For example, a step­by­step list of how to clean the kitchen can be a useful reminder, so that no steps are missed.

C­07 Use discrimination training procedures.

This is a good point to mention all four concepts of generalization and discrimination: Stimulus generalization, stimulus discrimination, response generalization and response discrimination.

Stimulus generalization ­ Behavior that is evoked when encountering similar stimuli.

Example: You say, “plane” when you see one, and your mother says, “Yes, that’s a plane.” When you see other flying objects you say “plane” ­ to which people say, “yes”, even if the planes are somewhat different.

Stimulus discrimination ­ Discriminative stimuli versus stimulus deltas.

Example: As you grow up, people reinforce you saying “plane” when you see a plane (Discriminative stimulus). They correct you when you say, “plane” when you see a helicopter, balloon or bird. These are all stimulus deltas for saying “plane”. That is, seeing a helicopter and saying “plane” receives no reinforcement or receives punishment. Through this process, you learn to discriminate between “plane”, “helicopter” and “bird”.

Response generalization ­ Emitting functionally­equivalent untrained responses.

Example: When you are a young adult, you not only say “plane”, but “airplane”, “aircraft”, “airliner” and “airship” to communicate the same thing.

Response discrimination ­ Differential reinforcement versus extinction.

Example: As an adult traveler, you tell people that you “need to catch a plane.” When you are on the plane, you read a brochure about the “airliner” that you are taking, and the pilot announces that the “aircraft” will be departing soon. As a result of hearing others speak and from being looked at strangely when you say “airbus”, you rarely say “airplane”, and almost never say “airship.” Some of the responses in this class continue to receive reinforcement, while others do not (extinction).

C­08 Implement stimulus control transfer procedures

Stimulus control refers to the idea that a particular condition evokes a particular response. For example, if someone shows you two pictures and asks you which one is green (stimulus condition), you choose the green one (correct response). In this case, we say that the presentation of pictures and the question that was asked control the response, because each time you present these, the correct answer is given. Sometimes something other than the stimulus that we want, is controlling the response. An example of this would be given the stimulus condition above. When two pictures are presented, a client may grab the one that she likes before a question is even asked. She might even choose the correct picture, but the response is not controlled by the correct stimulus (i.e. the question “Show me green”). Given a situation such as this, the behavior technician needs to transfer the control of the response to the correct stimulus. Instead of grabbing the picture they want, the client will need to wait until the question is asked and choose the answer to that question. This can be done in many different specific ways that will be part of an intervention plan. With the example above, a therapist or technician might move the pictures out of the client’s reach, repeat the trial and give the correct response along with handing over the green picture. Trials of this nature would be repeated until the client appropriately waited and correctly responded to the question or questions presented. This would transfer control of the response from the simple presentation of the pictures to both the presentation of the pictures and the question asked by providing reinforcement only for correct responding. Again, the specific procedures for transferring control will need to be developed by your supervisor and yourself, so that the details accomplish the stated goal. C­09 Implement stimulus fading procedures

Ideally, a student/client might respond correctly to a worksheet with numbers written on the left side of the paper and blank spaces following where the client copies these numbers by writing them on the blanks. Initially, however, most clients cannot just be presented with a worksheet such as this without any additional instruction. Using the example of a number­copying worksheet, a behavior technician might write all of the numbers out with or for the client and have the client trace over these numbers. Then, the behavior technician might write the numbers using dotted lines and ask the client to trace them again. Next, the RBT could write only two examples using fainter dotted lines and require the client to trace and reproduce several numbers on their own. Finally, the client would be required to copy the one given number example by reproducing it on a line next to the example. In this kind of procedure, the stimulus offers guidance to the client. This guidance is faded as the client shows mastery of the skill, such that they can eventually complete the task without any other stimulus, except giving them a sheet to complete.

Stimulus prompts operate on the stimulus or stimuli in order to cue the correct response. These are usually considered to be positioning, movement or pairing dimensions of the correct response with the stimulus.

Positioning: Putting the stimulus closer to the client in order to make it more obvious to choose.

Movement: Looking at the correct response, tapping close to it or on it or otherwise calling attention to the correct answer by movement of the stimulus or external movement.

Pairing: Making the stimulus and the correct answer the same color or shape, or some other similarity.

Example: I like to think of Las Vegas. It gives an excellent example of signs with stimulus prompts. The “message” of these signs is, of course, “Look at me!” A digital sign that begins with a small dot in the middle and expands to fill a giant screen, the sliding of a message in from one side, lights flashing on the perimeter of the sign, a change of text colors and fonts, etc. These are all examples of stimulus prompts to encourage looking at and reading the sign.

Stimulus prompt fading is a gradual change in the stimulus prompt that eventually fades the prompt. Examples might be having a student trace a letter, trace a dotted­line letter and finally write the letter without the stimulus prompt. Again, the point of fading is to transfer control of the behavior from the stimulus prompt to a more “natural” prompt such as someone verbally prompting you to write a letter.

C­10 Implement prompt and prompt fading procedures

Prompts are usually divided into two main categories. These are response prompts and stimulus prompts.

Response prompts do just what they sound like. They operate on the behavior that you want someone to perform. The three response prompts are:

1. Modeling

This is a visual demonstration of an entire task or part of a task that allows those with imitation skills to attempt to reproduce the modeled behavior. This prompt is often combined with others and repeated until the desired response is emitted. For example, showing someone how to Waltz may involve modeling the steps, observing the learners imitation, offering corrective feedback, repeating the modeling and/or offering physical guidance to obtain a correct response of proper Waltzing.

2. Physical guidance

Some partial or full physical guidance may be necessary in order to prompt correct behavior. I say “necessary”, because this method of prompting is the most intrusive into one’s personal space. This may be uncomfortable for some learners and some teachers. Physical prompting can lead to some confusion about the learner’s demonstration of skills, as well.

For instance, I read a progress report that indicated a student had mastered writing their name, although it was not noted that this involved full physical prompting! This led to confusion about the student’s actual skill (and was likely an inappropriate goal to begin with).

3. Verbal prompting

Verbal prompting is probably the most widely used prompt as it is most efficient, especially when a learner already has a skill in their repertoire. Verbal prompting not only includes spoken words, but signs, written instructions or pictures (such as in a picture schedule).

Examples of verbal prompts might be a verbal request such as, “Take out the garbage, please” or a picture of a person walking a dog with the caption “Please clean up after your pooch.” You can imagine that the efficiency of this type of prompt is compromised if a skill is not within a person’s current understanding. With the above examples, if a person isn’t physically able to take out the garbage or has never done so, they will not be able to demonstrate the skill without additional prompting. Also imagine that the picture of a person walking a dog has a caption written in Russian. Unless you can read it, the prompt does very little, to no good in helping one decipher the meaning of the picture.

Response prompt fading involves most­to­least prompting, least­to­most prompting, graduated guidance or a time delay. All of these fading procedures attempt to transfer control of the response from the prompt to the “natural” stimulus condition. Instead of taking the garbage out only when someone asks you to, for instance, you might be reminded by the stimulus condition of noting that today is Tuesday.

Most­to­least: The greatest amount of prompting necessary to assist the client in achieving mastery is used in the first trial and prompts are systematically faded until little to no prompting is necessary to elicit a correct response. This type of prompt fading is often associated with errorless learning and ensures a high degree of successful completion of the desired response throughout trials.

Least­to­most: This is used with a fixed time interval before prompts are offered. If a correct response is emitted within this time period, reinforcement is offered. If there is an incorrect response, or no response, the trial is a “do over.” That is, the stimulus is presented and

the least invasive prompt, such as a verbal prompt. If a correct response does not occur, then the trial is begun again with more prompting, such as a verbal and modeled prompt until a correct response is given. Remember this fading procedure as the “do over” method.

Graduated guidance: This fading method is used with physical prompts. The idea is to start with physical prompting and immediately fade to closely following and not touching the client. A common example with this method is to physically prompt a client’s hand, moving to the forearm, elbow, shoulder and then no touching at all. Being closely available and able to correct responses easily is what this fading procedure is about.

Time delay: This procedure increases the delay from a stimulus presentation to the response prompt. For example, a picture of a car is presented and a child is asked to name the item. At first, the stimulus is presented at the same time as the vocalization of the answer by the behavior analyst. That is, a 0­second trial. A 1­second delay in the prompt is then initiated so that the picture of a car is presented and 1 second passes before a verbal prompt of /k../ is given. Another second is allowed before prompting with the answer “car.”When a client correctly answers, reinforcement is offered. Incorrect responses are usually corrected and a trial is repeated with the previous level of delay before prompting. Using this procedure gives a client an increasing amount of time to correctly respond. That is, it removes the prompt further from the stimulus presentation.

C­11 Program for stimulus and response generalization.

I take this section to mean more than teaching these skills, but planning to both elicit them and understand what generalizations might occur as a result of your intervention; being prepared to explain or change them.

Stimulus generalization: Once you are taught to respond to one stimulus, you might respond in the same way given a different stimulus with some similar features (at least to you). For example, if you are taught to say, “Pleased to make your acquaintance” when introduced to an adult, you might meet a fellow child on the playground and say “pleased to meet your acquaintance”. In this example, the behavior analyst would have thought about the fact that certain conditions may be appropriate for the response “pleased to make your acquaintance”, but not others, such as meeting peers. Intervention could include different stimulus conditions such as meeting peers, formal occasions, casual occasions and the like. A behavior analyst could, instead, encourage a more generalizable or useful response for greeting others.

Response generalization: Once you are taught to respond in one way, you may respond differently to a similar situation. Using the above example, you might be initially corrected by your peers and greet them with, “What’s up” or “Hey” rather than what you were taught. The response changes, but still meets generally the same function. A behavior analyst would know this and plan the intervention to include a variety of responses that could be used to greet others.

Using these two ideas together in planning an intervention, assists in developing skills and teaching stimulus conditions that your client is likely to experience in his or her particular environment. Think about your client’s life after your intervention is completed.

Program for maintenance.

Speaking of life after intervention...maintenance. From the defining of a target behavior to the functional assessment, to the intervention planning; maintenance should be on your mind. When you approach test questions, look for information that discusses the client’s natural environment (i.e., outside the intervention sessions and after intervention is terminated). Support in the natural environment is necessary for your replacement behaviors to have a long­lasting effect (generality).

C­12 Assist with the training of stakeholders

In order to plan for maintenance, you and your supervisor will have to work on training those stakeholders (parents, teachers, guardians, supervisors, coworkers, etc.) to take over the intervention to the greatest extent possible. The major idea of this training is to continue to provide reinforcement for newly mastered skills in order for these skills to continue. Remember that if a skill is not reinforced, it will not continue. This is especially true if the “old” ways of behaving are again allowed to be reinforced instead of the newly developed skill.

D­Behavior Reduction D­01 Identify the essential components of a behavior reduction plan.

Any behavior reduction plan MUST include a replacement behavior that will be reinforced, and behavior analysts should consider using reinforcement as part of any intervention plan. D­02 Describe common functions of behavior

The most basic conceptualization of “why” people do what they do is to: 1. avoid something or 2. to get something.

This “why” is known as the function of behavior, which really means how the behavior’s consequences have shaped that behavior, either through reinforcement, punishment or extinction. This simple explanation of behavior can be used with lay people to more easily explain the concept. For most behavior analysts, these two basic concepts are expanded to four functions of behavior:

1. Attention 2. Escape/Avoidance 3. Sensory Stimulation 4. Access (to a preferred item or activity or condition)

Sometimes even these basic concepts are difficult to understand and misunderstood. The functions are not meant to be judgments about a person’s character flaws or necessarily conscious desires. Instead, these are determined by analysis of what generally occurs as a consequence of behavior that tends to increase or maintain the behavior. I’ll attempt to illustrate each with examples.

1. Attention ­ a. The consequence of hitting the desk with fist is for others to look at you. b. The consequence of saying, “You just don’t care” is for one to verbally deny this

(argue). c. The consequence of running away is that someone chases you. d. The consequence of wearing a new suit is for others to notice and comment.

2. Escape/Avoidance ­

a. The consequence of daydreaming and doodling is to avoid work completion. b. The consequence of completing a task is to remove that item from your “to do”

list. c. The consequence of putting your hat on is to avoid the cold. d. The consequence of having a boyfriend/girlfriend is that you feel less lonely.

3. Sensory Stimulation ­

a. The consequence of pinching yourself is to feel both pain and pleasure. b. The consequence of running a comb through your hair is the feeling of relaxation. c. The consequence of humming a song is the feeling of your lips vibrating that you

find pleasurable. d. The consequence of hitting your head is a high magnitude of nerve reactions.

4. Access ­

a. The consequence of whining and complaining is that you get the ring that you wanted.

b. The consequence of asking politely is that you are allowed to engage in the activity.

c. The consequence of using a key is that the door opens. d. The consequence of wearing new clothes is that you increase your social status.

Note: Many behaviors have more than one function. In the examples above, running a comb through your hair may be stimulating and it may escape some anxiety. You may wear new clothes in order to gain social status (access) and to get attention. You can probably also see that some behaviors require others to meet their functions and others do not. For example, to gain attention you need another person to notice or respond to your actions. In order to pinch yourself, you do not need another person at all.

The distinction between behavioral functions that are direct access and those that are socially mediated, as well as the general function(s) of behavior should be discussed with your supervisor. Knowing (or making educated guesses about) the function(s) of behavior will drive the type of intervention that is necessary and likely to be most effective. D­03 Implement interventions based on the modification of antecedents such as motivating/establishing operations and discriminative stimulus

This is very often difficult to do without some serious thought and discussion with your supervisor. It’s helpful to think of a discriminative stimulus as a “signal” that reinforcement is (or may be) available. In contrast, motivating operations increase or decrease the power of a reinforcer in value and in behavior that has led to obtaining reinforcement in the past. It is also helpful to think of motivating operations as states of satiation or deprivation in which a person acts to obtain something because they want it more than they might otherwise, or want something less because they’ve had enough.

As always, an example might help to illustrate this idea:

I wake up in the morning after only 4 hours of sleep. I want more (establishing operation because of sleep deprivation ­ value altering effect). I look at my e­mails and read that my first two appointments have been cancelled (discriminative stimulus that signals the availability of more sleep). I set my alarm for two more hours of sleep (establishing operation because of sleep deprivation ­ behavior altering effect).

In the last part of the example, note that I could set my alarm for two more hours at any time of day. I have the ability to do this whenever my alarm is near. However, the current motivating operation in effect makes the reinforcer (more sleep) more valuable. This is because I’m relatively deprived of sleep. In addition, my meetings were cancelled. I could go back to sleep anytime, but now the reasons for getting up have been removed (reinforcements for a job well­done, punishments for not making the meeting, etc.). In short, going back to sleep is not more available; it’s more desirable.

Take away: Motivating Operations ­ Look for deprivation or satiation. Also, ask yourself if a behavior could be performed at any time, but is momentarily more desirable. This is an MO.

Discriminative Stimuli ­ A “signal” or “cue” that has, in the past, indicated that reinforcement is coming or currently available.

Changing motivating operations usually involves satiation or deprivation. That means that when you restrict access to items or activities, they may become more reinforcing. Conversely, when the environment is saturated with an item or activity, it may reduce unwanted behaviors by making the behavioral consequence less reinforcing.

Example: Allowing any student to leave class and run a lap around a predesignated area at any time may reduce elopement, leaving one’s seat or fidgeting. If students are running more, they

are likely to “want” to sit and attend more. Conversely, when a young student has been sitting for an hour, the idea of recess outside is a more powerful reinforcer.

Changing SDs can involve removing ones that have tended to trigger undesirable behavior, morphing the SD enough so that it no longer elicits the target behavior, or adding new discriminative stimuli to encourage other behaviors.

Example: Say that you notice a student who resists writing by saying, “This is stupid” and ripping his paper. The teacher normally gives the student a piece of paper and a prompt, “You’d better get some writing down today if you want any recess time, mister.” If you were to remove the SD altogether (no writing prompt) you would likely not see the target behavior (teachers hate it when I mention that idea). Instead of removing the discriminative stimulus altogether, though, what if you change the approach to: “I want you to write in invisible ink today about your favorite animal.” This change, when repeated, may become an SD for writing behavior, rather than a likely aversive condition. Adding to a stimulus condition such as changing the desk, verbal prompt, the tone in which it is said, the addition of a smile and light touch, a new notebook, etc. may encourage a different response.

One personal example of adding new SDs is to encourage going to the gym by inserting a new stimulus condition. Previously, I’d come home and instantly feel too tired to consider going to the gym. In order to change this, I’d put out my gym clothes the night before, take them to work and change into them before leaving the office. At first, I would have to set an alarm to remind myself to get out my clothes and change. The alarm and the presence of my gym clothes then resulted in a visit to the gym. Through repetition, these stimuli became a habituated “signal” that it was time to workout.

Developing new habits in the way described above is much easier when one or more discriminative stimuli are presented and a behavior follows.

D­04 Implement differential reinforcement procedures (e.g. DRA, DRO)

Differential Reinforcement of OTHER behaviors (DRO)­

Guidelines:

­ A time interval is usually set in which reinforcement is contingent on the target behavior not being observed. Particular moments can be used instead of a time interval, much like momentary time sampling

­ initial time interval should be set at least as frequently as the target behavior occurrence, so that reinforcement for non­engagement is contacted

­ be careful not to reinforce other inappropriate behavior accidentally

­ gradually increase interval when problem behavior begins to decrease

­ decrease the interval in settings or times when the target behavior is more of a problem (i.e., increase the reinforcement for engagement in behavior other than the target behavior)

­ extend DRO to other settings and times of day after problem behavior is reduced significantly in treatment settings

­ reset the interval if the target behavior occurs during the interval

­ provide reinforcement when the target behavior does not occur during the interval

Example: A girl who picks the skin of her palm is given a token (good for 1 minute of Candy Crush ­ a video game) for every 15 minutes that she doesn’t pick her palm.

Differential Reinforcement of ALTERNATIVE behaviors (DRA)­

­ Should be a functional alternative to the target behavior ­ Reinforcement of the alternative behavior is usually begun with a continuous schedule

and faded later ­ Probably the most common differential reinforcement technique ­ Often paired with extinction of the target behavior

Differential Reinforcement of INCOMPATIBLE behaviors (DRI)­

­ DRI is actually a DRA procedure that uses a replacement behavior that cannot be performed at the same time as the target behavior.

Example: A client has a habit of yelling out phrases from his favorite movies about 10 times per hour. An incompatible behavior might be to have the client whisper the movie phrases, act them out quietly or hum the parts.

DRI/DRA Guidelines:

­select behaviors already in the learner’s repertoire (when possible), behaviors should require equal or less effort than the problem behavior, behavior should be emitted prior to intervention with sufficient frequency and the DRA/DRI should be likely to produce reinforcement in the natural environment when the intervention is completed.

­Select potent reinforcers with the same consequence (function) maintaining the target behavior

­ reinforce the alternative/ incompatible behavior on a continuous schedule initially and gradually thin

­ place problem behavior on extinction

D­05 Implement extinction procedures

Extinction­ to terminate reinforcement of a behavior to reduce and eventually extinguish a target behavior. That is, if a behavior no longer receives reinforcement, it is likely to decrease over time.

Successful extinction requires correctly identifying the reinforcer maintaining the behavior/ function of the behavior and eliminating reinforcement. This often involves ignoring attention­seeking behavior, blocking sensory reinforcement, restricting access to items/activities or eliminating escape/avoidance for target behavior occurrence.

Extinction burst­ a behavior may get worse before it gets better, the burst refers to a temporary increase in frequency, magnitude or variability of the behavior

Resistance to extinction ­ This can be confused with an extinction burst, but is a persistence of the target behavior over time, even when no reinforcement is given. In resistance to extinction, you might see a behavior continue at a similar or lower rate than before. This is how to contrast with the increase seen in an extinction burst.

Resistance to extinction can be affected by:

­intermittent schedules (the thinner the schedule the greater the resistance to extinction)­ variable schedules are more resistant than fixed­ strong EO for withheld reinforcer­ the number, magnitude and quality of reinforcer­ number of previous extinction trials­effort required of the response

The worst example?: A behavior such as gambling at a slot machine has received intermittent reinforcement every 3000 “button pushes” (occasional big wins of $1000 or more), extinction has only just begun and “button pushing” takes very little effort. Since this behavior is essentially on a VR3000 ratio, extinction will be very difficult. Once the gambler has reached 3000 “button pushes” with no reinforcement, the establishing operation (EO) makes the magnitude of winning very powerful. It may take many more thousands of “button pushes” with no reinforcement, in order to eliminate this behavior.

D­06 Implement crisis/emergency procedures according to protocol

Protocols are often part of a plan when self­injurious behaviors are involved or harm to others has either been observed or suspected from the intervention plan. You should not only discuss these plans with your supervisor, but role­play the application of them as they may involve physical redirection or restraint. You may require specific training in these procedures and all possible contingencies should be considered and covered by the emergency plan. It is important to review this procedure with your supervisor at least weekly if the emergency procedure is used during the previous week and at least monthly in order that the emergency intervention remains clear to you, your supervisor, clients and/or caregivers.

E. Documenting and Reporting

E­01 Report other variables that might affect the client (e.g., illness, relocation, medication).

This means to report to your supervisor, along with data collected and notes taken, any other factors that may be important in how the client is responding to treatment(s). If a client, for example, is very sluggish because of medication side­effects, this should be reflected in your notes and report to your supervisor. E­02 Generate objective session notes by describing what occurred during sessions.

E­03 Effectively communicate with supervisor.

You’ll be required to meet with your supervisor at least twice per month, although weekly meetings are likely to be necessary in most cases. It is important that you be able to communicate relevant information to your supervisor. The reasons for this are that your supervisor is responsible for the service that you provide ­ if they don’t know something, they may still be held liable. Another reason is that changes in client behavior (better or worse) can be used to adjust the intervention plan as necessary. In order to ensure effective communication, the use of paraphrasing should be often employed such as, “So, you placed the milk out of reach and the client said, /nuke/ approximating “milk?” E­04 Comply with applicable legal, regulatory, and workplace requirements (e.g., mandatory abuse and neglect reporting).

It may be confusing to understand and follow all applicable laws when they may include your current agency, insurance requirements, responsibilities to the ethical code and the BACB in addition to state and federal laws. If you have questions about your ethical or legal responsibilities you may always consult the ethical code and ask your supervisor for assistance. E­05 Comply with applicable legal, regulatory, and workplace requirements for data collection, storage, and transportation.

The BACB requires that client data be kept for a minimum of 7 years. This may be longer, in some jurisdictions where state or local laws apply. In general, these data need to be kept confidentially and only released in certain circumstances to those other than the client and/or their surrogate. Check with your supervisor if there is any confusion on what to keep, where to keep it, or how to store these documents.

F. Professional and Scope of Practice F­01 Describe the role of the RBT in the service delivery system.

An RBT is most often on the “front line”. That is, they are working most directly with clients and surrogates in the implementation of the behavioral intervention. Because they are most often seen by both the clients and parents/surrogates, RBTs often have a more developed rapport with parents than any other behavior professional. Because of this frequent contact it is important to do two things:

1. Relay relevant information to your supervisor about the level of support from caregivers and the progress of the client.

2. To develop and maintain a rapport with clients and their familiies that facilitates the inclusion of all stakeholders, especially your supervisor’s involvement.

3. F­02 Respond appropriately to feedback and maintain or improve performance accordingly

It is aversive for many to hear criticism. Although it may be difficult, however, criticism should always be constructive and meant to assist you in making improvements to your service delivery that will ultimately benefit the ultimate beneficiary of behavioral services. Your supervisor should allow you the opportunities to receive training and practice in order to meet expectations. F­03 Communicate with stakeholders (e.g. family, caregivers, other professionals) as authorized.

When communicating to stakeholders, it is important to retain the understanding of your role as the behavioral services provider. Do not attempt to suggest changes to the intervention plan or discuss issues that should be referred to your supervisor. Again, your supervisor is the person who is ultimately responsible for ensuring that the services that you provide are true­to­plan and completed correctly. F­04 Maintain Professional boundaries (e.g. avoid dual relationships, conflicts of interest, social media contacts).

The professional and ethical code (section 1.06) deals with this very topic. Multiple or dual relationships involve a professional relationships and some other kind of relationship (e.g., friendship, romantic, family or business relationships other than the behavioral service that you are hired for. It is important to maintain professionalism by avoiding these conflicts of interest. Do not accept gifts or allow working closely with one another to become a relationship in which professional and friendship boundaries are difficult to define. Instead, maintain only a professional relationship, as others can endanger the client­centered model. F­05 Maintain client dignity

Maintaining client dignity has much in common with communicating respect for the client and respect for confidentiality. Share client information only with clients, caregivers and your supervisor ­unless requested to do so in compliance with legal or ethical obligations. Communicating respect for the client, in general, means that you treat clients as you would any respected member of the community and maintain care and patience.