CHOOSING TARGETS AND TRACKING PROGRESS
Given the nature of a typical verbal behavior program, the choosing of targets goes hand in hand with the way you keep data. As one skill is mastered, it is often used to transfer to another verbal language function (operant). For example, mastered requests (mands) can then be transferred to labeling (tact) targets. The "behavior" involved in reaching toward and touching an item when manding can be transferred to a receptive response (touch, get). Mastered motor imitation targets can also be transferred to receptive targets. See the sections on teaching individual skills within each verbal operant for additional suggestions of transfer procedures.
When choosing target objectives as well as individual target responses, the following should be kept in mind:
1. The targets should be functional for the child.
2. Use skills the child previously has mastered to teach new skills.
3. The targets should be developmentally appropriate for the child.
4. Targets should be chosen based on the priorities the family has for their child.
5. Move across verbal operants rather than just choosing the next skill within each operant.
6. Make sure the child has the pre-requisite skills needed for the target skill.
The first time the ABLLS is completed, it is recommended that conservative estimates be used. If you are not sure if the child is able to perform the skill fluently (i.e. quickly and easily with no prompting), take some probes. It is better to underestimate the skills of the child and keep teaching "easy" rather than to overestimate and frustrate the child. Be aware that for some skills, the child may produce the desired "behavior" on occasion but not under "instructional control". For example, a child might be observed to imitate children in his favorite video but not be able to imitate you when you say, "Do this". It's also important to recognize if the target behavior is under stimulus control of specific questions. For example, the child might be able to see a train and say, "The blue train fell off the track." But not be able to respond when asked, "What's this?", "What color is the train?", or "What happened to the train?"
Once the ABLLS is completed, look at the next 2 objectives within each skill area for possible objectives. Then, look across verbal operants to see if the child has a mastered skill in another area that might be used to teach this new skill. For example, if a child is able to request (mand) items when asked, "What do you want?" we can use this skill to teach him to label (tact) items when asked, "What's this?" Also look for any "imbalances" in the ABLLS profile and focus your teaching on the "weak" areas. If the child has very strong labeling (tacting) skills but fewer requesting (manding) skills, you'd want to focus a great deal of instruction on the weaker skills.
Make sure the child the skill is developmentally appropriate. There are some skills that children are just not developmentally ready for. They may not have the pre-requisite skills. For example, it may not be wise to try to teach a child who is unable to imitate three words task G29 "Uses carrier phrase when labeling nouns with verbs and adjectives." Trying to teach skills that the child does not have the pre-requisites for can be frustrating for both the child and the instructors!
Choosing Individual Targets
The request (mand) is the first verbal operant taught because it allows the child to learn the function of language that will give him "power" over his environment. Review the section on manding to help choose these targets. The main point to remember is that you can't teach a child to request something they don't want! The main job of instructors of an early "mander" is to continually build the number of reinforcing items and teach the child to mand for them. Targets within other skill areas (verbal operants) might be closely related to the items they request. For example, if a child were manding for a car, a fish and a chip, these would be the same targets/stimuli used to teach the child the receptive tasks of "Follows instruction to look at a reinforcing item" (C3) or "Follows instruction to touch a reinforcing item in various positions" (C6). On the other hand, if a child is having difficulty with a given skill and isn't responding to prompting, that response may be chosen as a target within another skill area (verbal operant). For example, if a child is not able to tact a specific item and isn't responding to echoic prompting, the target may be chosen as an imitation target. Let's say the child tacts "refrigerator" as "diderator". While this is obviously an articulation problem rather than the fact that the child is unable to label the refrigerator, it could be chosen as an "echoic" or imitation target and be taught, syllable by syllable, to "say" at least a closer approximation to "refrigerator". Look at the child's current program to determine what to teach as targets for imitation tasks rather than having the child imitate "random" words or sentences. By the same token, if the child is signing, use parts or whole sign movements as targets for imitation rather than teaching the child to imitate random movements. All targets should be functional for the individual child and should relate to his current life situation.
The way you keep data can either help or hinder the process of transferring a response from one verbal operant to another. It is important for instructors to see the "big picture" as well as being able to keep track of the child's individual target responses. It's important that instructors understand that while these are individual target responses that we are teaching, they should all be "taught" from a previously mastered skill rather than being constantly "drilled" and "corrected".
It is important that the methods of data keeping allow instructors to focus on teaching rather than testing. Data keeping should never interfere with teaching. The goal of data keeping is to:
1. Monitor the effectiveness of teaching
2. Serve as a communication tool between instructors
3. Determine when individual targets have been mastered
4. Assist in determining when objectives have been mastered
5. Monitor the retention of maintenance targets
6. Give a "picture" of a child's learning history so any "dips" or decreases can be noted and addressed.
Rather than taking data on each individual response, most proponents of VB suggest taking data by probing the current targets. A probe is essentially a test. The idea is that if the child responds to the particular SD (discriminative stimulus= fill-in/question/command/direction + any visual stimulus present) on the first trial of the day quickly and with no prompting, the response is considered "Independent" or correct. If not, the response is corrected using previously discussed correction procedures. This daily probe is then used to determine WHAT to teach during the session. After the probe, the rest of the session is used to teach those responses that required prompting as well as any new targets chosen. These "targets" are mixed in with previously mastered targets across all verbal operants. The goal is to keep the child successful during teaching. The child's responses during the probe allow us to see which targets are still difficult and require more teaching.
The goal in "errorless" learning is to get the desired response in any way in which the child is successful then to take "advantage" of the fact that he will be more likely to repeat the same behavior by "sneaking in" a new condition under which the response will occur and thus be reinforced. We then gradually "separate" the new condition from the one that we used to transfer or prompt the response. We won't transfer to a new condition until the child is able to respond correctly, without prompting, under the current condition. In addition, we will continue to "run transfers" on all responses that are correct even if the child's actual target is not the "transferred" skill. In this way, by the time the target is added to the new operant, the child has had lots of experience responding correctly, even before the target was actually "chosen" in that operant class.
For example, let's a child was able to mand for "ball" and the current target is that he requests the ball when asked, "What do you want?". For the first session that day (probe trail) he had to be prompted with an echoic (ball? R= ball What do you want? R= ball) Throughout the rest of the session, multiple echoic to mand transfers were conducted. By the 3rd trial, the child no longer required the echoic prompt and was able to ask for the ball whenever he wanted it. Then, transfers were begun to the tact (label). EX; ball Inst: What's this? R= ball Inst: What do you want? R= ball. The response is still under control of the mand (he wants it!) and "ball" is still a mand target but the instructor is "practicing" having the child respond when asked, "What's this?". Later, when he wants an elephant, the instructor first asks for the tact of "ball" then the child asks for what he wants. EX: "What's this?" R= ball Inst: What do you want? R= elephant. "Ball" would still be a mand (request) target until the child has requested it on 3 "cold probes" with no prompting but other transfers can be run whenever the child is responding correctly on a consistent basis within a teaching situation.
Actual data isn't being taken on the "new condition" at this point. Once the child has met the criteria set for "mastery", the target is then added to the new condition to be included in the probe. Lot's of teaching has already occurred but the goal is to then see if he can respond under the new condition without needing the transfer. In the previous example, once "ball" was chosen as a labeling (tact) target, the instructor would ask, "What's this?" (R= ball) on the first "cold probe" of the day, (and when the child wanted something other than the ball) before any transfers or prompting had occurred. If the child responded correctly, and "I" (independent), check mark or "+" sign would be placed beside the target on the probe sheet and the instructor would know they didn't need to devote a great deal of time during the session to "teach" the child (using transfer procedures) to label the ball. If the child responds incorrectly or needs to be prompted, a "P" (prompted), or "-" would be placed on the probe sheet and the instructor would know that further teaching was indeed needed.
We want to keep learning fun! We all enjoy doing things we are good at so we want to be sure that the child finds he is good at learning! (contacting reinforcement frequently) The way to insure this is to continually mix in previously mastered responses during probing. Behaviorally, we know that a child is more likely to engage in a "difficult" behavior if the requirement for that behavior follows multiple "easy" or more likely behaviors. As related to language, this means the child would be more likely to respond to a difficult question after being asked some "easy" questions or given "easy" directions. This behavioral momentum that is built can be used to our advantage during teaching and probing. Depending on the child's response to frustration, a ratio of 80/20 to 70/30 of "easy" to "hard" responses is the target. Current targets are considered "hard" and mastered responses are considered "easy".
For other children, it may be fine to probe all the current targets at one time or a specific skill area on different days of the week. This will depend on the relative ease with which the child is acquiring new skills as well as the individual child's response to frustration. It would not be recommended for a child who demonstrates signs of frustration quite easily.
The amount of probing done and criterion for mastery varies across programs and can largely be determined by the individual needs of each child and teaching team. For example, if the parent is the only one working with the child, it might be easier for them to keep track of what they are teaching the child without a great deal of data but if other people are working with the child, it will be important to be able to communicate what was taught and how the child responded to that teaching.
The criteria for mastery should be determined by the learning history of the child. One team might find that the child masters a target in one setting and has no difficulty retaining it over a long period. For this child, correct responding on 2 probes may be adequate. For other children, it may take many more transfers and much more teaching for them to respond correctly or the team may find that the child frequently "loses" previously mastered targets. For this child it may be necessary to get 5 "correct" probes before considering the target "mastered" to fluency (responding quickly with no prompting).
Any time a target is probed and the child doesn't respond within 2-3 seconds, the answer is provided, the target behavior is elicited and the SD is presented again. (See "correction procedures). Even if the team "knows the child knows the answer", it is best not to wait any longer than this to prompt the correct response. "Strong" responses are produced quickly, with no hesitancy, and it is best to continue teaching until the responses are fluent. This will help insure the child retains the response over a long period of time.
Data Collection Methods
There are a variety of data collection methods used by individual teams. Whatever works best for the teaching team and allows the necessary data to be collected should be chosen. All teams typically have a notebook divided into sections into the areas covered by the ABLLS or at least those areas in which the child currently has goals and objectives. Some teams have all the "FFC" (Features, Functions, Classes) targets together in one section to allow for easy data keeping of transfers from one skill to another. In these sections are individual goal sheets with the target "behavior" (What you want the child to do) and a description of the teaching procedures to be used to teach the skill. Behind this sheet are data sheets with the current targets within that skill area. The data from the daily probes is kept on these sheets. The number of goal sheets in each section will depend on the number of objectives taken from the ABLLS in each area.
Many teams also keep a "Current Targets" board so they can easily see, at a glance, all of the child's current targets in each area. These sheets can be covered in sheet protectors and a grease pen can be used to mark the probe data for the day. Then, the sheet can be kept handy during teaching to allow the instructors to know which targets need continued teaching. The probe data is transferred to the notebook at the end of the session. These "target boards" can be especially helpful in keeping the instructors focused on the "big picture" as well as making data keeping for a child whose primary teaching is occurring in the natural environment quite simple.
Other teams keep all of the child's current targets on individual laminated cards. An apron can be worn with the cards sorted into "prompted" and "independent" pockets as the child responds to the probe. The "prompted" targets are then taught throughout the teaching session. The data on the cards is transferred to the data sheets at the end of the session.
Many families use "white board" or sheets that can be stuck to the walls to write the child's current targets. Current targets and special reminders are written on these white boards with dry erase markers. This allows for all those working with the child in the natural environment to remember the targets.
Another "handy tool" for a VB program is a counter, similar to those used by umpires at a baseball game. These are often used to count a specific behavior, such as requesting (manding). The goal is to get 1,000 requests a day! Often, having the instructors count how many mands are occurring helps them to focus their attention on this important skill! In addition, it may also be necessary to count "prompted" vs. "unprompted" requests. This may be important if it appears the child is becoming "prompt dependent" or if it appears the instructors may not be adequately fading their prompts. Remember, the goal in manding is that the child be able to request whatever he wants with no prompting. In other words, the child talks first!
If targets are well chosen, they should be those in which the child is in daily contact so "practice" with the targets would be continually occurring. As the child's program progresses and intensive teaching sessions begin to take place, the number of mastered targets will get quite large. During intensive teaching, these targets are those that are mixed in while teaching new targets to help keep the child successful. This also helps insure these targets are being maintained by "reviewing" and continually giving the child reinforcement for responding correctly.
If a child is not responding to "mastered" targets correctly, these should be focused on during the teaching situation and "re-taught" until the child is able to respond correctly. Be sure to look at both the questions being asked as well as the particular response the child is giving to determine what is causing the child to have difficulty discriminating. The more the child learns about a given item or a class of items, the more conditional discriminations they are required to make in order to respond correctly. Use the information from your probes and teaching sessions to determine what discriminations have to be taught.
Tracy Vail, 8/02