Dr. Mark L. Sundberg, Ph.D., BCBA
Dec. 4th and 5th, 2004
Analyzing and Ameliorating Language Acquisition Barriers for Children with Autism
Hey Guys, sorry this took me so long! Life is just a bit too hectic lately. But Happy New Year! This was an excellent conference and tough to write notes on, just really had to be there. Dr. Sundberg was wonderful, really inspired you to want to learn more about the behavior analysis of language and behavior analysis in general. When I was learning from him, I kept thinking how little I really know, or more specifically, how little I really know about Verbal Behavior and how to analyze it the same way you do other behaviors. To the interventionists who were able to come and spent their Sat and Sunday learning with other highly motivated parents (of which there were many, awesome all you moms and dads!), thank you for your dedication to this field and to our children.
Dr. Sundberg was generous with information (and encouraged us to be with it as well.) He gave some powerful tips and assessment tools in our handouts on how to analyze and fix what is “defective” in your child’s language repertoire that may be preventing and/or interfering with language acquisition. One interesting thing was the video he showed to support what he was talking about often caused much anxiety among attendees. Questions like, “well, why are they doing it that way?” “Why aren’t they doing this?” “I would have done this” “Shouldn’t the teacher be doing more pairing first?” We had to remind ourselves that he wasn’t showing us video of how to do it right in the first place, this conference was about when things go wrong and how to analyze and fix it. (You know, the REAL WORLD, when you start applying what you’ve learned in the intro VB conferences and it just doesn’t go like the video clips? It is so important that interventionists and parents understand that just because you see one or two conferences, do not think that now you know what it’s all about. Sorry but, I’m a little annoyed that some interventionists in our area didn’t come to this conference because they already “know all about Verbal Behavior”)
As a mom just trying to be at the very least, an equal contributor to my son’s education, I’d like to pass on some key things we learned during this 16 hour 2 day conference. My notes here are so far removed, and only touch on a small fraction of what we learned. My intent is to 1) help myself review 2) contrive MO’s to attend these important conferences that will inspire you to learn more! :0) Ok, here we go, not exactly in order of how he presented but this first information is so significant. As Dr. Sundberg said “To understand the whole, you need to understand the parts”..
Dr. Sundberg talked about how ABA is not just procedures and techniques; it’s the ANALYSIS that is extremely important. The good news is that ABA has become popular, the bad news is that not everyone can do it :0) ~It really is rocket science! Board Certification of Behavior Analysts exists (BCBA or BCaBa www.bacb.com) Know that this credentialing does not make someone an expert, but rather, is a way to know that someone has some of the minimal competencies necessary. Discrete trail training, ABA, and Verbal Behavior are ALL based on the work of B.F. Skinner and developments in the area of psychology known as Behavior Analysis (www.abainternational.com )
Dr. Sundberg helped us understand three aspects of ABA:
1. PROCEDURES = BEHAVIOR MODIFICATION (tip: you should know and be familiar with the research!---eg. Shaping, chaining, prompting, fading, modeling, fluency, pairing, Differential Reinforcement =DRO, DRI, DRL, Intermittent Reinforcement Procedures= FR, VR, FI, VI, Extinction procedures, Punishment Procedures (time-out), generalization and discrimination training, errorless learning, transfer of stimulus, token economy, etc, etc…
2. ANALYSIS = PRINCIPLES OF BEHAVIOR
Basic principles of operant behavior; Relationship between stimulus control EO/MO and the Response (important tip: Analysis is critical yet often neglected---Need to know motivators right now; Behavior analysis is ongoing and constant)
3. EXPIREMENTAL (control data)
Discrete Trial Teaching is good behavior modification, NOT behavior analysis. In many cases, DTT has become the WHOLE intervention. (ugg..soo sad, so true~no analysis, just cookie cutter programming)
Language intervention is a major goal of most programs for children with autism and is critical. What doesn’t match is that most often the ABA/DTT language intervention programs for our learners are based on NON-BEHAVIORAL theories of language (A.B)
Instead of language programs being based on BEHAVIORAL theories of language(C)
NON-BEHAVIORL THEORIES OF LANGUAGE
A. BIOLOGICAL- (Chomsky, Lennberg, Pinker) language is innate
Common terms: language acquisition device, language synthesizing, deep and surface structure; Brain - words
B. COGNITIVE- (Piaget, Brown, Pinker ) Language is the result of a cognitive processing system--- (dominating/overwhelming approach)
Common terms: Processing, storage, retrieval, coding encoding, expressive, receptive; Cognition - Words
BEHAVIORAL THEORY OF LANGUAGE
C. ENVIRONMENTAL-(Skinner, Kantor, Bijou)
Language is learned behavior controlled by environmental variables
Common terms: reinforcement, motivation, stimuli, mands, tacts, intraverbal;
Stimuli/motivation/words/consequences
(isn’t that so true! It doesn’t make sense. Many of our children are in “ABA” programs yet the language intervention part is based on non-behavioral theories of language! I know that was true for my son until for me, ELIJA came along, thanks girls ;0)
KEY POINTS:
Language is a LEARNED behavior under the functional control of environmental variables (if you look at what causes behavior, it would be the same analysis of language)
**The elementary Verbal Operants are separate repertoires and each must be taught**.
**Speaker and Listener skills are separate repertoires and both must be taught**.
After a quick review of the behavioral classifications of language (Receptive, Echoic, Imitation, Tact, Mand, RFFC, Intraverbal, Textual, Writing—refer to the book Teaching Language to Children with Autism and other Developmental Disabilities) we were reminded of the power of behavior analysis as a conceptual tool and how the real power of Verbal Behavior is in the analysis--- understand why we say the things we say, what words under what circumstances.
Along with knowing the behavior classifications of Skinner’s analysis of Verbal Behavior, to start being able to analyze language, you need to know the 3 different main sources of control and the verbal relation:
Type of Control: Verbal Relation
Motivation ---------------------------- > Mand
(known as establishing operation=EO more recently known as motivational operation=MO)
Nonverbal Stimulus (nonvervbal Sd) --------------------------> Tact (Things around our world;)
Verbal Sd (the verbal stimulus a.echoics b.intraverbals
and the response are different)
Textual (reading)
Transcriptive (spelling)
Often, when you start mixing together these sources of control and the verbal relation, your learner starts having problems. Even if you think you are doing “good” teaching, sometimes there are some important prerequisite skills that need more time and focus to properly develop like matching skills or good scanning skills (these skills need more emphasis for our early learners before moving on! Dr. Sundberg gave some great info in our handouts about assessments for Matching to Sample and RD nonverbal Array assessments) Maybe our good teaching isn’t as good as we thought because we are moving our learners along to quickly, that the intraverbals and skills learned become rote responding? Maybe the learner has a defective manding repertoire? Maybe the learner is prompt dependant?…
More generally remember: Analyze! Analyze! Analzye!: Some Quotable Sundberg Quotes :0) : “looks like a good response, but under wrong control” (An example would be something like eye contact learned from eye contact drills) … “The form of the response isn’t as critical as the function”.. “The real power of Verbal Behavior is the Analysis”.
A great tool is the *ABLLS which serves 3 purposes:
a.task analysis of child’s skills b. tracking system c. general IEP guide
(Dr. Sundberg did talk about how there are many components missing from the ABLLS and is working on an improved ABLLS that should be released in the near future)
Dr. Sundberg gave us strategies (and in our handout assessments) that will be helpful tools in analyzing “Language Acquisition Barriers”.
Quick Assessment part 1: The first part of assessing for a “ballpark” profile, would be utilizing the “Behavioral Language Assessment Form” (it is in Chapter 2 of Teaching Language to Children with Autism or Other Developmental Disabilities)
Quick Assessment part 2: Would be identifying, rating and graphing on a problem scale from 0=not a problem to 4=significant problem if the learner has:
1. A DEFECTIVE MANDING REPERTOIRE~ When they can’t ask for something without another sort of prompt; just a few “potential causes of a defective mand” are: a. motivation does not control the response (what your child asks for isn’t really what they want) b.satiation c.not assessing current motivation d. response requirement to high and weakens the relevant MO e.self stim or obsessive behaviors compete with MO and have a lower response requirement f. Prompt dependent with echoic, imitative, or verbal prompts g. negative behavior functions as mands h. single topography mands (eg, more, please) i. scrolling gets reinforced j. not enough mand trials k. mands are only required/reinforced in specific settings l. free access to reinforcers without manding m. fading out the object/nonverbal stimuls too soon n. nonverbal stimulus acquires control and blocks MO control o. verbal stimulus acquires control and blocks MO control p. small group of mands have strong history of reinforcement (no variety always only one or two items) Reminder: IF THER IS NO MOTIVATION, THERE IS NO MAND
2. ROTE RESPONDING (from memory, usually without any understanding)
3. ESCAPE/AVOIDANCE (issues of compliance and instructional control)
4. ECHOLALIA
5. SELF-STIMULATION
6. RESPONSE REQUIREMENT WEAKENS ESTABLISHING
7. OPERATION/MO (Does the demand extinguish value of reinforcement?)
8. PROMPT DEPENDENT (this is a huge barrier)
9. SCROLLING/CHAINING (significant problem, learners scrolling through and chaining responses)
10. REINFORCEMENT DEPENDENT (will only work for continuous reinforcement, not intermittent)
11. LEARNER DOESN’T ATTEND TO PEOPLE
12. LEARNER DOESN’T ATTEND TO MATERIALS
13. LEARNER HAS OBSSESSIVE COMPULSIVE BEHAVIOR
Quick Assessment part 3:
Would be scoring the following motivation and reinforcement on a scale of 0-4:
1. Types of Motivators:
a. Unlearned EO’s (UEO’s)
b. Learned EO’s (CEO’s)
2. Reinforcement Schedule:
a. CRF (continuous reinforcement), VR
3. Type and number of reinforcers
We always need to know and be aware of what is motivating RIGHT NOW! Behavior analysis is ONGOING. Make sure we are not limiting the opportunities for the learner by using reinforces that are just convenient for us to use.
Other helpful assessments included in our handout were:
Motivation Operation (MO) Assessment
Does the mand occur without verbal or nonverbal source of control present? Is there variation in the mands? Test the MO with a choice procedure. Does the word match the item chosen? Is there a short latency to accessing the manded item? Is the item consumer or manipulated quickly? Does satiation alter the evocative effect? (my note, evocative effect means stimulus that evokes a response) Does the mand trial increase a target behavior? Does negative behavior occur after the mand? Does the student emit scrolled responses? If the desired item is removed from choice menu does the child search for it? After the choice procedure present mand board or what do you want?
RD Nonverbal Array Assessment
Broken into 5 levels (My~Jennica’s note~ this and the Match to Sample Assessment was both so significant to me. Are we making sure our learners have enough generalized receptive discrimination skills before we are moving them on to more advanced skills?
If your learner has SD discrimination problems, or if your learners’ responding only happens in an array of 3, read on!!!)
LEVEL 1~small array; different stimuli; objects vs. pictures
a.Two very different objects b.Three very different objects (array in a row) c.Four very different objects (scattered array) d.Six very different objects (scattered array) e.Two very different pictures f.Three very different pictures g.Four very different pictures h.Six plus very different pictures
LEVEL 2~Generalization
a.Three versions of the object (stimulus generalization) b.Three versions of the picture c.Six plus versions of the object d.Six plus versions of the picture
LEVEL 3~Finer stimulus generalization
a.Two similar stimuli in a n array of three b.Three similar stimuli in an array of three c. Three similar stimuli in an array of six d.Six similar stimuli in an array of six
LEVEL 4~Size
a.Known stimuli smaller than 2 inches (generalized) b.Known stimuli smaller than 1 inches (generalized) c.Known stimuli samller than ½ inches generalized
LEVEL 5~Background, context, natural environment
a.Known stimuli in varied background b.Known stimuli in Contextual Background c.Small known stimuli in contextual background d.Known stimuli in the natural environment
Matching to Sample
The goals of M-T-S include: 1. Establish conditional discrimination skills 2.Establish Scanning Skills 3.Establish Specific Visual Perception Skills 4.Establish Stimulus Classes/ Pair Spoken Words with Visual Stimuli 5. Facilitate Receptive Discrimination
6. Facilitate Tacting 7.Facilitate RFFC 8. Facilitate more complex visual perception skills such as block designs, puzzles, assembly, play skills association of MTS
LEVEL 1~ Identical Pictures (Exact match/ array of 3)
a.Identical pictures-different positions and orientation b. Identical pictures-Different Size c.Identical Pictures-Different Background d. Identical pictures- Different Size and Background e. Increase array size and position of pictures f. Include similar comparison stimuli g. Comparison in Natural Context/book
LEVEL 2~ Vary Irrelevant Feature of Target Items
a. Non-Identical Pictures-Different Color (including Black and white) b. Non-Identical Pictures-Different color/size c. Non-identical Pictures-Different color/size/background d. Non-identical Pictures-Different color/size/background/Orientation e. Increase array size and position include similar comparison stimuli
LEVEL 3~ Vary More Salient Irrelevant Feature of Target Items
a. Non-Identical Pictures- multiple differences b. Large Array c. Include similar comparison stimuli (have the learner identify a pencil in a field including a broom, a paintbrush, a stick etc)
Dr. Sundberg also included a section on Task Analysis and Levels of Receptive by Function, Feature, and Class (RFFC) 6 different Levels (Listen, I know this is a lot of info, but I’ve got to include some of this, this is where it all comes together!!! This RFFC section will really help you begin to understand prerequisites and suggested learner competencies necessary. It’s like, we rush our kids. We get them manding, some tacts, some simple intraverbal fill-ins and we start rushing those FFC’s without focusing on core foundation skills necessary then blame them for not “getting it”)
Task Analysis and Levels of Receptive by Function, Feature, and Class (RFFC)
LEVEL 1
Student Entry Skills~100 plus tacts and receptive discriminations, strong mand, echoic, imitative and M-T-S repertoires, good generalization skills and good scanning skill
Verbal SDs~Song fill-ins, Animal sounds, fill-ins involving fun activities, Noun-noun associations and fill-ins (shoes+socks), Reverse associations and fill-ins, Vary carrier phrases, tone, rate of speech, pitch, intonation
Nonverbal SDs~ 3-6 very different stimuli, Constantly changing array, Several different examples of the same stimulus
Teaching Objective~ Develop advanced listener repertoires by establishing conditional discriminations involving a wide variety of verbal and nonverbal stimuli. The key element of the training is that the verbal stimulus does NOT contain the specifiec name of the item in the nonverbal array. In addition constant variation of the verbal and nonverbal SD’s are critical
Data Collection: Probe data on the first five trials of the target verbal stimuli n the target array configuration
LEVEL 2
Student Entry skills~ Successful performance on level 1 RFFC activities. Continued acquisition of new tacts and receptive vocabulary. Emerging intraverbal skills.
Verbal SDs~ same verbal tasks as level 1
Nonverbal SDs~Similar stimuli in array of 3-6; small stimuli in array of 3-6; Similar stimuli in a large array (7 or more) Small stimuli in a large array (7 or more); In-context (NET) stimuli
Teaching Objective~ Make the conditional discriminations harder by make the nonverbal aspect of the task more complex. Establish better scanning skills. Transfer RFFC skills to the natural environment
Data Collection: Probe data on the first five trials in the specific array targeted. Responses per minute samples (fluency measures)
RED FLAGS: Rote Responding; Not scanning; Negative behavior/ escape and avoidance during training; High rate of forgetting; High error rate/mixing of learned responses; Long Latencies; Large number of teaching trials for each new relation; failure to generalize; failure to emit RFFC behavior in the natural environment
LEVEL 3
Student entry skills: Successful performance on level 1 and 2. Minimal red flags. Successful performance on RFFC to IV transfer tasks. Strong noun and verb (two component) tact and receptive discrimination reportoires
Verbal SDs:Continue level 1 and 2 activities including verbal variation; “eat” and “drink” categorization (classes) ; Verb-noun fill-ins (functions) “What” questions with verbs and nouns
Nonverbal SDs: Small array of different stimuli (3-6 items) ; Similar stimuli in array of 3-6; Small stimuli in array of 3-6; Similar stimuli in a large array (7 or more); Small stimuli in a large array (7 or more) ; In context (NET) stimuli
Teaching Objective:Increase the complexity of the verbal stimulus by moving on to verb-noun combinations. Begin to break free from fill-ins by using “What” questions
Data Collection: Probe data on the first 5 trials of each target verbal SD in each specific array. Response per minute data
LEVEL 4
Student entry skills: Successful performance eon level 1,2, and 3 RFFC activities. Several hundred tact and receptive vocabulary words. Generalization very quick. Mean Length of utterances easily around 2 to 3 responses consisting mainly of noun and verbs.
Verbal SD’s : Continue level 1,2,3 activities including verbal variation; Where questions; Who questions; Beginning verbal categorizations (classes) Beginning feature questions
Nonverbal SDs: Same as Level 3
Teaching Objective: Establishing more complex listener reperotires by increasing the complexity of the verbal conditional discriminations. Introduce a wide range of context into the sessions, both verbal and nonverbal
Data Collection: Same as Level 3
LEVEL 5
Student Entry Skills: Successful performance on level 1,2,3 and 4 RFFC activities. Emerging adjective and preposition skills
(My~Jennica’s note~OMG!~How many 3 year olds do you know in ABA programming where they are working on prepostions~PUT IN, PUT ON, PUT UNDER~ It is suggested that Level 1,2,3, and 4 be successful before prepositions emerge, unbelievable! Many children don’t even have a strong or any manding repertoire and some already have them working on prepositions!)
Verbal SDs: Continue level 1,2,3, and 4 activities including verbal variation; Multiple items (verbal conditional discrimination); Two component multiple noun-noun verb stimuli (verbal conditional discrimination); Two component multiple verb-noun verbal stimuli (verbal conditional discrimination)
Nonverbal SDs: Small array of multiple items (3-6 items): Contrasting items relevant to verbal conditional discrimination; In-context (NET) stimuli
Teaching Objective: Begin to establish conditional discriminations within the verbal SD. Developing stronger listener skills. Moving into academic content.
Data collection: same as level 3 and 4
LEVEL 6
Student entry skills: Successful performance on levels 1-5 RFFC activities. Ready for more time in inclusion. High rate of spontaneous verbal behavior. Quick acquisition of new verbal responses. Ready for academic content
Verbal SDs:Continue level 1-5 activities, including verbal variation; Multiple verbal SDs evoking multiple verbal responses; (the following Multiple, two, three and four component verbal stimuli are verbal conditional discriminations)
Multiple items; Two component multiple adjective-noun verbal stimuli; Two component multiple verb-adverb verbal stimuli; Two component multiple preposition-noun verbal stimuli; Two component multiple pronoun-noun verbal stimuli; Three component multiple verbal stimuli; Four component multiple verbal stimuli; When questions; How questions; Academic content questions
Nonverbal SDs: Similar stimuli in a large array (7 or more stimuli); Small stimuli in a large array of 7 or more stimuli; In-context (NET) stimuli
Data Collection: Same as level 3,4,5
Teaching Objective: Beginning academic development, teaching complex multiple conditional discrimination; Much more development of natural environment verbal conditional discrimination in combination with nonverbal conditional discriminations. Reach the kindergarten academic, social, and linguistic level.
Dr. Sundberg also had a section and went over Intraverbal Task Analysis and Levels but for the sake of the fact my notes here are already 8 pages long, I would be more then happy to fax you this section if you e-mail me. (Or you can just harass Dr. Sundberg to get his revised publications out there! :0)
Can you believe? All of this info and we also got Dr. Sundberg’s chapter from the not yet available new version of the Cooper, Heron, and Heward Applied Behavior Analysis 2nd edition, , a glossary of technical terms, as well as a booklet “The Benefits of Skinner’s Analysis of Verbal Behavior for Children with Autism” (reprinted from Behavior Modification Volume 25 No.5, October 2001 pp. 698-724).
In summary, he gave some ideas and video exemplars on how to address much of what is discussed here. Ideas including a textual wheel for aiding in SD discrimination (Visual control is easier to get then Verbal –auditory- control); Offering choice prompts of many items when you realize that the child is only manding for one thing when others aren’t present; If you don’t have an EO, you don’t have a mand (My, jennica’s note~if the child is manding without an EO, that is just a request~you know, how often do you do that? You ask someone how they are, do you really want to know? Do you really have an EO for that info? :0)…How for some learners, Prompt level data may be more significant then acquisition data; Perhaps when the learner isn’t getting RFFC’s, pull back an little and focus on more receptive skills like scanning and M-T-S; Establish strong and generalized mand, tact and receptive repertoires prior to moving onto multiple responses and complex concepts (like adjectives, pronouns etc); Use typical language development as a guide for advancement in the programming; Choose your battles with correction procedures (pay more attention to which behavior you are targeting); Basically, always conduct Verbal Behavior Analysis of language barriers encountered along the way.
Training should be made as FUN as possible by:
Using lots of mand trials; pairing yourself with free reinforcement; errorless learning procedures; Vary pacing, materials, setting, tone of voice, people, context etc; minimize the use of aversives and punishment; REMEMBER, THEY ARE KIDS!!!!!
COMMON MISTAKES AND TEACHING ERRORS:
Failure to do mand training; 2.Failure to have an EO at strength (must vary reinforcers); 3.Not enough trials, not enough mands, no variation, too much focus on other repertoires;4.No instructional control; 5.Child is prompt bound ; 6.Reinforcement is provided before behavior; 7.Failure to actively shape successive approximations; 8.Failure to actively use transfer of stimulus control procedures; 9.Selecting the wrong words/concepts/parts of speech to teach; 10. Use of aversive control in training (CEO-R~ learned EO’s the reflexive type) 11. Failure to implement generalization procedures 12.Not consistently bringing known responses under new forms of stimulus control 13.Failure to do a VB analysis of verbal problems/barriers (eg. Rote responding) 14. Teaching out of developmental sequence 15.Teaching verbal responses that are useless to the child (my –jennica’s note~ SOO IMPORTANT keep it meaningful and functional for the child!) 16.Staying at one level too long 17. NO gradual change in the SD, Response, or Consequence 18. Failure to work together with all involved with the child 19.Too rigid of a teaching style 20. Failure to realize that learners are very young KIDS!! 20. Failure to realize that learners are very young KIDS!! 21. Assuming praise is a reinforcer 22.Allowing escape and avoidance related to language tasks 23. Accidental reinforcement of negative behavior 24.Failure to require VB, especially in the natural environment 25.Failure to sign TO a signing child 26.Moving ahead too quickly 27.Training becoming aversive
Ok guys, that’s it, I’m out!
Take care, if Dr. Sundberg is in your area, take advantage to learn from the source! ELIJA, thank you so much for bringing him to LI!
Jennica 1/17/05
For more info about publications and other info go to: www.behavioranalysts.com