Unit 2 Flashcards
Skinner on. Functional Analysis
- The external variables of which Bx is a function provide for what maybe called a casual/ functional analysis
- The relation btw bx & the Env provide “a comprehensive pic of the organism as a Behaving system”
Functional Analysis
- Process of identifying Environmental events that are Functionally related to Bx by systematically manipulating environmental variables
- AKA experimental assessment
Functional Analysis (con…t)
- Assessment in which:
- Antecedents &/or consequences are experimentally manipulated
- Bx is measured under these different conditions
- Functional relations are identified through this systematic manipulation
- Can take many different forms
Functional Analysis Purpose
- Identify Functional relations btw Bx & the Env
- Identify maintaining variables for Bx
Functional Analysis Assumptions
- Problem Bx is:
- Learned
- A function of environmental contingencies
Learned Functions of Bx
- Positive reinforcement
- Socially mediated
- Automatic
- Negative reinforcement
- Socially mediated
- Automatic
Functional Analysis Characteristics
- Direct observation
- Manipulation of antecedent & consequent conditions
- Measurement of Bx under test & control conditions
Functional Analysis E.g.
*Is Bx (screaming) maintained by socially mediated positive Rx in the form attention (reprimands) ?
-Condition A (test):
Scream - attention (reprimand)
-Condition B (Control):
-Scream - No attention
*Higher levels of Bx in test when compared to control suggests attention may be a reinforcer for this bx
Functional Analysis as a Process
- Functional analysis, also known as experimental assessment is NOT a prescribed set of procedures!
- It is a process of identifying independent variables that are Functionally related to Bx
———–reinforcement results from the addition of a(n)———– stimulus after the behavior.
- Positive
2. Appetitive (Aversive)
Lovass et al. (1965)
- Investigated social positive Rx (attention) on self-injurious Bx (SIB)
- Results: Rates of SIB increased when contingent attention was provided for Behavior
- Function of SIB: Socially mediated positive reinforcement
Carr et al. (1976, 1980)
- Functional analysis of antecedents
- DVs: SIB & GGRESSION
- IV: Antecedent demand Vs. no demand
- No programmed consequences following Behavior
Carr et al. (1976, 1980): Results
- SIB & aggression higher in demand condition
- Suggests prob Bx maintained by social negative Rx (escape from demand)
- Not confirmed cuz consequences not manipulated
Berkson and Mason (1963, 1964)
- DV: Stereotypy
- IV: Enriched (Activities present) Vs. impoverished (no activities present) env
- Results: Stereotypy higher in impoverished env
Early Studies: Common Ground
- Evaluated function through systematic manipulations
- Demonstrated control by a single contingency
- Took a Functional approach to the evaluation of problem Bx
A-B Functional Analysis
*Functional analysis in which one the antecedents to Bx are manipulated
A-B Functional Analysis Goal
-Identify antecedent conditions to Bx
A-B Fun tonal Analysis Model Development
Carr & Durand (1985)
- Manipulate antecedents only
- No programmed consequences for problem bx
A-B Functional Analysis Conditions
- Test condition= Condition in which the specific EO for an individual Reinforcer is present
- Control condition = Condition in which the specific Eos for tested reinforcers are absent
- Responding in test conditions is compared to responding in control
Remember
- Functional analysis is a process, NOT 1 specific set of procedures
- However, the term “Functional analysis” is often associated with a procedure first described by Iwata et al.
Analog Functional Analysis (FA)
- Assessment in which antecedents & consequences consistent with the 4 main functions of Bx are manipulated systematically in different conditions & the effect on Bx is measured
- AKA “Standard” functional analysis
Analog Functional Analysis (Con..t)
*Iwata et al. (1982/ 94) Created the general model for concurrently assessing sensitivity to different contingencies
Analog Functional Analysis Purpose
*To identify functional relations btw Bx & the Environment
Functional Analysis: When to Use
- When descriptive assessment does not yield enough info to form reasonable hypotheses
- When intervention based on descriptive assessment is not producing the desired change
- When much time & many resources will go into training staff on the intervention procedures
Analog FA in the Literature: Iwata et al. (1982/ 1994)
- 9 participants, all diagnosed with developmental disabilities, who engaged in varying topographies of Self-injurious behavior (SIB)
- DVs: SIB recording using partial interval recording
- IVs : Structured env conditions
Analog FA in the Literature: Iwata et al. (1982/ 1994)
Cont…d)
- Each test condition contains EO & Sd for Reinforcement
- Specific consequence follows prob Bx in test conditions
- Often conducted in analog setting
- Responding in test conditions compared to control condition
Analog Functional Analysis Experimental Design
- Typically multielement design
- Multiple test conditions repeatedly administered
- Session are typically 10 min in length
- Prob bx response rates measured in each condition
Multielement Design
- Experimental design in which there is rapid alternation of 2 or more independent variables or IV values
- Each session is a different condition
- Visually depicted on a line graph with each data path representing a different condition
Analog Functional Analysis Conditions
- Play (serves as control)
- Attention
- Escape (sometimes called “task demand or demand)
- Alone (sometimes called “ignore”)
- Tangible (not always included)
Learned Functions of Behavior
- Positive Reinforcement
- Socially mediated
- Automatic
- Negative reinforcement
- Socially mediated
- Automatic
Purpose of Conditions in a Standard Functional Analysis
- Play - Control
- Attention - Test for social positive reinforcement (attention)
- Escape - Test for Social negative Rx
- Alone - Test for automatic Rx
- Tangible - Test for Social positive Rx (specific item)
Interesting Facts About Functional Analysis of Problem Behavior
- We want the prob bc to occur
- We set-up the env so that it is likely to occur
- We deliver Rx contingent on the target bx
- We ignore all other bx (even appropriate bx)
Play Condition
- AKA: Control
- Not one of the “test” conditions
- Serves as the control
- Condition to which rates of bx from other FA conditions are compared
Play Condition (Con..t)
- Contingency tested: No contingency tested; serves as control to which test conditions are compared
- Antecedent: No Eos for Sr+ or Sr-
- No demands (AO for Sr-)
- Noncontingent attention (AO for Sr+)
- Enriched Env (AO for Sr+)
- Consequence: No programmed consequences for target Bx
Running a Play Condition: Antecedents
- Attention freely available
- Reduces EO for attention
- Tangibles/ Activities freely available
- Reduces EO for “things” & activities
- No demands
- Reduced EO for Escape
Play Condition Steps
- Start timer
- Present preferred items and initiate play (E.g. “Here are some fun things”)
- Remain within a few feet of the client and respond to hands or intraverbals related to attention
- Provide vocal/ physical attention on a fixed-time schedule
Play Condition Steps (Con…t)
- Ignore instances of target problem Bx
* Insert short delay (e.g. 5s) for attention delivery of target Bx occurs at the end of the FT interval
Identify the characteristics of the play condition in a standard functional analysis:
(check all that apply)
- No task demands are presented
- Attention is given on a time-based schedule
- Preferred items are available
- Serves as a control
Attention Condition
-Contingency tested: Social positive Rx in the form of attention
-Antecedent: Therapist present but not provided attention
-Consequence: Attention provided contingent on target Bx
E.g. Statement of concern, reprimand
Running an Attention Condition:
Antecedents
- Want EO for attention to be high
- Therapist engages in another activity
- Therapist ignores all bx, except the target Bx
Running an Attention Condition: Consequences
- Consequence for all problem bx is attention
- Attention is provided, if & only if problem bx occurs
- No attention provided for any other bx
Attention Condition Steps
- Client & therapist in room with moderately preferred items
- Start timer for session
- Therapist says, “I have some work to do” & moves away from client
Attention Condition Steps (Con..t)
E.g. of attention: Reprimands, Reactions or facial expressions, comfort or motivational statements, physical touch, eye contact
- Contingent on target bx, therapist provides 3-5 sec of attention
- After attention delivery, therapist returns to their work material
- Note: All non target bx is ignored
During the attention condition of a functional analysis, the therapist should:
(check all that apply)
- Provide attention contingent on the target bx
- Ignore all other behaviors
- Start by providing no attention
Escape Condition
- AKA: Demand or task demand
- Contingency tested: Social negative Rx (removal of demand)
- Attention: Demand presented e.g. task “do this”
- Consequence: Removal of demand contingent on bx (Escape)
Running as Escape Condition:
Antecedents
-EO for escape should be high
Therapist presents demands over & over
-Noncompliance followed by 3-step guided compliance
-Therapist ignores everything the client says/ does, unless prob bx occurs
Escape Condition Antecedent
- Should mimic real life:
- Academic tasks
- Vocational tasks
- Daily living skills
Running an Escape Condition:
Consequences
- Consequence for problem bx is a brief escape from the demand
- All materials removed, stop delivering demands
Escape Condition Steps
- Start timer for session
- Repeatedly present moderately challenging tasks
- Praise for correct response (e.g. “yes”)
- Guided compliance
Escape Condition Steps (cont..d)
- Contingent on target bx, remove materials & turn away for a fixed time after the end of the prob bx
- Note: All non target bx is ignored
Identify the characteristics of the escape condition in a standard functional analysis:
(check all that apply)
a. Starts with the therapist presenting a demand
b. Demand is removed contingent on the target bx
c. Demands continue for all other bxs
Alone Condition
- AKA: Ignore
- Contingency tested: Automatic Rx (removes social consequences)
- Antecedent: No stimuli (austere env)
- Consequence: No programmed consequence for Bx
Running an Alone Condition: Antecedents
*Improverished Env:
E.g. low preferred or no toys, food, activities, attention
*Therapist absent from room (or in room, but ignoring)
*Safety note: Client observed the whole time
Running an Alone Condition:
Consequences
-No programmed consequences
Alone Condition Steps
- Bring client into barren room then leave e.g. I will be back soon, you stay here
- Start timer for session
- Return at the End of the session
Note: Observe client for safety throughout the session
Tangible Condition
- Not always included
- Contingency tested: Social positive Rx (preferred item)
- Antecedent: Removal of preferred item
- Consequence: Preferred item delivered contingent on target Bx
Running a Tangible Condition:
Antecedents
- Want EO for items to be high
- Therapist has client’s preferred item(s) but withholds them
- No access to items unless prob bx occurs
Running a Tangible Condition:
Consequences
- Contingent on prob bx, tangible item is delivered
- Brief access provided & then removed e.g. therapist say “My turn” and removes item
- All other Bx is ignored
Tangibles Items/ activities
-Use items identified in a descriptive assessment
E.g. Foods, Toys, Table-top activities
Tangible Condition Steps
- Prior to beginning session, provide preferred item to client
- Start session timer
- Remove tangible items
- Contingent on target Bx, provide brief access to the tangible items
Note: All nontarget bx is ignored
Identify the characteristics of the “tangible” condition in a standard functional analysis:
(check all that apply)
- Starts with removal of a preferred stimulus
- Item is presented contingent on target bx
- Tests for Bx maintained by access to a tangible stimulus
What is a relevant feature of the antecedent in the “tangible” condition?
-Removal of the stimulus may function as an EO for positive Rx
How do therapists determine the type of stimulus to provide during the tangible condition of a functional analysis?
“Completing a direct descriptive assessment can help determine which stimulus to provide”
Tyler’s target behavior is SIB. The therapist is running a functional analysis. Each time Tyler engages in SIB, the therapist says, “Don’t do that, you will hurt yourself.” This is an example of which condition?
- Alone
- Escape
- Attention
- Attention
The therapist wants to assess Tara’s hand-flapping behavior. She puts Tara in a room with no toys, food, or activities and sits in the corner giving no attention to Tara. This is an example of which condition?
A. Alone
Bobby engages in aggressive behavior at school. The therapist places Bobby in academic task demands and removes the task each time he is aggressive This is an example of which condition?
Escape
Selecting Topographies to Target in Analog Functional Analysis
- Topography: Form of the response
- Many clients engage in multiple forms of prob bx
- Topographies Maybe evaluated together or separately
Evaluating Combined Topographies
- Multiple topographies maybe evaluated together
- All identified topographies result in the consequence of the condition E.g. In demand condition escape provided for SIB, aggression or property destruction
Selecting Topographies to Target in Analog Functional Analysis
- Topography: Form of the response
- Many clients engage in multiple forms of problem bx
- Topographies may be evaluated together or separately
Evaluating Combined Topographies
- Multiple topographies maybe evaluated together
- All identified topographies result in the consequence of the condition
- E.g: In demand condition escape provided for SIB, aggression or property destruction
Evaluating Combined Topographies: Advantage
*Most efficient to evaluate topographies together (i.e. reinforce multiple topographies)
Evaluating Combined Topographies: Limitations
- Problematic if different topographies have different functions
- Bx may conform to a hierarchy so other topographies are not seen
- May require another type of analysis
Evaluating Combined Topographies: Recommendations
- Only reinforce multiple topographies if descriptive assessment suggests the same function for all topographies
- If necessary, place topographies on extinction until all Topographies are seen
Evaluating Topographies Separately
- Potentially less efficient, but may provide clearer results
- Separate Topographies that are suspected to have different behavioral functions
- Most disruptive topography first
Selecting Topographies: Summary
- Most efficient to focus on all topographies
- Always graph each topography separately
- If necessary, place some topographies on extinction until all topographies emerge
What are the limitations of including multiple response topographies in a single functional analysis?
- May not observe all topographies during the assessment
- Potentially less efficient
- Different topographies could have different functions
- Behavior may conform ti a hierarchy
When analyzing topographies separately, which topography should be selected first?
-The one with the greatest social validity
Functional Analysis Procedural Guidelines: Pre-session
- Rule out biological or medical events/ causes
- Limit pre-session access to reinforcers
Selection of Materials
- Determination from ur indirect & descriptive assessment
- Play (Control): Items they like
- Demand: Aversive tasks that mimic natural demand context
- Tangible: Items the individual likes, but that are not of really high value
Sequence of Sessions
- Options to consider:
- Random order
- Fixed sequence
- Alone, attention, play, demand
Identify the following things to consider before choosing to conduct a functional analysis:
(check all that apply)
- Rule out biological or medical issues
- Limit access to reinforcers that will be included in test conditions
Functional Analysis Experimental Design Options
- Multielement
- Reversal
- Pairwise
Functional Analysis Experimental Design: Multielement
- Most common
- Rapidly alternate btw conditions
- Possible lack of discrimination btw conditions
Functional Analysis Experimental Design: Reversal
- One condition run at a time
- Time-consuming so less common
- May help with discrimination issues
Functional Analysis Experimental Design: Pairwise
- Fairly common
- More efficient than reversal
- One test condition alternated with control
- May assist in discriminability of conditions
What is a limitation of using the multielement design?
Quickly establishing discrimination btw conditions may be difficult
FA Procedural Guidelines: During Session
-Arrange consequences for a single topography when possible
*Enhance MO when possible
Condition sequence
Programmed precession deprivation
FA Procedural Guidelines: During Session (Con..t)
-Use different Sds in each condition to enhance discriminability e.g different rooms, therapists, shirts
-Allow prob bx to subside btw sessions
Reduces potential carryover effects
When analyzing the data from a functional analysis, how do we determine the maintaining variable?
-Conditions with high rates compared to the control are likely maintaining bx
Collecting Data
- Who should collect data?
- Someone that is trained but is not conducting the session
- Primary data collector, secondary data collector for IOA
Paper & pencil data collection advantages
- Easy to train staff
- Data sheets are low effort to create, low cost &accessible
- Allows for flexibility
Paper & pencil data collection
Disadvantages
- Must be designed
- Must be printed
- Must be stored
- Must be transferred to some other format for analysis
Electronic Data Collection Advantages
- Can be stored & analyzed electronically
- Can be recorded at the time it occurs
- Can record a wider range of data & variety of behaviors
- Improved & faster communication with staff members
Electronic Data Collection Limitations
- Can be expensive
- May not capture exactly what you are intending to (lack of flexibility)
- Technical difficulties
Variables to Measure
- Target bx(s): Separate Topographies
- Antecedents & consequences (onset-offset)
- Nontarget appropriate or problem bx
Dimensions to Measure
- Continuous recording is preferred as it captures every instance of the target bx
- Continuous recording methods typically used in FAs:
- Rate
- Latency
Data on different topographies should be recorded separately so they can be analyzed separately.
True
Identify the continuous measures commonly used in a functional analysis:
- Latency
- Rate
Analog Functional Analysis Graphs
- Equal-interval line graphs
- Typically, multi element design
- Reversal or pairwise also options
- Multielement: Each condition is represented by a separate data path
- Measure of problem bx on y-axis, sessions on x-axis
Separate Vs. Combined Topographies
- Ideally separate functional analyses for each topography
- Often not practical
- Graph together only if suspected to be maintained by same contingency
- Generally speaking, better to graph separately
Alternatives to Separate Functional Analyses for Each Topography
- Conduct one functional analysis of multiple topographies, graph as aggregate & separate
- Reduces time, cost and inconclusive results
- Conduct an extinction analysis
Aggregate Graphing
*Combining data from multiple response topographies & graphing them as a single dependent variable
Which type of graph is used to display data collected in a functional analysis?
-Equal-interval line graph
Derby et ak. (2000): Recommendation
- Graph each Topography separately in addition to aggregated
- This could improve accuracy of visual analysis interpretation
Extinction Analysis
Modification of an analog functional analysis in which extinction is provided for the predominant response & Rx effects are measured in other responses
Extinction Analysis in the literature: Richman et al. (1999) Phase 1
- Analysis of prob Bx in response class hierarchies
- Phase 1: Standard Functional analysis conducted for 3 participants
- All Topographies would have produced reinforcers
- Only identified functions for less severe topographies
Extinction Analysis in the literature: Richman et al. (1999) Phase 2
- Only 1 topography was reinforced, all others placed on Extinction
- If reinforcer was not provided right after initial/ mild problem bx, the client would try a different (typically more severe) topography of prob bx
Take-Home Points
- When consequences are provided for multiple topographies of problem bx, interpretation can be difficult
- It is best to reinforce only 1 topography per functional analysis when possible; if this is not practical it is recommended to always Graph topographies separately
The Language We Use and What it Means
- “The bx is maintained by attention”
* “Attention is the function of the bx”
Attention
- Attention can function as a type of socially mediated positive reinforcement
- Behavior results in access to attention from others
Attention as a Function:
*Problem bx maintained by attention maybe likely to occur:
- When bx not getting enough attention (infrequent attention)
- When others are preoccupied (divided attention)
- When bx lacks functional communication (difficulty communicating wants & needs)
Escape
- Type of socially mediated negative Rx
- Bx results in the termination of something aversive
Escape as a Function
- The Bx terminates
- Unwanted social situation(s)/ ppl
- New, non-preferred, or difficult tasks or activities
- Appropriate responses ignored or delay to termination
- Lacking appropriate communication methods
Automatic Reinforcement
- Tested by alone/ ignore condition which rules out social reinforcement
- Type of automatic positive or negative Rx
- Does not depend on the actions of others
- Some sensory stimulation is produced or removed as a result of the bx
Automatic Reinforcement as a Function
- Appetitive stimulus produced (feels good)
- Aversive stimulation is terminated
- Problem Bx may produce/ remove sensations
Tangible
- Type of socially mediated positive Rx
- Bx results in access to “stuff”
Tangible as a Function
- Appropriate requests don’t receive the reinforcer quickly or reliably
- Appropriate requests not in the repertoire
Visual Analysis of Analog Functional Analysis
- Line graphs used to visually compare test condition with the control condition
- Controlling variables when the rate of Bx in the test is consistently higher than control
- Looking for separation
Identifying the Learned Functions of Bx:
Looking for level of bx elevated in 1 or more test conditions when compared to bx in the control
- Attention
- Escape
- Tangible
- Automatic Rx
- Some combination of these
We would say that the behavior is maintained by socially mediated negative reinforcement when the data path is elevated in the
——–condition compared to control.
-Escape
If behavior were maintained by socially mediated positive reinforcement in the form of a juice box, we would expect to see an elevated line in the———-condition compared to control.
-Tangible
We would expect the data path in the _____ condition to be elevated compared to control if the behavior were maintained by automatic reinforcement
-Alone
Undifferentiated Response Patterns
-Lack of differences in level of responding btw conditions
Other Possible Causes of Undifferentiated Response Patterns
- Discrimination difficulties
- Relevant variables not tested
- Hierarchical responding
multiple control
when bx is maintained by more than 1 function, Can be evidenced in a functional analysis by high rates in comparison to control in more than one condition
Visual Analysis Summary
- Compare control to test conditions
- Condition(s) elevated relative to control indicative of Function
- Automatic Rx potentially indicated if responding:
- Elevated in alone conditions
- Undifferentiated across all conditions
- Can be more than 1 function
Hagopian et al. (1997)
- Many studies have shown visual inspection to be subjective
- Purpose: To develop structured criteria for visual inspection of multielement functional analysis data results
Modified Visual-Inspection Criteria
- Modified the structured criteria from Hagopian et al to account for Functional analyses that varies in length
- Converted the visual-inspection criteria to to a % of data points
Modified Visual-Inspection Criteria Primary Rule
*Rule: If 50% or more data points fell above upper criterion line then fell below lower criterion line, participants conclude that the condition was differentiated
