Chapter 15 o Multiple Stimulus Assessment (aka multiple
Differential Reinforcement of Alternative Behavior stimuli without replacement MSWO): several
Differential Reinforcement: use reinforcement and potential reinforcers presented and the
extinction to increase occurrence of a desirable target researcher records which one is approached
behavior or decrease occurrence of undesirable first, then second, then third…
behavior. Assess reinforcer by making each potential one
Differential Reinforcement of Alternative Behavior contingent on an operant response. If frequency
(DRA): used to increase frequency of a desirable or duration increases when stimulus is contingent
behavior and decrease frequency of undesirable on response, the stimulus is a reinforcer (e.g.
behavior. The desirable behavior is reinforced each time pressing a button to hear music more than
it occurs, while undesired ones are not reinforced to pressing button to turn fan on, shows that music
decrease them. is a reinforcer)
When to use DRA: 3. Reinforce desired behavior immediately and
You need to want to increase the desirable behavior consistently.
The behavior needs to be already occurring 4. Eliminate reinforcement for desirable behavior
occasionally. If not, use shaping or prompting to evoke 5. Use intermittent reinforcement to maintain the
it, and then use DRA. target behavior (continuous is only early on). This
You need to have a reinforcer that can be used each makes the behavior more resistant to extinction.
time the desirable behavior occurs. If you have no 6. Program for Generalization. It should be reinforced
control on reinforcer/have no reinforcer, then you in as many relevant situations as possible.
can’t use DRA Different variations:
Steps: Differential Reinforcement of incompatible behavior
1. Define desirable and undesirable behaviors. It (DRI): alternative behavior is physically incompatible
helps to determine whether treatment is successful with problem behavior, so the 2 can’t happen
2. Identify reinforcer. You must be able to determine together. E.G: to prevent the problem behavior of
it because they may vary across people. Ways: head-slapping, the researchers can reinforce any other
Reinforcer can be what reinforces the undesirable behavior that needs the hands
behavior. Differential Reinforcement of Communication (aka
Observing/asking the person (Premack principle: functional communication training): individual with
use high-frequency behavior as a reinforcer for a problem behavior learns to make communication
responses that are functionally equivalent to problem
Preference Assessment: try a variety of stimuli behavior. E.G: when a person engages in problem
and see which ones are reinforces to the person, behavior to get attention, he can instead ask for
and then use reinforcer assessment (to attention without engaging in that bad behavior, since
determine that item is the reinforcer, deliver it both will produce the same end result.
contingent on behavior and show that the Differential Reinforcement of Other Behavior
behavior increased). Ways to conduct: Differential Reinforcement of other behavior (DRO):
o Single Stimulus assessment: each potential reinforcer is contingent on the absence of the problem
reinforcer is presented, one at a time to see behavior. Although the name suggests that you will
which one will be reached. The number of reinforce another behavior, you are actually just
times the reinforcer has been approached is reinforcing the absence of the problem behavior.
recorded, and the one with most is it. Steps:
o Paired Stimulus assessment (aka forced 1. Identify reinforcer for the problem behavior; you
choice): 2 potential reinforcers are presented need to use functional assessments. Like DRA, if
and the one that’s reached is recorded. you can’t use extinction for the problem behavior,
you shouldn’t use DRO (exception: if reinforcer for absence is more powerful than reinforcer for Chapter 16
existence of problem behavior). Defining Antecedent Control Procedures
2. Identify reinforcer to use in DRO procedure. Antecedent control procedures: antecedent stimuli are
3. Choose the initial DRO time interval. As the manipulated to evoke desirable behaviors so that they
frequency of problem behavior decreases, DRO can be differentially reinforced, and to decrease
intervals get lengthened gradually. undesirable behaviors that interfere with the desirable
4. Eliminate reinforcer for problem behavior and behaviors. They involve manipulating some aspect of
deliver reinforcer for absence of the problem the physical/social world. Competing behaviors are
behavior. concurrent operants reinforced on concurrent
5. Reset the interval if the problem behavior occurs. schedules of reinforcement (i.e. doing bad in school
6. Gradually increase the interval length. because you’d rather party than study). There are 6
Two types of DRO: whole-interval (whole-behavior different antecedent manipulation procedures, but they
must be absent for the whole interval for the reinforcer should always be used with differential reinforcement:
to be delivered; more effective) and momentary D D
Present the S or cues for the desired behavior (e.g. S
(problem behavior must be absent at the end of the for eating healthily is having healthy food present).
interval; not effective alone, needs to act with whole- Arranging Establishing Operations for the Desirable
interval to maintain the behavior). Behavior. By increasing the reinforcing value of the
Differential Reinforcement of Low Responding Rates consequence of a behavior, you make it more likely to
Differential reinforcement of low rates of responding
occur (e.g. buying a healthy cookbook would make it
(DRL): reinforcer is delivered contingent on a lower rate more likely to eat healthily because your food would
of responding during a period of time. be tastier).
Variations: full-session DRL (reinforcement is delivered Decreasing Response Effort for the Desirable Behavior.
if fewer than a specified number of responses occurs in So, arrange antecedent conditions that require less
Increase Desirable Behavior
a period of time), spaced-responding DRL (specified effort to engage in the behavior to make it more likely
amount of time between responses for the reinforcer to for that behavior to occur (people are inherently lazy).
be delivered), and interval DRL (dividing a session into
Remove discriminative Stimulus for undesirable
intervals and providing the reinforcer if not more than behavior (e.g. presence of unhealthy food is a stimulus
one response occurred in each interval; similar to for eating unhealthy foods)
spaced). Presenting abolishing operations for undesirable
Steps: behaviors. This isn’t always possible. For example,
1. Determine whether DRL is appropriate procedure.
when grocery shopping, you’re more likely to buy junk
Use it if you want to decrease but not eliminate. foods when you’re hungry than when you’re not. So to
2. Determine acceptable level of behavior (how many eat healthily, eat before grocery shopping, so your
responses per session, interval of time between desire to buy junk food is eliminated.
each behavior…) Increase response effort for undesirable behaviors. If
3. Decide whether to use full-session or spaced (this is
competing behaviors take more effort, they’re less
more appropriate if time matters vs. just likely to interfere with desirable behavior. First you’d
decreasing overall rate). have to remove undesirable behavior then make it
4. Inform client about procedure so that she knows harder to get to again.
about schedules Research on Antecedent Control Strategies
5. Implement DRL procedure and give instructions as
Antecedent control strategies can be conducted by
well as feedback. manipulating discriminative stimuli (social or physical
Stereotypic behavior (aka self-stimulatory behavior): environment), response effort, and manipulating
repetitive behavior that does not serve any social motivating operations (i.e. eliminating escape by
function for the person. making tasks less aversive so escape would no longer be
Interresponse Time (IRT): time between responses.
reinforcing). Using Antecedent Control Strategies take the people to. The time-out room must be barren
Differential reinforcement and extinction are used in (except from one chair), well-lit, and without locks. (3) Is
conjunction with these strategies. the room safe? (4) Is the time-out period brief? Time-
1. Identify and define desirable/undesirable behavior out should be brief, but if the problem behavior
that you want to change (can response effort be persists, continue with contingent delay (extend the
manipulated?) time period by 10 sec to 1 min). (5) Can escape from
2. Analyze antecedent situations related to the time-out be prevented? If refraining the escape would
desirable/undesirable behavior. not be successful, don’t use time-out, because then it
3. Identify reinforcer for desirable/undesirable behavior would be pointless since the escape will negatively
These control strategies, alongside differential reinforce the aggressive/escaping behavior. (6) Can
reinforcement and extinction are known as functional interactions be avoided during time-out? Time-out
interventions (they’re functional because they decrease should be implemented without emotions from the
problem behaviors and increase desirable ones by change agent. (7) Is time-out acceptable here?
modifying the A and C). These are non-aversive (since Contingent Observation: contingent on the occurrence
they don’t depend on punishment). They’re always the of the problem behavior, the child has to sit and watch
1 step to treatments because the address function of the other children play appropriately.
behavior (consequences) and the antecedents (what Response Cost
evokes it) Response Cost: removal of a specified amount of a
reinforcer (usually money or a privilege) contingent on
Chapter 17 the occurrence of a problem behavior. It’s a negative
Punishment is implemented only after functional punishment procedure. Differential reinforcement
interventions (differential reinforcement, extinction or should also be used in conjunction.
antecedent control) have been tried. The processes used to decrease problem behaviors are:
Time-Out extinction, time-out and response cost. With extinction,
Time-out (from positive reinforcement): being removed problem behavior is not followed by a reinforcing event
from a reinforcing activity for a few minutes contingent that had maintained the behavior before. With time-
on the instance of the problem behavior. Types: out, the person is removed from access to sources of
Nonexclusionary time-out: can stay in room, but has reinforcement when problem behavior occurs. With
to sit across from where the other people who are response cost, a reinforcer is removed (by quantity)
participating in the positively reinforcing behavior are; after problem behavior.
used when the person can be removed from the Considerations: (1) Which reinforcer will be removed?
activity without the need of leaving the room, and if (2) Is the reinforcer loss immediate or delayed?
his presence isn’t disruptive. Reinforcers can be delayed and would still be effective
Exclusionary time-out: taken out of the room, so that only if you present a verbal statement that they will lose
reinforcer is not available. the reinforcer soon. Delays are usually not appropriate
Time-out should always be used with differential for people with intellectual disabilities. (3) Is the loss of
reinforcement; otherwise the behavior will re-emerge reinforcers ethical? (4) Is response cost practical and
after the treatment session. acceptable?
Considerations: (1) what’s the function of the problem Chapter 18
behavior (time-in environment: where the problem Application of Aversive Activities
behavior occurs)? It’s not appropriate to use time-out Punishment through application of aversive activities:
for negative reinforcement (sensory stimulation or contingent on the problem behavior, a person will be
automatic reinforcement). (2) Is time-out practical made to engage in an aversive activity)low-probability
here? So, the change agent may needs to be able to behavior the person would not choose to engage in).
control the person, and there must be a room (for This follows the Premack principle that when high-
exclusionary time-out) without positive reinforcers to probability behavior is followed by low-probability ones, the high-probability behavior decrease in occurrence. Physical Restraint: contingent on problem behavior,
Aversive activities are punishers in the form of change agent holds immobile the part of the client’s
behaviors. The change agent is usually required to use body that’s involved in the behavior, such that agent is
physical guidance to keep the person working on the physically restrained from doing problem behavior.
aversive activity, but eventually, the person will engage Some people have physical restraint as a punisher,
in the activity on command to avoid physical guidance. others as a reinforcer, so it`s important to know which
Types of Aversive Activities: (physical restraint will only work for those that see it
Overcorrection: developed to decrease aggressive and as a punisher). This can be used with pica behavior
disruptive behavior for people with intellectual (ingesting non-food items).
disabilities. Here, the client is required to engage in an o Response Blocking: change agent prevents
effortful behavior for an extended period contingent occurrence of a problem behavior by physically
on each instance of the problem behavior. 2 types: blocking the response.
o Positive Practice: client has to engage in correct Application of Aversive Stimulation
forms of relevant behavior contingent on an Bruxism: behavior in which a person grinds upper and
instance of problem behavior (the student who lower teeth together.
spells a word incorrectly is required to spell it out Aversive Stimulation: delivering aversive stimuli after
10 times). problem behavior to make it less likely to occur. Types
o Restitution: procedure in which, contingent on of aversive stimulation: Electric shocks, aromatic
each instance of the problem behavior, the client ammonia (used for self-injurious behavior by breaking
must correct the environmental effects of problem an ammonia capsule and waving it under client`s nose –
behavior and restore environment to a condition like TV show Dexter), spray mist of water on face, facial
better than that which existed before (when a girl screening (touch the face), alarm sounds, and
scribbles on one house wall, restitution would be reprimands (these decreased problem behaviors in
to make her clean that wall, plus an extra one). those who received them and those who didn’t).
Contingent Exercise: client is made to engage in some Positive Punishment (Last Resort)
form of physical exercise contingent on an instance of Positive punishment procedures are used only as a last
problem behavior. It’s different than overcorrection resort, after functional interventions have been used.
because in overcorrection, the aversive activity is Negative punishment (time-out and response cost)
based on what the problem behavior is (you have to procedures are more accepted and common than the
restore/perform the correction of behavior). For positive ones.
contingent exercise, you perform an unrelated Considerations in Using Positive Punishment
exercise that the client is capable of (e.g. to stop a Use functional interventions first.
child from swearing in front of family, each time he Implement differential reinforcement w/ punishment
swore, his father would make him clean windows for Consider function of problem behavior (for example,
10 minutes). positive punishment through applications of aversive
Guided Compliance: person is guided physically activities would not work if the problem behavior’s
through the requested activity contingent on the occurs to get attention, unless change agent doesn’t
occurrence of the problem behavior. To use this, display any emotions).
physical guidance must be a punisher for the client; Choose aversive stimulus with care.
otherwise, it shouldn’t be used. It serves 2 functions: Collect data to make treatment decisions (punishment
(1) it’s a positive punishment of problem behavior procedure should produce a rapid decrease in
(aversive situation happens) and (2) negative problem behavior.
reinforcer for compliance (when you comply, aversive
Address ethical considerations in use of punishment.
situation is removed). It’s equally effective as time-out Ethics of Punishment
in children. The client/parents must fully understand punishment
procedure, rationale, how/when it’ll be used, intended effects or side effects, and possible alternatives. stimulus exemplars it will take to generalize the
Informed consent must be given by adult in charge of behavior. One strategy is with general case
client (client himself, parent, or legal representative). programming (using multiple training examples that
Punishment should never result in harm, and should be sample the range of relevant stimulus situations and
reserved for more severe problems. There must be response variations).
written guidelines to eliminate ambiguity. The Incorporating Common Stimuli: put stimuli from
punishment procedure must be peer-reviewed (before generalization environment (target situation) into the
or after implementation). Also, everybody involved in training. So, use physical stimuli in the training
implementing the procedure must be held accountable alongside the stimulus exemplar/situation.
for its success. Teaching a Range of Functionally Equivalent
Chapter 19 Responses: teaching client a variety of responses that
Programming for generalization increases likelihood all achieve the same outcome (functionally equivalent
that behavior change will occur in all relevant situations responses are the responses that produce the same
Generalization: occurrence of behavior in presence of Incorporating Self-Generated Mediators of
stimuli that is similar in some way to the S that was Generalization: mediator of generalization is a
present during training, or occurrence of behavior in stimulus that’s maintained and transported by client
presence of all relevant stimuli outside training situation as part of the treatment (e.g. parents can go to a
Strategies: Promoting Behavior Change Generalization
lecture on parenting skills and take notes. These notes
Reinforcing Occurrences of Generalization: reinforce are self-generated mediators because they can use
the behavior when it occurs outside of training them in a home setting, thus making the lecture
situations in presence of relevant stimuli. One generalizable to the home too). These can also be
drawback is that it isn’t always possible to provide physical mediators (‘good luck charms’ can be used in
reinforcement for the behavior. If you can’t reinforce, the training situation; then when taken to the target
use a different strategy. situation, will help facilitate generalization). Also, self-
Training Skills that Contact Natural Contingencies of recording can be used (when you have a chart for you
Reinforcement: if you can’t provide reinforcement to mark when you perform the target behavior, it will
when behavior occurs, you should make sure there’s a make it more likely for you to perform the behavior).
natural reinforcer present (e.g. teach students to Also, self-instruction is a mediator.
‘recruit’ attention from teachers by asking “how’s my Implementing Strategies to Promote Generalization
work?” The reinforcement will come from teacher’s Guidelines:
response, and the good academic behavior would Identify target stimulus situations for behavior
have generalized). Sometimes skills are not natural Identi