미국에서

Direct Instruction

• uses assessment of each student's skills and then they get placed into small groups based on their skill level to help the efficiency of learning.
• Direct instruction uses fast-paced teaching to teach more content in less time and help students focus. (teach skills in a shorter time frame)
• Direct instruction consists of 6 guiding principles.
• A foundational principle of the direct instruction teaching method is that all children can learn. #presumecompetence
• Signals are given by the teacher during DI to signal when to participate in a lesson with choral responding - vocally respond in unison #chorus #singinginchorus.

Precision Teaching(PT)

• Ogden Lindsey was the founder of Precision Teaching Method. It is an individualized teaching method that emphsizes fluency, charting performance, and designing and implementing teaching that reinforces the emission of each specific behavior under all conditions in which it's expected to occur. 
• the guiding principle that the student is always right. It does not put the blame on the student, the teacher is the responsible party for making necessary changes.
• uses a semi-logarithmic scale of the y-axis to chart academic and social behaviors.
• PT tracks data on the frequency of responses and accuracy, not percentage of responses. It focuses on building fluency. PT builds fluency which is accuracy and speed of the response rate. It attempts to gain the correct response rate while moving towards the highest rate of responding possible.
• Past behavior changes inform the future learning of the studens in precision teaching. #historyrepeatsitself
• PT uses continuous direct measurement. #frequency #rate

Personalized system of instruction(PSI)

• In PSI, a proctor is a student that is further along in the program, they grade tests and provide feedback for the other students.
• PSI uses lectures as reinforcement and the role of the professor is administative, they do not use lectures to teach material. Professiors lectures' are used as reinforcers for students achieving a predetermined mastery criterion, they are fun and interesting. 
• Example, Julie is in college and is currently in a self-paced ethics course that requires her to receive a score to move to the next lession. This scenario shows the guiding principles of PSI. PSI has 5 guiding principles: "Personalized"/Self-pacing, unit mastery, written materials, proctors, and lectures used as reinforcers/role of the professor #KellerPlan
• PSI wse developed by Fred Keller and was first used in college classrooms but has expanded to other levels like high school. 

DTT

• There are 5 components of DTT: the Sd, prompt, response, reinforcement, inter-trial interval.
• A restricted operant is a controlled operant, that usually occurs during DTT, there must be some condition in place for the behavior to occur. It is when a discrete response occurs, when an opportunity is presented. So, you're teaching one thing at a time in a contrived setting. For example, When teaching a client to tact their colors, each color you teach or ask them to label trial.
• A mass trial is when you expect a target response repetitively when the client is in the beginning stages of learning. We move away from mass trials once they have mastered this type of response to increase generalization. RBT running with same SD in multiple trials.
• A Block trial, which involves: Giving a block of 1 SD followed by a block of another SD. The target bx changes one time during the series of trial.  Ex, touch nose, touch nose, touch head, touch head
• An expanded trial involves adding or inserting distractor trials between the SDs that are currently worked on for acquisition. Make sure the distractor trials are targets that have been mastered already(Think of high-P targets for distracters.
• DTT uses contingecies that are very discriminated to teach in a contrived setting with manipulated variables to yield the correct response.

Incidental Teaching

• Incidental teaching uses MO's to contrive opportunities to help a child build language to ask for items. THis is done is a naturalistic wasy by arranging the environment to limit AOs for the target behavior and emphasize EOs for the target behavior so that the clinet is highly motivated to engage in the desired behaviors. 
• requires the use of objects the amount of learning opportunities in the natural environment. 
• Incidental teaching does not use scripts, it uses naturalistic methods to generalized skills and increase opportunities for learning using preferred stimuli
• Incidental teaching requires the use of objects that are of interest to them to increase the amount of learning opportunities in the natural environment
• Incidental teaching uses an intermittent schedule of reinforcement which help maintain behavior with either ratios or time intervals to deliver reinforcers contingent upon behavior.
• Incidental teaching does not use very discriminated contingencies as the teaching is naturalistic and variables are not manipulated/controlled as stringently. 

Free-operant

• A free-operant is a response class UNLIMITED by constraints or any prompts so that it can be produced freely and can be repeated. There is NO SD required for a response to be emitted. So in this case, the client is allowed to interact with all the items within the park, there are no restrictions posed. 
• Free operant arrangement is kept natural, there is no manipulation of variable in the environment-it is not contrived. There are no SDs presented, as there are no discrete trials.
• Free operant data is measured/reported by the rate of responding and DTT data reports the percent correct of trials.

Discrete trial

• There's a clear distinction in between trails there should be a brief pause, but we don't want to wait too long so we don't lose momentum.
• When conductin  a discrete trial, prompts should always be programmed to be faded out or removed, so the learner can eventually independently or spontaneously produce the response under the conditions procided. It is important that the prompts are temporary.
• The response is the behavior that the client emits after being presented the Sd under the certain conditions set up by the instructor. Responses are instances of behavior.
• Reinforcement should be delivered immediately after the desired target response. 

• Backward chaining occurs when you teach a behavior chain starting with the last step of the chain. Every other behavior in the chain would be prompted/taught and the client will come into contact with the terminal reinforcer from the chain each trial.
• Backwards chanining with leaps ahead has some steps that get skipped because the learner already knows/mastered those steps. It can decrease training time when learning the skill.
• Forward chaining occurs when the learner cannot complete any of the steps in chain, so they are taught the first step first and then the BCBA helps with the rest and as each step in the chain is mastered indepedently, a step gets added until they are able to do the whole chain independently. Advantage of using forward chaning is easy to implement, easy to use in classroom setting, the chaning procedure can be link smaller chains to larger ones.
• Total task chaining occurs when all the steps in a chain are taught to the client to successful complete the task.
• The single opportunitiy method assesses the ability to perform each behavior in the TA in the right order and ends the assessement and chain if the client is unable to perform a step. It is more conservative method. The instructor should stop at the first step that the client cannot complete independently.
• The multiple opportunity method assesses the level of mastery across all the behaviors in the TA. The client will be completing the steps on their own when assessed by the instructor. The client will be completing the steps on their own when assessd by the instructor. It takes longer to complete, but it increases the chances that learning will occur.
• Stimulus variation, response variation, and schedule of reinforcement effect the performance of a behavior chain.
• Response variation is caused by varing stimuli
• When stimulus variation is a factor of performance when using a behavior chain it is important to introduce all variations of the SD.
• The longer the behavior chain, the more time it will take for a learn to learn it.

Purpose of creating and confirming a task analysis

• Task analysis is needed to implement the chaining procedure
• Task analysis is sequence of steps that breaks down complex tasks.
• Each step must be teachable, easily attained, observable. 
• to create a task analysis you can use a few different methods: you can perform the task yourself and see the sequence of steps you needed to complete it, consult with a expert of person that is highly knowledgeable in the particular task, and you can observe a competent individual/expert performing the task and observe the sequence they use. 
• More complex behaviors can be broken down into smaller steps or multiple chains of behavior.
• It determines the sequence the behaviors that is crucial and acceptable need to be ordered in to complete the task efficiently.
• For example, teaching a TikTok dance routine would have a set of dance moves that go in a particular order to complete the whole routine. 

chaining step 

• assess the baseline level of the mastered steps.
• create and confirm
• decide the behavior chain procedure

Shaping across topographies

• is when the topography of the behavior changes during shaping.
• When we reinforce the successive approximations of a response that look different as the client masters each step.
• For example, when you're teaching a baby to hold a bottle, you reinforce them grabbing the bottle, then once that's mastered you reinforce them holding the bottle with one hand, and then once that's mastered you reinfore them holding it with 2 hands. 

Shaping within topographies  

• is when another measurable dimension of behavior is changed/when the topography of behavior remains constant.  when we reinforce the successive approximations of a response that look the same as the client masters each step.
• when we reinforce the successive approximations of the duration of the same response to increase the time spent engaged in the behavior. If the topography of the behavior isn't changing, this is shping WITHIN topographies.
• For example, when you're teaching a child to lengthen the time of their hand washing behavior, you reinforce the behavior for different increasing time intervals until they reach the mastery criterion. 

Successive approximations

• are what we use in a shaping procedure, they are any behavior that is similar to a target behavior, these behaviors are a part of a series that are differentially reinforced toward the end goal of producing the target behavior. 
• When the sequence of new response classes develops during the shaping process as a product of differential reinforcement

• Differential reinforcement is shaping, reinforcing the target behavior you want to see and then fading out the schedule of reinforcement to natural conditions. 
• Response differentiation is a behavior change that is produced by differential reinforcement, you are develiping a contingency for reinforcement of a particular type of response vs another(maladaptive) response.
• Shaping teaches novel behaviors #shapenewbx
• Shaping across response topographies is a type of shaping method. #shaperesponse.
• Chaning criterion during shaping results in new reponse classes due to the different desired behaviors to achieve mastery of each successive criterion.
• Clicker training is used to shape behavior, it pairs a sound with a reinforcer to establish the clicker sound as a conditioned reinforcer when a desired behavior exhibited. 
• Changing the response requirements is to shaping as chaning the antecedent stimuli is to fading.

• Terminal behavior is when the behavior reaches the end goal of the performance you previously determined according to certain criteira. #endgoal
• When the terminal behavior has been achieved, reinforcement should continue to maintain the response (so that the behavior continues to occur)

Rule-governed behavior (AKA rule governance)

• Deadlines in rule-giverned behavior help change the behavior. Deadline imposed upon reinforcement contingencies converts them into avoidance contingencies.
• The time before the deadline is either an SD(discriminated avoidance contingency) which tells us reinforcement is available or the time when there's an opportunitiy to respond(S-delta). After the deadline is passed, reinforcement is no longer available.
• The behavior is governed by the rule in rule-governed behavior, we know that under certain conditions, the behavior must occur a certain way. 
• Rule-governed behaviors are learned indirectly without direct contact with the consequence. Whenever you have an indirectly-acting contingency, you have rule-governed behavior - rules control the behavior because the reinforcer is too delayed to reinforce or punish the response. 
• is behavior that is controlled by a set of consequences creating a contingency between the behavior occurring due to a certain set of antecedent conditions. Reinforcers are often delayed until the rule requirements have been met. This is usually presented with a verbal or written decription. If x happen, then y will occur.
• Consequences in rule-governed behavior are also known as reinforcers. For example, mom tells her son, Bobby, that if he takes out the trash this week, he can go to the moives with his friends otherwise he canot-bobby takes out the trash and his mom allows him to go to the movies. His behavior was reinfored due to the contingency of the fule being met. 
• example, a sign states that the coffee maker is broken at work so you don't make coffee. 

Contingency-shaped behaviors

• Require a previously directly contacted consequence which often creates a rule to avoid or contact the consequence in the future. 
• is learned because of the reinforcement or punishment that the individual experience as a result of the response. Their future actions are altered based on that past experience of reinforcement and punishment. 
• Example, A sign states that the coffee maker is broken at work. You make coffee anyway and the coffee maker sipts water all over the counter. Next time you see this sign, you don't make coffee. 

Compare Rule-giverned behavior VS Contingency-shaped behaviors

• Rule-giverned behavior is more efficient than contingency-shaped behavior as most of these behaviors involve deadlines.

Imitation

• Imitation can be functionally defined using the behavior-environment relations of formal similarity, model, immediacy.
• During the assessing and teaching phase of imitation training it is important to reduce challenging behaviors, determine gross and fine motor skills, and attending skills which are prerequisites for the target behaviors. 
• The goal of imitation training is to teach the client to follow what the model does.
• One of the guidelines for imitation training is to record data on the behavior.
• Immitation training sessions should be during short intervals.
• It is important when using prompts during imitation training that the BCBA does Fading out prompts so the learner does not become dependent on the prompts and is able to indepently engage in the target behavior until they achieve mastery. 
• When sequencing during imitation training it is important to arrange the models using behavioral momentum to increase the likelihood of the novel behaviors to occur successfully when arranged after mastered tasks #highptolowp
• The first step to implementing imitation training is pre-assessment of the imitation behavior to see what the client can do without being taugh to get a baseline.
• Select model: they are given a model for the correct behavior that shows them what to do and how to complete the behavior.

Modeling

• A Model that occurs in everyday life is considered unplanned as it is naturally occuring/incidental
• The two types of models:planned and unplanned models.
• A model that's pre-arranged is a planned model. They are contrived and not naturally occurring in everyday life.
• Model is to stimulus as imitation is to behavior. The model is the stimulus and the behavior would be to imitate the model. 

Four of  behavior-environment : Formal similarity, Immediacy, Model, Controlled relations

• Controlled relations : The model serve as the SD for the imitation behavior shows a controlled relation.
• Immediacy : if something is within a few seconds, we are looking at the IMMEDIACY of the response. 
• Model : SD is presented in an effort to evoke imitation.
• Formal similarity: the model and the behavior have to look alike and be in the same mode #mirror #lookthesame

Errorless learning

• the aim is to avoid errors, so that the correct behaviors are learned as efficiently as possible
• An instruction method that is designed to prevent or minimize a student's errors that are used to teach discriminations. #nomistake
• Most-to-least prompting is relevant when teaching new skills using errorless learning.
• procedures try to prevent errors before they happen to minimize the chance that the client learns the wrong way to respond because they're more likely to repeat the error in the future.
• Graduated guidance is never used in errorless teaching, it is a seperate teaching method. 
• Prompts are removed gradually faded out/removed during errorless teaching, to provide a natural transition toward the independent response criterion.
• During errorless learning, a client is hindered from making an incorrect response: You want to prevent wrong/incorrect responses ASAP. As soon as you see the possibility of an error, block/prevent it from occuring.
• Disadvantage of errorless learning : expensive, high effort, require a lot of time. 

Prompts

• 2 different types of prompt: Response prompts and stimulus prompts
• Stimulus prompts : putting a sticker on a card you'd like a client to choose is a stimulus prompt because you are acting on the antecedent.
• A stimulus prompt acts on the antecedent stimuli, NOT on the response. They call attention to the simuli that is in need of help, a few types are: movement, position, redundancy.
• Response prompt acts on the response itself, NOT on the actecedent stimuli.  A response prompt acts on the response. 3 forms of response prompt: verbal, modeling, and physical prompting. 
• Stimulus fading : is the systematic and gradual fading of intrusive prompts until the control transfers to the natural stimulus, Example, when teaching someone to read the word BLUE, you can write the word is in BLUE and then graduallay fade the blue color aways.
• Stimulus fading and stimulus shaping are ways to remove stimulus prompts. 
• Stimulus shaping: Transforming the shape of the stimulus. These are gradual transformations of the physical shape. example: taking the image of a star and gradually changing it to form the letters of the word "star" until it looks like the typed word "star" 
• Least to Most prompting: the method that results in the least amount of over-prompting . It starts with minimal cues and gradually increase until resulting in the correct response. The prompts become more intrusive ONLY if it's necessary.
• When teaching staff to prompt a client, its best to use the least intrusive method.
• Physical, paried and positioned are all types of prompts.
• Prompts should not be provided after the response.
• Graduated guidance : utilizes a minimal amount pf physical response and then gradually reducing physical prompts. The tech is using graduated guidance to face out the level of prompting needed. Graduated guidance is never used in errorless teaching, it is a seperate teaching method.  Ex, when teaching a client how to brush their teeth and the prompting starts by using hand over and then gradually moves away form the hand and up the arm
• Effective step when using a prompt: present the antecedent stimulus, prompt the correct bx, reinforce the correct BX whether a prompt was used or not. 

Unconditioned punishers

• are not taught, they'er things that effect our biological systems, we're born with the responses to these punishers. 
• Example, bright lights, extreme hear/cold, loud noises, pain, etc.

Unconditioned reinforcers.

• Food, water, heat/cold, loud noises, pain, and sex are all unconditioned reinforcers - they are NOT taught, we have a biological need for these
• are producs of phylogeny

Generalized conditioned reinforcer (GCSRs)

• a conditioned reinforcer that is backed up by many other sources of reinforcement, they were previoulsy paired with many other reinforcers that don't depend on an EO for any particular reinforcement for its effectiveness.
• a GCR is based on primary reinforcers, it's effectiveness depends less on the EOs appropriate to any single primary reinforcer. It is effective because the person is usually deprived of at least one of the backup reinforcers.
• They can be collected/used to earn backup reinforcers.
• Praise and attention from your spouse. they are conditioned reinforcers that have been taught through pairing with previouly established and primary reinforcers. Thy are not specific to an EO
• Example, money, tokens, stickers

Conditioned reinforcer = CR = Secondary reinforcer = Learned reinforcer

• Conditioned reinforcers are products of Ontogeny

Conditioned punishers= CP= secondary punisher = learned punisher

• example, time-out, fines, verbal reprimands

Generalized conditioned punisher

• a stimulus change that was paired with various unconditioned and conditioned punishers.
• a stimulus that as a result of having been paired with many other punishers, functions as punishment under most conditions because it is free from the control of the motivating conditions for specific types of punishement.
• For example, yelling, reprimands and social disapproval are generalized conditioned punishers because they were paired with both conditioned and unconditioned punishers.

MO는 아래 글을 참고해주세요.
https://americanlifeinca.tistory.com/m/607

SD(Discriminative stimulus) = signals the availability of reinforcement
 
S-Delta= signals the non- availability of reinforcement
MOs

• An abolishing operation(AO) in the value-altering effect that reduces the valude of a consequence(reinforcer). it is going to cause the consequence less valuable/desirable and less likely to cause you to engage in a particular behavior(#abate ba). #AO It is not valuable in that #MOment. If you're offered salty pretzels after a 10 mile run, you're thirsty AF! So NO, you don't want pretzels. There's an AO here. 
• AO decreases the effectiveness of the consequence as a punisher, so the punisher is not effective at reducing a particular behavior, so as a result it has an evocative effect for the bx. 
• The EO, or establishing operation, increases the value of a consequence(as an effective reinforcer). It is going to help increase the likelihood of the behavior occurring (#evoke) to obtain that reinforcer. 

SDs and MOs 

• SDs and MOs are both antecedents, as they are present before behavior occurs.
• MO's change the value of reinforcement and SD's signal the availability of reinforcement; both of these interact and result in changes in behavior.
• Discriminative stimulus and motivating operations share tow similarities.

          1. They both occur before the behavior which makes them antecedents.
          2. They both evocative functions which bring out behaviors. 

• A motivating operation alters the value of reinforcement and the current frequency of behavior. 

non-contingent reinforcement(NCR)

• Delivering non-contingent reinforcement of a particular reinforder that is directly related to (maintain) the function of the problem behaviors.
• If the reinforcement they're seeking by engaging in the problem behaviors is freely available and often, they have less need to seek it out by engaging in the behaviors.
• When you provided reinforcers on a fixed time or variable-time interval scheduel, regardless of whether behaviors occur or not. So you're going to give Timmy an M&M every 5 minutes or an average of 5 minutes, whether he's doing anything or not in the session. Non-contingent=not contingent upon anything. 
• a fixed-time schedule is different than a variable-time schedule because fixed-time uses the same time-interval to determine adminstration of the reinforcement, while varible-time interval uses varing time intervals that average out to the specified interval. 
• The problem behaviors signal that there is a need not being met, or there is deprivation, of a certain reinforcer - so making that reinforcement more abundant freely, will meet that need. #NCR #filltheircup
• A disadvantage of NCR is that free access to the NCR may reduce motivation to engage in the desired behavior. This is because they are at a level where their MOs are satisfied and they are not motivated to engage in the desired behavior. There is not as much value placed on the reinforcers because they are available all the time. 

High-probability request sequence(behavior momentum)

•  an antecedent intervention in which a few easy tasks with a known history of learner success are presented in quick succession immediately before the more difficult target task, or low-p request. This helps build behavioral momentum-which is a rate of responding and its resitance to change following an alteration in reinforcement conditions. 

Antecedent intervention

• High-probability request sequence, NCR, Functional cummunication training(FCT)

Functional cummunication training(FCT)

• an application fo DRA. Differential reinforcement of an alternative behavior than the target problem behavior which serves as a functional replacement.
• establishes an appropriate communicative behavior to compete with problem behaviors evoked by an establishing operation.
• When first implementing FCT, the alternative communicative response should produce reinforcement on a continuous schedule to establish a functional relationship for the reinforcement. This increases the likelihood of the alternative beahvior occuring in the future. 

Ecological intervention

• recognized the relation between behavior and the environment, the teacher moving the student that kicks others' chairs to an isolated desk next to hers is altering the environment to improve the behaviors. 

Matching Law

• Behavior goes where reinforcement flows, baby! The #matchinglaw is what we're looking at because people go where their level of effort matches the level of reward. A person will match their responses to the proportion of the payoff
• Matching law is when two or more concurrent-interval schedules are available, relative rate of response matches(or equals) relative rate of reinforcement. 
• when two or more concurrent-internval schedules are available, relative rate of response matches(or equeals) relative rate of reinforcement. The ratio formula can be used as a general guide as a tool for matching law, as it shows a proportion. 

Negative reinforcement

• Behavior is negatively reinforced as there is an #aversive being removed(teacher's monotone voice/boredom) by engaging in the doodling and daydream behavior which he continues to #escapte #relief
• A process that occurs when a response is immediately followed by the removel of stimulus that increase the future frequency of the behavior in similar conditions. 
• Example: Jane has a stomachache. She takes Pepto Bismol. She no longer has a stmachache. In the future, whenever Jane has a stomach, she takes Pepto Bismol. Stomachache=aversive; engaging in behavior of taking the Pepto Bismol REMOVES the aversive.

Behavior Momentum

• Behavioral momentum is the use of a series of interspersed hign-probability requests to increase compliance with lower-probability requests, 2~4 mastered pre-tasks should be presented before presenting the new, more difficult tasks. #highplowp

Positive reinforcement(AKA Type 1 Reinforcement)

• Reinforcement occurs when the future frequency of a behavior increases. #increase in behavior.
• A process that occurs when a response is immediately followed by the presentation of a stimulus that increases the future frequency of the behavior on similar conditions. 
• Reinforcement can strengthen all of those dimensions(magnitude, duration, topography) of behavior. It can increase a response length, intensity and appearance
• Temproal Contiguity = direct contact in time - so the immediacy of the reinforcer after the response is CRUCIAL.
• Reinforcement uses the 3 term contingency(ABC) to establish the conditions that a behavior will occur again in the future in similar conditions. It is the fundamental building block of operant behavior.  
• Reinforcement makes future antecedent stimulus conditions relevant.

Premack Principle

• The Premack Principle states that making the opportunity to engage in a high probability behavior contingent on the occurrence of a low frequency behavior will function as reinforement for the low frequency behavior. It uses, "First this, then that." 
• Presenting undesired activities first, then desired activities is an example 

Grandma's law

• Grandma's law is an AKA for the Relativity Theory of reinforcement and the Premack Principle
• It states that making the opportunity to engage in a high probability behavior contingent on the occurrence of a low frequency behavior will function as reinforcement for the low frequency behavior. 

Unconditioned reinforcer

• is a primary reinforcer, something that is not learned. You are #bornwithit, these include:water, food, sleep, shelter, sex, activity, removal of pain, warmth, Warmth is not a learned reinforcer it is something that you biologically need, it is a positive reinforcer because you are adding something to the environment to cause the behavior occur again in the future. 

• A scatter plot (AKA pattern analysis) is also known as a pattern analysis for recording the extent to which a target behavior occurs more often at particular times than others, it tells us when the bx occurred at a certain point in time #scatterplot #pointintime.

         - When using this to take data, the day is divided into different increments of time to indicate when the bx occurs

         - It is the most subjective descriptive FBA as it involves subjective interpretation- looking for patterns of bx, nothing any kind of correlation of bx and some other factor. 

• ABC narrative recording( AKA sequence analysis or AKA ABA desriptive narrative data) is a method in which data are collected only when behaviors of interest are observed and the recording is open-ended(any events that immediately precede and follow the target behavior are recorded). This measure has more false positive outputs as it only records data when the problem bx is present. 

• In ABC continuous recording, the BCBA records all occurrences of the target problem bx and selected environmental events in the natural routine during a period of time, using partial interval, momentary time sampling, or frequency recording. The data is recorded for a minimum of 20-30 minutes in the natural environment during a specified time period.

• The 3 main ways to record data for descriptive FBAs are ABC continuous recording, ABC narrative recording, and Scatter plots.