The terms a BCBA® or RBT® candidate needs, each with a short original definition, an everyday example, the term it is most often confused with, and a link to practice it.
Terms
223
Outline areas
9
Diagrams
15
Also in Spanish, with each English term in brackets.
ABC data (antecedent–behavior–consequence recording)
Area F · C
A descriptive record of each occurrence of a behavior with what happened just before it (antecedent) and just after it (consequence), written as observable events.
Example: 10:05 — A: Instructor asks Leo to put the blocks away. B: Leo drops to the floor and cries. C: Instructor puts the blocks away herself.
Diagram
Two rows of ABC data
TimeAntecedentBehaviorConsequence
10:05
Antecedent: The Instructor asks Leo to put the blocks away.
Behavior: Leo drops to the floor and cries for about 40 seconds.
Consequence: The Instructor puts the blocks away herself.
10:40
Antecedent: The Instructor talks with another adult; Leo is drawing alone.
Behavior: Leo taps the Instructor’s arm and says “look.”
Consequence: The Instructor turns and comments on his drawing.
Each row records observable events, not interpretations such as “he was upset” or “she wanted attention.” Rows like these can suggest a hypothesis about the function; adding more rows does not confirm it. A functional analysis tests it experimentally.
A motivating operation that decreases the effectiveness of a consequence. For a reinforcer, it abates behavior that has produced it; for a punisher, it can increase behavior that the punisher had suppressed. Satiation is a common example. It can reduce responding even though the reinforcer is still available, which is not extinction.
Example: After a big plate of fries, a child stops asking for more fries even though the bowl is right there.
The extent to which recorded values match the true values of the behavior. It is checked against an independent, trusted record such as a video scored with care.
Example: A data sheet says 14 requests; a carefully scored video of the session shows 14. The data were accurate for that session.
A line drawn on a graph from the current level of performance to the goal by the target date, showing the progress needed. For a behavior to increase, data below it for several sessions suggest the plan needs changing; for a behavior to decrease, data above it do.
Example: A straight line from 10 words read correctly per minute today to 40 words read correctly per minute in 15 weeks represents an average weekly increase of 2 words read correctly per minute.
The branch of behavior analysis that uses principles discovered in research to change behavior that matters in a person’s daily life, and that measures whether the change happened and why.
Example: A team teaches a young adult to ride the city bus alone, records each step he completes on every trip, and adjusts the teaching when the data stop improving.
Agreement from a person who cannot give legal consent, such as a child, shown in words or behavior. Signs of withdrawn assent, such as turning away or pushing materials away, are respected and acted on.
Example: A child comes to the table and reaches for the materials; later he pushes them away and walks off, and the Instructor pauses to change the activity.
Verbal behavior that depends on other verbal behavior of the same speaker and modifies its effect on the listener, for example by signaling certainty, quantity or ordering.
Example: In “I think the store is closed,” the words “I think” tell the listener how far to trust the tact that follows.
Reinforcement produced directly by the behavior itself, with no other person delivering it. It can be positive (stimulation the behavior produces) or negative (discomfort it relieves).
Example: Humming a tune while alone, kept going by the sound it makes; rubbing a stiff neck, kept going by the relief.
Negative reinforcement in which the response prevents or postpones a stimulus that has not arrived yet.
Example: Leaving for work ten minutes early has repeatedly meant missing the traffic jam that builds after eight o’clock, so leaving early has become the habit.
The Instructor completes all steps except the last, which the learner is taught first; then the last two, and so on. The learner contacts the end of the chain, and its reinforcer, from the start.
Example: The Instructor helps with every step of putting on pants except pulling them up the last few inches, which the child does himself.
A condition in which behavior is measured without the intervention being evaluated, usually before it is introduced and sometimes again in later phases. It provides the comparison against which change is judged; a difference from baseline alone does not prove the intervention caused it.
Example: For two weeks, staff count how often a resident makes his bed without changing anything else; those counts are the baseline.
The reasoning behind single-case designs: stable baseline data predict future behavior; a return toward baseline when the intervention is withdrawn, or no change in an untreated baseline, verifies that prediction; and repeating the effect replicates it.
Example: In a multiple baseline, the untreated behaviors stay flat while each treated one changes in turn, verifying and then replicating the effect.
Anything an organism does, including actions others can see and private activity such as thinking in words. In practice it is specified as something the person actively does, not as a label or an absence.
Example: “Writes her name on the worksheet” describes behavior; “is cooperative” is a label, and “does not leave the table” is an absence.
An ordered sequence in which a response produces the stimulus conditions for the next response. Intermediate outcomes can both reinforce the preceding link and cue the next one; completion produces the terminal reinforcer, which need not be conditioned.
Example: Putting bread in the toaster, pressing the lever, waiting for the pop, buttering the toast and eating it.
A written plan, based on assessment, that states the target behaviors and the antecedent strategies, replacement behaviors, consequence strategies and data collection everyone will use.
Example: The plan tells every staff member to offer choices before tasks, prompt the “break” card, and record each outburst with its duration.
A behavior that, once learned, gives access to new environments, reinforcers and learning opportunities, so its value goes far beyond the behavior itself.
Example: Learning to ride a bike lets a child go to a friend’s house, join neighborhood games and run errands.
An account of response persistence: behavior in a context rich in reinforcement tends to be more resistant to disruption than behavior in a leaner context. Persistence is measured relative to baseline when something disrupts responding.
Example: A skill practiced in a lively classroom with frequent praise holds up better on a day full of distractions than one practiced where praise was rare.
A teaching package of four steps: instructions, modeling, rehearsal and feedback, repeated until the skill meets a set criterion. Widely used to train staff and caregivers.
Example: A new staff member reads how to run a preference assessment, watches the supervisor do one, does one herself, and gets specific feedback until she runs it correctly.
Two or more basic schedules occur in a fixed order, each signaled by its own stimulus. Completing one produces the stimulus of the next, and the terminal reinforcer follows completion of the last.
Example: With a blue light on, five button presses turn the light yellow. Under yellow, the first press after 30 seconds earns a token and the light returns to blue: a chain of FR 5 and FI 30 seconds.
Teaching the steps of a task analysis so they link into one smooth sequence. The three common methods are forward, backward and total-task chaining.
Example: Teaching a teen to do laundry by linking sorting, loading, adding detergent, choosing the cycle and starting the machine.
Diagram
Three ways to teach the same chain
1Wet hands
2Soap and rub
3Rinse
4Water off
5Dry
Forward.Step 1 is taught; the Instructor completes the rest.
Backward.Step 5 is taught; the Instructor completes the steps before it.
Total task.All steps, every session, with prompts where needed.
Taught in this phase
Completed or helped by the Instructor
Done by the learner, prompted as needed
A simplified five-step handwashing sequence, shown at the first teaching phase. Forward chaining teaches the first step first; backward chaining teaches the last step first, so the learner finishes the chain and contacts its reinforcer from the start; total-task chaining practices every step every time. A real task analysis sets the steps and the help each learner needs.
A design in which the performance requirement for reinforcement is changed in steps, and control is shown when behavior closely follows each new criterion.
Example: A man earns a reward for 10, then 15, then 20 minutes on an exercise bike; his daily minutes rise and level off at each new target.
A discrimination in which an additional, conditional stimulus changes which stimulus–response relation is reinforced. In matching-to-sample, the sample has this role: the same comparison can be correct with one sample and incorrect with another.
Example: Hearing “cup,” the child touches the cup picture; hearing “spoon,” she touches the spoon, with both pictures on the table each time.
A motivating operation whose value-altering effect depends on a learning history. The three types usually distinguished are surrogate, reflexive and transitive.
Example: A commuter has learned that an access code unlocks an audiobook she wants to hear. When the app requires the code, obtaining it becomes reinforcing and evokes a learned request for it. The person who can supply the code is equally available before and after that requirement appears; what changes is the code’s value. This is a transitive CMO.
A stimulus that has become a reinforcer through a learning history with reinforcers that already worked. Pairing alone is not proof; it must increase behavior when delivered contingently.
Example: A phone game’s ding has gone with bonus points many times. Later the ding alone, after a new move, makes that move more frequent: it now works as a reinforcer in its own right.
Keeping client information private and secure. Disclosure needs informed permission or a specific exception under applicable legal and professional rules, and is limited to information needed for its purpose. Clients are told the limits of confidentiality in advance.
Example: For a training presentation, a behavior analyst uses a graph with invented data instead of assuming that removing a child’s name and photo makes a real record anonymous. Any real client information is used only with the required consent.
A situation in which a professional’s personal, financial or other interests could influence decisions that should be made only in the client’s interest.
Example: A supervisor would earn a commission on a product she might recommend to families. She discloses the interest and steps out of choosing the product; disclosure alone would not remove its influence on her judgment.
A change in the environment that follows a behavior. Whether it reinforces, punishes or does nothing is shown by what happens to that behavior afterward.
Example: A friend laughs after a joke; if the person tells jokes more often around that friend, the laugh worked as a reinforcer.
Prompting first with a 0-second delay (the prompt comes with the cue), then on later trials waiting a fixed number of seconds before prompting, so the learner can respond before the prompt.
Example: “What color?” and the immediate answer “red” at first; later, “What color?” followed by a 4-second wait before the Instructor says “red.”
An ongoing practice of examining one’s own cultural assumptions, learning from the people served about their values and practices, and addressing power imbalances in the relationship.
Example: Instead of assuming that eating at a table is the goal, a behavior analyst asks how meals happen in the family’s home and builds the plan around that.
A graph in which each response adds to the running total, so the line can never go down. A steeper slope means a higher rate; a flat stretch means no responding.
Example: A student’s total words read across the week climbs steeply on Monday and Tuesday and stays flat on Wednesday, when he read nothing.
A quick check on a target: if a dead man could “do it,” it is not behavior. Being quiet, not hitting or staying still fail the test and should be restated as something the person does.
Example: “Will not call out” fails the test; “raises a hand and waits to be called on” passes it.
The three levels of understanding a science of behavior aims for: describing what happens, predicting when it will happen from relations that recur, and showing control by changing one variable and seeing the behavior change.
Example: Noting that a child bites (description), noticing that biting follows crowded transitions (prediction), and reducing biting when transitions are spaced out, then seeing it return when they are not (control).
Directly observing a behavior in the person’s usual environment and recording what happens before and after it, without changing anything. It shows correlations, not causes.
Example: An observer notes that spitting usually follows a peer taking a toy and is usually followed by the teacher coming over.
The assumption that behavior is lawful: it occurs in orderly relation to other events, so it can be studied, predicted and influenced.
Example: A behavior analyst who looks for the conditions under which a child’s screaming starts, instead of calling it random, is working from determinism.
Reinforcing one response, or one dimension of responding, more than others: the others receive no reinforcement or less of it. It is the basis of discrimination training, shaping and many reduction procedures.
Example: An Instructor answers a child’s quiet “excuse me” right away and does not respond to tugging on her sleeve.
Differential reinforcement of alternative behavior
DRAArea G
Reinforcing a specific appropriate behavior that is an alternative to the problem behavior, while the problem behavior is no longer reinforced (or is reinforced much less). The alternative often produces the same reinforcer the problem behavior used to.
Example: A child who grabbed toys now gets a turn when she says “my turn, please,” and grabbing no longer gets her the toy.
Differential reinforcement of incompatible behavior
DRIArea G
A form of DRA in which the reinforced behavior physically cannot occur at the same time as the problem behavior.
Example: A student who bites his nails earns points for keeping both palms flat on the desk during lectures; with his hands there, he cannot bring a nail to his mouth.
Reinforcement is delivered when the behavior occurs at or below a set rate, or after a minimum time since the last response. It is for behavior that is fine in itself but happens too often.
Example: A student who asks a question every minute earns praise if he asks no more than three questions during a 30-minute lesson.
Reinforcement is delivered when the problem behavior has not occurred for a set interval (or at a set moment). No particular replacement behavior is taught.
Example: A child earns a sticker for each 5 minutes of circle time without spitting; spitting resets the timer.
Seven defining features of applied work: applied (socially important), behavioral (measures real behavior), analytic (shows a functional relation), technological (procedures described well enough to replicate), conceptually systematic (tied to principles), effective (produces meaningful change) and showing generality (change that lasts and spreads).
Example: A written protocol so precise that a new staff member can run it exactly as written meets the technological dimension.
Measurement that samples the observation period, so some instances may be missed and the result is an estimate. Interval recording and time sampling are the common forms.
Example: Checking a classroom every 5 minutes to see whether each student is on task, instead of watching continuously.
Structured teaching in short, repeated trials, each with a clear start and end: an instruction, the learner’s response (with a prompt if needed), a consequence, and a brief pause before the next trial.
Example: “Touch blue.” The child touches the blue block. “Yes, blue!” and a token. Three seconds later, the next trial begins.
A cue that comes to occasion a particular response through a history of more reliable reinforcement in its presence than in its absence. Its discriminative function concerns the opportunity for reinforcement; the same stimulus can separately acquire respondent or motivating functions.
Example: An “Open” sign on a shop door: knocking or pushing the door has paid off when the sign is up.
Diagram
Stimulus control is a difference you can see
SDDog
Touching has been reinforced
Touches per minute
S-deltaCat
Touching has not been reinforced
Touches per minute
Illustrative bars, not recorded data. Touching has been reinforced when the dog appears and not when the cat appears. Different rates provide evidence of control by these stimulus conditions. Equal rates would provide no evidence of control in this measure, without ruling out differences in other dimensions or control by other features.
Arranging prompts and their fading so that the learner makes few or no errors while acquiring a skill.
Example: On early trials the Instructor gives the answer immediately with the question (0-second delay), so the child can respond correctly from the start. The delay is then lengthened gradually, keeping errors rare while the child begins to answer alone.
Negative reinforcement in which the response ends a stimulus that is already present.
Example: Each time a student tears up a page of hard problems, the page is taken away, and from then on tearing happens more often when hard pages are handed out.
Discontinuing the escape consequence that maintains a response: that response no longer produces the removal, reduction or postponement of the relevant demand or situation. The procedure requires individualized planning and monitoring and is often combined with teaching and reinforcing an alternative response.
Example: Under an assessed plan, shouting previously made the Instructor remove a worksheet. Shouting now leaves the worksheet and its requirements unchanged, while a taught break request produces a brief break. The withdrawal of the shouting-to-escape contingency is the escape-extinction component.
A motivating operation that increases the effectiveness of a consequence. For a reinforcer, it evokes behavior that has produced it; for a punisher, it abates behavior that has been followed by it. Deprivation is a common example.
Example: After two hours in the sun with nothing to drink, water is worth more and asking for water becomes more likely.
Recording each occurrence of a behavior as it happens during the observation. Best for behaviors with a clear start and end that are not too fast to count.
Example: A clicker is pressed every time a student blurts out an answer without being called on.
Basic research that studies how behavior relates to the environment under tightly controlled conditions, without the requirement that the behavior studied be socially important.
Example: A laboratory study compares how fast pigeons peck a key under two reinforcement schedules to describe the patterns each schedule produces.
Discontinuing the reinforcement that a previously reinforced response used to produce, so the response no longer pays off. The process is the decrease in responding that follows; it can be gradual.
Example: A child used to get candy at the checkout by whining; the parent stops buying candy when he whines, and over several trips the whining fades.
A temporary increase in the frequency, intensity or duration of a response soon after extinction begins. It can happen but is not guaranteed, and it does not mean the procedure is failing.
Example: When the TV remote’s batteries die, a person presses the button harder and faster for a while before giving up.
Diagram
One possible pattern during operant extinction
↑ Responses per minute
ReinforcementExtinction
Extinction burstSpontaneous recovery
Sessions →
Illustrative values, not observations from a real case. This example shows a temporary increase after reinforcement is discontinued, followed by declining responding and a brief return after a break. Bursts and spontaneous recovery can occur, but neither is inevitable or a defining requirement of extinction.
Responding controlled by an irrelevant feature of the situation instead of the stimulus the teaching was meant to bring into control.
Example: A learner picks the correct card every time, but only because the correct card is always on the left; when positions are mixed, accuracy drops to chance.
The first response after a set amount of time has passed is reinforced. A response is still required; responding typically speeds up as the end of the interval approaches.
Example: After a token is delivered, a two-minute timer starts. The first button press after those two minutes earns the next token and restarts the timer; earlier presses earn nothing. This is FI 2 minutes.
Schedules that deliver a stimulus after a fixed or varying amount of time, with no response required. They differ from interval schedules, which still require a response.
Example: A garden sprinkler that switches on every 30 minutes, whatever anyone does, runs on a fixed-time schedule.
Teaching the first step of the chain first; once it is mastered, the first two steps, and so on. The Instructor completes or helps with the remaining steps.
Example: A child learns to pick up the toothbrush first; the Instructor does the rest. Next week, she picks it up and wets it.
The reinforcer that maintains a behavior, commonly grouped as social attention, access to tangible items, escape from demands or situations, and automatic reinforcement.
Example: A boy’s head-banging occurs mostly when chores are presented and stops when they are removed: its likely function is escape.
An experimental assessment that manipulates the variables suspected of maintaining a behavior and compares responding across test and control conditions. Most test conditions deliver one possible social reinforcer contingent on the behavior; an alone or no-interaction condition checks whether it persists without social consequences, which suggests automatic reinforcement.
Example: Self-injury occurs far more in the condition where it ends a task than in the attention, alone or play conditions, pointing to escape.
The overall process of identifying the variables that predict and maintain a behavior, using indirect assessment, descriptive assessment and, when appropriate, a functional analysis.
Example: A team interviews the teacher, records ABC data for a week, then tests the most likely function before writing a plan.
A form of DRA that teaches a communication response that produces the same reinforcer the problem behavior produced, identified by functional assessment.
Example: A child who screamed to escape loud rooms is taught to hand over a “break” card, which gets him a few quiet minutes outside.
Two behaviors are functionally equivalent when they produce the same reinforcer. A replacement behavior should be functionally equivalent to the problem behavior it replaces.
Example: Raising a hand and shouting both get the teacher’s attention, so they are functionally equivalent for that student.
Experimental evidence that manipulating an independent variable reliably produces a change in a dependent variable. Repeated demonstrations and adequate controls make other explanations implausible.
Example: Hand-raising rises each time praise for raising a hand is introduced and falls each time it is withdrawn, across several repetitions.
After reinforcement has established imitation across varied actions, the learner also imitates newly demonstrated actions without direct training or reinforcement for imitating those particular actions.
Example: After learning to copy dozens of actions, a child copies a brand-new silly dance the first time she sees it.
A decrease in an elicited response across repeated presentations of the eliciting stimulus, beyond a decrease explained by sensory adaptation or motor fatigue. Unlike respondent extinction, habituation does not require discontinuing an established pairing between a conditioned and an unconditioned stimulus.
Example: A new neighbor jumps at the first train that passes at night; by the second week she barely notices them.
A procedure that presents several quick requests the person reliably follows, each followed by reinforcement, immediately before a request they are less likely to follow.
Example: A learner reliably follows “Touch your nose,” “Give me five” and “Show me the ball,” but rarely puts the ball away. The Instructor gives the three easy requests quickly, praising each response, then right away says: “Put the ball in the box.”
Doing what someone has just shown: another person’s behavior sets the occasion, the response comes right after it and has the same form. Copying long after the model, or copying cued by something else, is not imitation.
Example: An Instructor taps the table twice and the child taps the table twice right after.
Gathering information about a behavior and its possible function from people who know the person, through interviews, checklists or rating scales, without observing the behavior directly.
Example: A parent answers questions about when, where and with whom her daughter’s tantrums usually happen.
Measurement obtained through reports about behavior, such as interviews, questionnaires or rating scales, rather than direct observation of the target behavior. Recall and interpretation can affect the results, so its accuracy should be checked rather than assumed.
Example: A parent rates on a 1–5 scale how often her son argued at dinner over the past month.
A voluntary decision by someone legally entitled to consent, or their legally authorized representative, after understandable information and an opportunity to ask questions. The person understands the proposed procedures, relevant risks, benefits and alternatives, and the right to refuse or withdraw; a signature alone does not establish informed consent.
Example: Before assessment, a child’s legally authorized parent explains in her own words what the assessment involves and its main risks, asks questions about alternatives, and freely agrees. Her signature documents that decision, and she knows she can withdraw consent.
The extent to which a study shows that the independent variable, and not something else, produced the change in the dependent variable.
Example: Repeating the effect three times across three settings, at three different times, makes a coincidence an unlikely explanation and strengthens internal validity.
IOA describes how closely independently collected records match for the same observed events. The calculation must be named because different methods can give different percentages; matching records can still share an error, so agreement does not establish accuracy.
Example: Two staff members score the same 10 intervals; they agree on 8, giving 80% interval-by-interval IOA.
Diagram
Same totals, different intervals
Interval
1
2
3
4
5
6
7
8
9
10
Observer 1
scored
scored
not scored
scored
scored
not scored
scored
scored
not scored
scored
Observer 2
scored
scored
scored
scored
not scored
not scored
scored
scored
not scored
scored
Agreement
agree
agree
disagree
agree
disagree
agree
agree
agree
agree
agree
Totals compared: 7 ÷ 7 = 100%
Interval by interval: 8 agreements ÷ 10 intervals = 80%
Illustrative records, not real data. Both observers scored 7 of the 10 intervals, so comparing totals alone suggests perfect agreement. Interval by interval, they disagree on two intervals.
The time between the end of one response and the start of the next instance of the same behavior.
Example: A student finishes one bite at 12:00:10 and starts the next at 12:00:50: an IRT of 40 seconds, measured from the end of one response to the start of the next. Lengthening it can slow fast eating.
A verbal operant evoked by a verbal stimulus that it does not match point to point, as in answering questions, conversation or filling in a familiar phrase.
Example: “What do you wear on your feet?” “Shoes.”
A case of multiple control in which the same response form is evoked at the same moment by two sources, such as a word being rehearsed and the tact of an item, and that match controls a further response such as selecting the item.
Example: Sent for “oregano,” a cook repeats the word while scanning jars; when the label she reads matches the word she is repeating, she takes the jar.
Choosing, among procedures likely to work, the one that places the fewest limits on the person’s freedom and access to reinforcement, and favoring reinforcement-based options.
Example: Assessment and data indicate that two safe plans can meet an agreed goal for leaving the activity area. The team selects the plan with scheduled breaks and a taught break request because it allows more freedom of movement, then monitors whether it remains effective.
On each trial the learner is first given a chance to respond independently, and prompts become more intrusive only as needed until the correct response occurs.
Example: The Instructor waits, then says “zipper,” then points, then guides the hand only if the earlier prompts did not work.
Where the data sit on the vertical axis within a phase, often summarized by the mean or median. A change in level between phases is one sign of an effect.
Example: Correct responses hover around 30% in baseline and around 75% during intervention.
Responding under the control of another person’s verbal behavior, such as following instructions or selecting a named item. In teaching receptive skills the term refers to this nonverbal responding; a spoken reply is analyzed as a verbal operant of its own.
Example: Asked “Get your coat,” a child walks to the hook and takes her coat.
A behavior continues to occur over time after some or all of the intervention that produced it has been withdrawn. It is assessed by checking the behavior after the supports are gone.
Example: Three months after daily practice ended, a student still uses the subtraction strategy she was taught.
A verbal operant controlled by a motivating operation and reinforced by a characteristic consequence that the response specifies. Requests and questions are often mands, but the form alone does not decide: what makes a response a mand is that these variables control it.
Example: Cold in the car, a child says “heat on” and the heater is turned on.
A duty imposed by applicable law on specified people to report suspected abuse, neglect or other covered harm. Who must report, the reporting threshold, recipient and deadline depend on the jurisdiction and role. These duties can limit confidentiality.
Example: An Instructor observes injuries and disclosures that meet the applicable legal reporting threshold. She ensures the required report reaches the designated authority within the legal deadline; notifying her employer alone does not satisfy an additional duty to report to that authority.
The performance standard, set in advance, at which a skill is considered acquired, such as a minimum accuracy across a set number of sessions or Instructors.
Example: “80% or more correct, without prompts, in three consecutive sessions with two different Instructors.”
A description of choice under concurrent schedules: the share of responses allocated to an option tends to approximate the share of reinforcement obtained from it. Actual allocation also depends on reinforcer quality, delay, effort and bias.
Example: Two available buttons produce the same token under separate variable-interval schedules. Across a stable observation, about two thirds of the tokens come from the left button and about two thirds of the presses are allocated to it: relative responding approximates relative obtained reinforcement.
A trial format for conditional discriminations: a sample stimulus is presented, and the learner selects the comparison that goes with it from two or more options. Positions of the comparisons should vary across trials.
Example: A photo of a dog is shown above three pictures (dog, car, apple); the child touches the dog.
Teaching selected combinations from a grid of elements, such as actions and objects, so that untaught combinations of the same elements are performed correctly.
Example: After learning “wash the car” and “dry the plate,” a child correctly does “wash the plate” and “dry the car” without being taught them.
An explanation that gives a presumed mental agent, trait or mechanism the job of causing behavior, rather than accounting for the behavior through its history and current conditions. Merely discussing thoughts or feelings is not mentalism; the issue is the explanatory role assigned to them.
Example: “She hoards toys because she is possessive” names a mental cause; the only evidence for “possessive” is the hoarding itself.
Two or more schedules alternate exactly as in a multiple schedule, but with no stimulus signaling which one is in effect.
Example: In a programmed teaching session, periods in which every button press earns a token alternate with periods in which presses earn none. The screen looks the same throughout, with no cue identifying continuous reinforcement versus extinction.
Demonstrating the target behavior so the learner can imitate it. A model alone works only to the extent that the learner already imitates; otherwise imitation itself may need to be taught first, or other prompts added.
Example: The Instructor shows how to fold a towel in half twice, then hands the towel to the learner.
The behavior is recorded only if it is occurring at the exact moment each interval ends. It can over- or underestimate, and longer intervals make the estimate rougher.
Example: A timer vibrates every 2 minutes; at that instant the observer notes whether the child is playing with a peer.
Teaching starts with the most intrusive prompt needed for success, and prompts become less intrusive across trials or sessions as the learner succeeds.
Example: Full hand-over-hand guidance to zip a coat on day one, then a light touch at the elbow, then a pointing gesture, then nothing.
An environmental variable that temporarily changes how effective a consequence is as a reinforcer or punisher, and with it how often behavior that has produced that consequence occurs right now. It alters current value and frequency; it does not teach new behavior.
Example: Skipping lunch makes food more valuable for the afternoon and makes checking the fridge more frequent.
Diagram
A motivating operation has two effects at once
Establishing operation
Three hours of hiking, no water
Value of water as a reinforcer goes up
Looking for a tap, asking for water goes up
Abolishing operation
Just drank a liter
Value of water as a reinforcer goes down
Looking for a tap, asking for water goes down
Deprivation and satiation change how much water is worth right now (value-altering effect) and, with it, how often behavior that has produced water happens right now (behavior-altering effect). Nothing new is learned.
A design that rapidly alternates two or more conditions, such as by session or by day, and compares the separate data paths for each. It needs no withdrawal phase but can produce sequence effects.
Example: Two ways of teaching spelling are alternated in random order across sessions; the data path for one method climbs well above the other.
A design that introduces an intervention at staggered points across behaviors, settings or people. Concurrent baselines overlap in calendar time; nonconcurrent baselines need not. A functional relation is supported when changes repeatedly follow intervention introduction, with suitable controls for competing explanations. Design quality depends on those controls, not on the concurrent or nonconcurrent label alone.
Example: Self-monitoring is introduced in math, then two weeks later in reading, then two weeks later in science; on-task behavior rises in each class only when its turn comes.
Most verbal responses have more than one source of control. Convergent: several variables control one response. Divergent: one variable controls several different responses.
Example: A hungry child who sees pizza says “pizza” (convergent: hunger and the sight of pizza). Hearing “beach” evokes “sand,” “waves” or “sun” (divergent).
Overlapping roles with a client, stakeholder or colleague, such as providing professional services while also sharing a friendship or business partnership. The overlap can create competing obligations, conflicts of interest and pressure within the professional relationship.
Example: A behavior analyst is asked to provide services to the child of a close friend. Taking the case would combine a professional role with an existing friendship, so she arranges a referral to another qualified analyst.
Two or more schedules alternate, and each is signaled by its own stimulus.
Example: During a green-card component, each request for attention receives a brief response from the Instructor. During a red-card component, requests receive no attention. The components alternate, and the cards identify continuous reinforcement versus extinction.
Several items are presented together, and after each choice the chosen item goes back into the array while unchosen items may be swapped for new ones. A strong favorite can be picked again and again.
Example: A child picks the toy train on all five presentations, so the train stays on top and little is learned about how the other items rank.
Punishment in which a stimulus, or access to reinforcement, is removed after the response, and the response becomes less frequent. Response cost and time-out are the common forms.
Example: A teen loses an hour of phone time each time she misses curfew, and late arrivals become rare.
Reinforcement in which a stimulus is removed, reduced or postponed after the response, and the response becomes more frequent. It increases behavior, so it is never a kind of punishment.
Example: Putting on sunglasses cuts the glare, and you reach for them more often on bright days.
Delivering a reinforcer on a time-based schedule, regardless of behavior. When the reinforcer is the one maintaining a problem behavior, freely available access can reduce the motivation for that behavior.
Example: A child who acts out for attention gets a brief, friendly check-in from the Instructor every 3 minutes, whatever he is doing.
A change in behavior after watching another person’s behavior and its consequences. Unlike imitation, the new behavior need not look like what was observed or follow it closely in time.
Example: A boy sees his sister get scolded for climbing the bookshelf, and he stops climbing it himself.
Learning in which behavior changes because of the consequences it has produced in the past. Antecedents then signal when those consequences are available.
Example: A child who gets a turn on the tablet after saying “my turn” starts saying it more often when the tablet is out.
A description of a target behavior in observable, measurable terms, with clear boundaries (when an instance starts and ends) and examples and nonexamples, so that two observers would score it the same way.
Example: “Hitting: any contact of an open or closed hand with another person’s body with enough force to make a sound. High-fives during games are nonexamples.”
A positive punishment procedure in which the person, after a behavior, restores the environment beyond its original state (restitutional) or repeatedly practices the correct form of the behavior (positive practice).
Example: A student who scribbles on a desk cleans that desk and the two next to it (restitution); a student who runs in the hall walks it correctly five times (positive practice). In both cases the behavior becomes less frequent.
Favoring the simplest explanation that fits the data. A more complex or speculative account is entertained only when simpler ones, such as a testable environmental variable, cannot explain the results.
Example: Before attributing a sudden rise in crying to a new diagnosis, a team first checks whether a recent change in staff or routine could account for it.
The observation is divided into intervals, and an interval is scored if the behavior occurs at any point in it. It tends to overestimate how much of the time the behavior occurs.
Example: In 10-second intervals, an interval is marked if the child hums even once, even for one second.
A ratio expressed per 100, such as correct responses out of all opportunities. It hides the number of opportunities and the time involved, so report those too.
Example: 9 correct answers out of 10 trials is 90%, and so is 90 out of 100. The percentages match, but the second rests on many more opportunities.
The person has demonstrated the skill but does not perform it adequately under the conditions of concern. Assessment examines cues, resources, effort and competing consequences; intervention usually changes those conditions, although practice or instruction may also be useful.
Example: An employee files reports correctly when the manager asks on the spot, but rarely does so on her own.
A function-based assessment of a staff performance problem that looks at antecedents (instructions, training), equipment and processes, knowledge and skills, and consequences, to choose an intervention that fits the cause.
Example: Staff skip data entry; the assessment finds the tablets are often uncharged, so the fix is a charging station, not retraining.
A lasting output used to measure the behavior that produced it after the behavior has occurred. Before relying on that output, check that the target behavior reliably produces the defined product and that other actions or people cannot produce an indistinguishable one.
Example: Counting the number of correctly solved problems on a worksheet after the student hands it in.
A structured way to identify which items or activities a person prefers, as candidates for reinforcers. A preferred item is not proven to be a reinforcer until it increases behavior.
Example: Before a session, the Instructor lets a child choose among a bubble wand, a car and a book several times and ranks the choices.
An activity that occurs more often when both activities are freely available can reinforce a less probable activity when access to the former depends on performing the latter. The comparison concerns baseline response probability, rather than preference inferred from a label or a first–then instruction alone.
Example: When both options are freely available, a child spends much more time riding her bike than putting clothes away. Making bike riding available after she puts her clothes away increases clothes-putting-away: the higher-probability activity reinforces the lower-probability one.
Behavior and stimulation that occur inside the skin and are accessible mainly to the person experiencing them, such as thinking in words or feeling a headache. Behavior analysis treats them as natural events subject to the same principles as public behavior: they can take part in controlling other behavior, but their own origins also need explaining.
Example: Silently counting to ten before answering a rude email is a private event, and it can be shaped by consequences like any other behavior.
The extent to which an intervention is carried out as written. Without it, you cannot tell whether poor results come from the plan or from how it was delivered.
Example: A checklist of the 8 steps of a teaching procedure shows that a staff member completed 6 of 8 steps correctly: 75% integrity.
Extra help added before or during a response to make the correct response more likely while a skill is being learned. Prompts are planned to be faded so the natural cue takes over.
Example: While asking “Where is the cup?”, the Instructor points to the cup.
Diagram
Two directions through a prompt hierarchy
Most-to-least
Across trials or sessions, fades from a strong, effective prompt toward independence.
Full physical guidanceStart
Partial physical guidance
Model
Gesture
Verbal prompt
Independent
Least-to-most
On each trial, allows an independent response first, then adds stronger prompts only as needed.
Full physical guidance
Partial physical guidance
Model
Gesture
Verbal prompt
IndependentStart
An example hierarchy from most to least intrusive. The levels used, and their order, vary with the learner and the skill; what defines each method is where it starts and which way it moves.
A pattern of relying on an extra cue to perform a response that teaching is intended to bring under the natural cue’s control. Independent responding may be absent or inconsistent; the pattern alone does not establish why stimulus control failed to develop or persist.
Example: A student writes his name only after the Instructor taps the paper, even though he knows every letter.
Claims or practices promoted as scientifically established while relying on misleading evidence or avoiding adequate tests and correction. Testimonials and impressive-sounding mechanisms may replace credible evidence. An openly identified, testable hypothesis is not pseudoscience merely because it is not yet supported.
Example: A product claims to “rewire the brain” in two weeks, citing only parent testimonials and no controlled data.
The process in which a consequence decreases the future likelihood of the response that produced it under similar conditions. An unpleasant consequence is not demonstrated to function as punishment unless it reduces that responding.
Example: Every time a man parks on a certain street he gets a ticket, and he stops parking there.
The philosophy of behavior analysis that treats everything a person does, including thinking and feeling, as behavior to be explained by its relations with the environment. Private events are included, not ruled out.
Example: Rather than saying a student skips homework “because he is unmotivated,” a radical behaviorist asks what happens before and after homework time that makes skipping more likely.
A change in behavior produced by the observation or measurement itself, so the data do not represent the behavior under usual conditions. Knowing one is being observed is a common cause, but not the only one.
Example: A staff member follows every step of a protocol while the supervisor is in the room and skips steps once she leaves.
A stimulus that has come before a worsening (or an improvement), so its own removal becomes reinforcing (or punishing), and it evokes (or abates) behavior that has ended it.
Example: A red folder has repeatedly preceded difficult work. After that history, removing the folder becomes reinforcing even before the work begins. A previously learned break request makes the Instructor put the folder away, and the learner requests a break more often when the folder appears: the warning cue establishes its own removal as reinforcement.
The process in which a consequence increases the future likelihood of the response that produced it under similar conditions. The consequence functioning in that process is a reinforcer; whether it seems pleasant does not establish its function.
Example: A coworker always replies when you message her early in the morning, and you find yourself messaging her early more often.
Diagram
Four consequences, two questions
Stimulus added
Stimulus removed
Behavior increases
Behavior increases
Positive reinforcement
A child says "more" and gets another push on the swing; she says "more" more often.
Behavior increases
Negative reinforcement
Closing the window reduces drilling noise; on later drilling occasions, you close it more often.
Behavior decreases
Behavior decreases
Positive punishment
Touching an electric fence produces a jolt; touching it becomes rare.
Behavior decreases
Negative punishment
Parking in the loading zone results in a monetary loss; parking there becomes less frequent.
The column says whether a stimulus was added or removed; the row says what happened to the behavior afterwards. Both answers come from observation, not from how the consequence feels.
A direct test of whether a stimulus actually functions as a reinforcer: does making it contingent on a response increase that response compared with when it is not?
Example: A child sorts more blocks when each sorted block earns a few seconds with a spinner than when the spinner is not available.
The return of a behavior reduced by extinction after the reinforcer that used to maintain it is delivered again independently of that behavior. If the behavior itself earns the reinforcer again, that is reacquisition, not reinstatement.
Example: A child had stopped whining for candy at home. One afternoon a relative hands out candy freely there, regardless of whining, and whining for candy returns even though it still earns none.
A behavioral account of language and cognition that analyzes derived relations beyond sameness, such as opposition, comparison and hierarchy, and how context controls which relation applies.
Example: Told that coin A is worth more than coin B and coin B more than coin C, a child picks coin A when asked for the one worth the most, though no one compared A and C directly.
The consistency of a measure: the same behavior, measured again, gives the same value. A measure can be reliable and still be inaccurate if it makes the same error every time.
Example: A scale that always reads 2 pounds heavy is reliable but not accurate.
The return of a behavior reduced by extinction when the context changes, for example back to the setting where it was first reinforced, or to any setting other than the one where it was reduced, even though it still produces no reinforcer.
Example: Tantrums for candy were reinforced at home, then stopped at the clinic, where they never produced candy. Back home, where the family now also never gives candy for them, the tantrums reappear.
An appropriate behavior selected to take the place of a problem behavior. It works best when it serves the same function and is at least as easy and quick to produce.
Example: Instead of pushing peers to get space, a student learns to say “I need a minute” and step to a quiet corner.
Repeating a condition or a whole study to see whether the same effect occurs again. Within a study it strengthens the case for a functional relation; across studies it shows generality.
Example: In an ABAB design, the second intervention phase replicates the effect seen in the first.
Learning in which a stimulus that did not elicit a particular response comes to elicit it after being paired with a stimulus that already does. It works on antecedents; no consequence of the response is involved.
Example: After several visits where the clinic smell came just before a shot, the smell alone makes a child’s heart race.
Diagram
Respondent conditioning, in three moments
1 Before
NSDrill whine
No tension
USSharp twinge
URTension
2 During pairing
NS + USDrill whine, then sharp twinge
URTension
3 After
CSDrill whine
CRTension
NS neutral stimulus · US unconditioned stimulus · UR unconditioned response · CS conditioned stimulus · CR conditioned response
A sound initially neutral with respect to tension can acquire the ability to elicit it after repeatedly preceding a stimulus that already elicits tension. The conditioning arrangement does not make any consequence contingent on tensing.
Presenting a conditioned stimulus repeatedly without the unconditioned stimulus, so the conditioned response weakens. Avoiding the conditioned stimulus does not produce it.
Example: A child who tenses at the dentist’s chair visits several times just to sit in it, with no treatment, and the tension fades.
A group of responses that produce the same effect on the environment, even if they look different. Members are interchangeable alternatives, not steps in a sequence.
Example: Saying “open,” handing over the jar and pointing at the lid all get an adult to open the jar for a child: one response class.
A design that alternates baseline (A) and intervention (B) phases, showing control when behavior changes each time the condition changes. It is a poor fit when the behavior cannot reverse or when withdrawing treatment would be harmful.
Example: Baseline, then a picture schedule, then removal of the schedule, then the schedule again: transitions improve each time the schedule is present.
Behavior controlled by a verbal description of a contingency, such as an instruction or a rule someone states to themselves, often before contact with the consequence. It can be slow to adjust when the real contingency changes.
Example: A hiker never eats wild berries because a guide once said some are poisonous, without ever having been sick from one.
A cue associated with nonreinforcement of a particular response, in contrast with conditions in which that response has been reinforced. This discrimination makes the response less likely in the cue’s presence. The cue is an antecedent, not a punishing consequence.
Example: A “Closed” sign on the same shop door: pushing the door has not worked when that sign is up.
The rule that states which responses will be reinforced, based on the number of responses, the time elapsed, or both. Each schedule tends to produce a typical pattern of responding.
Example: A coffee card that gives a free drink after 8 purchases is a schedule: every eighth purchase pays off.
Diagram
Four simple schedules, four typical records
FRFixed ratio
Pause after each reinforcer, then a fast run
VRVariable ratio
High, steady rate with few pauses
FIFixed interval
Slow after each reinforcer, faster toward the end: the scallop
VIVariable interval
Moderate, steady rate
↑ Total responses · → Time Tick = reinforcer delivered
Sketches of cumulative records: the line climbs with every response, so a steeper line means faster responding and a flat stretch means a pause. Each short tick marks a reinforcer. These are textbook-typical shapes drawn for study, not recorded data.
Gradually reducing how often reinforcement is delivered, or increasing the requirement for it, so behavior can be maintained by a leaner, more natural schedule.
Example: A child earns a token for every correct answer, then for every second one, then for every fourth, without the accuracy dropping.
The populations, problems and procedures a behavior analyst is trained and experienced to handle. Outside it, the analyst seeks training, supervision or consultation, or refers.
Example: A behavior analyst experienced with young children is asked to treat a severe feeding disorder and arranges consultation with a feeding specialist.
An account of how consequences affect which variants persist over time: natural selection operates across generations, reinforcement selects behavior during an individual’s lifetime, and consequences can sustain or change cultural practices. Selection does not require foresight or a conscious decision.
Example: Of the many ways a toddler tries to open a lid, the twist that works is repeated and the others drop out.
Teaching a new behavior by reinforcing successive approximations of it: each step closer to the terminal behavior is reinforced, and earlier forms stop being reinforced.
Example: To teach a child to say “cup,” the Instructor first reinforces any sound, then only “k” sounds, then “kuh,” then “cup.”
Diagram
Shaping: the criterion moves step by step
Any vocal sound
No longer reinforced
A “k” sound
No longer reinforced
“Kuh”
Reinforced now
“Cup”
Terminal behavior
An example of shaping a first word, shown midway. Each form is reinforced until the learner produces it reliably and a closer form is within reach; then the criterion moves up and the earlier form stops being reinforced. The steps depend on the learner and are set from what they already do.
A research design in which each participant serves as their own control: the same person’s behavior is measured repeatedly across conditions, and effects are judged for that individual.
Example: One child’s rate of requests is compared across baseline and intervention phases, with no comparison group.
The extent to which the goals, procedures and results of an intervention are judged acceptable and important by the person served and those around them.
Example: A family is asked whether the new bedtime routine is easy to follow and whether the change in sleep matters to them.
A brief return of a response that had decreased during extinction, typically at the start of a later session or after a break. It is possible, not inevitable, and usually fades if extinction continues.
Example: After a week of no response, a child’s calls after bedtime had stopped. On Monday night, with still no response, he calls out once again, then stops.
A set of stimuli grouped by a shared criterion: common physical features (a feature class), the same timing relative to behavior, or the same effect on behavior. In an arbitrary class, membership comes from learning rather than physical resemblance.
Example: All the different chairs a child has sat on form a feature class. The written word “three,” the numeral 3 and three claps share no form but can become an arbitrary class.
What is shown when a response occurs differently in the presence of different antecedent stimuli because of a learning history, usually differential reinforcement. The difference can be in rate, latency, duration or another dimension.
Example: A driver brakes at red lights and not at green ones; the color of the light controls braking.
Responding differently in the presence of different stimuli, as a result of a history in which the response was reinforced in the presence of one and not the other.
Example: A toddler says “dog” to dogs and not to cats after adults have confirmed “dog” only for dogs.
The emergence, after some conditional discriminations are trained, of derived relations that were not trained, showing reflexivity, symmetry and transitivity among the stimuli. The stimuli then function as members of one class.
Example: After learning spoken “four” → numeral 4 and spoken “four” → the written word FOUR, a learner matches 4 to FOUR and FOUR to 4 without being taught.
Diagram
Trained relations and probes of untrained relations
"dog"ASpoken word BPhoto CDrawing Photograph of a dog. Simple drawing of a dog.
Trained: A→B and A→C Probed without direct training: B→C and C→B
Solid arrows show trained A→B and A→C relations. Dashed arrows show probes of B→C and C→B without directly training those relations. Each combines symmetry with transitivity: B→A with A→C, or C→A with A→B. Reflexivity and reverse-to-A tests are not drawn. This dog example does not by itself demonstrate equivalence, because visual resemblance could also explain photo–drawing matching. Physically dissimilar, arbitrarily related stimuli help distinguish derived relations from that alternative explanation.
The trained response occurs in the presence of stimuli that were not part of training but share features with the trained ones, such as new people, places or materials.
Example: A child taught to wash her hands at the school sink also washes them at grandma’s sink, which has different taps.
Diagram
Generalization gradients: narrow and broad
Trained SD
← LowerPitch of the test toneHigher →
Narrow: less generalization across pitch Broad: more generalization across pitch ↑ Responses
Illustrative gradients across test-tone pitch. The narrow curve shows responding concentrated near the trained pitch; the broad curve shows more responding at other pitches. Both curves show some control by pitch. Neither represents recorded data, and the broad curve is not a completely flat gradient.
An alteration or addition to the teaching stimuli that makes the correct response more likely. A within-stimulus prompt changes a feature of the task stimulus itself, such as its size or contrast; an extra-stimulus prompt supplies a separate cue, such as an added marker.
Example: The correct word card is printed in bold while the others are in light gray.
Ongoing, structured oversight of another person’s work that builds their skills, checks the quality of the services they deliver, and protects clients. In behavior analysis it relies on direct observation, data and feedback.
Example: Each week a supervisor watches part of an RBT®’s session, scores the teaching steps, and reviews the results with her afterward.
After A → B is trained, B → A emerges without being trained: the trained relation is reversed.
Example: Taught to select a picture of a key when shown the printed word KEY, a learner selects KEY when shown the key picture, with no training in that direction.
A verbal operant controlled by a nonverbal stimulus, such as an object, event or property, and typically maintained by generalized conditioned reinforcement.
Example: Seeing a plane overhead, a child says “airplane!” and her dad answers, “Yes, a big one.”
Breaking a complex skill into a sequence of smaller, teachable steps, in order. It is the basis for chaining and for measuring which steps a learner can already do.
Example: Brushing teeth: pick up brush, wet brush, put on toothpaste, brush top teeth, brush bottom teeth, rinse, put brush away.
A verbal operant in which written words evoke vocal responses that correspond to them point to point, without formal similarity. Reading aloud, with or without understanding.
Example: A child reads the word EXIT aloud from a sign.
The basic unit of operant analysis: an antecedent condition, a response, and a consequence that depends on that response. It explains why a behavior occurs in some situations and not in others.
Example: The ice-cream truck’s music plays (antecedent), a child asks for money (response), a parent hands over a coin (consequence).
Diagram
The three-term contingency
A
Antecedent
What is present before
The vending machine shows "Ready"
B
Behavior
What the person does
You press the button for a snack
C
Consequence
What changes after
The snack drops into the tray
If obtaining the snack reinforces pressing, pressing becomes more likely when the machine next shows “Ready”.
The three-term operant contingency relates antecedent conditions, a response and a consequence contingent on that response. This example assumes that obtaining the snack reinforces pressing; temporal order alone does not demonstrate reinforcement.
A form of negative punishment in which a behavior is followed by loss of access to positive reinforcement for a short, specified time. The person may stay in the setting (nonexclusionary) or leave it (exclusionary). It only works if time-in is reinforcing.
Example: A child who hits during a board game sits out of the game for two minutes while it continues without him, and hitting during games becomes less frequent.
A system in which specified behaviors earn tokens that can later be exchanged for backup reinforcers. It states which behaviors earn tokens, how many, and what tokens can buy and when.
Example: A student earns a star for each finished task; five stars can be traded for ten minutes of a favorite game.
The physical form of a behavior: what it looks like, sounds like, or how it moves through space. Two behaviors with the same topography can serve different functions.
Example: Two children both throw a cup; for one it ends a meal, for the other it brings an adult over. Same topography, different functions.
An environmental condition that, through a learning history, changes another stimulus’s effectiveness as a reinforcer and changes the current likelihood of behavior that has obtained it. Often the effect is an increase because that stimulus is needed for another valued outcome; its effectiveness can also decrease when that need changes.
Example: A child is handed a bowl of cereal with no spoon; the spoon suddenly becomes valuable, and asking “spoon?” becomes likely.
After A → B and B → C are trained, A → C emerges without being trained: two trained relations are combined.
Example: A learner is taught to select a horse photo on hearing “horse” (A → B) and to select the printed word HORSE when shown the horse photo (B → C). Without further teaching, on hearing “horse” she selects HORSE (A → C).
A stimulus that functions as a reinforcer without any learning history, usually because it matters for survival. Its effectiveness still depends on motivating operations.
Example: Water for someone who has not drunk all afternoon; warmth for someone who is cold.
How much the data points within a phase bounce around. High variability makes an effect harder to see and usually calls for more data or a search for the variables causing it.
Example: One session shows 2 outbursts, the next 15, the next 4: too variable to judge a change yet.
The first response after an amount of time that varies around an average is reinforced. It tends to produce a moderate, steady rate.
Example: Checking for a reply from a friend who answers at unpredictable times, about every 20 minutes on average: only the first check after a reply has arrived pays off. That is roughly VI 20 minutes.
The functional units of verbal behavior, classified by their controlling variables and reinforcement history, not by the words said: mand, tact, echoic, intraverbal, textual, transcription and autoclitic.
Example: The same word, “ball,” can be a mand, a tact, an echoic or an intraverbal depending on what evoked it and what followed.
Diagram
The same word, different verbal operants
OperantBeforeResponseAfter
Mand
Before: Thirst (motivating operation)
Response: "Water"
After: Gets a glass of water
Tact
Before: Sees a puddle (nonverbal)
Response: "Water"
After: "Yes, it rained" (generalized)
Echoic
Before: Hears "water"
Response: "Water" (same form)
After: "Good!" (generalized)
Intraverbal
Before: Hears "What do fish swim in?"
Response: "Water"
After: “That is right” (generalized conditioned reinforcement)
Textual
Before: Sees the written word WATER
Response: "Water" (spoken)
After: "Well read" (generalized)
Transcription: taking dictation
Before: Hears "water"
Response: Writes water
After: "Spelled right" (generalized)
These rows illustrate established verbal operants. Assume that approval has acquired a generalized conditioned reinforcing function and that the relevant response histories are in place. Classify each operant by its controlling variables and history, not by “water” or by the wording of praise alone. Here, transcription means taking dictation.
Judging graphed data by inspecting level, trend and variability within phases, and the immediacy of change, overlap and consistency of data patterns across similar phases, rather than relying on a statistical test. These features are read together with the design to judge whether the intervention had an effect.
Example: A supervisor looks at the graph and notes that the drop in self-injury came right after the phase change and that no intervention point overlaps with baseline.
An interval is scored only if the behavior occurs throughout the entire interval. The percentage of scored intervals tends to underestimate how much of the observation time the behavior occupied, so it is often chosen for behaviors to increase, such as staying on task.
Example: A 15-second interval is marked “on task” only if the student worked for all 15 seconds.
Independent study resource. INTERLAZA is not affiliated with or endorsed by the BACB. BCBA® and RBT® are registered marks of the BACB. The definitions and examples are original, are not official exam content and do not reproduce any exam item.
