6437 FINAL

False

A BCBA does a descriptive assessment and determined that Timmy is tantruming for attention. Is it true or false that she demonstrated a functional relation.

Functional Relation

A ________ can be determined when the treatment phase has a different level, trend, and latency to change than the baseline.

bidirectional change

A ________ is when you reduce the criterion to an earlier phase in order to strengthn control.

Correlations; experimental control

Single case design is different than a case history because a case history derives its conclusions based on __________; whereas, single case design derives its conclusions by ______________.

The Law of Effect; the learning of novel behavior

The ___________ was discovered by Thorndike in 1898 and demonstrated ______________.

A-B-A-B DESIGNS

• Strengths - Allow researchers to compare two conditions - The number of replications is a function of the clarity of the data - Frequently used by researchers because of its flexibility (B-A-B, A- B-C-B) •Limitations - Not all behavior is reversible - May not be ethical or safe - May be too complex in educational settings (too many conditions)

B-A-B Designs:

- The intervention phase comes before the baseline phase - Helpful for situations in which behavior is dangerous, when a skill needs to be targeted immediately, or in the analysis of previously introduced interventions - Additional phases may strengthen the argument that the IV was responsible for changes in the DV

Questions

"any experiment worth its salt (i.e.., worthy of investigation) will raise more _______ than its answers" (Sidman, 1960 p.8).

Results (what was found)

- The results section of a research article is where all the information collected, using the previously described method, is reported. This section should be very mater-of-fact, with absolutely no interpretation. All information collecting processes described in the method should have a corresponding sub-heading in the results. In short, the two sections should clearly match. There shouldn't be information reported in the results that weren't first described in the data collection process in the method, and vice versa.

Wolf's three general domains for subjective analysis

1)Goals = targets of an intervention, including individuals, settings, and specific behaviors 2)Procedures = techniques used to change behavior 3)Outcomes = behavioral changes produced by an intervention (direct and indirect)

social validity

1)Subjective evaluation 2)Normative comparison 3)Sustainability

Correlation; functional relation

ABC data recording can demonstrate _______. A fuctional analysis can deonstrate possible ________.

Correlation; causation

ABC data recording can demonstrate _________. A functional analysis can demonstrate possible _________.

Changing Criterion Design

ADVANTAGES•No counter-therapeutic behavior change•Need only 1 target behavior & 1 participant•Goal reached in manageable steps•DISADVANTAGES•Limited to consequences for behaviors in repertoire•Predictions of criterion levels somewhat subjective

The experimenters willingness to follow the data

According to Skinner, many experimentars have answered profound questions they did not ask, due to what practice?

Measurement

Accurate measurement can help you determine: If treatment is necessary (or not) What the best treatment options might be Whether or not the target behavior changed after the treatment was implemented. Depending on your area of focus, you may use operational definitions or pinpoints to provide a precise way of defining target behaviors. Your textbook utilizes the term "behavioral definition", but the goal here is to expose you to different terms that are utilized in practice.

Operational definition

An operational definition provides a topographical (observable and measurable) description of the behavior that is targeted. For example, someone may request for "aggression" to be targeted for behavior reduction. What does "aggression" look like? Aggression could consist of a variety of behaviors such as (but are not limited to): hit others with a closed fist, hit any surface with an open fist, using negative language, speaking at a high volume, pushing others, kicking surfaces or people, and the list could go on. Because people can interpret vague descriptions of behavior in several ways, it is important to use precise operational definitions (such as: hit any surface with an open fist) when targeting behavior for measurement. The operational definition allows the entire treatment team (which may include BCBAs, BCaBAs, RBTs, Caregivers, and others) the ability to understand what the behavior looks like and collect data on the occurrences of that specific response. Operational definitions work the same with skill acquisition when targeting behaviors for increase (such as tact capital letters, or write single digit numbers). Without observable and measurable operational definitions, accurate measurements will not occur, which will ultimately effect the outcome of our interventions. The ultimate goal of a good definition is to collect accurate data on the target response. Wallace, Sulzer-Azaroff, and Meyer (2019) describe four characteristics of a good measurement system: Sensitive - reflects subtle changes in the target response. Objective - a publicly verifiable and observable measurement. Reliable - consistent measurement, regardless of who measured or where the behavior was measured. Valid - the correct response was measured with an appropriate measurement procedure.

Assent

Assent Let's talk about assent. This may be a term you haven't heard before. In short, assent is the expression of approval or agreement by someone not legally able to give consent. An example of assent, for an 8-year-old individual, would be first having the parent of that 8-year-old sign an informed consent but then explaining the process of participating in today's activities and asking the 8-year-old if they would like to participate. Assent Can be given vocally/verbally, can be given in contract form, can be given by behaviors, and, most importantly, can be revoked at any time. What is the difference between assent and consent? Consent is usually legally required, it is given by individuals who are legally able, may often be given by someone who is not your client, depending on their legal ability to consent. This may be affected by their age and guardianship status. Consent can be revoked at any time without penalty. In general, consent is usually given for unemancipated minors, so individuals under 18 by a legal guardian. This can be different if the client who is over 18 has a legal guardian appointed. Assent is different because it is not usually legally necessary, but it is ethically necessary. It is given or withdrawn by the person participating the activity. Assent should be obtained in addition to legal consent and can be revoked at any time without penalty. So if a client says stop with their words or behaviors, we stop. These behaviors may look different for every client. For one client it might look like saying "NO" when asked to come to the workspace, for another client it might look like shutting down or stopping responding, for another client it might look like trying to elope or leave the workspace, and for another client it might look like flopping to the ground while doing an activity.

True

Baseline is optional in an alternating treatment design.

True

Baseline only needs to be long enough to determine a stable state of responding.

Task analysis for reversal designs:

Define target behavior and check for accuracy of recording. Determine measure. Collect baseline data when environmental variables are held constant. Implement intervention. Collect data until an established pattern appears. Withdraw treatment Collect baseline again. Reintroduce treatment. Repeat as needed.

Task Analysis of Multiple Baseline

Determine if it will be people, settings, or behaviors Start baseline condition for all Most stable baseline will receive treatment first, and so on Design concludes with treatment is constant across

Discussion

Discussion Limitations. The limitations of a study are the flaws in a study. Every study has them. Flaws in the design, participant attrition, history, maturation, experimenter bias... they go on and on.... If a researcher recognizes a flaw in their study, they are obligated to explain that it is there in the discussion section. Future Research Questions. This is the researcher's opportunity to pose future ideas that will build on or compliment the study they have described. This is a way to continue the process of research by building ideas from one investigation to another. Implications. The implications of research tell us what is important about the findings, why it is important, and whom it is important to.

Empirical research

Empirical research could be explained as a systematic intentional inquiry involving collection, analysis, and interpretation of data. Empirical research is not someone's "opinion" about what works or a story about how they did something or a case study in which anecdotal evidence is reported about one instance of an event (different from single-case design). Empirical research should be based on theory. It is very evident today that a lot of conspiracy theories and anti-science rhetoric are maintained by our affection to anecdotal evidence. Wikipedia defines anecdotal evidence as evidence collected in a casual or informal manner and relying heavily or entirely on personal testimony. We tend to believe what we experience more than what we do not explicitly experience. We are all susceptible to irrational fears and knowledge. Watch this short video from Richard Dawkins, showing how we are impressed by watching a law of physics that has been understood for 100s of years

continuous measurement

Frequency or Event Recording (Count) - How many times a behavior occurred. (There is a debate in behavior analysis regarding whether frequency is a synonym to count or rate, and if you are reading this, please email your instructor that Skinner defined frequency to be a measurement of behavior, but the BACB has chosen rate on the task list for your points). Rate - How many times a behavior occurred over time. It is almost always recommended to include a time component when counting behavior. Duration - How long did the behavior occur? Latency - How long did it take for the behavior to occur after the stimulus or prompt? Magnitude (or intensity) - How much force, energy, or exertion was utilized? (This is more challenging to measure and may require a measurement instrument or scale) Interresponse Time (IRT) - How much time occurred between responses? It is highly recommended that you utilize continuous measurement in order to obtain the most accurate information about occurrence of a target behavior, however, we also have discontinuous measures available to utilize when continuous measurement may not be feasible.

When the behavior meets the preselected goal

How does a changing criterion design demonstrate a functional relationship betweent the variables?

False

If you summarize literature in your own words, you do not have to cite it.

intervention

In ABA design, what does the B stand for?

Intervention 2

In an ABABABCBCA design, what does the C stand for?

AB design

In the AB design, there is an independent variable that has two levels (A versus B). In single case designs, the independent variable is administered in phases. So you will have an A phase and a B phase. The A phase is typically referred to as the baseline phase or the control phase. It represents the control group or the comparison group some may be familiar with in a group design. After establishing a baseline phase, the independent variable is manipulated and data are collected during the intervention, or B phase.

Inter-Observer Agreement (IOA)

Inter-Observer Agreement is the measure of how well two observers agree when measuring behavior by comparing how well the results of both observers match when recording the same behavior at the same time. This is an important way to evaluate treatment integrity, which refers to the accuracy with which the intervention or treatment is implemented. The text describes a couple variations of IOA, but here are the most common that you will encounter: Formula for Trial-by Trial IOA = # of trials agreement/total # of trials x 100 Formula for Scored Interval IOA = # of intervals both observers recorded occurrence of behavior/total # of intervals at least one observers recorded occurrence of behavior x 100 Formula for Unscored Interval IOA = # of intervals both observers recorded non-occurrence of behavior/total # of intervals at least one observer recorded non-occurrence of behavior x 100 The standard to establish reliability is to agree more than 80% of the time on at least 33% of the trials.

Introduction

Introduction Purpose. The purpose of the research is explicitly stated towards the end of this section. There should be no question to the reader why the research chose the topic, questions, and variables. Research question. The research question(s) is/are the specific question(s) that the researcher attempted to answer by conducting the study. It is a narrowed focus that is specifically addressed/answered using the Methodology described in the next section of the paper. We will discuss the types of questions in single case design in a bit. Independent and Dependent Variables. Remember, to determine a functional relation, we need to be certain that the change in the dependent variable is attributed to the independent variable. These will be explicitly stated in the introduction.

Philosophical Doubt

Jerry, a BCBA recently read that social stories have no effect in skill acquisition and discounts this information because he has a good experience using them in his clinic. What attitude of science is he violating?

alternating treatments design

Key Concepts Compares two conditions to determine which impacts behavior more. FA and problem behavior Control may or may not be present Level is the primary analytic tactic Good when you what to know what works better Have to be careful to randomly arrange the order so the learner does not predict a pattern.

Changing Criterion design

Key concepts Changing Criterion design is a series of phases with a predetermined criterion for each phase. A functional relation occurs when the behavior meets the criterion in each phase. After watching the video, email the instructor with the behavior used in the example. Bidirectional change is a procedure in which the following phase has a drop in criterion.

Graph formats utilized in Applied Behavior Analysis (Wallace et. al 2019):

Line Graph - Most frequently used in ABA. Displays data that can be scaled along some dimension, such as time or the order of responses in a sequence Standard Celeration Charts (SCC) - A semi-logarithmic chart that shows the frequency of performance and its celeration, i.e., growth of learning across time. Bar Graphs - Do not represent response sequences but are used to compare data or summarize performance with a condition or group of individuals. Scatterplot - An interval recording method that can help you discover patterns related to a problem behavior and specific time periods. Cumulative Records - Display rates of responses in the form of changes in the slope or curve of the record.

correlation

Maggie goes to the doctor for her fever. She is given antibiotics. Later that evening, her fever goes down. This has been her experience with antibiotics when she is sick. Based on the manner in which this was investigated, we would say this is an example of __________.

True

Marcus completes a functional analysis and it indicated that attention is the function for Timmy's tantrumming. Is it possible he demonstrated a functional relation?

True

Megan has a rash. The doctor prescribes a cream to apply eachhour to reduce itching. Every hour Megan applies the cream, the itch is reduced. Every hour Megan forgot to apply the cream, the rash is very itchy. We can discern a functional relationship based on these events.

Methods

Methods Participants. The participants are the individuals who contributed to the research by supplying information. These individuals are typically identified by demographic information in the written article. In Single Subject Designs, typically the participants are observed multiple times across phases over time. Design. The design is a map of the study. Single case designs describe the phases that participants will experience (Independent Variables) as well as the behavioral measures taken on individuals over time across phases (Dependent Variables). There are an unending number of Single Subject Designs because phases can be added to create increasingly complex designs. Some basic formats that govern how design phases are put together include Multiple Baseline Design, Reversal, Withdrawal, and Alternating Treatment Designs. There are more, but this will get you started. Measurement. Operational Definitions of all observation data. An operation definition includes both a specific description of what the observer is looking for as well as a means to measure that behavior including a timing period floor. Validity. External validity concerns the extent to which the results can be generalized to individuals outside of the sample we tested. Internal validity concerns the extent to which changes in phases (Independent Variables) caused changes observed in our measures (Dependent Variables). Social validity concerns the extent to which people view the relationship between the variables as important. And at this moment, we want to see which students find the written lecture important - if you are reading this, email your instructor with the phrase: written lectures are important. But back to the importance of the variables. If we find that our Independent Variable did cause a change in our Dependent Variable (high internal validity) then so what? What does it mean to our participants? Was it a socially significant/meaningful change? Reliability of Data. When using data taken from direct observations it is imperative to gain inter-observer reliability. Two independent observers take data for at least 33% of the observations. Then, their data are compared to assess their agreement. This establishes the reliability of the observation measure. Fidelity of Data. Inter-rater reliability data can be taken on the fidelity of the implementation of the Independent Variable - assuring that it was applied in the manner described in the Method. Using the same process, two independent observers take data for at least 33% of the observations of the implementation of the IV. Then, their data are compared to assess their agreement. This establishes the fidelity of intervention.

All of these: settings, people, behaviors

Multiple baseline can include:

People

Nonconcurrent multiple baselines can only be used when which variable is being measured?

Reversal/Withdrawal Design

One way to demonstrate internal validity is to simply repeat the AB design over and over so that we can show the data jump when we tell it to. These designs are called reversal designs in some cases and withdraw designs in others. These designs can become more complicated as you add variables. For instance, if you wanted to compare more than two levels of a dependent variable you would add a C phase into the mix ABACABAC design with a baseline in-between each intervention phase. You can see how quickly this will get complicated. Watch this short video on withdrawal/reversal designs.

analytic

Our ability to demonstrate causation over mere correlation is defined by what charactristic of Applied Behavior Analysis (Baer et al. 1968).

False

Principles of operant conditioning were first used with humans with mental illnesses such as schizophrenia, as opposed to developmental disabilities.

Reactivity

Reactivity is the phenomenon in which the process of recording behavior causes the behavior to change even before treatment is implemented. Imagine that you are hired to conduct a behavior assessment on a student, which includes a school observation. You walk into a full classroom with a clipboard. The students and teacher see you taking data, but they don't know what kind of data you are taking. Can you be sure that this observation is representative of typical behavior in the classroom, or do you think that everyone in the classroom might act differently because they being monitored? Either is possible, and you can reduce reactivity by finding less intrusive ways to observe (such as through a camera) or conduct multiple observations so that the client(s) become accustomed to the observer. Reactivity is also common when individuals engage in self-monitoring.

Research Summary

Research is not mere information gathering. Many of us may offhandedly say that we are going to the library to do some research. Most likely we are actually going to get information or collect facts of some sort. This may be information discovery but it is not research. Research is not watching conspiracy theories on YouTube and simply Googling the confirmation bias in the search bar. When we Google an idea, the algorithm is designed to produce results with the highest number of matches to your key words - it is NOT based on accuracy. Today, we face an increasing problem with "science" being disseminated without doing the science part. Most of us are guilty for getting our news from social media - some of us are better at filtering the noise. While this video is palatable for all audiences, it does a great job of highlighting some key areas to spot bad reporting where we could all pay attention to - it will be helpful for your discussion post this week!

Results

Results In a single case design, data interpretation is done by visual analysis of graphs displaying multiple data points within subjects and between phases using level, trend, and variability of data. We will learn more about data analysis as we progress.

Correlation

Sally uses a scatterplot to determine what time of day Timmy tantrums. A scatterplot is one way to determine.

True

Starting with baseline is not a requirment in reversal design.

Repeated acquisition design

Strength-Allows for the study of behaviors that cannot be reversed•Limitation•-Establishment and demonstration of item suitability/equivalence among stimuli

multiple baseline design

Strengths - Simple and flexible, and therefore most used - Helpful for use with behavior that cannot be reversed - Helpful for cases where IV withdrawal is not ethical or ideal •Limitation - Comparison, component or parametric analyses are difficult to conduct because of no reversal - Some students/behaviors/settings can be in baseline for extended periods of time

Discussion (why it's important).

The final section of a research article is the discussion section. This is where the author summarizes the finding for the reader and is allowed to make educated interpretations of the findings. The author typically interprets the findings in light of previous research by tying the section back to the introduction. A good author will highlight the strengths and weaknesses of their own study while interpreting the findings. Finally, authors will often offer future research questions that other researchers may want to investigate. In a sense, the article comes full circle with the discussion section in that it is tied to the content of the introduction.

Introduction (what and why).

The introduction of a research article: Introduces the topic of interest Defines the purpose of the research Provides a rationale for why the research is important (based on previous knowledge in the field of study) Proposes specific research questions that will help add to the body of knowledge in that field. These concepts should be made explicitly obvious in the introduction. There shouldn't be any guessing as to what the research is about, what the research questions are specifically, why the study was conducted, why the study is important to the field, or what contribution the author hopes to make. It is the authors responsibility to clearly convey these ideas to the reader, and thus it will be your responsibility when you write your paper for this class.

Methods (how).

The methods section of a research article describes in detail the methodology used to answer the research questions that were posed in the introduction. The introduction leads the reader to understand there are specific research questions that are important to be answered, so the methods section explains how those research questions will be answered specifically. While some may skim the methods section of a paper, it is important that students do not skim the lessons in the course. If you are reading this right now, please email your instructor with the phrase "Does reading lessons carefully increase exam scores at the end of the course" for bonus points. In this section of a research article you can find out who the participants were in the study, what kind of information was collected, what variables were manipulated or tested, what design was used to orchestrate the study, and what data was collected in addition to the main variables to insure that the study is valid and reliable.

True

The more times you remove the intervention, the stronger the evidence that the intervention caused the behavior change.

Correlation is not Causation

This is another threat to science - when we see things co-occur, we naturally want to assume that one event caused the other event to happen. But this is not so - even if we "observe" what looks like causation. In science, we still avoid using the word "causation" or "proves", because thanks to philosophical doubt, we are always amicable to new information changing previously held beliefs. In behavior analysis, you should be well aware of the functional relation. So, we can substitute our terminology in this phrase to include "correlation is not a functional relation". Watch this short video discussing the differences between correlation and causation (Links to an external site.) to learn about this concept in more depth. transcript of short video discussing the differences between correlation and causation

parsimony

We can assume functional relations demonstrate causation because of what attitude of science?

True

We should assume all interventions used in the alternating treatment design are effective in some capactity.

technological

What dimension of ABA is most important for accurate replications?

stopwatch

What recording system would you use to document duration?

Permanent Product

What recording system would you use to document if the dishwasher was emptied?

stopwatch

What recording system would you use to document the duration of time it took for students to exit after the fire alarm was sounded?

Mutiple Baseline Design

Which Design would you use for a behavior that cannot be reversed?

level

Which aspect of visual analysis refers to the average response?

Latency to Change

Which aspect of visual analysis refers to the change in the last data point of baseline to the data point in intervention?

Variability

Which aspect of visual analysis refers to the difference between the data points within a phase?

trend

Which aspect of visual analysis refers to the difference in direction of the data between phases?

All of these: variability, level, trend

Which aspect of visual analysis requires more than two data points?

Variability

Which aspect of visual analysis will demonstrate whether there is a steady-state of responding?

Multielement design

Which design is most often used for a functional analysis (or variations of a functional analysis?)

self-injurious behavior

Which is the only appropriate behavior for a withdrawal/reversal design - control aspects of the design?

When the behavior is dangerous, such as SIB

Which of the following scenarios would be a reason to skip the first baseline condition?

Introduction

Which section of a peer reviewed paper would you find the following; previous knowledge in the field

Reliability

Which section of methods would you find the following: IOA was 80% and was taken for 33% of the sessions.

Validity

Which section of the methods would you find the following: Description of how the research accounted for the extent i which changes in the intervention cause a change in the measured behavior and not other variables.

design

Which section of the methods would you find the following: This study utilized multiple baseline across participants.

Parametric

Which type of experimental question is the following: What effect will verbal praise have on responding behavior when faded from 1 minute to 5 minutes to 10 minutes

comparison

Which type of experimental question is the following: Will functional communication training alone or functional training with a multiple schedule of reinforcement decrease escape behavior?

demonstration

Which type of experimental questions is the following: Will a token economy increase responding in the classroom?

True

You must use the same person in MB design across settings

Baseline Logic

_________ suggests that if no intervention occurs, the observed behavior will continue on a similar level or trend.

Withdrawal/Reversal

__________ is the strongest design in single case design.

Baseline

___________ serves as the companion to evaluate the effects of an intervention.

Experiementation

____________ involves a clear experimental question, a plan for measuring and recording the variables, analysis of the results, and critical peer review.

Experimentation

_____________ involves clear experimental question, a plan for measuring and recording the variables, analysis of the results, and the critical peer review.

Functional Relation

_____________ is when you can say with confidence that event A likely cause event B to occur.

Correlation

is when there is a predicted relationship between two variables observed.

PROCEDURAL INTEGRITY

•Create a task analysis of your procedures •Go through your procedures and write down each step•Create a table with the step in one column and a yes and no box in the second column •Train a second observer on the procedures of your intervention so they will be able to identify if you completed the step correctly •Have a second observer observe you 20%-33% of the time and record if you completed each step as designed •At the end of the session divide the number of steps completed correctly by the total number of steps to arrive at a percentage

MULTIELEMENTDESIGNS

•Strengths-Experimental control can be demonstrated as long as response differentiation occurs in at least one condition-Two or more experimental conditions can be analyzed using a single participant-More rapid results than other designs-Can be used with behaviors that are not reversible -Do not need to take treatment away •Limitations-Multiple treatment interference -Beware of interaction effects-Competing treatments may make it hard for the participant to discriminate between them