The First set of steps of ABA is for the purpose of Understanding the System. We do so with the hopes of eventually being able to intervene in the system to change the system’s behavior. This intervention could take the form of a direct action or recommendations for action.

Step 1.1 ST Loop Analysis, Test Your Mental Model or Prototype

Regularly Test Your Mental Model/Prototype throughout the ABA Process

Throughout the ABA process you should be thinking about the ST Loop.

In particular about the need for testing your mental model or ‘prototype’ in the real world. Although this should be occurring all the time, there will be critical points where it is important and you should do it formally and purposefully. Those include:

During the DSRP Mapping step, you should be checking your map against reality at least 2-3 times in the creation of your map. See this FAQ for ways to test your mental models.

This is especially true when taking Perspectives. Be sure to check that the Perspectives you are taking would agree with your synopsis.

When you get to the Intervention/Recommendation Rubric and the Specific Recommendations steps. Think of these as prototypes for field use. You’ll want to check with your audience (whoever is doing the intervention) to see if the prototype you designed is the one they need/want.

Below, keep a record of your St Loop Checks. Items in bold are the critical ones mentioned above but feel free to do your ow checks or multiple checks along the way. Return to this form later in the process as you do checks.

Test #, Name, Date

With Whom, On What (sample, etc.)

KMMM Method

(Observe, Question, Experiment, Synthesize)

Feedback Received

How it Changed Your Mental Model or Prototype

Example: #1 Initial DSRP Analysis Test

06/23/2020

Experts and People in the field

Observation + Questioning

Mostly accurate, needed to tweak a few terms and was missing some critical relationships

Changed distinctions, added relationships, RDS’d a few Rs

Add Your Test Here.

Add more rows as necessary.

** We recommend that at the least you get real-world feedback at Step 1.2, 2.1, and 2.2 but ideally you would test and receive feedback all along the way

Step 1.2 DSRP Analysis

DSRP Mapping of Your System in a Structure Determines Behavior (SdB) Map

The first step in ABA is to engage the ST Loop and map your system of interest for the purpose of understanding. Avoid ‘problem solving’ at this stage and just be genuinely curious about how and why the system behaves the way it does. This step is the bulk of an ABA analysis and takes the most time of all the other steps combined. In fact, this step is really a set of steps detailed in this step-by-step worksheet. Go to this worksheet, make a copy of your own and walk through it.

Insert the link or image of your DSRP Map here

Step 1.3 SdB Analysis

Place your DSRP map in an SdB Template

No matter how complex your system is, the system structure that is contained in your DSRP Analysis (Map) is your best guess as to why the system behaves the way it does. So, in this special case, you’re going to take the DSRP Map of Your System and insert it into the right side of this template-map where the system structure determines the behavior of the system.

First, simply replace the link to the map in the template with the link to your map.

Next, fill in the System Behavior box with a list of behaviors

You can do this in the template map, or in the table below, or both.

Make a copy of this Map template here. You can add a link from your systems map to the Systems Structure block.

System Structure	System Behaviors
Insert the link or image of your SdB Map here
Add your DSRP Analysis Map link here

Step 1.4 POSIWID Analysis

Purpose Of a System Is What It Does (POSIWID)

Systems and management scientist Stafford Beer developed an important and popular systems thinking heuristic known by the acronym POSIWID: “The purpose of a system is what it does.” When assessing alignment, we need to focus on what the system actually does rather than its stated, ostensible, original, or ideal purpose (since these often do not match). You can think of POSIWID as the ‘goal state’ of the system or the place that it will naturally end up left to its own. The POSIWID of a system is also synonymous with its emergent property or ‘emergence.’ This step involves two substeps:

Step 1.4.1 POSIWID Comparison

Differentiate the current POSIWID from the desired [future] POSIWID

In the following table, you can see the value of POSIWID thinking. It flips the system and its purpose on its head. Instead of looking at the results of a system as problematic, you look at the results of the system as by design. Instead of the system being ‘badly designed’ to serve a ‘good’ outcome, it is ‘brilliantly designed’ at bringing about a ‘bad outcome’. The worse the result, the more clear the value of POSIWID thinking. Read the examples, and then try it yourself!

Table A
1	2	3
Identify a system that you think is failing or isn’t working well enough	Describe the specific results of the failure	Turn your description on its head with POSIWID
Example: The U.S. education system	Too many disengaged kids	The U.S. education system is exceptionally well designed and good at its purpose of controlling and boring kids so that they disengage.
Example: Acme's software product	Too much churn	Acme's software product is extremely well designed and effective at its purpose of getting customers interested enough to sign up and then leave.
Your turn: Feel free to delete the examples to create more space.	Your turn:	_____________ is exceptionally well entry from column 1 (system) designed and good at its purpose of ______________________. entry from column 2 (results)

Now describe the POSIWID for what you would like to see in the future

Table B
1	2	3
Identify a future system	Describe the specific future results/behavior you would like to see	Describe the System’s POSIWID
Example: The U.S. education system	Kids engaged; prepared for life (rather than school)	The U.S. education system is exceptionally well designed and good at its purpose of engaging kids and preparing them for life.
Your turn:	Your turn:	_____________ is exceptionally well entry from column 1 (system) designed and good at its purpose of ______________________. entry from column 2 (results)

Step 1.4.2 Root-Difference Analysis

Identify the Root-Difference between the current POSIWID and the desired [future] POSIWID

Now we want to compare and contrast the two POSIWIDs (current and future state) to identify what is fundamentally different about them. We use the table below to summarize what we did above and then look at the core differences:

CURRENT System POSIWID

FUTURE System POSIWID

Root Difference

The U.S. education system is exceptionally well designed and good at its purpose of controlling and boring kids so that they disengage.

The U.S. education system is exceptionally well designed and good at its purpose of engaging kids and preparing them for life.

At its core, POSIWID difference is:

learner centered vs. teacher centered;
thinking centered vs. content centered;
Authentic engagement vs. strategic or ritual compliance
Goal orientation is life success vs. school success

Your answer from Table A Column 3

Your answer from Table B Column 3

Add your analysis of how they are different here. Feel free to delete the examples to create more space.

Step 1.5 CAS Analysis

Identify the CAS Elements of the System

The purpose of the CAS analysis is to identify the various elements that make up your CAS and to use these elements to identify where you can be most effective in your intervention or recommendations (below the line in the green area).

Step 1.5.1 General CAS Analysis

Describing Complex Adaptive Systems (CAS) and its Agents and Simple Rules

Complete the table below based on the examples given.

	Flock of Birds	COVID	Your CAS
Who/What are the agents?	Individual birds	COVID is hoax/scam people COVID is real/deadly people	Add your analysis here. Feel free to delete the examples to create more space.
What is the agent’s perspective?	Local: paying attention to their nearest neighbors	Local: paying attention to their nearest neighbors
What are the simple rules?	Direction Distance Avoidance	hoax/scam Life as usual real/deadly Distance Quarantine Mask vaccine
What is the system-level behavior (emergent properties)?	Flock; superorganism	hoax/scam Death toll/other bad metrics *climb* real/deadly Death toll/other bad metrics *fall*
How do the items above lead to the system-level behavior?	Uncountable number of interactions lead individual birds to adapt locally causing flock to adapt globally	Uncountable number of interactions lead to more/less infection

Step 1.5.2 Agents/Rules Analysis

List and Detail the Agents/Rules in your CAS

From the table above, fill in the table below to identify and isolate the agents. We will use this list later when devising interventions/recommendations. Be sure to give your agents specific names that are clear and won’t cause confusion. If need be provide a description to further distinguish one type of agent from another.

Agent Type Name	Description	Simple Interaction Rules
Example:
Anti-science people	People who believe COVID is hoax/scam and who are anti-vaccine, anti-mask, anti-distancing, anti-quarantine, anti-shutdown, anti-adapting to data, etc	Life as usual
Pro-science people	People who believe COVID is real/deadly and who are pro-vaccine, pro-mask, pro-distancing, pro-quarantine, pro-shutdown, pro-adapting to data, etc.	Distance Quarantine Mask Vaccine Data Repeat
Add your agents here. Add a row for each agent. Feel free to delete the examples to create more space.

Step 2.0 Design Interventions/Recommendations

Step 2.1 Recommendation-Rubric Analysis

Develop Intervention/Recommendation Design Specs or “Rec-Rubric”

Come up with 3-10 “Recommendation Principles” that all future recommendations must meet. This list acts as design principles (and as a backcheck) for proposed recommendations to see if they pass this litmus. In other words, there may be a nearly infinite number of creative, specific recommendations that could be made for any given problem, system, or issue. And, you may need to adapt and adopt new recommendations based on new learning. However, this step requires that you look beneath the specific recommendations to identify the "pattern(s) that connects" any and all recommendations (current and future). For example, if the desire is to change a system, one can imagine many specific recommendations. But a recommendation principle would be: "Cannot maintain, reinforce, or engender the status quo." Therefore, if we make a recommendation that violates said principle, it should be rejected.

Often this rubric will be heavily informed by the Root Difference Analysis that you did. For example

Root Difference Analysis for U.S. Education System

Possible Rec-Rubric

At its core, POSIWID difference is:

learner centered vs. teacher centered;
thinking centered vs. content centered;
Authentic engagement vs. strategic or ritual compliance
Goal orientation is life success vs. school success

No specific recommendation should violate these rules:

Must be learner centered rather than teacher centered;
Must be thinking centered rather than content centered;
Must increase or incentive authentic engagement rather than strategic or ritual compliance
Must reinforce goal-orientation of life success rather than school success

Add your Root Difference Analysis here.

Add your Rec-Rubric items here. Add rows if needed. Feel free to delete the example to create more space.

Step 2.2 Specific Recommendations

Develop and Backcheck Specific Recommendations

Identify 3-10 recommendations. Be wildly creative but constrained by the principles in the Rec-Rubric. Any recommendation that adheres to the principles in the Rec-Rubric is fair game. The structure of this step is to allow for structured or constrained creativity. Thus, the actual recommendations can be wildly creative and “out of the box” with the one condition that they adhere to the design principles provided by the Rec-Rubric.

Example:

Rec-Rubric
No specific recommendation should violate these rules: Must be learner centered rather than teacher centered; Must be thinking centered rather than content centered; Must increase or incentive authentic engagement rather than strategic or ritual compliance Must reinforce goal-orientation of life success rather than school success
Specific Interventions/Recommendations	Backcheck (Does it violate any of the Rubric items?)
Intervention/Recommendation #1: Base testing, measurements and assessments for learners, teachers, administrators, schools, districts, and States on what was learned rather than what was taught by using portfolio assessments and developing the technology to support it.
Intervention/Recommendation #2: Incentivize and measure curriculum based on structural cognition rather than specific content by redesigning the classroom as a reification or a redesigned lesson architecture.
Intervention/Recommendation #3 Incentivize authentic engagement and disincentivize strategic or ritual compliance by making these ideas explicit and providing examples of behaviors associated with each.
Intervention/Recommendation #4 Redesign curriculum to focus content-learning on life-skills that are constantly being revised based on feedback from focus groups that represent parents, business, government, NGOs, scientist, and research sectors.

Now do your own. We suggest that this step be done in a creative environment, possibly including others in a brainstorm utilizing the Rec-Rubric as guide rails.

Rec-Rubric
Add your Rec-Rubric here for easy reference
Specific Interventions/Recommendations	Backcheck (Does it violate any of the Rubric items?)
Intervention/Recommendation #1:
Intervention/Recommendation #2:
Intervention/Recommendation #3
Intervention/Recommendation #4

It is important that you return to Step 1.1.1 to “loop” your Rec-Rubric specific Recommendations. A recommendation is a kind of ‘prototype’ that should be tested with stakeholders and feedback should be used to adapt mental models and the prototype.

Congratulations!

You’ve now completed your Agent Based Approach (ABA).