Model Procrastination Loops in LOOPY

Build a LOOPY feedback model and test a tiny-start intervention.

Introduction

30 Second Summary

An uncomfortable task can make avoidance feel useful for a moment. Later, less time raises stress before the same pattern starts again.

In this project, you will build an interactive habit laboratory in LOOPY. The simulation lets you compare a one-way explanation with a feedback loop before you test a Tiny start intervention.

What You'll Build

Raise Task discomfort in your finished model to watch the habit feed itself before Tiny start pushes Avoidance downward.

By the end of this project, you'll have:

A linear model where you can raise Task discomfort and watch the signal stop at Time pressure.
A reinforcing loop where signals repeatedly circulate through Task discomfort, Avoidance, Time pressure, and Stress.
A tested balancing intervention that lets you trigger Tiny start and observe how it changes Avoidance.
Secret Mission: Add a delay to the intervention. Test whether Stress rises before the benefit reaches Avoidance.

Are there any prerequisites?

You need a Windows device with Microsoft Edge plus browser access to LOOPY v1.1. Prior experience making system maps is helpful, while no account or paid service is required.

Before We Start

Before any hands-on work, this is your moment to use systems thinking to examine procrastination as interacting causes and effects. Naming one usual trigger and one desired outcome gives that investigation a clear focus.

Open Your Systems Thinking Lab

A static map cannot show how a habit changes over time. You need a working simulation canvas before you can test the procrastination pattern.

In this step, you will confirm that Microsoft Edge can run LOOPY v1.1. You will load its editable starter model without creating an account.

In this step, get ready to:

Confirm that Microsoft Edge is available on Windows.
Load the LOOPY v1.1 editable starter model.
Confirm that the full editing interface is visible.

Open Microsoft Edge

Microsoft Edge is the Windows browser for this lab. A successful launch confirms the first part of your setup.

Press the Windows key to open Windows search.
Type Microsoft Edge into the search field.
Press Enter to launch the browser.

✔️ Microsoft Edge opens

Good, Microsoft Edge is ready to run your simulation.

ⓧ Microsoft Edge is unavailable

Microsoft provides an official Windows installation flow. You can use another installed browser to reach it.

Visit the official Download Microsoft Edge page in another installed browser.
Follow Microsoft's Windows installation flow.
Approve the Microsoft installer if Windows asks for permission.
Return to Windows search after the installation finishes.
Type Microsoft Edge into the search field.
Press Enter to launch the installed browser.

Microsoft Edge Still Unavailable?

An incomplete installation can keep Microsoft Edge out of the search results.

Help me troubleshoot the Microsoft Edge installation on Windows.

Load LOOPY v1.1

LOOPY is a browser-based tool for thinking in systems. Its starter model confirms that the simulation canvas is ready for editing.

Why Use a Browser Lab?

LOOPY opens directly into an editable simulation. You do not need to create or sign into an account.

Click the address bar at the top of Microsoft Edge.
Enter https://ncase.me/loopy/v1.1/ into the address bar.
Press Enter to load LOOPY v1.1.

That gets your systems lab online. You should see an editable starter model filling the canvas.

Starter Model Not Loading?

An incomplete page load can leave the canvas blank.

Confirm that the address bar shows https://ncase.me/loopy/v1.1/.
Select the circular reload icon beside the address bar.

Help me troubleshoot a blank LOOPY canvas in Microsoft Edge.

Confirm the lab controls

The left toolbar contains the editing tools. The control below the canvas starts simulation playback.

Before you inspect the interface, do you think the complete editing workspace loaded?

Hover over the left toolbar to reveal its labels.
Confirm that the toolbar includes PE(N)CIL.
Confirm that the toolbar includes (T)EXT.
Confirm that the toolbar includes MO(V)E.
Confirm that the toolbar includes (E)RASE.
Locate the Play control below the canvas.

✔️ All controls are visible

Your systems thinking lab is ready for experiments.

The editable starter model is loaded in Microsoft Edge.
The left toolbar shows all four required editing controls.
The Play control is visible below the canvas.

ⓧ A control is missing

A partially loaded interface can hide part of the toolbar or the playback area.

Select the circular reload icon beside the address bar.
Wait for the editable starter model to reappear.
Hover over the left toolbar again.
Repeat the control check above.

Control Still Missing?

The page may still be loading incompletely if a control remains missing after a reload.

Help me find a missing LOOPY control in Microsoft Edge.

Your simulation canvas is ready. Next, you will replace the starter model with a three-variable habit chain that visibly stops.

Build the Linear Model

Your LOOPY canvas is ready in Microsoft Edge. You can now turn one procrastination episode into a model you can run.

A linear explanation captures one episode of procrastination. It hides the feedback that makes the habit recur.

In this step, get ready to:

Replace the starter content with three named habit variables.
Connect the variables with positive causal relationships.
Run the linear model to test how its signal behaves.

Clear the starter model and add variables

This first model uses three circles to represent the parts of one procrastination episode. Their left-to-right layout makes the sequence easy to follow.

Switch back to the LOOPY tab in Microsoft Edge.
Select (E)RASE from the left toolbar.
Drag across each piece of starter content until the canvas is blank.

You should now see an empty canvas ready for your habit model.

Select PE(N)CIL from the left toolbar.
Draw a circle on the left side of the canvas.
Enter Task discomfort in the circle's Name: field.
Draw a second circle in the center of the canvas.
Enter Avoidance in the second circle's Name: field.
Draw a third circle on the right side of the canvas.
Enter Time pressure in the third circle's Name: field.

Good progress. Your canvas now shows the three variables in the order that one procrastination episode can unfold.

Connect the positive relationships

Causal polarity describes the direction of influence between two variables. A positive relationship pushes the destination in the same direction as the source.

Select PE(N)CIL from the left toolbar.
Draw one stroke from inside Task discomfort to inside Avoidance.
Set the new arrow's Relationship: control to the positive side.

What does positive polarity mean?

A positive causal relationship sends a change in the same direction. If Task discomfort rises, it pushes Avoidance upward.

Select PE(N)CIL from the left toolbar.
Draw one stroke from inside Avoidance to inside Time pressure.
Set the second arrow's Relationship: control to the positive side.

You should see one positive arrow from Task discomfort to Avoidance. A second positive arrow should run from Avoidance to Time pressure.

Run the linear model

A simulation turns a map into behavior over time. A consistent one-click increase lets you test what this structure can explain.

Before you press Play, do you think one increase will keep circulating or reach an endpoint?

Select Play below the canvas.
Point to the upper half of Task discomfort.
Click the upper half once to raise Task discomfort.

You'll see the signal move from Task discomfort through Avoidance to Time pressure. It stops there.

That shortfall is intentional. The structure explains one event chain without showing a recurring habit.

Signal moving the wrong way?

Select Stop to return to editing mode.
Check that the arrowheads point from Task discomfort to Avoidance.
Check that the arrowheads point from Avoidance to Time pressure.
Check that both Relationship: controls use the positive side.

Ask for help diagnosing the model if the signal still behaves differently: help me troubleshoot my LOOPY signal path

Select Stop below the canvas.
Select (T)EXT from the left toolbar.
Add If nothing feeds back, this is an event chain, not a recurring habit. to an empty area of the canvas.

You should now be back in editing mode. Your three-variable chain should remain visible beside the prediction text.

✔️ My linear model is complete

Strong result. You have tested a model that visibly reaches its endpoint instead of explaining recurrence.

ⓧ I'd like to double check the model

Compare your canvas with this completed state.

The editable LOOPY canvas contains none of the starter content.
A circle named Task discomfort appears on the canvas.
A circle named Avoidance appears on the canvas.
A circle named Time pressure appears on the canvas.
A positive causal arrow runs from Task discomfort to Avoidance.
A positive causal arrow runs from Avoidance to Time pressure.
The canvas displays If nothing feeds back, this is an event chain, not a recurring habit..
The model is back in editing mode after the simulation test.

Your model now proves what a straight chain leaves out. Next, you will connect the consequences back to the trigger and test whether the behavior repeats.

Reveal the Reinforcing Loop

Your linear LOOPY v1.1 model captured one procrastination episode. The signal reached Time pressure before the story stopped.

A recurring habit needs consequences to feed back into an earlier variable. You will add Stress to close a reinforcing loop that can keep the pattern moving.

In this step, get ready to:

Add Stress to the existing event chain.
Draw two positive relationships that close the feedback path.
Demonstrate the named loop with one controlled run.

Close the loop with Stress

The missing feedback sits between Time pressure and Task discomfort. Adding Stress makes that consequence visible before it feeds back into the trigger.

Leave your event-chain prediction from the previous step in place.
Select PE(N)CIL in the left toolbar.
Draw one large circle in an empty area beside Time pressure.
Select the new circle.
Enter Stress in the Name: field.

Good progress. Your model now has a fourth variable that captures the pressure created by delay.

Return to PE(N)CIL in the left toolbar.
Draw an arrow from Time pressure to Stress.
Select the new arrow.
Set the Relationship: control to the positive side.
Draw an arrow from Stress to Task discomfort.
Select the second new arrow.
Set its Relationship: control to the positive side.

That closes the structure. You can now follow arrows from Task discomfort through all four variables before returning to the starting point.

Why do positive links reinforce the loop?

A positive relationship sends change in the same direction. An increase at the source pushes the destination upward.

Every relationship in this closed path is positive. Each trip around the loop can compound the original increase.

Arrow pointing the wrong way?

Use (E)RASE to remove an arrow that points to the wrong circle.
Redraw the relationship with PE(N)CIL from the source circle to the destination circle.
Select the replacement arrow.
Choose the positive side of the Relationship: control.

Help me check the direction and polarity of my new relationships.

Name the recurring pattern

A loop label gives the structure a short name that you can use while explaining its behavior. The complete story should return to the variable where it began.

Point to Task discomfort on the canvas.
Say this story aloud: More Task discomfort leads to more Avoidance. More Avoidance leads to more Time pressure. More Time pressure leads to more Stress. More Stress leads to more Task discomfort.
Select (T)EXT in the left toolbar.
Add R1: Avoidance spiral beside the closed path.

The loop now has a name that captures its behavior. Its final relationship reconnects the consequence to the original trigger.

What makes this pattern recur?

The signal can return to its starting point because Stress affects Task discomfort. That return path allows another trip around the loop.

Each positive relationship carries the increase forward. Repeated circulation can intensify the pattern over time.

Run the closed loop

A controlled simulation reveals whether the new structure behaves like a recurring pattern. You will use the same one-click increase from the previous test.

Before you select Play, consider whether the signal will stop at Stress or return to Task discomfort.

Select Play below the canvas.
Point to the upper half of Task discomfort.
Click once to raise Task discomfort.
Watch the signal travel around the closed path.

That is the structural shift you were looking for. You should see signals repeatedly circulate through all four variables.

Select Reset to restore the starting amounts.
Confirm that all four circles return to their starting amounts.
Select Stop to return to editing mode.

The left editing toolbar is visible again. All four circles show their starting amounts.

The canvas keeps four positive relationships. The original event-chain prediction remains in place.

The R1: Avoidance spiral label sits beside the closed path.

Signal still stops?

Check that an arrow connects Time pressure to Stress.
Check that another arrow connects Stress to Task discomfort.
Select each new arrow.
Confirm that its Relationship: control uses the positive side.

Help me find why the signal stops in my closed loop.

Your model now exposes the feedback structure that can make procrastination repeat. Next, you will add a balancing intervention that acts directly on Avoidance.

Test a Balancing Intervention

Your closed reinforcing loop now carries a change all the way back to Task discomfort. You watched that signal circulate repeatedly.

Describing the trap still leaves you without a way to test change. You will add an intervention that acts on Avoidance.

In this step, get ready to:

Add the Tiny start intervention to the existing feedback structure.
Test how its negative relationship changes Avoidance.
Save the completed model as a link for a repeatable demonstration.

Build the balancing path

A balancing loop counters change through an opposite-direction influence. In LOOPY v1.1, Tiny start creates that pressure on Avoidance.

Use PE(N)CIL from the left toolbar to draw a fifth circle beside the existing path.
Set the circle's Name: field to Tiny start.

You should see a fifth circle named Tiny start on the canvas.

Use PE(N)CIL to draw an arrow from Task discomfort to Tiny start.
Select the new arrow to show its settings.
Set the Relationship: control to the positive side.

You should see a positive causal path from Task discomfort to Tiny start.

Use PE(N)CIL to draw an arrow from Tiny start to Avoidance.
Select the new arrow to show its settings.
Set the Relationship: control to the negative side.

You should see a negative causal path from Tiny start to Avoidance.

How does the negative link work?

A negative relationship multiplies the arriving signal by -1 when it reaches Avoidance. The direction flips.

The new closed path has one opposite-direction link. That makes its overall behavior balancing.

Can't make the link negative?

Select the arrow from Tiny start to Avoidance before changing Relationship:. Reselect that arrow if a different relationship changes.

Help me fix the Tiny start relationship.

Test the intervention

A controlled simulation changes one structure while keeping the stimulus consistent. Resetting the starting amounts makes the comparison with your reinforcing-only run meaningful.

Select (T)EXT from the left toolbar.
Add the label B1: Start before ready beside the new path.

You should see B1: Start before ready beside the balancing path.

Before you run the test, do you expect the intervention signal to move Avoidance in the same direction as the spiral?

Select Reset below the canvas.
Select Play below the canvas.
Point to the upper half of Task discomfort.
Click once to raise Task discomfort.
Watch the signal pass through Tiny start.
Compare the response in Avoidance with your reinforcing-only run.
Select Stop after you observe the intervention.

The increase in Task discomfort should send a signal through Tiny start. The negative relationship should then push Avoidance downward.

That is the key shift: your model now tests a response that counteracts the spiral.

What did the comparison show?

The model is a hypothesis about the habit's structure. A consistent one-click stimulus makes the two runs comparable.

The changed response shows how a structural intervention can alter behavior inside the system. It does not claim to forecast an exact real-world result.

Save the model as a link

A saved link preserves the model for another demonstration. Opening it in a new Microsoft Edge tab checks that the structure survives outside the current editing session.

Select save as link in the editing sidebar.
Locate the generated URL beneath copy your link:.
Copy the generated URL.
Record the copied URL here: your saved LOOPY model link.
Use the browser's tab bar to open a new Microsoft Edge tab.
Paste your saved LOOPY model link into the address bar.
Press Enter to load the saved model.

You should see the Tiny start node in the reopened canvas. The negative path should still connect it to Avoidance.

The label R1: Avoidance spiral should remain visible. The label B1: Start before ready should remain visible.

The original event-chain prediction should also remain on the canvas.

Before you replay the saved copy, do you expect the intervention to behave like it did in the original tab?

Select Play in the reopened model.
Point to the upper half of Task discomfort.
Click once to raise Task discomfort.
Watch the signal pass through Tiny start.
Select Stop after the negative intervention signal reaches Avoidance.

The reopened model should play successfully. You should see the negative intervention signal reach Avoidance again.

You now have a shareable demonstration that preserves the tested feedback structure.

Saved link missing your changes?

A saved link captures the model state present when you generate it.

Return to the original editing tab.
Confirm that the negative arrow reaches Avoidance.
Generate a fresh link with save as link.
Open the fresh URL in the new tab.

Help me fix my saved LOOPY model.

Secret mission

Make the Intervention Arrive Late

An immediate Tiny start can look more powerful than it feels in real life. Add a delayed intervention path to discover what Stress does while the benefit is still travelling through the system.

Clean Up Your Resources

Everything in this project ran inside LOOPY v1.1 in Microsoft Edge. This project has no ongoing costs. Choose whether to keep the delayed model ready, close it for later, or remove its saved link.

Resources you used:

Microsoft Edge tabs containing the delayed LOOPY model.
A generated LOOPY link that reopens the delayed model.

Keep everything running

No action needed. Choose this if you want the delayed model ready for another demonstration.

Keep the verified Microsoft Edge tab containing the delayed model open.
Keep the generated LOOPY link in a location you control.
Reuse the generated link whenever you want to demonstrate the delayed intervention.

Pause - I'll come back to this later

Close the active model to clear your browser workspace. Keep its generated link so you can continue later.

Paste the generated LOOPY link into a note you control.
Close each Microsoft Edge tab that contains the LOOPY model.
Reopen the generated link when you are ready to continue testing.

Delete - I don't want to use this again

Remove the active model plus every retained link. Follow each action below.

Open Microsoft Edge's browsing history page.
Find the reopened delayed-model URL in the history list.
Delete the delayed-model URL from the history list.
Copy the word cleared from this instruction to replace the generated URL in your Windows clipboard.
Delete every additional copy of the generated LOOPY link that you stored.
Close every Microsoft Edge tab containing the delayed model.

The project leaves no server, account, or paid resource running after these actions.

Nice Work!

You did it! Your interactive LOOPY model now shows how procrastination can reinforce itself. It also demonstrates how a Tiny start intervention can change that behavior.

You've learned how to:

Built a visible event chain from Task discomfort through Avoidance to Time pressure. You observed why its signal stopped.
Closed the chain into a reinforcing loop labeled R1: Avoidance spiral. You watched the habit repeatedly feed itself.
Tested a balancing intervention with the same one-click input. You saved a link that reopens the completed model for demonstration.
Completed the optional Secret Mission by redrawing the negative Tiny start relationship as a delayed path. You observed whether Stress rose before the benefit reached Avoidance.

Ready to quiz yourself?