As designers and researchers, we constantly face a challenge: how do we measure something that’s inherently personal? We can track clicks, time-on-task, and success rates. But how do you quantify "immersion," "plausibility," or "comfort"? What do you do when a dozen different factors—from the visual design to the control scheme—all mix together to create a single, subjective feeling?

Traditional methods like A/B testing fall short when you have more than a few variables. Asking users to compare 50 different combinations in a survey is a recipe for confusion and unreliable data.

Enter the Subjective Matching Method, a clever and user-centric approach to untangle this complexity. It turns the evaluation process into a game where the user actively builds their own perfect experience. Let's break down what it is, why it's so powerful, and how you can use it.

The Core Concept: Adjusting the Knobs to Find the "Sweet Spot"

Imagine you're trying to create the perfect listening experience on a high-end audio system. You have knobs for bass, treble, balance, and volume. Do you listen to 50 pre-set combinations and try to rank them from memory? No. You start with a reference track, then you adjust each knob, one by one, until it sounds just right.

That's the Subjective Matching Method in a nutshell. Instead of presenting users with a rigid set of options to compare, you give them the controls and ask them to adjust the system until it "matches" a target sensation.

Methodology Background: This technique has its roots in psychophysics and was popularized in the field of Virtual Reality by researchers like Mel Slater to evaluate the concepts of Place Illusion (the feeling of being there) and Plausibility Illusion (the feeling that the scenario is actually happening). The goal is to move beyond simple "I like it" ratings to understand the specific ingredients that make an experience feel real and believable.

Why and When to Use Subjective Matching

This method isn't for every project, but it shines when your design goals involve:

1. Taming Complexity: It's perfect for experiences with many interacting factors. Instead of a "combinatorial explosion" of conditions to test, the user navigates the possibility space themselves, making the study manageable.
2. Uncovering User Priorities: The sequence of adjustments is just as important as the final result. What do users change first? This reveals their priorities. If most users immediately fix the controls before touching the visuals, you've learned something critical about what drives their experience.
3. Measuring the "Unmeasurable": It excels at evaluating fuzzy, holistic qualities like plausibility, presence, engagement, or comfort. The final "matched" configuration is a concrete recipe for achieving that subjective feeling.
4. Reducing Cognitive Load on Participants: The method is intuitive. It relies on immediate, in-the-moment judgment ("Does this feel better?") rather than asking participants to recall and compare dozens of past experiences from memory.

How to Use It: A Step-by-Step Guide

Implementing this method requires careful setup. Here’s a practical workflow:

Step 1: Define Your Factors and Levels
Identify all the "knobs" a user can turn. These are your independent variables. For each factor, define incremental levels of improvement (e.g., low, medium, high).

Step 2: Establish the "Gold Standard" Experience
Create what you theorize is the "optimal" configuration by setting all factors to their highest level. The first thing a participant does is experience this. This gives them a clear benchmark—a mental target of the "best" feeling they are trying to match.

Step 3: Start from a Basic Configuration
The experiment begins with all factors at their lowest setting. To ensure the starting point doesn't bias the outcome, it's good practice to run trials starting from different initial configurations.

Step 4: Run the Iterative Matching Task
This is the core loop:

• The user performs a task with the current configuration.
• They are then allowed to make one single upgrade to one factor.
• They perform the task again with the upgraded configuration.
• This cycle repeats until the user declares, "This feels as good as, or better than, the gold standard I experienced at the start."

Step 5: Record the Results
The two key pieces of data you collect are:

• The Matched Configuration: The final combination of factor levels the user settled on.
• The Transition Path: The chronological sequence of upgrades the user made to get there.

A Real-World Example: Plausible Walking in VR

A recent study by Gao et al. (2024) provides a perfect illustration of this method in action. The researchers wanted to know what makes body-centric locomotion in VR feel plausibly like real walking.

They identified five key factors:

1. Point of View (PoV): Third-person vs. first-person.
2. Avatar Appearance: Abstract spheres, a stickman, or a personalized realistic avatar.
3. Body Part for Control: Knees, arms, or head.
4. Transfer Function: The math that turns movement into speed (Power, Piecewise, or Linear).
5. Speed Coefficient: A multiplier for normal vs. fast walking speed.

Following the method, participants first experienced the "optimal" configuration (first-person, realistic avatar, etc.). Then, they started from a basic setup and iteratively upgraded one factor at a time until the walking sensation felt plausible and engaging.

What did they find? By analyzing the transition paths, they discovered that Point of View was the most critical factor. Participants almost always upgraded from third-person to first-person immediately. Furthermore, the most frequently chosen "matched configuration" was not the theoretical "optimal" one, revealing a more nuanced combination of arm-swinging control and natural walking speed that users found most believable. This is a deep insight they might have missed with a simple questionnaire.

Final Thoughts

The Subjective Matching Method is a powerful tool for any designer or researcher working on complex, interactive experiences. It respects the user's subjective reality and empowers them to show you what matters, rather than just telling you. By turning evaluation into a process of discovery, it provides richer, more actionable insights into how to create experiences that don't just work well, but feel right.

---

Reference:

Gao, B., Zheng, H., Zhao, J., Tu, H., Kim, H., & Duh, H. B.-L. (2024). Evaluating Plausible Preference of Body-Centric Locomotion using Subjective Matching in Virtual Reality. 2024 IEEE Conference on Virtual Reality and 3D User Interfaces (VR), 1-10. doi: 10.1109/VR58804.2024.00124.