Resonances extracted in truncated partial-wave analysis are effective mixtures of angular momenta

This paper demonstrates that in truncated partial-wave analysis, the extracted coefficients are not direct projections of the full amplitude but rather effective mixtures of angular momenta arising from the nonlinear coupling of bilinear interference terms within the restricted fit space.

The Gist

The Big Idea: The "Blurry Lens" Problem

Imagine you are trying to identify a specific person in a crowded room by looking at their shadow on a wall.

In the world of particle physics, scientists try to find "resonances" (which are like short-lived particles or baryon resonances) by analyzing how particles scatter off each other. They do this using a mathematical tool called Partial-Wave Analysis (PWA). Think of this tool as a set of lenses that can focus on different "shapes" or "spins" of the scattering pattern.

Usually, scientists can't look at every possible shape because the math gets too hard. So, they use a Truncated Partial-Wave Analysis (TPWA). This means they only look at the first few shapes (the "low-order" waves) and ignore the rest, hoping the rest don't matter too much.

The paper's main warning is this:
When you ignore the higher shapes, you aren't just "missing some details." You are actually changing the rules of the game. The numbers you calculate for the "simple" shapes aren't just the true simple shapes; they are a fake mixture of the simple shapes plus a distorted version of the complex shapes you threw away.

---

The Analogy: The Smoothie and the Recipe

To understand why this happens, let's use a Smoothie Analogy.

1. The Real World (The Full Amplitude)

Imagine a perfect smoothie made of three ingredients:

• Strawberries (Low energy, simple shape)
• Bananas (Medium energy)
• Blueberries (High energy, complex shape)

In physics, the "ingredients" are the mathematical coefficients ($a_0, a_1, a_2$). The "taste" of the smoothie is what we actually measure in the lab (the observables).

2. The Measurement (The Bilinear Fit)

Here is the tricky part: We don't taste the ingredients directly. We only taste the smoothie itself.
In physics, the "taste" (the observable) is a bilinear mix. This means the taste depends on how the ingredients interact with each other.

• It's not just "Strawberry + Banana."
• It's "Strawberry $\times$ Strawberry" + "Strawberry $\times$ Banana" + "Banana $\times$ Blueberry," etc.

The interaction between the Strawberry and the Blueberry creates a unique flavor note that you can't get from just the Strawberry alone.

3. The Truncation (The Mistake)

Now, imagine you are a chef trying to recreate this smoothie, but you are only allowed to use Strawberries and Bananas. You are forbidden from using Blueberries.

You try to make a smoothie that tastes exactly like the original one using only Strawberries and Bananas.

• The Naive View: You might think, "I'll just put in the right amount of Strawberry and Banana to match the Strawberry and Banana flavors of the original."
• The Reality (The Paper's Point): Because the original taste included the interaction between Strawberries and Blueberries, you can't just ignore the Blueberries. To fake that specific flavor note using only Strawberries and Bananas, you have to change the recipe.

You might have to add more Strawberry or mix the Banana in a weird way to mimic the "Blueberry-Strawberry" flavor.

Result: The "Strawberry" amount in your new, simplified smoothie is not the same as the "Strawberry" amount in the original. It is a mixture. It contains the true Strawberry flavor plus a "ghost" of the Blueberry flavor that you tried to fake.

What This Means for Physics

The author, Alfred Švarc, is saying that when physicists extract "resonances" (the particles) from these simplified, truncated models:

1. They aren't seeing the "True" Particle: The number they get for a specific particle isn't a direct measurement of that particle.
2. It's a "Truncation-Dependent Mixture": The number is a blend of the real particle and the influence of the higher-energy particles they ignored.
3. Changing the Rules Changes the Result: If you decide to include one more "shape" (increase the truncation limit), you aren't just getting a "sharper" picture of the same thing. You are fundamentally changing the mathematical recipe. The "Strawberry" number you get at the end might be totally different from the "Strawberry" number you got when you only looked at two shapes.

The "Toy Model" Proof

The author proves this with a simple math example (the "Toy Model").

• He takes a complex wave (up to order 2) and tries to fit it with a simple wave (order 1).
• He shows that the numbers for the simple wave mathematically depend on the numbers of the complex wave.
• Even if the complex wave has a "resonance" (a special spike), that spike gets mixed into the simple wave's numbers. The simple wave's numbers are no longer pure; they are "contaminated" by the complex wave's influence.

The Takeaway for Everyone

Think of it like listening to a song on a low-quality speaker.

• Old thinking: "If I turn up the volume, I'll hear the bass better."
• New thinking (this paper): "If I turn up the volume on a broken speaker, the bass doesn't just get louder; it starts to sound like a mix of the bass and the treble because the speaker is distorting the frequencies."

The Conclusion:
When scientists find a new particle using these truncated methods, they must be very careful. They cannot say, "We found a particle with mass X." They should say, "We found an effective mixture that looks like a particle with mass X, but this result depends heavily on how much of the data we decided to ignore."

It's a call for humility: The "resonances" we see might be real, but they are also partly an artifact of the mathematical "lens" we used to look at them.

Technical Summary

1. Problem Statement

The paper addresses a fundamental conceptual issue in Truncated Partial-Wave Analysis (TPWA), a standard method used to extract baryon resonance information from meson photoproduction data.

• The Standard Practice: In TPWA, experimental observables (angular distributions, polarization) are fitted using a finite set of multipoles (partial waves) up to a maximum orbital angular momentum $\ell_{max}$.
• The Misconception: It is often implicitly assumed that the coefficients extracted in a truncated analysis are simple projections of the coefficients of the full, infinite-amplitude problem. Consequently, it is assumed that changing the truncation order ($\ell_{max}$) merely refines the precision of the same underlying physical solution.
• The Core Issue: The author argues that this assumption is incorrect because observables are bilinear functionals of the scattering amplitudes (e.g., $|f|^2$), not linear functionals. Therefore, truncating the amplitude basis does not merely discard higher-order terms; it fundamentally alters the set of bilinear interference terms allowed in the fit. This leads to a situation where extracted low-order coefficients are not independent projections but are coupled, nonlinear mixtures of the full amplitude's coefficients.

2. Methodology

The paper employs a rigorous algebraic approach to demonstrate the mechanism of this mixing, utilizing a minimal scalar toy model.

• Theoretical Framework:

  • The scattering amplitude $f(W, x)$ is expanded in Legendre polynomials $P_l(x)$ with coefficients $a_l(W)$.
  • The observable (Hermitian bilinear) is $B_f = f \cdot f^*$.
  • A truncated approximation $g(W, x)$ uses coefficients $b_m(W)$ up to order $M < N$ (where $N$ is the order of the "true" amplitude).
  • The observable for the truncated model is $B_g = g \cdot g^*$.

• Mathematical Formulation:

  • The author expands both the true bilinear $B_f$ and the truncated bilinear $B_g$ into Legendre series.
  • The coefficients of the bilinear expansion ($F_l$ for the true case, $G_l$ for the truncated case) are shown to be quadratic combinations of the amplitude coefficients (e.g., $F_l \sim \sum a_i a_j^*$).
  • The fitting process is defined as minimizing the least-squares error between $B_f$ and $B_g$.
  • Crucially, the optimization problem for the truncated coefficients $b_m$ is nonlinear and coupled. The value of a low-order coefficient $b_m$ depends on the entire set of retained bilinear moments ($F_0, \dots, F_{2M}$), which themselves contain contributions from higher-order amplitude coefficients ($a_{l>M}$).

• Toy Model Demonstration:

  • The author constructs a specific case where a true amplitude of order 2 ($f = a_0 P_0 + a_1 P_1 + a_2 P_2$) is approximated by a truncated amplitude of order 1 ($u = \alpha_0 P_0 + \alpha_1 P_1$).
  • The paper derives explicit analytical relations showing that the fitted coefficients $\alpha_0$ and $\alpha_1$ depend on bilinear combinations of all three original coefficients ($a_0, a_1, a_2$).
  • Specifically, the fitted low-order terms inherit contributions from the higher-order $a_2$ term through the bilinear structure of the fit.

3. Key Contributions

• Identification of the Mechanism: The paper explicitly identifies that truncation in a bilinear fitting problem induces angular-momentum mixing. The fitted coefficients are not projections but "effective mixtures" generated by the restricted bilinear space.
• Non-Linearity of Truncation: It establishes that lowering the truncation order changes the algebraic structure of the fit itself. The map from the exact amplitude to the truncated coefficients is a coupled nonlinear optimization, not a linear projection.
• Reinterpretation of TPWA Results: The paper argues that resonances extracted at different truncation orders ($\ell_{max}$) do not necessarily represent the same physical object with varying precision. Instead, they may represent different effective mixtures of the true resonant and non-resonant contributions.
• Generalization: While derived using scalar scattering, the author demonstrates that the algebraic mechanism applies generally to Legendre-moment analyses and photoproduction observables, as these are also bilinear in the underlying amplitudes.

4. Results

• Analytical Proof: In the toy model, the fitted coefficient $\alpha_1$ (representing the P-wave in the truncated model) is shown to depend on terms like $Re(a_0 a_2)$ and $|a_2|^2$. This proves that the "P-wave" extracted in a truncated fit contains information from the "D-wave" ($a_2$) of the full theory.
• Dependence on Truncation Order: The extracted quantities are shown to be truncation-dependent. Changing $\ell_{max}$ does not just add corrections; it redefines the effective parameters being inferred from the data.
• Pole Mixing: Since the original coefficients $a_l(W)$ may contain poles (resonances), the mixing occurs at the level of pole-bearing quantities. The "effective" resonance extracted in a truncated fit is a superposition of poles from different angular momenta in the full theory.

5. Significance and Implications

• Caution in Interpretation: The primary significance is a warning against the naive interpretation of TPWA results. Resonances extracted from a truncated fit should not be identified directly with the "true" poles of the infinite problem. They are effective structures specific to the chosen truncation.
• Comparison of Results: Comparing resonances extracted at different $\ell_{max}$ values as if they were approximations converging to a single value is fundamentally flawed. They may correspond to different physical mixtures.
• Methodological Shift: The paper suggests that the community must distinguish between the true resonance content of the full theory and the truncation-dependent effective resonances extracted by the fit.
• Future Work: The author notes that while the mixing mechanism is established at the level of fitted bilinear quantities, the detailed analytic continuation of these reconstructed truncated amplitudes to find resonance poles is a separate, complex step not covered in this short note, but the mixing mechanism established here is a prerequisite for understanding that step.

In summary, the paper fundamentally challenges the standard interpretation of TPWA by proving that the truncation of a bilinear fit creates a new, coupled mathematical object where low-order coefficients are inextricably linked to higher-order physics, rendering extracted resonances as "effective mixtures" rather than direct projections of the true spectrum.