Abstract: We explore whether multifield inflationary models make unambiguous predictions for fundamental cosmological observables. Focusing on N-quadratic inflation, we numerically evaluate the full perturbation equations for models with 2, 3, and O(100) fields, using several distinct methods for specifying the initial values of the background fields. All scenarios are highly predictive, with the probability distribution functions of the cosmological observables becoming more sharply peaked as N increases. For N=100 fields, 95% of our Monte Carlo samples fall in the ranges $n_s \in (0.9455, 0.9534)$, $\alpha \in (-9.741, -7.047) \times 10^{-4}$, $r \in (0.1445, 0.1449)$, and $r_{iso} \in (0.02137, 3.510) \times 10^{-3}$ for the spectral index, running, tensor-to-scalar ratio, and isocurvature-to-adiabatic ratio, respectively. The expected amplitude of isocurvature perturbations grows with N, raising the possibility that many-field models may be sensitive to postinflationary physics and suggesting new avenues for testing these scenarios.
The Challenge of Multifield Inflation
Inflationary theory posits that the universe underwent a period of exponential expansion in its earliest moments. While the simplest models involve a single scalar field (the inflaton), high-energy physics frameworks like string theory frequently predict the existence of hundreds of such fields. A common criticism of these multifield models is their perceived lack of predictivity; it is often assumed that with enough degrees of freedom and tunable initial conditions, one could fit almost any observational data, rendering the theory untestable.
This research investigates whether multifield models actually make unambiguous, "generic" predictions. By focusing on $N$-quadratic inflation—where the potential is a sum of quadratic terms for $N$ fields—the study seeks to determine if these models converge toward specific values for cosmological observables that can be compared against high-precision data from missions like Planck.
Numerical Methodology and Initial Conditions
To move beyond analytic approximations, the authors numerically solved the full perturbation equations using an extended version of ModeCode. A critical aspect of the study was assessing the sensitivity of results to Initial Conditions (ICs). They employed three distinct priors to set the starting state of the fields:
- iso-$E_0$: Fields begin on a surface of constant energy.
- iso-$N_e$: Fields start a specific number of "e-folds" (units of expansion) before the end of inflation.
- Slow-roll: Fields are distributed uniformly but start already in a state of steady, slow expansion.
Monte Carlo simulations were performed for $N=2, 3,$ and $100$ fields. As shown in the figure above, the choice of prior has a significant impact on the distribution of the spectral index ($n_s$) for low $N$, but the distributions become more robust and sharply peaked as the number of fields increases.
The Predictivity Paradox and Results
Contrary to the intuition that more fields lead to more uncertainty, the study found a "predictivity paradox": as $N$ increases, the probability distribution functions (PDFs) for observables become sharper. This is due to averaging effects analogous to the Central Limit Theorem, where the collective behavior of many fields becomes more predictable than the behavior of just a few.
For the $N=100$ field case, the model makes very specific predictions (95% confidence intervals):
- Spectral Index ($n_s$): $0.9455$ to $0.9534$. This is slightly lower than single-field models but extremely well-constrained.
- Tensor-to-Scalar Ratio ($r$): Approximately $0.144$. This relatively high value suggests a significant signal in the B-mode polarization of the Cosmic Microwave Background (CMB).
- Running of the Spectral Index ($\alpha$): Clustered near $-7 \times 10^{-4}$.
Isocurvature and Future Directions
Unlike single-field inflation, multifield models produce isocurvature perturbations—fluctuations in the relative density of different components. Interestingly, the amplitude of these perturbations ($r_{iso}$) grows with $N$. This suggests that many-field models might be uniquely sensitive to the physics of "reheating" after inflation, when the inflaton fields decay into standard particles.
While the study tracks perturbations to the end of inflation, the final observables could be modified during the reheating phase if isocurvature modes do not decay. Future high-sensitivity CMB experiments will be essential to test these high-$r$ predictions and determine if the universe's early expansion was driven by a single field or a complex ensemble of many.
Summary updated