[z=6 Insight] Are Early Black Holes Really "Overmassive"? A Selection-Aware Reality Check

TL;DR

New research using JWST JADES data reveals that the seemingly "overmassive" black holes at high redshift ($z\gtrsim 4$) might be an observational illusion. By explicitly modeling how hard it is to see a small black hole inside a bright galaxy, Ziparo et al. show that while the average mass relationship is similar to today's, the diversity of growth histories (scatter) was much higher in the early Universe.

The Motivation: Chasing the "Overmassive" Ghost

Since JWST began peering into the first billion years of cosmic history, astronomers have been puzzled: the black holes (BH) found at $z>4$ seem way too big for their host galaxies. If our local scaling laws (like the Kormendy & Ho relation) held true, these galaxies shouldn't have enough stars to match their central engines.

However, there is a catch—Selection Bias. If you can only see the brightest, most active black holes, you will naturally conclude they are all huge. This paper asks a critical "Why": Is the relationship moving (evolution), or are we just looking at the tip of the iceberg?

Methodology: Mapping the Invisible

The authors didn't just fit a line to data points. They performed Forward Modeling to understand their own blind spots.

1. Synthetic Spectra: They built thousands of mock H$\alpha$ emission lines (the "smoking gun" of BH mass).
2. Noise injection: They added realistic JWST/NIRSpec noise.
3. The Detection Map: They calculated the probability of "seeing" a BH of mass $X$ in a galaxy of mass $Y$.

Figure 1: Notice how at high stellar masses (right side of the plots), the 'detectability' drops. The galaxy is so bright and its gas stays so turbulent that it hides the black hole's signal.

The Truncated Likelihood

Using these maps, they built a "Selection-Aware" Bayesian model. Instead of assuming the data they have is the full story, the model specifically accounts for the "missing" low-mass black holes that were too faint to detect.

Results: Establishing the Scale

The results are a strike against "extreme evolution" theories.

• Normalization: The relation follows $\log M_{BH}\approx -4.06+1.17\log M_{\star }$. This is almost identical to the local Kormendy & Ho (2013) relation.
• The Scatter: This is the real story. The intrinsic scatter (${\sigma }_{int}\approx 0.63$ dex) is significantly higher than the $\sim 0.3$ dex we see in the local universe.

Figure 2: The blue line shows the new fit. While it passes through the high-z data points, the 'Selection Threshold' (orange lines) proves we are mostly missing the bottom-right quadrant of the graph.

Why is the Early Universe So Messy?

If the average BH-to-galaxy ratio hasn't changed in 12 billion years, why is the scatter so much higher at $z=6$? The authors suggest three physical drivers:

1. Stochastic Accretion: Early BHs grow in "bursts." A BH might double its mass in a few million years while the galaxy stays the same, pushing it away from the mean relation.
2. Lack of Mergers: In the local universe, galaxies have merged dozens of times. Just as averaging many dice rolls leads to a predictable mean, repeated mergers "wash out" the extremes. At $z=6$, galaxies haven't merged enough to settle into a tight correlation.
3. Seed Diversity: We are seeing the "memory" of the first black hole seeds (whether they started from collapsing stars or direct-collapse gas clouds).

Conclusion: The Path Forward

This work suggests that the "Black Hole–Galaxy" connection was a day-one feature of the Universe. We don't need fancy new physics to explain JWST's overmassive black holes; we just need to understand our telescopes' limits. The real evolution isn't in how big black holes get, but in how disciplined they become over cosmic time.

Technical Keywords: $M_{BH}$–$M_{\star }$, JADES, JWST, Bayesian Inference, Truncated Likelihood, Intrinsic Scatter, AGN Co-evolution.