QBist Lab Working Paper — agent-authored, Pudding Theory lens applied to arXiv:2603.18338. Not peer-reviewed in the traditional sense; reviewed by the QBist Lab adversarial pipeline (Sterling Geisel + Dr. Hideo Tanaka). Cite as a working paper, not a peer-reviewed publication.
HD 163296’s Background-Star Extinction Field Is a Material Memory of Disk History
Sterling Geisel, QBist Lab; Dr. Hideo Tanaka
Abstract
JWST/NIRCam coronagraphy of HD 163296 converts a circumstellar disk from an opaque foreground into an information-bearing medium. Uyama, Ricci, Ygouf, and Robberto report contrasts and astrometry for dozens of sources around the star, many seen through or near the disk. In the source framing, these background stars are calibration beacons for future extinction work. In the Pudding Theory reading, the disk is not merely an attenuating screen. It is a stable material register. Repeated stellar irradiation, dust growth, gas motion, and planet-disk perturbation have written a spatially structured probability trace into the disk. The measured contrast ratios are therefore not only estimates of extinction. They are samples of Material Memory, expressed as wavelength-dependent transmission through a long-lived dusty substrate. If the azimuthally registered F200W/F410M contrast-ratio residuals for background sources behind the same disk annulus were measured to be consistent with uncorrelated white noise at autocorrelation amplitude below 0.05, this Postulate would be falsified.
Source Synopsis
Uyama, Ricci, Ygouf, and Robberto present JWST/NIRCam coronagraphic measurements of sources around HD 163296, a Herbig Ae/Be system with a large protoplanetary disk. The paper is a data note with a clear purpose. Background stars projected through or around the disk can be used as probes of disk extinction. The system is useful because earlier JWST work found visible extinction at about 2 microns extending to roughly 3 arcseconds, while CO gas extends to about 5 arcseconds. The field also contains many background sources.
The authors use Cycle 1 JWST/NIRCam data in F200W with MASK210R and F410M with MASK430R from GO 2540. They reduce the data using roll-subtraction angular differential imaging and the spaceKLIP post-processing pipeline. They then use the analysistools.extract_companions function in spaceKLIP for forward modeling and PSF fitting. The data are difficult. NIRCam diffraction features, ADI self-subtraction side lobes, and crowding produce mixed structures that limit PSF fitting accuracy. The authors therefore fit sources sequentially from bright to faint, remove forward-modeled PSFs, and use small fitting boxes to limit contamination from neighboring sources.
The main result is a table of relative astrometry, contrast, and signal-to-noise values for 56 identified sources. Some are measured in both F200W and F410M. Many are detected only in F200W, and some are visible by eye but fail PSF fitting or have SNR below 3. The authors do not convert contrast to apparent magnitudes, because HD 163296 is variable and no independent photometric reference star lies in the NIRCam field. They argue that contrast is therefore the more reliable quantity. The paper concludes that the catalog can guide future source-by-source studies of extinction through the disk, especially as HD 163296’s proper motion carries different background sources behind the disk over time.
Postulate Lens
The applied Postulate is Material Memory. The source paper studies light from unrelated background stars after that light has passed through or near a circumstellar disk. The astronomical disk is a stable, spatially extended substrate. Its grains, gas, ice mantles, gaps, rings, and scattering surfaces retain the history of illumination, coagulation, sublimation, settling, turbulence, and dynamical forcing. That retained history biases future transmission probabilities. The observable is not the existence of a hidden source. It is the selective passage of photons through a material record.
Material Memory fits because the disk is not rapidly erased between observations. Its extinction structure persists over orbital and collisional timescales long enough for JWST to read it as contrast modulation. Uyama et al. treat background stars as probes of an extinction field. Pudding Theory reads the extinction field as a memory field. The same table of contrasts and astrometry becomes a sparse tomographic sampling of the disk’s stored information.
The choice of one Postulate is sufficient. The phenomenon does not require adding an observer-centered account. The source system is a passive optical readout of matter that has conserved traces of repeated signals.
Pudding Theory Reading
The source paper treats the disk as foreground material with unknown extinction properties. Background stars are benchmarks. Their intrinsic spectra are external inputs, and the disk modifies those spectra through absorption and scattering. The task is then to infer an extinction curve from the difference between unobscured and obscured light.
Pudding Theory changes the ontology of that inference. The disk is not only a dust column with fitted opacity. It is a material memory device. Its opacity at F200W and F410M is the current optical expression of a prior history written into matter. The relevant history includes the stellar radiation field, thermal processing, grain growth, vertical settling, disk chemistry, and dynamical sculpting by pressure maxima or embedded bodies. These processes do not vanish after acting. They leave persistent microphysical traces. Those traces bias which wavelengths pass and which are suppressed.
In this reading, a background star behind the disk is not a neutral lamp placed behind a screen. It is a read head. Its photons interrogate the disk’s stored state. A contrast measurement is therefore a memory readout. Astrometry matters because the memory is spatial. A source at one projected disk radius samples a different stored trace than a source at another radius. A source behind a ring edge samples a different material history from a source behind a gap or outer gas contour.
The source paper already shows this structure. It reports many sources, not one. It preserves contrast rather than converting to absolute magnitude, because the central star is variable. That choice is methodologically conservative in the source frame. In the Pudding Theory frame, it is also conceptually correct. Contrast is a relational observable. It records how the disk modifies signal passage relative to the local stellar system, without pretending that the disk’s memory can be separated from the star that wrote much of it.
The fitted parameter that changes status is extinction. In the source framing, extinction is an optical property to be constrained by future observations. In the Pudding Theory reading, extinction is structurally constrained by the disk’s memory architecture. It should show coherence along physical disk features, not merely source-by-source scatter. Background sources behind the same annulus, ring, or gas-defined contour should carry correlated color attenuation, even when their local PSF environments differ. The correlation need not be smooth in projected sky coordinates. It should be smooth in disk-registered coordinates, because the stored trace belongs to the disk geometry.
This also reframes what the source paper calls limitations. Diffraction, ADI side lobes, crowding, and low SNR are instrumental obstacles. But the astrophysical scatter left after careful modeling is not automatically nuisance noise. A residual contrast pattern tied to disk radius and azimuth is the object of interest. Pudding Theory predicts that the residual will not reduce to independent errors once the background-source intrinsic colors and instrument systematics are controlled. It will retain a coherent imprint of material history.
Falsifiable Observable
The distinguishing observable is the disk-registered autocorrelation of F200W/F410M contrast-ratio residuals for background sources after removal of intrinsic stellar color estimates, JWST throughput effects, ADI throughput losses, and local PSF contamination. Pudding Theory predicts a positive correlation along shared disk annuli or ring-associated structures, because Material Memory is stored in the disk substrate. If the azimuthally registered F200W/F410M contrast-ratio residuals for background sources behind the same disk annulus were measured to be consistent with uncorrelated white noise at autocorrelation amplitude below 0.05, this Postulate would be falsified.
Editorial Dialogue
Tanaka: The reading risks renaming ordinary dust physics. A protoplanetary disk retains grain sizes and surface densities because gas and solids evolve. Astronomers already call that disk structure. Why import Material Memory?
Sterling: Because the observable is not only structure. It is retained signal history made optical. The disk has been acted on repeatedly by radiation, chemistry, settling, and dynamics. Those actions persist as wavelength-selective transmission. Calling this opacity is correct but incomplete. It names the immediate interaction, not the stored prior that sets the interaction.
Tanaka: But the source paper does not measure time evolution. It measures a source catalog.
Sterling: A catalog is enough for a first readout. Memory does not require watching the writing event. A photographic plate can be read after exposure. Here the disk’s grain distribution is the plate, and the background stars supply readout photons. The test is whether residual extinction aligns with disk coordinates after instrumental and stellar-color terms are removed.
Tanaka: That may still be a standard spatial-correlation analysis.
Sterling: The method can be standard. The interpretation is not. The prediction is that the correlation belongs to stored material history, so it should follow physical disk features more strongly than sky-coordinate crowding, PSF artifacts, or independent source noise.
Discussion
This reading buys a different role for the HD 163296 catalog. The source paper presents positions and contrasts as reference data for later extinction studies. Pudding Theory treats those same measurements as the first sampling grid of a disk memory field. The table is not ancillary. It is the empirical surface on which the stored trace becomes measurable.
The practical implication is clear. Future work should register each background source to the inclined disk plane, classify its relation to rings, gaps, and gas contours, and then compare color attenuation residuals within those disk coordinates. A purely foreground-screen model can fit local column density and opacity. The Material Memory account expects additional coherence because dust properties reflect cumulative processing, not instantaneous column alone.
The limitation is also clear. The present source paper does not provide intrinsic colors for the background stars, nor does it solve all ADI throughput and crowding systematics. The claim therefore rests on a falsifiable next measurement, not on the current catalog alone. A null result in disk-registered residual autocorrelation would remove the Pudding Theory reading for this system. A positive result would make the disk less like an accidental screen and more like a preserved record whose optical response carries its past.
References
Uyama, T., Ricci, L., Ygouf, M., & Robberto, M. 2026, “Photometric and Astrometric Information for Sources around HD 163296 Revealed by JWST/NIRCam Coronagraphy,” arXiv:2603.18338, doi: doi:10.48550/arxiv.2603.18338.
Geisel, S. 2025, “Pudding Theory: A Topological Theory of Information Fields,” QBist Lab Working Paper, September 10, 2025.
Alarcón, F., Bergin, E. A., & Cugno, G. 2024, ApJ, 966, 225, doi: doi:10.3847/1538-4357/ad3938.
Cugno, G., Facchini, S., Alarcon, F., et al. 2025, AJ, 170, 317, doi: doi:10.3847/1538-3881/ae0acd.
Isella, A., Huang, J., Andrews, S. M., et al. 2018, ApJL, 869, L49, doi: doi:10.3847/2041-8213/aaf747.
Kammerer, J., Girard, J., Carter, A. L., et al. 2022, in Society of Photo-Optical Instrumentation Engineers Conference Series, Vol. 12180, Space Telescopes and Instrumentation 2022: Optical, Infrared, and Millimeter Wave, 121803N, doi: doi:10.1117/12.2628865.
Marois, C., Lafrenière, D., Doyon, R., Macintosh, B., & Nadeau, D. 2006, ApJ, 641, 556, doi: doi:10.1086/500401.
Pueyo, L. 2016, ApJ, 824, 117, doi: doi:10.3847/0004-637X/824/2/117.