Science1 distinct publisher2 min readPublished
Maurizio Forte and Jacqueline Ortoleva treat two painted chambers at Tarquinia as spaces to be walked rather than pictures to be decoded. The measurable part of that ambition ends with the acoustics; the rest is modelled.
The Scientist · Science desk

Compiled by The ScientistSomething wrong?How this is made
Forte and Ortoleva report that Tomba del Gallo's visual and acoustic properties correspond with alpha-band dominance at 8 to 12 Hz, a band linked to relaxed alertness, sustained visual attention and suppression of irrelevant stimuli [15]. That claim moves from measured sound to brain rhythm, and the pipeline behind it is a computer vision saliency model, an architectural model, a handclap survey and simulated agents tracing plausible routes [4][8][19]. The same move appears in their treatment of frequency, where sounds below 1000 Hz are read as bodily rather than merely audible: capable of inducing vibration and disorientation, with potential effects on cognitive states [7]. That is a statement about human physiology fastened to a measurement of a room.
The burial chamber is 3.31 metres by 2.50 metres [10], which is 8.3 square metres of floor [18]. That figure carries more of the calm reading than the paintings do. The authors note the room is easy to light with candles because it is small [12], and they report short reverberation times, with sound propagating up and out of the tomb rather than accumulating [14]. What would settle the acoustic question is a comparison space of similar volume that wasn't designed for clarity, and no such space appears in the account. As written, the chamber both fails to accumulate sound and reverses that tendency on crossing the threshold from the corridor [14], a pairing that needs the full paper to reconcile.
The design inference rests on a contrast between two rooms [20]. Tomba del Gallo reads as orderly and visually led, with saliency clustering at the paintings and at the high-curvature zones where walls meet [13]; the Tomba dei Demoni Azzurri is offered as the disorienting counterpart [17]. The conclusion that these tombs were built to work on the living and not only to hold the dead [16] depends on the pair diverging, and a contrast between two rooms is not yet a pattern.
What earns confidence is the protocol. Recorded handclaps at fixed positions, with decay and travel measured through each tomb [6], is a procedure another team can repeat in the same chambers and either reproduce or fail to reproduce. That option does not exist for a reading of painted images, which is how the Tarquinia tombs, celebrated for their scenes of music, feasting and procession [2], have mostly been approached [3]. The saliency maps and the alpha-band correspondence sit further out from the evidence: modelled where the acoustics were measured.
Ranked by verification strength, evidence, and original report placement.
Archaeologists Maurizio Forte and Jacqueline Ortoleva published a study in the Journal of Archaeological Method and Theory on what it might have felt like to experience two Etruscan painted tombs at Tarquinia, Italy.
The Etruscan painted tombs in Tarquinia, Italy, date to the first millennium BCE and are known for vivid scenes of music, feasting, processions and the afterlife.
Archaeologists have typically looked at the Tarquinia paintings as images to decode rather than as parts of an experience that people moved through.
The study integrates AI computer vision-based visual analysis, architectural modelling, acoustic data and agent-based perceptual trajectories to describe where attention may have been drawn and how sound reverberated in different areas of the tombs.
The researchers made detailed 3D models of both tombs and conducted standardized acoustic surveys.
The acoustic surveys involved playing recorded handclaps at different locations and measuring how sound decayed and travelled through each tomb.
Distinct publishers with included, body-backed reporting in this cluster.
1 article · September 4, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
science
243 jars, 11 sites: a shape model for Edo pottery that inherits the dates it replaces1 distinct publisher
science
What you expect from your own old age shows up a decade later in who you still see1 distinct publisher
science
NASA's two Sun papers ask paleoclimate to carry an astrophysical term2 distinct publishers
science
Researchers trace a third of Central Valley groundwater loss to 122 named producers1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Measured sound, modelled looking
The handclap surveys are the one part of this reconstruction that produced numbers of its own, alongside chamber dimensions given to the centimetre. The gaze anchors, the ritual affordances and the alpha-band reading all come out of the saliency model and agent-based trajectories, which phys.org relays without any measurement taken from an observer. Grading the acoustics is possible on what has been published; grading the experience is not.
Nobody else shown running it
The reporting shows no second team using this pipeline, no released code or model, and no sign that Tarquinia's site managers or any museum has taken up the sensory reconstruction. One journal paper covered once gives us nothing to measure uptake against.
Brain rhythms inferred from stone
The headline credits AI with revealing what these tombs felt like, and the most specific claim in the piece is a match to alpha-band dominance at 8-12 Hz. No one's brain was recorded. What was recorded is genuine and useful: an 8-square-metre chamber that lets sound escape against a 6.2-metre one that holds it, with reverberation measured in both. The neurological and design-intent readings sit on top of that contrast, drawn from two tombs, and the write-up passes them along at the same confidence as the measurements.
Funding and affiliations undisclosed
phys.org names the two authors and the journal and stops there. Funders, institutions and any stake in the modelling tools go unmentioned, so scoring pressure on this work would mean inventing it.
One account, precise but unchecked
Figures quoted are exact and the two chambers' behaviour is internally consistent, which is what lets us grade the acoustic side at all. Working against that: a single publisher, the paper itself represented by a few quoted passages, and no independent reading of the chain that runs from saliency map to ritual affordance to alpha band.