Attribution

Credits

Last updated: September 6, 2026

Cultural Capital images

The Cultural Capital task uses real museum pieces, not stock photography or illustration. Every image below is public domain or CC0, checked against the holding institution’s own per-object rights record rather than a policy page — a collection can be open access in general and still hold objects that are not. Each plate carries its full object credit and a link to the institution’s own listing, so you can verify the piece yourself.

The bank holds 50 objects across 3 registers, tagged per Bourdieu’s Distinction. The tag is used in sampling and in scoring, so it is a measurement decision and not a caption.

Register	Objects	What it can test	Scoring
Highbrow	28	Which consecrated traditions you were taught to recognise	Warmth counted as shown
Folk	11	Whether you find something in an object nobody framed for you	Warmth counted as shown
Contested	11	Bourdieu’s diagnostic subjects — beauty in bark, in a sunset	Signed; three plates reversed
Pop	0	Mass circulation now — which is what copyright is the shadow of	Kept empty on purpose

Why folk was emptied and refilled

Six objects shipped tagged Folk: a Benin royal palace plaque, an Ottoman sultan’s imperial monogram, an elite Wari tunic, a kente prestige cloth, a signed master calligraphy, a monastic icon. Every one is its own society’s consecrated court or ecclesiastical art, and all six were retagged Highbrow in August 2026.

Calling them Folk had turned the register into a proxy for Europe-versus-everywhere-else, in the language of a taste hierarchy. It also meant the bank had never held folk art of any origin, European included. The eleven vernacular objects that replaced them the next day — things made to be used, worn out and eaten off — are all Cleveland Museum of Art, all CC0, four of them European or Euro-American, six three-dimensional, four made after 1889.

Why contested is signed

Bourdieu found some of his diagnostic subjects rise with education and some fall — the bark of a tree from 15.5 to 60.5 per cent answering “beautiful”, a sunset over the sea from 88.5 down to 65. Averaged unsigned, the two cancel. So eleven objects carry the sign and three are scored in reverse. Four are paintings rather than photographs, because the American documentary archives supplying the rest cannot hold a first communion or a bolero.

One object removed

A north Queensland Aboriginal rainforest shield was CC0 and clean, but shields of that class are initiation-restricted and their designs clan-owned; the community’s own published account records a living norm under which reuse is authorised by the oldest clan member. A CC0 dedication releases a museum’s copyright and cannot grant that permission, and this project has no way to seek it. It is gone.

Where the placements come from

The plates below are only one of the banks this battery draws on. Media Diet places outlets into registers, and the music and film passes use published genre structures — and a reader is scored against those placements, so whose judgement they are is not a detail. Every source the outlet and genre banks rest on is listed here, rendered from the same records the instrument reads rather than transcribed, so this list cannot fall behind the bank.

Pew Research Center (September 2025). Many Americans who generally distrust national news organizations still express trust in certain outlets.

American Trends Panel, 10-16 March 2025, n = 9,482 U.S. adults, 30 outlets

https://www.pewresearch.org/short-reads/2025/09/02/many-americans-who-generally-distrust-national-news-organizations-still-express-trust-in-certain-outlets/

Pew Research Center (January 2020). U.S. Media Polarization and the 2020 Election: A Nation Divided.

n = 12,043 U.S. adults, trust and distrust across 30 news sources

An audience is called left- or right-leaning when it holds at least two-thirds more liberal Democrats than conservative Republicans, or the reverse.

https://www.pewresearch.org/journalism/2020/01/24/u-s-media-polarization-and-the-2020-election-a-nation-divided/

Pew Research Center (June 2025). The political gap in Americans’ news sources.

American Trends Panel, 10-16 March 2025, n = 9,482 U.S. adults, 30 outlets

Not the two-thirds rule of the 2020 report. Pew places the average member of each outlet’s regular audience against the average U.S. adult, and names The Atlantic, HuffPost, NPR, The Guardian and Axios among the audiences that are “particularly liberal and Democratic”.

https://www.pewresearch.org/journalism/2025/06/10/the-political-gap-in-americans-news-sources/

Redhead, M. (2019). Music preferences of those on the left and right of the UK political spectrum: What can be deduced about individuals’ music libraries from their political attitudes? DURMS, Volume 2, Autumn 2019, Durham University.

103 UK citizens, grouped left or right by the party they would vote for; liking rated for genres and analysed as five dimensions

UK, one study, 103 people, and the genre groups are its own five-dimension solution rather than Rentfrow and Gosling’s four.

One significant difference: the left-wing group liked R&B (jazz, soul/funk, blues) more than the right-wing group did. Upbeat, Traditional, Reflective and Rebellious did not differ, and on three of those four the right-wing group’s mean was the higher.

https://musicscience.net/wp-content/uploads/2019/11/redhead.pdf

Xu, X., Mar, R. A., & Peterson, J. B. (2013). Does cultural exposure partially explain the association between personality and political orientation? Personality and Social Psychology Bulletin, 39(11), 1497-1517.

Print exposure (Study 1), film exposure (Study 2) and knowledge of American history (Study 3) as mediators between Openness/Conscientiousness and political orientation. Exposure breadth, never genre preference.

https://doi.org/10.1177/0146167213499235

AllSides Media Bias Chart and per-outlet Editorial Reviews / Blind Bias Surveys, ratings current as of September 2026.

Content bias, rated by panels drawn from left, center and right against 16 named bias types. Not audience composition.

https://www.allsides.com/media-bias

Ad Fontes Media, per-outlet bias and reliability scores, current as of September 2026.

Content bias (-42 to +42) and reliability (0-64), scored by analyst panels drawn from left, center and right. Not audience composition.

https://adfontesmedia.com/

Kiss, B. L., Deak, A., Veszprémi, M. D., Blénessy, A., & Zsido, A. N. (2024). The role of excitement and enjoyment through subjective evaluation of horror film scenes. Scientific Reports, 14, 2987.

n = 558; morbid curiosity, sensation seeking, disgust sensitivity against horror enjoyment

Disgust sensitivity was NOT associated with horror enjoyment. Morbid curiosity was. The obvious prediction is the wrong one.

https://www.nature.com/articles/s41598-024-53533-y

The named works

Beyond the genres, the music, film, visual and literature passes ask about named works and artists — 2051 musicians, 1316 films and series, 488 visual artists and 1366 authors and poets. None carries a published placement, by construction: audience data for individual artists and titles is commercial and unpublishable, so nothing here is read against a baseline and nothing here is scored.

They exist to catch one thing a genre list cannot — the distance between the genre somebody claims and the work they can actually name. The classifications behind them are this project’s own and unsourced, which is acceptable because a misfiled artist cannot move a number. Where a work is Iranian it carries its name in Persian as well; a recognition task that shows only a transliteration is a transliteration quiz.

Music has a rung between the genres and the names: 140 subgenres on the General track and 54 on the Iran one. Each subgenre entry names the reference works it leans on — genre histories, AllMusic’s style pages, the Persian music sources — and carries a standing of canon, popular or dismissed, so that a listener of what critics deride is measured as a listener rather than as an absence. The names under them are placed by this project.

Film & TV has the same rung: 105 subgenres on the General track and 44 on the Iran one, with the titles under them. 21 of the General subgenres are television forms rather than film genres, because the bank has always held series. On the Iran track the rungs are the categories Persian film discourse actually sorts by — festival cinema, the home network, the Ramadan slot, Sacred Defence — rather than translations of the General ones, and every title carries its Persian name.

Fine Arts has it too: 43 subgenres on the General track and 36 on the Iran one — schools, movements, forms and scenes under each register, the mall-gallery painter and the tattoo shop beside the Old Masters — with the artists under them. Each entry names the reference works it leans on, Gombrich and Grove Art and a museum’s own department names, and on the Iran track Pakbaz’s history of Iranian painting and Seif on the coffeehouse painters. Every Iranian subgenre and artist carries its Persian name.

Visual artists	Count	Asked as
Public domain	200	Plates — the work is shown
In copyright	288	Names — reproduces nothing, needs no clearance

Three of the six Iranian registers hold fewer than five public-domain artists — مکتب سقاخانه (4 of 20), نقاشی‌خط (3 of 20), مدرن و آوانگارد (1 of 55) — so a plates-only Iranian visual task could show the classical half and almost none of the modern.

Literature has it too: 107 shelves on the General track and 112 on the Iran one, poetry divided as finely as fiction and the airport thriller and the Instagram poet at full strength beside the classics, with 748 authors under the General shelves and 618 under the Iranian ones. Each General shelf names the heading it leans on — a Thema or BISAC code, a prize remit, the Penguin Classics list, a bookseller’s own table — and 61 of the 107 say plainly that the cut is this project’s own, because no trade heading draws the line there. Every Iranian shelf and author carries its Persian name.

Three elements were added on 6 September 2026 without a ladder of their own. Sport is 28 lines on the General track and 24 on the Iran one, Games 19 and 12; both are asked at Surface only, what is followed or played and never who, since athletes, teams and streamers are where a taste question turns into a loyalty question. Podcasts & creators has two rungs: 15 categories on the General track and 12 on the Iran one, with 126 and 70 shows and channels under them — podcast, YouTube or both — and no third rung, because a show is already a name. It carries no political hosts, a rule read on the premise of the show rather than the biography of the host, and the categories, placements and standings are this project’s own. Every Iranian line, category and show carries its Persian name.

Object credit, plate by plate

“Isfandiyar’s Third Course: He Slays a Dragon”, Folio 434v from the Shahnama (Book of Kings) of Shah Tahmasp, ca. 1530, Iran

HighbrowPersian

“Isfandiyar’s Third Course: He Slays a Dragon”, Folio 434v from the Shahnama (Book of Kings) of Shah Tahmasp. ca. 1530. Opaque watercolor, ink, silver, and gold on paper. Gift of Arthur A. Houghton Jr., 1970