Memory Twin Framework

Definition

What Is a Memory Twin?

The Memory Twin Framework is a cultural heritage digitisation methodology that extends the Digital Twin. Where a Digital Twin reproduces the measurable geometry of a monument, site or object, a Memory Twin binds that geometry to structured metadata, to paradata recording how and why the record was made, and to the intangible cultural values, uses and testimonies that give the heritage asset its meaning.

How does a Memory Twin differ from a Digital Twin?

A Digital Twin of a heritage asset is a geometric and often sensor-linked replica: it records surfaces, dimensions, materials and, in engineering contexts, live operational state. A Memory Twin, defined within the Memory Twin Framework for cultural heritage digitisation, keeps that geometry but treats it as only one of three layers. Alongside the measured data it carries metadata describing the record, and paradata describing the decisions, instruments, uncertainties and interpretations behind it, together with the intangible cultural values attached to the place or object.

The practical difference is durability of meaning. A Digital Twin answers what a monument looks like; a Memory Twin also answers why it was documented that way, what was uncertain, and what the asset means to the community that holds it.

What is paradata and why does the Memory Twin require it?

Paradata is the documented reasoning behind a digital heritage record: the survey methods and instruments used, the conditions on site, the choices made during processing, the reconstruction hypotheses adopted, and the uncertainties that remain unresolved. It differs from metadata, which describes the resulting file, and from data, which is the record itself.

The Memory Twin Framework requires paradata because a cultural heritage model without it cannot be verified, reproduced or responsibly reused. Decades after a scan is captured, the survey team disperses and the reasoning disappears while the file survives. Paradata makes that reasoning an auditable part of the asset, protecting scientific defensibility and guarding against later systems, including generative AI, silently filling gaps that were never measured.

How does the Memory Twin relate to HHBIM?

Heritage HBIM (HHBIM) structures a historic building as a semantically rich parametric model, associating each element with construction phase, material, condition and intervention history. It is the natural carrier for a Memory Twin of a built heritage asset, and the Memory Twin Framework uses it as such.

The Memory Twin Framework extends HHBIM in two directions. It attaches paradata to model elements, so the evidence and uncertainty behind each reconstructed component travel with the geometry. It also binds intangible cultural values, oral testimony, ritual use, community association, to the same objects, so that adaptive reuse and conservation decisions made through the model are informed by meaning as well as by fabric. Villa Portelli in Malta is the framework's principal demonstration of this approach.

How does the Memory Twin apply FAIR and CARE principles?

The Memory Twin Framework treats FAIR and CARE as complementary obligations for cultural heritage digitisation. FAIR governs the technical record: persistent identifiers and DOIs make heritage assets findable, open licensing and standard formats make them accessible, CIDOC-CRM alignment makes them interoperable, and certified metadata and paradata make them genuinely reusable rather than merely downloadable.

CARE governs authority over the record. Communities associated with a heritage place retain a voice in how their memory is encoded, who benefits from it, what is published and what remains restricted, and how their intangible practices are represented. Applying both together prevents a familiar failure: a dataset that is technically exemplary yet extracted from the people whose memory it holds.

Where has the Memory Twin framework been applied?

The Memory Twin Framework has been demonstrated across several cultural heritage settings. In Cyprus, the Lambousa and Fikardou cases grounded the original formulation of the data, metadata and paradata architecture. In Malta, Villa Portelli served as the detailed built-heritage case study, combining survey, archival research, oral testimony and heritage HBIM ahead of adaptive reuse.

Heritage Malta's digitisation programme and its interpretive productions provide the operational grounding for cultural tourism applications, while the ENIGMA and AURORA work applies the framework to provenance evidence and the fight against illicit trafficking of cultural goods. European infrastructure projects, including HERITALISE and EUreka3D, connect the framework to shared standards for long-term preservation of 3D heritage records.

Who developed the Memory Twin framework?

The Memory Twin Framework was developed by Anthony Cassar with Marinos Ioannides and Drew Baker, working through the UNESCO Chair on Digital Cultural Heritage at the Cyprus University of Technology, with Heritage Malta providing the operational and curatorial grounding, and with support from the EU-funded HERITALISE project on trustworthy digitisation of cultural heritage.

Memory Twin in cultural heritage: a note on terminology

The Memory Twin described here is a cultural heritage digitisation framework, developed in the context of the UNESCO Chair on Digital Cultural Heritage at the Cyprus University of Technology and the EU-funded HERITALISE project, and applied with Heritage Malta. It concerns the documentation of monuments, sites and objects through 3D records bound to metadata, paradata and intangible cultural values.

It is unrelated to similarly-named commercial products in other sectors, including conversational-AI assistants, personal memory chatbots and digital-memorial or afterlife services. Those offerings share a name but not a purpose, a method or a standards base. Any reference on this site to a Memory Twin means the cultural heritage framework and its peer-reviewed publications.