Faculty of Arts Digital Research Seed Fund

AI Pre-Screening of Research Ethics Applications: Developing, Evaluating, and Governing an LLM Tool for a University Ethics Committee

Lindsey Te Ata o Tū MacDonald and Lui Holder-Pearson (Engineering)

Faculty of Arts Digital Research Seed Fund Project 2026

Human research ethics committees (HRECs) are under pressure. Growing submission volumes mean delays that affect researchers, participants, and the institution. This project asks: can an AI pre-screening tool reduce that pressure, and if so, how should it be built and governed?

The project will develop, pilot, and evaluate an AI pre-screening tool for the UC Human Ethics Committee. The tool will be sandboxed within UC’s IT infrastructure, trained on anonymised past ethics applications, and used to triage incoming submissions before they reach the ethics advisor to the committee. It will not make approval decisions. It will check completeness, flag likely issues, and route applications to the appropriate review pathway.

There is a rapidly-growing literature about the possibilities of AI use in research ethics review. Empirical studies are cautiously encouraging: LLMs identify around nine ethical issues per protocol on average, and specialist fine-tuned systems outperform general models on consistency. The critical gap is the absence of any live pilot in a real ethics committee. Every existing study is conceptual, architectural, or prototype-based. This project fills that gap and will generate the first empirical evidence of what actually happens — to timelines, submission quality, and workload — when an AI pre-screening tool operates in a functioning institutional setting. 

Image: ElkeDebrie, CC BY-SA 4.0, via Wikimedia Commons

Better Judgement Through Confidence Calibration

Doug Campbell, Joshua Wilson Black, Randy Grace (Psychology) and Nicky Morton (Psychology)

Faculty of Arts Digital Research Seed Fund Project 2026

A very valuable life-skill for good decision-making is that of being good at judging the probabilities of different claims about the world being true. However, it is well established that people’s probabilistic judgements tend to be very poorly calibrated. People commonly suffer from ‘over confidence bias’, which is to say, when they judge something is likely to be true, they tend to assign an excessively high chance to it being true. For example, typically when people say they are 80% confident that a claim is true, the claim at issue is true only about 60% of the time.

Fortunately, research also shows that overconfidence bias can be readily reduced through feedback training, in which people get to see how overconfident they are being so that they can then recalibrate their confidence levels downwards accordingly.

This aim of this project is to create a website on which anyone in the world can play a confidence calibration game, in the form of a trivia quiz in which the player must not only guess the answer to a question but also assign a likelihood to their guess being correct. A very simple Bayesian computation will be used to update players’ scores in such a way that the player with the best calibrated confidence levels wins. In addition to allowing people to play the game, the website will allow collection of data on how users played the game, to examine the efficacy of confidence calibration training in improving the success rate of real-world decision-making, and whether such training can affect peoples’ dispositions to accept conspiracy theories and other extreme positions.

Image: Philippe Capouillez, CC BY-SA 4.0, via Wikimedia Commons

Desiderium – What Stood Here: Exploring immersive documentary practice through 3DGS and land art

Ellie Adams and Hamuera Kahi

Faculty of Arts Digital Research Seed Fund Project 2026

Desiderium is an augmented reality land art installation combining a physical artefact, practical lighting, and a real-time 3D AR environment to visualise a representation of one of New Zealand’s least-known environmental atrocities, the loss of 99% of the ngãhere/forest in Banks Peninsula. 

Before European settlement, Te Pātaka o Rākaihautū | Banks Peninsula was covered in dense podocarp forest, with some tree specimens being nearly a thousand years old. Within fifty years of colonisation, 99% of this forest was gone. It was felled by one of the dozens of sawmills that erupted on Banks Peninsula to build homes in Christchurch and beyond, or it was simply burned to make way for farmland pastures.

By 1920, less than one percent of the original forest cover remained. The transformation was so complete that the loss has become nearly invisible, there is almost nothing but a scattering of stumps to compare the current landscape against. The installation will centre on a single stump recovered from the peninsula, incorporated into the installation as a physical AR marker. A ‘regeneration’ of the forest with gaussian splat captures will emulate the native ecosystem that would have once surrounded the tree before it was felled.

The exhibition will be displayed publicly within Õtautahi later this year.

Image: Alexander Gerst, CC BY-SA 2.0, via Wikimedia Commons

Lincoln, Clay and Jefferson: A Meta-Analysis of Proper Nouns (Names) in the Emergence of Lincolnian Statesmanship

Peter Field

Faculty of Arts Digital Research Seed Fund Project 2026

Abraham Lincoln has long been synonymous with statesmanship. He served as American president, managed to hold the Union together during a civil war, and, in the midst of that war, signed the Emancipation Proclamation, the greatest executive order in the nation’s history. Unsurprisingly, no American has been more studied. The consensus agrees that he evolved from a local politician into a statesman through the 1840s and 1850s in large measure because of his rhetorical gifts. There is less agreement on how to characterize his evolution.

Virtually all Lincoln’s public writings are digitized and available. These include his Collected Works (Roy Basler, ed.), his correspondence, his legal writings, as well as his pseudonymous political journalism. This project seeks to explore how Lincoln alluded to prominent American political figures in his public utterances as a means of assessing his shifting public focus, by combining quantitative and qualitative analyses of references to names over time, and cross-refencing the names with other key terms (such as slavery and citizenship).

Image: © Frank Schulenburg, CC BY-SA 4.0, via Wikimedia Commons

Quantifying the translation process of te Paipera Tapu: comparing the Greek and Māori Text of Te Kawenata Hou (The Māori New Testament)

Forrest Panther and Joshua Wilson Black

Faculty of Arts Digital Research Seed Fund Project 2026

This project aims to understand the translation practices used in the 19^th century translation of the Paipera Tapu (the Māori Bible). From the beginning of translation efforts, the translations of the biblical text were printed, with the first full edition of the Paipera Tapu completed and printed in 1868. The translators involved in this project included figures such as Robert Maunsell and William Williams, who were literate in the original Biblical languages of Hebrew (the language of the Hebrew Bible, or the Christian Old Testament) and Greek (the language of the New Testament) respectively. It is understood that the Māori text was translated directly from these languages, rather than the predominant English text of that time, the King James Version.

This project will create and analyse an alignment between the underlying Greek text of the New Testament and the Māori text of the 1868 Paipera Tapu. The aim is to determine what type of translation strategy was used: do Māori words consistently correspond to Greek words in a predictable way, or did the translators allow for more freedom in how they translated this? How does this correspond to philosophies around translation in the 19th century?

Online ukulele forums: Musicking on the small web in the era of social media

Francis Yapp

Faculty of Arts Digital Research Seed Fund Project 2026

The ukulele movement of the past three decades can be seen to embody the values of Community Music (CM), even though it developed independently of the theoretical paradigm in which CM initially evolved. Meanwhile, considerable research has revealed that online music communities can foster access and active participation in music-making. These two worlds come together in a popular ukulele forum for players of all levels, luthiers, and others with an interest in the instrument.

It presents a unique case study both of how an online community has fostered active and inclusive participation in music, and of how conflict, difference and dialogue can and must be negotiated in such a space. It promotes accessibility and inclusion; its virtual gatherings and song ‘challenges’ provide a space for both synchronous and asynchronous sharing of participants’ playing and singing; and the forum continues to be an active, vibrant community in an era where the very forum format has otherwise seen a decline in use. However, the community has negotiated some serious challenges in recent years, caused by a polarised political landscape, competition from algorithm-driven commercial social media, and an internet increasingly flooded with AI-generated content. These challenges postdate more optimistic, earlier scholarly analyses of online music communities.

By analysing posts from the past 20 years and triangulating with data from surveys, interviews, and focus groups, this project will shed light on challenges and opportunities the ukulele movement faces in the post-lockdown era, and how the forum might continue to contribute to the ongoing global ukulele community.

Image: Derek van Vliet, CC BY-SA 2.0, via Wikimedia Commons

When a location is more than a location: Connecting physical locations and social meaning in the Quake Box corpus

Meredith Marra and Robert Fromont

QuakeCoRE CEISMIC Digital Research Seed Fund Project 2026

The QuakeBox corpus offers a unique data set of narratives that both record the events and lived experiences of those who were in Christchurch at the time of the 2011/2012 earthquakes, as well as offering an invaluable resource for learning about our community, resilience and identity.

There is a ‘field’ of meanings associated with place names which are differentially available to audiences. For example, Sumner may simply signal a suburb to outsiders, but locals recognise a much wider array of information: distance from the city; history around the impact of the various earthquakes; the proximity and danger associated with cliffs and beaches; a set of ideologies and socioeconomic information regarding people who lived in the area at the time of the quakes and who live there now.

This project will create a database of the salient social meanings associated with locations referred to in the QuakeBox corpus, and map the trajectories described in the QB narratives using physical location markers, determining which extra information is needed to make sense of the pathways. This will enhance the value of the data for all users regardless of the paradigm within which they work.

The project will also explore how this coding can delimitate communities and how community focus changes over time. The kinds of community to which the recorded Cantabrians refer gives us evidence of those groupings which are used to shore up resilience in the immediate, short and long term. 

Image: BeckerFraserPhotos, CC BY-NC https://quakestudies.canterbury.ac.nz/store/object/188499

Community Notes on Social Media: Democratic Epistemology in Practice

Elizabeth Stewart

Faculty of Arts Digital Research Seed Fund Project 2025

Misinformation online presents a particularly thorny challenge for social media platforms, who must wrestle with the competing values of free expression and accurate information. How this tension is resolved has important implications for democracy, human rights, health, and climate change, among other things. For several years, Facebook relied on third-party fact checkers to provide analyses of content flagged as misinformation. The resulting fact checks would be appended to posts which, if found containing misinformation, would be ranked lower on users’ feeds. However, researchers and the public alike have criticized professional fact checking for checking unverifiable claims and for biased verdicts. In light of perceived bias, Meta has recently announced important changes to their content moderation practices in the US, shifting from third-party fact checkers to a Community Notes model, which X currently employs. This model relies on crowdsourced fact checks from platform users, rather than on professional fact checkers. Users enrolled in the Community Notes program can write a note about any post, but in order for that note to be published a sufficient number of enrolled users must vote to approve it.  

Given the scale of misinformation on social media, there are pressing questions regarding whether and how these content moderation methods address misinformation differently. This project is an initial examination of the epistemic practices of crowdsourced fact checkers.

Image: Jason Howie, CC BY 2.0 https://creativecommons.org/licenses/by/2.0, via Wikimedia Commons

Germany and the Pacific Islands: Media Framing and Discursive Construction 

Mathew Doidge and Serena Kelly

Faculty of Arts Digital Research Seed Fund Project 2025

German foreign policy has historically been tied to a core set of tenets, rooted in the historical experiences of the Second World War and of the post-1989 reunification process. These have included: an emphasis on soft power mechanisms (e.g. ‘Wandel durch Handel’) and an aversion to the application of hard power; a commitment to multilateralism and a rules-based international order; and a reluctance to engage in geopolitical contestation. Recent events, however, have placed a strain on this values framework. The emergence of the geopolitically-charged Indo-Pacific context, and the 2022 Russian invasion of Ukraine (recognised by Olaf Scholz as a Zeitenwende to which Germany would need to respond), have forced Germany to consider its role and place on the global stage. Since 2020, this has included an apparent rediscovery of the Pacific Island space, a long neglected area in Germany’s external relations.  

This project constitutes a first step in a broader project examining Germany–Pacific engagement. Through a study of news media, its intent is to establish a baseline understanding of the way in which Germany is framed and understood in the Pacific Island space (here defined as the membership of the Pacific Islands Forum [PIF], excluding Australia and New Zealand), and correspondingly the way in which the Pacific Islands are framed and understood in German media.

Image: Kahuroa, Public domain, via Wikimedia Commons

Māori/Iwi Infrastructure Mapping

Garrick Cooper and Hamuera Kahi

Faculty of Arts Digital Research Seed Fund Project 2025

Infrastructure development, often discussed and analysed at a nation state level, has received increasing scholarly attention over the last 10 years in what is sometimes referred to as the “infrastructure push” or “ turn”.

That the “infrastructure turn” in the Global South (in decolonial literature indigenous peoples of Global North nations are seen as part of the Global South) is increasingly being discussed, particularly with the greater visibility of BRICS (a grouping of Global South nations) in geo-political discourse, raises the question to what extent can Māori/iwi aspirations for greater autonomy be advanced with more focussed attention on infrastructural needs. By circumstance and perhaps by necessity Māori/iwi aspirations have largely leant on the machinery of the New Zealand government.  Yet, as has long been demonstrated, particularly in service delivery areas like health, education and social welfare, yet alone the legal system, Māori have not fared well. On the back of iwi settlements, but not only, the Māori economy has grown exponentially over the last few decades. If internationally infrastructure investment has been demonstrated to aid and stimulate economic development, it would follow that a more tailored and bespoke approach to infrastructure that meets the need of Māori/iwi would contribute likewise to Māori economic aspirations. 

This project looks to map out existing activities and services to reveal future infrastructure investment possibilities.  

Image: David Palmer, Creative Commons Attribution-Share Alike 4.0 International via Wikimedia Commons

Uncovering hidden voices: digitising New Zealand’s nineteenth-century shipboard diaries to explore language, identity and experience

Sarah van Eyndhoven and Lyndon Fraser

Faculty of Arts Digital Research Seed Fund Project 2025

This project seeks to digitise the shipboard diaries of 19^th century Scottish, Irish and English settlers immigrating to New Zealand. It will apply a variety of digitisation and text mark-up methods to historical collections held by the Canterbury and Otago museums, to transform these materials into text-searchable and geo-socially tagged documents, incorporate these into an existing historical linguistic corpus, and produce a public-facing searchable resource. This will enable complex searches into the geographic origins and destinations of these settlers, quantitative investigations into the frequencies of different linguistic variants and topics discussed in the diaries, and qualitative analyses of historical and sociological themes of interest, such as identity, materiality, community and death. Converting these historical diaries into an accessible and searchable output offers a significant contribution to New Zealand history, linguistic and genealogy research.   

With advances in historical document digitisation in specialised software platforms such as Transkribus, precious documents can be accurately converted to a text-searchable format, with various mark-up options that capture a fine-grained level of detail and enable intricate investigations into their contents. This can feed forward into the application of clustering and modelling techniques to extract linguistic features of interest, as well as text analytics, which offer new possibilities for tracking changes in language, thought and sentiment in large populations. This is thus an excellent opportunity to demonstrate the value of new digital methods. 

Image: Public domain. Copyright 1900 by Charles E. Bolles, Brooklyn, N. Y. Source: Library of Congress

Tracking the changing forms of racism directed at Muslims and Islam in online immigration discourse in Aotearoa

Donald Matheson and Zahra Emamzadeh

Faculty of Arts Digital Research Seed Fund Project 2025

This project is the second stage of research into Kiwiblog and The Standard, two major sites of public commentary in Aotearoa, studying how racialising talk about Muslims and Islam has changed since the terrorist attacks in Ōtautahi Christchurch. Racist talk is difficult to study because of the ways it changes in its justifications and other discursive resources. Studying the discourse here will tell us about Aotearoa New Zealand public discourse but also contribute to research on the fluidity of racialising discourse.

The project will use corpus analysis of immigration-related discourse to extend previous work tracking the justification of racism against Muslims over the 2003-2019 period, particularly the preoccupation of right-wing political discourse with the topic, tactics of deferral, denial and debatability used and the broadening of the discourse to a cultural discourse as anti-Muslim talk became less acceptable in 2019.  

Image: Michal Klajban, CC BY-SA 4.0 https://creativecommons.org/licenses/by-sa/4.0, via Wikimedia Commons

Exploring the vā and lā of Aotearoa’s climate journalism

Tara Ross

Faculty of Arts Digital Research Seed Fund Project 2024

This project explores Aotearoa’s climate journalism to gauge how well it accounts for Pacific understandings of interconnected space and relationships. The scholarship relating to climate journalism and media coverage of the climate crisis in the Pacific comprises several key threads, mostly about roles and responsibilities, media framing and representation, and the importance of incorporating Indigenous voices and perspectives into climate discourse. Little of this work has been done by Pacific scholars (these are mostly dominant western accounts) and there is a need for media research that is grounded in Pacific perspectives and worldviews.

This study looks toward filling that gap by exploring climate journalism in Aotearoa through Pacific frameworks of vā and lā – relationships/relationality and shared ocean space – and an understanding of a collective Moana. It examines the ways in which Aotearoa news media understand and report the climate crisis in Pacific terms and/or serve the interests of Moana peoples more widely. In what ways do they represent Pacific perspectives and advocate for Pacific interests in their reporting? And how does their work connect and square with Pacific climate discourses in other spaces?

Tracking the Development of the Paipera Tapu through the 19th Century

Forrest Panther and Joshua Wilson-Black

Faculty of Arts Digital Research Seed Fund Project 2024

The first printing press arrived in Aotearoa in 1834. There was printed Māori text even before this time but the arrival of a printing press precipitated the large-scale printing of literature in Māori.

Much Māori text of the 19^th century is not currently digitized in a way that is machine-readable. There are images and PDF documents, but attempts at OCR often produce mistakes. This is due to the quality of the scans, the combination of different font sizes and styles that is often difficult for OCR tools to handle, and a printing process that often produces incomplete characters. Further complications include the fact that OCR tools often make use of predictive tools based on English, rather than Māori. Accurate OCR for te reo Māori, which is sensitive to a range of historical usages and local dialectal variation, is vital for the historical disciplines in Aotearoa.

This project systematically investigates and attempts to improve the performance of OCR algorithms for use with 19^th Century documents in te reo Māori.

Criminalisation and Deportation of Pacific Peoples by New Zealand and Australia

Lin Mussell and Josephine Varghese

Faculty of Arts Digital Research Seed Fund Project 2024

Last year, a harrowing incident unfolded in Auckland when authorities forcibly removed a construction worker and “overstayer” from his residence in the presence of his terrified children. This event resonates with historical injustices, such as the Dawn Raids targeting Pacific Peoples between 1974-1976 and the enactment of the Pacific Islander Labourers Act of 1901 in Australia, which mandated the mass deportation of Pacific Peoples in Australia. Despite being invited to address labour shortages in New Zealand (NZ) and Australia, Pacific Peoples endure heightened scrutiny from law enforcement and immigration authorities, leading to imprisonment and deportation.

This project will create a corpus of social media content published 2022-24 by select NZ and Australian actors, which is inclusive of one year preceding and following the ‘official end’ of the COVID-19 emergency. It draws on corpus-assisted discourse analysis techniques to comparatively analyse narratives about labour, detention, and deportation of Pacific Peoples across borders.

Image: New moon over Auckland at dawn. Jin Koo Niersbach. CC BY 2.0

The possibilities of Big Podcast Data

Zita Joyce

Faculty of Arts Digital Research Seed Fund Project 2024

It’s 20 years since Ben Hammersley, writing in The Guardian, coined the term ‘podcasts’ as a portmanteau of ipod and broadcast to describe a new form of audio blogging using RSS feeds. Now Listen Notes, a podcast database company, identifies that there are more than 193,139,807 podcast episodes on the internet.

For these 20 years, academic research on podcasts tends to be small-scale and qualitative, using textual analysis, interviews, and occasional surveys to explore things like narrative, journalistic strategies, advertising, and audiences.  There seems to be a very big gap in the academic literature on large scale analysis of podcast trends, though it should be possible, given that there are so many extant episodes, and all require standardised metadata to be indexed by podcast apps.

This project explores the possibilities of a) accessing / scraping large scale podcast metadata, and b) generating useful insights from it. This work will contribute to larger scale understandings of how podcasts have developed over 20 years and make methodological contributions to the field. 

Image: bettnet, CC BY-SA 4.0, via Wikimedia Commons

Empowering Indigenous Narratives: An Exploration of Video Games for Historical and Cultural Education

Chia-rong Wu

Faculty of Arts Digital Research Seed Fund Project 2024

This project revolves around the representation of Indigenous people and their rich history and culture in video games as a means of fostering historical and cultural education. It embarks on a substantial study that provides a broad perspective on the portrayal of Indigenous customs and practices across different cultural contexts. It thus offers insights into the common themes and unique approaches employed in representing Indigeneity.

The overarching aim is to shed light on the pivotal role that video games play in empowering Indigenous narratives and serving as educational tools for preserving and sharing cultural heritage and enhancing the visibility of Indigeneity in the digital world. These games serve as prime examples of how digital media can be utilised to showcase Indigenous stories, traditions, and values in an engaging and interactive format. By critically analysing the representation of Indigenous subjects in video games, this project aims to contribute to the ongoing discourse on cultural diversity and inclusivity in video gaming.

Mapping Public Lectures and Debates in 19th Century Christchurch Through Digitised Historical Newspapers

Joshua Wilson Black

Faculty of Arts Digital Research Seed Fund Project 2023

This project maps public lectures and debates in 19^th century Christchurch, by applying a variety of text analytic methods to historical newspaper advertisements to extract the ‘who’, ‘when’, ‘where’, and ‘what’ of these events and produce a public facing interactive visualisation. Unlocking this information from historical newspapers will be a significant contribution to both New Zealand history and to the emerging field of experimental history of philosophy.

This project contributes to the ADL’s ongoing work on Papers Past newspaper content. The investigation of workflows for OCR improvement is particularly important for future work as OCR errors are a significant limit on the success of text analytic methods.  

2013-2023 Electoral Immigration Detention Discourse in Aotearoa, Australia, and the UK

Lin Mussell and Sian Troath

Faculty of Arts Digital Research Seed Fund Project 2023

Immigration detention is a contested topic mobilized during elections to shore up political support and votes. These discourses reveal assumptions, values, and norms on the part of politicians, bureaucrats, pundits, and community stakeholders about social risk and deservingness. These discourses have real world impacts through policy decisions and on the lived reality of migrants.

This research project is creating a corpus of immigration detention discourse, drawing on corpus-assisted discourse analysis techniques to comparatively analyse narratives about immigration detention, specifically risk and deservingness; track how such narratives are transferred across borders (policy learning); and pin-point how narratives are reflected in subsequent policy.

Image: Takver, CC BY-SA 2.0, via Wikimedia Commons

Taniwha: A Cultural History – Digital Corpus Analysis

Kirsty Dunn and Madi Williams

Faculty of Arts Digital Research Seed Fund Project 2023

This project asks “who” rather than “what” are taniwha?  To whom are they related and how? Are there connections between taniwha narratives? What kinds of knowledge are embedded in these narratives? And how might taniwha narratives help us to understand and navigate current conflicts and challenges that are present both here in Aotearoa and beyond? 

We are building and analysing a digital corpus to assess and address ways in which taniwha have been represented in pūrākau, literature, visual art, news media, and a range of digital and web-based source material.  

Exploring digital approaches to film studies at UC

Erin Harrington

Faculty of Arts Digital Research Seed Fund Project 2023

Over the last fifteen years, researchers in film studies and associated visual analysis fields have been exploring the use of digital and computational tools in novel ways. One strand of inquiry, based more in film and arts practice, is how the creation of videographic essays might be a form of research and meaning-making itself, as well as a mode of dissemination and a pedagogical approach. Another is how digital tools might open up new ways of thinking about film as object and artefact. Approaches are almost staggering in their diversity.

This scoping project thus aims to understand what digital approaches to film studies might look like at UC – practically, ethically, creatively. The aims of this research project are broad, and are intended as first steps towards a better understanding of this field in our context.

Image: Archives New Zealand, CC BY 2.0, via Wikimedia Commons

Imperial reinforcements: MI7b propaganda in New Zealand and the Dominions during the First World War

David Monger

Faculty of Arts Digital Research Seed Fund Project 2023

During 1917 and 1918, a newly established branch of British military intelligence, MI7b, recruited a collection of more than 20 talented writers from the ranks of the army to write propaganda for circulation to newspapers in Britain and its imperial Dominions (Australia, Canada, Aotearoa New Zealand and South Africa). Among the recruits were such prominent authors as the fantasy writers Lord Dunsany and H.R. Wakefield, the humourist J.B. Morton, the Irish author Patrick MacGill and the journalist and author A.J. Dawson, as well as the internationally-renowned post-war children’s author, A.A. Milne.

This project aims to expand the exploration of digital newspaper repositories, by seeking further pieces by Milne and other MI7b propagandists. The research will shine new light on Britain’s self-promotion to itself and its Dominions during the war, and increase knowledge of British propaganda activities in Aotearoa.