Appendices

Glossary

Augmentative and Alternative Communication (AAC)

Systems, tools and devices that support people with communication challenges. This includes digital tools such as text-to-speech computer systems.

Artificial Intelligence (AI)

Computational systems capable of tasks that conventionally required human cognitive abilities. This includes tasks such as object recognition, text generation and problem-solving. AI systems are used widely in mainstream technologies such as search engines, navigation systems and virtual assistants.

Augmented reality devices or headsets

Devices (often worn on the head such as smart glasses) that ‘augment’ a user’s perception of the physical world (eg a user being able to magnify their surroundings).

Circular economy

A model aiming to extend the lifecycle of products through sharing, leasing, reusing, repairing, refurbishing and recycling existing products as long as possible. This can support more sustainable and affordable digital products.

Closed captioning

A process which displays audio information (such as speech or sounds) as text on a visual display (such as on a smartphone), where a user can choose whether the text is displayed.

Data linking

The process of joining together information from different datasets to gain a more comprehensive understanding. This can include combining information about the same individual whose data may be held in separate datasets.

Data minimisation

A principle that limits the collection and storage of data to what is strictly necessary for the purpose the data is being used for. For example, data minimisation could involve not collecting demographic information about a person if it’s not needed for a process (eg making an online purchase).

Digital assistive technology (DigAT)

A digital technology that processes information to help make people’s lives easier, such as screen-readers, speech-to-text software and smartphone apps. This definition was co-formulated with disabled people as part of focus groups conducted by the Research Institute for Disabled Consumers (RiDC) for this project.

Digital exclusion

The lack of access to digital technologies and tools. This includes barriers such as unreliable internet connections, unaffordable digital devices or lacking the skills to use digital technologies.

Digital literacy

The ability to use digital technologies to find, evaluate, share and create information and content.

Disability

A significant and long-term impairment which negatively impacts a person’s ability to perform personal daily activities and their participation in society. Two common ways of understanding disability are medical models, where disability is understood primarily as a health condition to be avoided and social models, where disability is due to societal barriers in the environment. Some models of disability (eg biopsychosocial) combine aspects of the medical and social models. Refer to the ‘Background’ section in the Introduction for more detail.

Disability data

Information regarding an individual’s disability, a disabled person’s other personal data and national or international information on disability prevalence within a population. Refer to chapter 2 for more detail.

Few-shot learning

A type of small data technique used in machine learning where models are trained to perform tasks using only a small number of examples (eg recognising images of cars based on a few images).

Generative AI

AI systems generating new text, images, audio, or video in response to user input using machine learning techniques.

Haptic suits

A type of wearable technology which provides tactile sensory feedback to the user (eg vibrations in response to live music or while playing a video game).

Interoperability

The ability of different data, devices or systems to communicate and work together with minimal effort.

Machine learning

A type of artificial intelligence (AI) involving algorithms that learn patterns from data and apply these findings to make predictions or generate content.

Meta-learning

A type of machine learning technique where models are trained to adapt to new tasks using prior knowledge from multiple datasets. This can be especially useful in small data contexts where several small datasets can be used to improve model performance.

Neuro-symbolic AI

A hybrid approach to developing AI systems that combines the pattern recognition capabilities of machine learning models with the structured reasoning approach of symbolic AI. This aims to combine the strengths of these two AI approaches and can be useful in small data contexts.

Obsolescence

In the context of technology, this refers to the process of technologies becoming no longer useful or obsolete due to the availability of newer technologies or lack of support for an older technology. Compare to ‘technology transience’ glossary entry.

Screen-readers

A type of digital assistive technology (DigAT) which supports blind or low vision users to read text by converting it into audio or Braille format.

Small data analysis/techniques

The use of tools and techniques for data analysis in settings where there are small datasets (ie limited amounts of data and information available). Examples of techniques useful in small data contexts include few-shot learning, meta-learning and neuro-symbolic AI. Refer to chapter 3 for more detail.

Smart home devices

Interconnected household devices that are controlled automatically or remotely by a user through a smartphone or computer (eg app controlled smart lighting and switches).

Technology transience

The temporary nature of digital technologies due to products quickly becoming outdated or obsolete. Also refer to ‘obsolescence’ glossary entry.

Text-to-speech

Software systems converting text information into speech, which can be useful for disabled people who need text read aloud (eg Blind or low vision people).

Virtual reality devices or headsets

Devices (often worn on the head) where users perceive and interact with a computer-generated 3D virtual environment (eg for people to explore a route and practice ahead of travelling or for site tours).

Voice assistant/voice-controlled assistant

Software systems (typically using AI) which respond to user’s voice commands (eg Amazon Alexa, Siri).

Voice-to-text

Software systems converting audio speech information into text, which can also be used by disabled people to control devices through voice (eg people with mobility issues).

Wearable technologies

Devices that are designed to be worn on a user’s body such as smart watches or sensors on prosthetics. These can be used to collect or monitor data from a user.

Details on methodology

Summary of research activities

This report draws on several research and evidence-gathering activities as described below.

• Five commissioned research and evidence-gathering projects including a policy analysis report on factors affecting digital assistive technology (DigAT) adoption, a YouGov survey exploring general public attitudes towards technology accessibility, a mixed-methods study on disabled people’s opinions and experiences of assistive technologies, a literature review on small data research methods and a literature review on disability data metrics and gaps.
• 35+ semi-structured interviews with experts in digital assistive technologies.
• Four roundtables and workshops on the topics of inclusive design of DigAT, technical and ethical challenges with DigAT for social care, DigAT in gaming and technology transience and obsolescence of DigAT.

Commissioned evidence-gathering and reviews

Hackenberg M, Nolde S, Kabus F, Backofen R, Köttgen A, Rohde A, Binder N, Brawner J, Markham E, Hardalupas M, Chowdhury A and Binder H, 2024.
Small data explainer – The impact of small data methods in everyday life.

Danemayer J, Holloway C, 2024. Disability and Assistive Technology in Population-Based Data.

YouGov survey
Where figures are from YouGov Plc., the total sample size was 2076 adults. Fieldwork was undertaken between 14th – 15th March 2024. The survey was carried out online. The figures have been weighted and are representative of all UK adults (aged 18+).

Cashman C, Chessell D, 2024.
Disability, data and digital assistive technologies (DigAT).

Baskerville J, Pan Y, Pham T and Sutton D, 2024.
Towards the Adoption of Digital Assistive Technologies in the UK: An International Comparison of Policy Factors.

Event and research activities

The Royal Society would like to thank all those who contributed to this project, through participation in the following events.

35+ interviews, October 2022 – July 2024

Royal Society staff interviewed scientists, researchers, industry professionals and civil society representatives on digital assistive technologies.

Roundtable on inclusive design of Digital Assistive Technologies (DigAT), June 2023

The Royal Society hosted a roundtable in Edinburgh as part of its Creating Connections event series, which hosts regional meetings addressing the scientific opportunities and challenges faced by the UK. The roundtable convened industry leaders, academics and civil society representatives in Edinburgh to discuss the potential and challenges of inclusive design of DigAT. The roundtable was chaired by Professor Jacques Fleuriot, Chair of Artificial Intelligence in the School of Informatics at the University of Edinburgh and Head of the AI Modelling Lab. The key topics discussed were barriers to industry development, co-design methodologies and ethical challenges.

Workshop on DigAT for social care, April 2024

The Royal Society and Policy Connect jointly organised a workshop on the inclusive design and deployment of smart home devices for social care and independent living. The workshop convened an interdisciplinary group of UK experts in data ethics, disability, social care and assistive technology development from both public and private sectors. Presentations from Clive Gilbert, Senior Policy and Research Manager at Policy Connect and Professor Lee-Ann Fenge, Professor of Social Care at Bournemouth University, framed the discussion around the opportunities and challenges of DigAT for social care. The workshop was chaired by Sir Bernard Silverman FRS, Emeritus Professor of Statistics at the University of Oxford. The key topics discussed were technical challenges, user design and co-production challenges, privacy and surveillance concerns and challenges to affordability of devices.

Roundtable on DigAT in gaming, July 2024

The Royal Society and PlayStation jointly organised a roundtable on DigAT in gaming to understand how DigAT can enhance accessibility and how insights from gaming DigAT could be adopted for daily life. The roundtable convened industry representatives developing DigAT for gaming and included remarks and case studies shared by Katy Minshall, Public Policy Director at PlayStation and Dr Kieren Mayers, Senior Director of Environment, Social and Governance at Sony Interactive Entertainment. The roundtable was chaired by Areeq Chowdhury, Head of Policy (Data and Digital Technologies) at the Royal Society. The key topics discussed included best practices for DigAT in gaming, challenges and limitations in developing and implementing DigAT and future trends for advancing DigAT in gaming and other sectors.

Roundtable on technology transience and obsolescence of DigAT, July 2024

The Royal Society organised a virtual roundtable on technology transience, obsolescence and user abandonment of DigAT. The roundtable convened an interdisciplinary and international group of experts to explore the drivers of technology transience and propose actionable solutions. These topics were introduced through invited presentations by Professor Tim Denison from the University of Oxford, Fernando Botelho representing UNICEF, Margaret Noonan from AT Suppliers’ Association and Professor John Naughton from the Minderoo Centre for Technology & Democracy at the University of Cambridge. The roundtable was chaired by Professor Michael Okun, Director of the Norman Fixel Institute for Neurological Diseases at the University of Florida. The key topics discussed were structural and policy changes for sustainable innovation in DigAT and opportunities and challenges with existing and alternative business models for development.

Acknowledgements

Steering Committee members

The members of the Steering Committee involved in this report are listed below. Members acted in an individual and not a representative capacity and declared any potential conflicts of interest. Members contributed to the project based on their own expertise and good judgement.

Royal Society staff

Many staff at the Royal Society contributed to the production of this report. The programme team is listed below.

Reviewers

This report has been reviewed by a panel of experts, who provided feedback on the report. The review panel members were not asked to endorse the conclusions or recommendations of the report, but to act as independent referees of its technical content and presentation. Panel members acted in a personal and not a representative capacity. The Royal Society gratefully acknowledges the contribution of the following reviewers.