The categories and definitions outlined below reflect the work undertaken in PRO-Ethics. They emerged as common references during the project, and are important for the implementation of ethical participatory processes, particularly in relation to the activities of research funding organizations.

Bias
In the context of our work, bias is relevant in two ways: First, as often unconscious preconceived opinions, beliefs, or attitudes that influence how stakeholders in R&I define and address problems, set up processes, and perceive and interpret data. This might entail a preference for the inclusion of particular stakeholder groups over others (or a perceived greater validity of some perspectives), a predilection for specific outcomes, and an overall skewing of the process towards certain – often hegemonic – power structures. Second, as systematic errors in the vein of statistical bias that distort the process as well as the collected data and its analysis. While often unintentional, flawed methodologies, selection biases and information biases can have significant ramifications for the quality of an R&I process and its outcomes, while also negatively impacting its participants. Thus, it is essential to be mindful of potential biases and take active steps to identify and address them in order to ensure the quality and equality of an R&I process.

Citizens
While „citizen“ is not a term that should be employed uncritically²⁹, we decided to draw on this category as an established umbrella term that includes the general public, lay people, and citizens as persons (or collectives) with civic expectations³⁰. Moreover, since end-users can be categorized as citizens as well, this distinction serves to underline the general dimension of involvement, referring to the broader sense of “public participation”.

Civil society organizations
Civil society organizations are not-for-profit organizations that may represent specific groups of citizens, but whose knowledge and leverage differs to that of individual citizens. They may defend interests, often professional interests (trade unions), or causes (e.g., animals, environmental issues), or rights (e.g., minorities, women).

Communication and dissemination
In the context of R&I processes, communication refers to the sharing of contents and results of R&I activity in an accessible manner, increasing its public visibility. It is distinguished from dissemination by its primary target groups, as dissemination is targeted more towards a scientific audience, but also policymakers and industry representatives. Both communication and dissemination tend to be one-way exchanges of information towards any type of stakeholder.

Co-creation
We use co-creation to encompass comprehensive collaboration between all stakeholders of an R&I process, from its inception to its conclusion. Hailing more from the context of an R&I project, co-creation covers all stages of the research cycle, from the definition of a research question to the evaluation of a project and assessment of its impact. While this has not yet been similarly established as a valid approach, this process can be mirrored in the context of the R&I funding cycle, starting from the development of the R&I funding strategy and ending with the evaluation and impact assessment of funded projects and the overall funding program. As an umbrella term, co-creation also covers the concepts of co-design (collaboratively defining a problem and its solutions by designing technologies, processes and solutions), co-production, and co-development.

Collaboration
While collaboration broadly covers the process of people or organizations working together to achieve a goal, in the context of our framework it is important that collaboration is equal and meaningful, i.e., allow all involved stakeholders to contribute towards and impact the process and its outcomes.

Consultation
Processes of engagement (see definition of ‘engagement’) where any group of citizens or stakeholders are asked to provide input on an issue, process, policies or programs. These inputs are not guaranteed to be taken up in a meaningful manner, in that they make an impact on processes and their outcomes.

Diversity, equality/equity, inclusion
Diversity as a term reflects the many different ways we understand and categorize people (e.g., according to gender and gender identity, sexual orientation, race and ethnicity, dis/ability, socioeconomic status, etc.). Inclusion is about providing equal access to take part in a process or activities. Thus, diversity in participation is about including a range of perspectives and experiences to create outcomes that work for more than just a few. In this context, equality refers to everyone being treated equally and receiving the same opportunities and resources, while equity is about addressing individuals’ specific needs. It is essential to address bias and discrimination to achieve both equality and equity for diverse groups of people, but also to ascertain the quality of R&I and mitigate potential harms it may create³¹

Duty of care
In tort law, a duty of care is a legal obligation prescribing adherence to a standard of reasonable care to avoid careless acts that could foreseeably harm others and lead to claims of negligence. A duty of care can apply in many different contexts, ranging from a school’s duty of care towards its students and an employer’s duty towards its employees, to businesses’ duty of care towards consumers who buy certain products. In the context of participatory processes, the initiators of these processes have the duty to take measures to prevent participants from being harmed by or as a result of their participation. A duty of care may be considered a formalization of the social contract, where a person or an entity who is in a position to prevent others from being harmed has the responsibility to do what is in their ability to prevent this harm from materializing.

Empowerment
Empowerment is about enabling individuals and groups to actively participate in, influence, and benefit from R&I processes and their outcomes. It aims to distribute power equitably and in turn have a positive impact both on the concrete processes at hand and wider society. Empowerment can entail the provision of knowledge, ensuring that people have access to and an understanding of scientific findings. And it can entail participation, which is about inviting a broad range of relevant stakeholders to the table – including also underrepresented and marginalized groups – while ensuring that all participants in a process are heard and have a hand in shaping the course of the R&I endeavor.

End-users
End-users are the (presumed) groups and individuals intended to make use of the end product (including solutions and services) of an R&I process. While the concrete (groups of) end-users cannot always be foreseen in their entirety, assumptions about their needs are always inscribed into a technology, process, service or solution. Engaging potential end-users in the development of R&I is intended to better meet their needs, and thus heighten the chance of the outputs to be useful and used.

Engagement
In this document, engagement is used as an umbrella term for different types of one-way and two-way exchange, as well as collaboration between stakeholders from R&I (such as professional researchers and research funding organizations) and stakeholders beyond the R&I system (such as citizens, end-users, civil society organizations, NGOs and so on). This may include forms of communication, consultation, or more intense approaches to participatory engagement such as co-design and co-creation.

Ethics
Ethics is the discussion of and reflection on moral values and norms (in short: morality). The adjective "moral" indicates that those values and norms have a special status, typically taking the form of obligations and prohibitions. Their special status is manifested by the fact that moral rules are accompanied by praise and blame, rewards and punishments, to motivate people to live according to these norms and values³². The adjectives ethical and moral are often used interchangeably.

Evaluation (of projects and programs)
This category encompasses several types of evaluation: evaluation of project proposals (i.e., the ethical and scientific evaluation) as part of the selection process intervening in funding schemes; the interim and ex-post evaluation for projects and programs that received funding; and program evaluation. Evaluation reflects on the implementation and results of an R&I endeavor to ascertain its overall quality, and can focus both on processes and outcomes/results. In contrast, impact assessment always focuses on the broader long-term effects of an R&I process³³.

Experts
This category serves to identify individuals who are enrolled as internal or external experts in R&I processes. In the context of this document, we include both lay and professional experts in this category. For instance, citizens may be involved in an R&I process as ‘experts by experience’, providing insights into their lifeworlds and value systems. Experts may also be individuals with any sectoral/disciplinary expertise (e.g., with a background in medicine, psychology, sociology, philosophy — among others). Consequently, participants might bring different kinds of expertise and different types of knowledge (e.g., tacit, formal, endogenous, living world, etc.) to an R&I process.

Explainability, interpretability, understandability
Current AI systems often involve deep learning. While deep learning has allowed for significant progress in artificial intelligence, in comparison to traditional machine learning methods such as decision trees and support vector machines, deep learning is comparably weak in explaining its inference processes and the way its decisions come about. For that reason, deep learning algorithms are typically considered a black box by both developers and users. This lack of transparency has led to the development of ‘explainable AI’ (also XAI), in which explainability is seen as a desirable feature or even ‘must have’ of AI systems³⁴. Explainability is considered especially relevant when AI is used as decision making tool in professional context, where it is important for its acceptance that people know how a decision came about³⁵. Notions that are related to explainability are interpretability, understandability, and to a lesser extent also transparency. While these notions may have slightly different meanings, they all refer to the need to break open the black box of AI system.

Fairness
As a concept related to diversity, equality/equity, and inclusion, fairness encompasses ensuring equal access to resources and opportunities, unbiased decision-making processes, and outcomes that don't unjustly advantage or disadvantage certain groups. Like other ethical principles covered here, the specific interpretation and implementation of fairness may vary depending on the context.

Human-in-the-loop design
The term human-in-the-loop (HITL) was developed in the context of artificial intelligence (AI). It can refer to the role of human intelligence in machine learning or simulation, but also in the use of otherwise autonomous systems. When used in machine learning, the role of the human in the loop is to select the most critical data that is needed to improve the performance of an AI system (i.e., to improve its learning). In machine learning without any human in the loop, data is selected via random sampling, which may not lead to the most effective and efficient learning of the algorithm. In simulation, HITL is also referred to as interactive simulation, where the physical simulation includes human operators, such as in a flight or a driving simulator. In the context of autonomous systems, HITL is meant as a safeguard to prevent that AI-based systems autonomously take decisions with high stakes, such as in the case of autonomous weapons or autonomous cars. In those contexts, HITL is operationalized via the principle of “meaningful human control”, according to which systems should be designed in such a way that humans and not computers and their algorithms should ultimately remain in control of, and are thus morally responsible for, relevant decisions about possibly lethal operations³⁶. This involves a twofold design requirement: (1) the autonomous system should be responsive to the relevant moral reasons of the humans designing and deploying the system and the relevant facts in the environment in which the system operates; and (2) the system should be designed in such a way as to grant the possibility to always trace back the outcome of its operations to at least one human along the chain of design and operation.

Human-led design / Human-centered design (HCD)
Human-centered design (HCD) is an approach which enables designers, developers, and engineers to focus their projects on the prospective users of the products or services they are working on. It emerged in the field of Information and Communication Technology (ICT) as an approach to avoid the design of products or services that people cannot or do not want to use. HCD can be used as an umbrella term to include a diverse range of approaches³⁷. There is also a standard for Human-Centered Design Processes for Interactive Systems from the International Organization for Standardization (ISO), which describes the following key principles: To start with an explicit understanding of prospective users and their tasks and environments; to involve prospective users throughout the process of design and development; to involve prospective users in timely and iterative evaluations and to let these evaluations drive and refine the process of design and development; to organize an iterative process; to view the user experience holistically, e.g., not as usability, but also as people’s aspirations and emotions; and to organize a multidisciplinary project-team.

Informed consent (IC)
Informed consent (IC) is one of the fundamental ethical principles for research involving human participants. The principle aims to ensure that no person can be made a research subject without free and voluntary consent and with full information about what it means for them to take part. The principle is also key in medical decisions, where patients should always give consent to treatment after being given information about the treatment or diagnosis and its potential risks. In both research involving human participants and in medicine, securing informed consent is a formalized requirement before research or treatment can take place. In recent decades, alternative approaches to IC have been suggested that better address the needs of participants and account for the social embeddedness of both the stakeholders and the processes of R&I. These include new and more accessible formats for mediating IC procedures (including movies and comics), but also approaching informed consent as a continuous process in need of adjustment due to the unpredictability of R&I projects. Apart from these more formalized procedures, informed consent is increasingly also referred to in contexts beyond research involving human participants or medicine, to warn against potentially exploitative technologies that are enforced upon people against their wish.

Intellectual Property (IP)
Intellectual property rights (IPR) are the rights given to persons over the creations of their minds, such as inventions, literary and artistic works, designs, but also symbols, names, and images used in commerce. IP can be protected in law e.g. via patents, copyright and trademarks, which enable people to earn recognition or financial benefit from what they invent or create. In the context of innovation, intellectual property rights aim to strike a balance between the interests of innovators and the wider public interest. Without a system of IPR, companies are allegedly less willing to invest in the development of new products, if the monetary benefits of these efforts are not somehow protected³⁸. In research with low Technology Readiness Levels, discussions on intellectual property mostly concern authorship of scientific publications. Common standards in research integrity prescribe that authorship should only be granted to a person if he or she has concretely contributed to the paper, often via a CRediT statement accompanying the paper³⁹. In the context of participation, it is not always easy to demarcate the participants’ contribution to the end product. Recognition of participants’ efforts may require that participants are granted authorship. A too strict view on who deserves authorship may misrecognize participants’ contribution, even if that contribution has not been primarily an intellectual one. In any case, it is important to set up clear procedures – and, where relevant, legal contracts – from the very start of a collaborative process, to ensure clarity and transparency over each contributor’s rights over a final product.

Impact assessment
Impact assessment focuses on the longer-term and broader effects of an R&I process. It entails the definition of specific qualitative and quantitative outcomes and indicators for achieving impact, as well as instruments to measure these indicators. Evidence is then gathered and analyzed to show concrete results. Depending on the concrete focus of an R&I process, it might aim to achieve societal, political, institutional, scientific, economic, environmental or technological impact. As impact necessarily unfolds over time, change usually unfolds beyond the lifetime of an R&I process, making it difficult to substantiate.

Monitoring
The systematic observation of the implementation of funded projects and their results in the context of RFO funding schemes. Monitoring is usually carried out internally with support by external experts, e.g., for interim or final reviews. Ex-post monitoring of results can also involve other stakeholders, in addition to the involvement (feedback) of program beneficiaries.

Participants
Participants are defined as persons who take part in participative processes. In this document, we primarily use this term for non-traditional stakeholders of an R&I process such as citizens in the broadest sense, (end)-users of a technology, residents of an area, people affected by an issue, entrepreneurs, project beneficiaries, and so on. Participants are engaged in such processes due to their specific knowledge, perspectives and/or interests they bring to the table. Participants can be individuals or representatives of institutions or groups and may include vulnerable groups such as patients, children, or older adults. Often, there is an overlap in categories of participants, as individuals may draw expertise from different fields and experiences. In each instance, it is important to develop an awareness of the characteristics of those participating, and address them in the design and implementation of a participatory process.

Participation
While there is no uniform definition of participation, the term is often described as a form of engagement that allows (potentially) affected stakeholders to partake in decision-making for R&I. There is a graduation to the intensity of participatory processes, ranging from limited and short-term involvement to a comprehensive collaboration between all stakeholders of an R&I process, from inception to completion. Truly participatory practices empower stakeholders to shape decisions in accordance with their own values and worldviews.

Privacy
Privacy is a fundamental right that refers to a person's ability to control their personal information and decide when, how, and to what extent such information is shared with others. There are clear regulations and compliance procedures structured by the GDPR and other relevant national and international law, which entail provisions for the collection, processing, dissemination and storage of personal data and the gathering of (informed) consent. In participatory R&I processes, informed consent is especially important as it opens up the conversation to consider the focus and context of a process, its goals, any associated risks, the intended scope of participation and roles of involved stakeholders, as well as the expectations people bring to the table. It is an essential tool for creating transparency from the very outset.

Program design
In the context of research and innovation, program design refers to the identification of program objectives and of R&I priorities, resulting in the definition of funding opportunities, while adhering to specific regulations.

Research ethics and research integrity
Research integrity refers to the process of conducting research in a way which allows others to have trust and confidence in the methodologies used and resultant findings. Data, methods, interpretation and presentation/reporting must meet established and appropriate scientific, legal and professional standards. Research ethics pertains to the moral issues that arise in research design and implementation, for instance in relation to the protection of humans, animals, the environment, data, as well as the proper protection of other objects⁴⁰.

Representation
The substitution of an individual or class in place of a person (such as a sibling of a severely ill person who is not able to express her/his own preferences). Representation needs to be fair, but the precise meaning of fairness is context dependent. This may mean that some contexts require additional efforts to include particular stakeholders.RFO activitiesIn the context of PRO-Ethics, RFO activities and processes refer to the entirety of the R&I funding cycle: 1) strategy development; 2) program/funding scheme development; 3) call topic development and call launch; 4) proposal evaluation; 5) project implementation and monitoring (which might include training and support of project beneficiaries); and 6) project/program evaluation and impact assessment.

Scientists and other research institute representatives
Scientists (of any scientific or technical field) or other representatives of research institutions involved in RFO processes, either for their individual expertise or for their affiliation to a research institution, but speaking primarily for themselves. This is separate to stakeholders involved explicitly as representatives of their organizations. This category also includes researchers, who are individuals working for a legal entity (SMEs, universities, research institutes) conducting public or privately funded research.

Stakeholders
In the context of R&I funding and program development, the term “stakeholder” usually refers to businesses, institutional representatives and other interest groups that are traditionally included in the processes of R&I funding. For PRO-Ethics, we have broadened this definition of stakeholders to anyone with a stake in an R&I process, i.e., anyone who may affect or be affected by it in any way. In the project, we focused mainly on the participation of “non-traditional” stakeholders, by which we mean anyone not usually included in the activities of research funding organizations and other R&I processes. Among these are citizens, in the broadest sense, residents of an area, end-users of a technology, people affected by an issue, beneficiaries of funding calls, entrepreneurs, and others. Due to the broad nature of our understanding of stakeholder, our use of the term throughout the document includes both traditional and non-traditional stakeholders, as well as anyone else involved in an R&I (funding) process such as research and program managers, scientists, experts, consultants and facilitators⁴¹.

Sustainability
Coming from broadly different schools of thought historically, sustainability is nowadays most commonly understood as the three-pillar goal for people to co-exist on earth for a long time. These three pillars or dimensions are: environmental, economic, and social⁴². Some conceptions of sustainability refer primarily to the environmental dimension, typically encompassing problems related to climate change, loss of biodiversity, loss of ecosystem services, land degradation, and air and water pollution. There is no unanimously accepted definition of sustainability, and in that sense the concept can function as a boundary object. However, the definition of sustainable development by the 1983 UN Commission on Environment and Development (Brundtland Commission) has been influential in the current three-pillar use of the term sustainability. In its report, the commission defined sustainable development as development that “meets the needs of the present without compromising the ability of future generations to meet their own needs”⁴³.

Transparency
Transparency, literally the quality of being easily seen through, is nowadays often used in the sense of being open and honest. It can refer to people, companies, and businesses, but also to governance practices. Together with accountability, transparency is seen as the cornerstone of integrity, allowing others to see what actions are performed⁴⁴. With the digitalization of society, transparency is increasingly seen as an important safeguard against black-box algorithmic decision-making. In this context, the Organization for Economic Co-operation and Development (OECD) has defined transparency as one of the three principles for a good digital government. More generally, transparency in participatory processes can refer to clarity and openness about the role and expectations of participants in these processes. Such aspects can be made explicit in informal conversations, but also more formally through codes of conduct, declarations of intent, and informed consent sheets.

Trustworthiness
Trustworthiness is often used in the context of digital technologies and particularly in AI. The European Commission’s High-Level Expert Group on AI (AI HELG) defines trustworthy AI as having three components: (1) it should be lawful, ensuring compliance with all applicable laws and regulations; (2) it should be ethical, demonstrating respect for, and ensure adherence to, ethical principles and values; and (3) it should be robust, both from a technical and social perspective, since, even with good intentions, AI systems can cause unintentional harm⁴⁵. Trustworthy AI concerns not only the trustworthiness of the AI system itself but also comprises the trustworthiness of all processes and actors that are part of the system’s life cycle. In their Guidelines on trustworthy AI, the AI HELG put forward seven key requirements that AI systems should meet in order to be deemed trustworthy: (1) human agency and oversight; (2) technical robustness and safety; (3) privacy and data governance; (4) transparency; (5) diversity, non-discrimination and fairness; (6) societal and environmental well-being; and (7) accountability.

Value-sensitive design (also value-led design)
Value-sensitive design or value-led design refers to a family of design approaches that aim to account for values in the design. Developed in the context of Information and Communication Technology (ICT), value-sensitive design aims to account for human values in a principled and comprehensive manner throughout the design process. It was originally introduced as a tripartite methodology that consists of conceptual, empirical, and technical investigations⁴⁶. The conceptual investigations identify the stakeholders and values that are relevant to the design at hand. They aim to clarify different ways of looking at values and to make the different interpretations explicit, which is further contextualized through empirical investigations which look at how an artifact is used by humans, also as a means of evaluating the success of a particular design. Lastly, technical investigations focus on the technology itself, looking at how technological properties may or may not allow for inclusion of relevant values identified in the conceptual and empirical investigations. Due to certain physical properties, some design options are not possible. However, the technical investigations go further than simply describing these physical constraints; they also include more pro-active investigations focusing on how the design should be changed so that the values identified in the conceptual and empirical investigations can be included in the design. Inspired by the value-sensitive design approach, Design-for-Values is a design approach that explicitly takes the challenges and dilemmas stemming from conflicting values as a driver for innovative design, but also giving design a much more prominent role in actually solving potential value conflicts⁴⁷. In this approach, innovative design strategies can open up new possibilities so that certain trade-offs between conflicting values no longer need to be made. The nature of design allows for creating new opportunities and making certain state-of-affairs that have hitherto been impossible now feasible or physically realizable.