What Is Human-Computer Interaction? Research Areas, Methods, and Career Paths

Human-computer interaction (HCI) is the interdisciplinary field that studies how people use computing technology and applies that understanding to design, build, and evaluate interactive systems that people can use effectively. It draws on computer science for the technology, on psychology and cognitive science for models of how people perceive, think, and act, and on design, the social sciences, and engineering for the methods used to find out whether a system actually works for the people it is meant to serve. Where computer science asks what can be computed and built, HCI asks a complementary question: given what is known about people, what should be built, and how can we tell whether it works for them?

What human-computer interaction studies

The unit of study in HCI is not the computer and not the person alone, but the interaction between them in a real context of use. Typical questions include how quickly and accurately people can complete a task with an interface, what mistakes a design invites, how a tool changes the way a group collaborates, and who is excluded by assumptions built into a product. International standard ISO 9241-11 frames usability in these terms, as the extent to which specified users can achieve specified goals with effectiveness, efficiency, and satisfaction in a specified context of use. ISO 9241-210 describes a human-centred design process for interactive systems. The key point in both is that usability is a property of a system used by particular people for particular purposes, not an abstract property of the software.

HCI research generally falls into three kinds of contribution. Empirical studies establish how people behave with existing or new technology. Artifact contributions present a new interface, device, or technique together with evidence about how it performs. Theoretical and methodological contributions offer models, design frameworks, or better ways of measuring interaction. A single paper at a major venue often combines two of these, for example a new input technique and a controlled study that evaluates it.

HCI, human factors, and related fields

HCI overlaps with several neighboring disciplines, and the boundaries are fuzzy by design. Human factors engineering is the older, broader tradition concerned with fitting all kinds of systems, physical and cognitive, to human capabilities; HCI can be seen as the part of that concern focused on computing, and many practitioners move between the two. Cognitive science supplies much of the theory about attention, memory, and decision-making that interface design relies on. Artificial intelligence has become increasingly entangled with HCI as interactive systems incorporate learned models whose behavior is less predictable than conventional software. User experience (UX) design is the industry practice most closely related to HCI; UX work tends to emphasize shipping products, while academic HCI emphasizes generalizable knowledge and rigorous evaluation, although people and methods flow freely between the two. Related disciplines such as psychology, communication studies, robotics (human-robot interaction), sociology, and anthropology all contribute concepts and methods.

Major subfields and research areas

• Interaction techniques and input/output devices – touch, pen, gesture, voice, gaze, and haptic interaction; text entry; and the evaluation of pointing and selection performance.
• Interface and interaction design – the design of layouts, workflows, visualizations, and feedback, including the study of design processes themselves.
• Information visualization and visual analytics – interfaces that help people explore and make sense of data.
• Computer-supported cooperative work (CSCW) and social computing – how groups communicate, coordinate, and share information through technology, from workplace collaboration tools to online communities.
• Ubiquitous, mobile, and wearable computing – interaction away from the desktop, including sensing and context-aware systems.
• Augmented, virtual, mixed, and extended reality (AR/VR/MR/XR) – immersive interaction and its perceptual and ergonomic consequences.
• Accessibility and inclusive design – technology for people with disabilities and for people of different ages, languages, and circumstances.
• Human-AI interaction – how people understand, direct, trust, and correct AI-enabled systems, including explanation, transparency, and appropriate reliance.
• Affective computing and brain-computer interfaces – systems that sense or respond to emotion or neural signals.
• Usable privacy and security – studying the human side of security decisions and privacy controls.
• Health, learning, and domain-specific interaction – interfaces for clinical settings, education, creative work, and other domains where context shapes what good design means.

A short history

The roots of HCI lie in the 1940s to 1960s work on human performance and in early interactive computing. Fitts’s work on the speed and accuracy of aimed movements (1954) later became the basis for models of pointing performance, and Ivan Sutherland’s Sketchpad (1963) is widely cited as an early demonstration of direct graphical interaction. Douglas Engelbart’s 1968 public demonstration showed a mouse-driven, windowed, collaborative computing environment well before such features were commonplace. The personal-computer era of the late 1970s and 1980s brought computers to non-specialists and created demand for a scientific basis for usability. In 1983 Stuart Card, Thomas Moran, and Allen Newell published The Psychology of Human-Computer Interaction, which applied cognitive psychology to interface design and helped establish the term “HCI” and the field’s identity. Since then the field has widened from the individual desktop user to networks, mobile and embedded devices, social platforms, immersive systems, and AI-enabled products.

Research methods in HCI

HCI is methodologically pluralistic. A single research program may combine laboratory experiments, field studies, qualitative inquiry, and building new systems. The sections below describe the most common approaches and where they fit in a project.

User research: finding out what people need

Early-stage studies aim to understand users, tasks, and contexts before a design exists. Common methods are semi-structured interviews, contextual inquiry (observing and interviewing people in the setting where they actually work), diary studies in which participants record experiences over days or weeks, focus groups, surveys, and ethnographic observation. These methods produce largely qualitative data, so HCI papers draw on the same analytic toolkit as other qualitative research; see CASRAI’s dictionary entries on qualitative research and qualitative data collection techniques, and its guide on data saturation and information power for how researchers justify the number of interviews. For large-sample attitudinal data, see survey research methods.

Design and prototyping methods

Designers and researchers externalize ideas as sketches, storyboards, paper prototypes, and interactive prototypes, then test them with people. Participatory and co-design approaches involve prospective users as partners in generating the design rather than only as subjects of evaluation. Because prototypes are cheap to change, this phase is typically iterative: design, test, revise.

Expert-based inspection

Heuristic evaluation, introduced by Jakob Nielsen and Rolf Molich around 1990, has a small number of evaluators examine an interface against a set of recognized usability principles. Cognitive walkthroughs step through a task from a new user’s perspective and check at each step whether the user would know what to do. These inspection methods require no participants, so they are fast and inexpensive, but they cannot substitute for observing real users.

Usability testing

In a usability test, representative participants attempt realistic tasks with a system while researchers observe, record outcomes, and ask about their experience. Typical measures are task success, time on task, errors, and subjective ratings; standardized questionnaires such as the System Usability Scale (SUS) are widely used for the latter, and the NASA Task Load Index (NASA-TLX) is a common instrument for perceived workload. Formative testing during design aims to find and fix problems, while summative testing evaluates a finished design against a benchmark. The think-aloud technique, in which participants narrate their thoughts while working, is a staple of formative testing; see CASRAI’s guide to think-aloud protocols and its guide to cognitive interviewing, a closely related technique for pretesting questionnaires. In regulated domains, usability testing is also a compliance activity: the FDA expects human factors and usability engineering evidence for many medical devices (see the FDA human factors and usability engineering guidance).

Controlled experiments and quantitative evaluation

When the question is causal or comparative, such as whether a new input technique is faster than an existing one, HCI researchers run controlled experiments with within-subjects or between-subjects designs, counterbalancing of conditions to manage order effects, and standard statistical analysis. Online experiments and large-scale A/B tests compare variants of a live system on real traffic. The design concepts are the general ones covered in CASRAI’s dictionary entry on experimental design. Instrumentation can include interaction logs, eye tracking, and physiological sensors.

Field deployment and longitudinal study

Some phenomena only appear over time or in natural settings: how a tool is adopted, appropriated, or abandoned. Field deployments place a prototype or product in real use for weeks or months and combine logs with interviews. They trade control for ecological validity.

Choosing and combining methods

No single method answers every question. A common pattern is to use interviews and observation to understand a problem, prototype and test to refine a design, an experiment to compare candidate solutions, and a field study to see how the result holds up in practice. Mixed-methods work is common, and reviewers expect the chosen method to match the research question and the claim being made.

Human-subjects research and IRB considerations in HCI

Most HCI studies involve people, so research administration questions arise early. In the United States, research funded or regulated under the Common Rule (45 CFR 46) requires review by an institutional review board (IRB) unless it falls outside the regulatory definition of human subjects research or qualifies for exemption; see CASRAI’s entries on the Common Rule and the IRB. Points that come up repeatedly for HCI researchers include:

• Is it “research” at all? The regulations apply to systematic investigations designed to develop or contribute to generalizable knowledge (see Research (45 CFR 46.102(l))). Routine product usability testing done only to improve a company’s own product is often not research in that sense, but an academic study intended for publication generally is. Institutions make this determination, not the researcher, so ask your IRB office.
• Exemptions. Many HCI studies are low risk. Under the revised Common Rule, one exemption category covers research that involves only educational tests, survey procedures, interview procedures, or observation of public behavior when specified conditions are met, such as recording information so that subjects cannot readily be identified, or confirming that disclosure of responses would not place subjects at risk, or an IRB conducting limited review. Interviews and surveys about interface use often fit this category; see exempt human subjects research and limited IRB review. Exemption determinations are made by the institution, and institutional policies vary, so do not assume a study is exempt without confirmation.
• Informed consent. Participants must understand what they are doing and that participation is voluntary; see informed consent. Online studies commonly use consent pages, and studies of deployed systems need a plan for obtaining consent from people who did not choose to be in a study.
• Data and privacy. Screen recordings, audio, eye-tracking, and interaction logs can be identifying even when no name is collected. Plan storage, retention, and de-identification before data collection, and state them in the protocol and consent form.
• Vulnerable and special populations. HCI frequently works with children, older adults, people with disabilities, and patients, each of which can trigger additional protections or accessibility needs for the consent process itself. For children, see 45 CFR 46 Subpart D.
• Deception and debriefing. Some experiments, such as Wizard-of-Oz studies in which a human secretly simulates an automated system, withhold information about how the system works; the IRB will expect justification and a debriefing plan.
• Crowdsourced and online participants. Platforms that recruit paid online participants raise questions about fair compensation, data handling, and whether workers are treated as research subjects; institutions typically expect review for studies intended for publication.

Outside the United States the same questions are handled by research ethics committees under national rules, and funder and journal requirements may apply regardless of country.

ACM SIGCHI and the CHI conference

The Association for Computing Machinery’s Special Interest Group on Computer-Human Interaction (SIGCHI) describes itself as the world’s largest association of professionals contributing to the research and practice of HCI, with more than half of its members outside the United States. The formation of SIGCHI was first publicly announced at the 1982 Human Factors in Computer Systems conference in Gaithersburg, Maryland, organized by Bill Curtis and Ben Shneiderman. That meeting is regarded as the first of the CHI conference series, formally the ACM Conference on Human Factors in Computing Systems, which has been held annually and is generally considered the most prestigious venue in the field.

For researchers, the practical consequence is a publication culture that differs from many disciplines: selective peer-reviewed conference papers, archived in the ACM Digital Library, carry weight comparable to or greater than journal articles in HCI and related areas of computing. This affects tenure files, grant biosketches, and author-credit conventions; see CASRAI’s discussion of conference versus journal authorship credit in computer science and the ACM policy on authorship. The ACM also runs an artifact review and badging program relevant to sharing study materials and software. Alongside CHI, the community publishes in journals such as ACM Transactions on Computer-Human Interaction (TOCHI), the International Journal of Human-Computer Studies, Human-Computer Interaction, and Interacting with Computers, and in specialist conferences on topics such as user interface software and technology, cooperative work, ubiquitous computing, and accessibility.

Who funds HCI research

In the United States, the National Science Foundation is the principal federal funder of core HCI research, through the Directorate for Computer and Information Science and Engineering (CISE). NSF’s Division of Information and Intelligent Systems (IIS) supports research on the interrelated roles of people, computers, and information, and its Human-Centered Computing (HCC) program has funded research on designing computing systems that amplify human capabilities and on assessing their benefits, effects, and risks. NSF reorganizes its program structure from time to time, so check the current CISE and IIS pages and active solicitations for the program names and deadlines that apply today. Proposals follow the Proposal and Award Policies and Procedures Guide and go through merit review; early-career researchers may be eligible for the CAREER award.

HCI work is also funded by other sources according to application area: health agencies and foundations for clinical and digital-health interfaces, education funders for learning technologies, defense and aerospace agencies for human-systems integration (see the funder list in the human factors engineering guide), and, substantially, industry research labs and corporate collaborations. Specific program names, award sizes, and success rates change; confirm them with the agency before planning a proposal.

Training and career pathways

HCI is taught in many different departments: computer science, information science and information schools, psychology and cognitive science, design, engineering, and communication, and some universities have dedicated HCI programs. Typical entry points are a bachelor’s degree in one of these fields, a master’s degree with a specialization in HCI or interaction design, and a PhD for academic and industrial research roles. A strong researcher usually combines technical skill (programming, prototyping, data analysis) with training in research design, statistics, and qualitative methods, which is why graduate programs emphasize both. Careers span academic faculty positions, industry UX research and design, and research roles at technology companies and government laboratories. Professional involvement typically runs through SIGCHI, and the ACM’s wider community of special interest groups, along with national and regional HCI societies.

HCI and research administration

HCI projects raise the same administrative questions as other human-subjects research, plus a few particular to technology: IRB review and reliance for multi-site studies, data management and retention for rich media recordings, intellectual property in software and interface designs, and open-science expectations for sharing materials, code, and anonymized data. For discipline-level context, see CASRAI’s guide to the branches of science.

Frequently asked questions

What is human-computer interaction in simple terms?

HCI is the study of how people use computers and other interactive technology, and the use of that knowledge to design technology that is easier, safer, and more useful for the people who rely on it.

What is the difference between HCI and UX?

The two overlap heavily. HCI usually refers to the academic research field, which aims at generalizable knowledge and rigorous evaluation, while UX (user experience) design usually refers to professional practice focused on the experience of a particular product. Many people work in both, and methods such as interviews and usability testing are shared.

What is the difference between HCI and human factors?

Human factors (or ergonomics) is the broader tradition concerned with fitting all kinds of systems and environments to human capabilities, including physical workstations, vehicles, and equipment. HCI is concerned specifically with interactive computing. The two communities share methods and often publish in related venues.

Do HCI studies need IRB approval?

Studies intended to produce generalizable knowledge with human participants usually require some form of IRB review at US institutions, and many HCI studies qualify for exemption or expedited handling, but the institution makes that determination. Testing done purely to improve an internal product may not count as research. When unsure, ask your IRB office before collecting data.

What is CHI?

CHI is the ACM Conference on Human Factors in Computing Systems, the flagship annual conference of ACM SIGCHI. The series dates to a 1982 meeting in Gaithersburg, Maryland, and is generally regarded as the field’s most prestigious venue.

Who funds HCI research?

In the US, chiefly NSF through its computing directorate, along with health, education, and defense agencies for application-specific work, and a large share from industry. Check current program announcements, because NSF program names and structures change.

What degree do you need to work in HCI?

There is no single required degree. Practitioners and researchers come from computer science, psychology, design, information science, and related fields. Research-track roles typically require a master’s degree or PhD; many applied UX roles require a bachelor’s degree plus a portfolio of work.

Related CASRAI resources

For neighboring disciplines, see human factors engineering, cognitive science, computer science, and artificial intelligence. For the ethics and compliance side, see what an IRB is and informed consent.