01 · The Question
Are the Disciplines Actually Integrated, or Just Working Beside One Another?
A research project includes education, psychology, and computer science. Researchers from all three fields participate. The literature review cites all three. The proposal therefore looks interdisciplinary.
But imagine removing the computer-science analysis. The educational and psychological studies could continue unchanged. Remove the educational component, and the computer-science analysis would still ask the same questions, use the same methods, and reach the same conclusions. The findings meet only near the end, where they are summarized together.
That may be valuable multidisciplinary research. It is not necessarily interdisciplinary.
The distinction matters because interdisciplinarity is not established by disciplinary variety alone. What makes the research genuinely interdisciplinary is integration: the disciplinary contributions interact sufficiently that the resulting question, design, analysis, or explanation becomes something the components could not produce independently.
03 · What You Need to Know
The Difference Is Not How Many Disciplines You Have but What Happens Between Them
Multidisciplinary research can be excellent research
The distinction should not be treated as a ranking in which interdisciplinary research is sophisticated and multidisciplinary research is somehow deficient.
Multidisciplinary work can be exactly what a problem requires. Several disciplinary specialists may investigate different dimensions of a shared problem and produce useful findings without needing to alter one another's conceptual or methodological approaches.
The important issue is accuracy of description.
A longstanding distinction in scholarship on research integration characterizes multidisciplinary work as involving components that can largely stand on their own, whereas interdisciplinary work contains substantive linkages among those components. More recent reviews similarly describe multidisciplinary research as comparatively additive or parallel and interdisciplinary research as interactive, with knowledge crossing disciplinary boundaries while the participating disciplines can retain their identities.
Parallel contributions are the clearest sign of multidisciplinarity
Imagine a project examining generative AI in universities.
An education researcher studies learning outcomes. A psychologist studies student attitudes. A computer scientist evaluates model accuracy. Each uses disciplinary theories and methods appropriate to the respective subproblem.
At the end, the project reports all three sets of findings.
Nothing is necessarily wrong with this design. The question is whether the components influence one another. If the education study would have been designed identically without the computer-science component, and the computational analysis would be interpreted identically without the learner evidence, the contributions remain largely parallel.
Multidisciplinary
Different disciplines contribute to a common topic or goal, but their analyses can remain substantially independent and are often juxtaposed or coordinated.
Interdisciplinary
Disciplinary insights are connected so that they become substantive inputs to one another and contribute to an integrated understanding of the problem.
Look for intellectual dependence between contributions
A useful diagnostic question is: does what happens in one disciplinary component matter to what another component asks, measures, analyzes, or concludes?
Integration becomes visible when the answer is yes.
For example, a computational analysis might identify characteristics of AI-generated explanations that educational researchers then use to formulate experimental conditions. Learner evidence might subsequently reveal that a technically important feature is educationally irrelevant, causing the computational researchers to reconsider which system characteristics deserve attention.
Now knowledge is moving in both directions.
This kind of intellectual interdependence is consistent with earlier work describing interdisciplinary integration as involving component analyses that serve as substantive inputs to each other rather than merely standing beside one another.
Integration can occur at several points in the research process
Researchers sometimes look for integration only in the conceptual framework. That is too narrow.
Interdisciplinary integration can occur in the formulation of the research problem, development of concepts, theoretical reasoning, choice or adaptation of methods, construction of data, analysis, interpretation, or synthesis of findings.
| Where integration occurs |
What it might look like |
| Problem formulation |
Different disciplinary perspectives jointly change what the project considers the central problem |
| Concepts |
Definitions are compared, translated, differentiated, or reconstructed to establish common ground |
| Theory |
Explanatory mechanisms from different fields are connected into a coherent account |
| Methods |
Methods or data from one field are adapted because of questions arising from another |
| Analysis |
Evidence generated in one component changes how another component is analyzed |
| Interpretation |
The meaning of one disciplinary finding depends on evidence or concepts from another |
Not every project needs integration at every point. The stronger test is whether integration occurs where it is necessary to answer the overarching question.
Integration should produce some cognitive advancement
One influential account of interdisciplinary understanding emphasizes integrating disciplinary knowledge and modes of thinking to produce an advancement that would have been unlikely through a single disciplinary approach.
This is a demanding but useful standard.
After integration, what can you understand that you could not understand from the disciplinary contributions independently?
Perhaps the integrated account explains why a relationship occurs. Perhaps it identifies a cross-level mechanism. Perhaps it reveals that apparently contradictory findings apply under different conditions. Perhaps it reframes the original problem.
If the final conclusion is essentially “Discipline A found this, while Discipline B found that,” integration may still be limited.
Integration does not require complete disciplinary fusion
Genuine interdisciplinary research does not necessarily erase disciplinary identities.
A recent review of interdisciplinary research distinguishes it from transdisciplinary approaches partly on this basis. Interdisciplinary work can integrate disciplinary perspectives while preserving their underlying disciplinarity. The objective is not necessarily to create a single homogeneous perspective.
This matters because researchers sometimes believe that successful integration requires every difference in terminology, theory, or method to disappear.
It does not.
Disciplines may retain different methods because they answer different portions of the problem. They may preserve distinct concepts where collapsing them would destroy meaningful distinctions. Integration concerns how those contributions are related, not whether they become indistinguishable.
Common ground is often necessary before integration
Disciplinary perspectives may use different definitions, assumptions, units of analysis, or standards of evidence. Before their findings can be integrated, researchers often need enough common ground to understand how the contributions relate.
Common ground does not necessarily mean agreement on everything. It can involve clarifying terminology, identifying overlaps and differences, establishing relationships among constructs, or determining which apparently conflicting insights operate under different conditions.
Empirical work on interdisciplinary teams has shown that explicitly discussing disciplinary differences and constructing common ground can support the integration process. This matters because disciplinary perspectives may be internally coherent yet not mutually coherent when first brought together.
If the perspectives rest on conflicting assumptions, researchers may first need to determine whether those disciplinary assumptions make the research question internally inconsistent.
One discipline can still provide the conceptual anchor
Integration does not require equal disciplinary weight.
A project may be organized primarily around an educational theory while integrating concepts from human-computer interaction that materially alter how the educational mechanism is understood. Another may use a psychological framework as its conceptual center while incorporating institutional processes that define important boundary conditions.
The relevant question is whether the secondary contribution changes the explanation in a necessary way.
This is why choosing which discipline should provide the primary conceptual framework is compatible with genuine interdisciplinarity. An anchor can organize integration without monopolizing it.
Collaboration does not guarantee integration
A team containing researchers from several departments can still produce multidisciplinary work. Conversely, interdisciplinary reasoning can sometimes be undertaken by an individual researcher with sufficient disciplinary competence.
Team composition and intellectual integration are related but distinct.
Research on interdisciplinary collaboration emphasizes full intellectual participation, ongoing interaction, and interconnection of knowledge throughout the research process. Simply dividing the project into disciplinary work packages and assembling the outputs later offers fewer opportunities for that integration.
If expertise from another discipline is essential, this is also why relevant collaborators should usually become involved while important decisions remain changeable.
The strongest integration is not always the most appropriate integration
Interdisciplinary integration can vary in degree.
Rossini and Porter distinguished weaker arrangements, in which components retain substantial independence, from stronger forms involving increasingly substantive relationships and, at the strongest end, a shared overarching theoretical framework.
Not every project needs the strongest possible integration. A well-defined research problem may require only a particular connection between two disciplinary analyses.
The appropriate question is therefore not “How can we integrate everything?” It is “What needs to be integrated for this research problem to be answered adequately?”
This protects interdisciplinary research from becoming integration for integration's sake, which is perhaps the scholarly equivalent of adding another conceptual arrow because there is still room in the figure.
The final product should reveal the integration
A genuinely interdisciplinary project should make its integration visible to the reader.
The research question should show why several perspectives are necessary. The conceptual framework should show their relationship. Methods should generate evidence capable of addressing that relationship. Results and discussion should return to the integrated problem rather than reporting separate disciplinary conclusions indefinitely.
If readers need the authors to assure them repeatedly that the study is interdisciplinary because they cannot locate the integration in the research itself, the design may deserve another look.
04 · A Practical Example
From Three Parallel AI Studies to One Integrated Research Problem
Hypothetical Example
Students are accepting inaccurate AI-generated feedback
A research team includes an educational researcher, a psychologist, and a human-computer interaction researcher. They want to understand why students sometimes accept incorrect feedback generated by an AI system.
The team first proposes three parallel components.
Education component Measure students' feedback literacy and learning outcomes.
Psychology component Measure students' trust and metacognitive confidence.
Technology component Analyze characteristics and accuracy of AI-generated responses.
Those components could produce a useful multidisciplinary project. Each discipline contributes information about the common topic.
To move toward interdisciplinary integration, the team changes the question: How do characteristics of AI-generated explanations interact with students' metacognitive judgments to influence acceptance of inaccurate academic feedback?
The technological analysis now identifies features used to construct or classify the feedback students encounter. Psychological concepts help explain how students interpret those features. Educational theory determines why acceptance or rejection matters within the feedback process. Findings concerning learner judgment can also change which technological characteristics are considered educationally consequential.
The disciplines no longer simply provide three views of AI feedback. Their contributions depend on one another to explain the relationship at the center of the research question.
