related work meaning in research

Student Academic Success Center

Related work / literature review / research review, download pdf handout:   literature reviews, watch video:   literature reviews.

A  literature review, research review,  or  related work   section compares, contrasts, synthesizes, and provides introspection about the available knowledge for a given topic or field. The two terms are sometimes used interchangeably (as they are here), but while both can refer to a section of a longer work, “literature review” can also describe a stand-alone paper.

When you start writing a literature review, the most straightforward course may be to compile all relevant sources and compare them, perhaps evaluating their strengths and weaknesses. While this is a good place to start, your literature review is incomplete unless it creates something new through these comparisons. Luckily, our resources can help you do this!

With these resources, you’ll learn:

• How to write a literature review that  contributes  rather than  summarizes
• Common mistakes to avoid
• Useful phrases to show agreement and disagreement between sources

Need one-on-one help with your literature review or research article? Schedule an appointment with one of our consultants now!

Schedule an Appointment

Quick Links

• Academic Calendar
• Academic Integrity
• Bias Reporting and Response
• Statement of Assurance
• Documents, Forms, and News [Internal Staff Only]

Other Helpful Departments

• Disability Resources
• Center for Student Diversity & Inclusion
• Graduate Education
• Office of International Education
• University Health Services

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Springer Nature - PMC COVID-19 Collection

What is a related work? A typology of relationships in research literature

Shayan doroudi.

School of Education, University of California, Irvine, 401 E. Peltason Drive, Suite 3200, Irvine, CA 92617 USA

Associated Data

An important part of research is situating one’s work in a body of existing literature, thereby connecting to existing ideas. Despite this, the various kinds of relationships that might exist among academic literature do not appear to have been formally studied. Here I present a graphical representation of academic work in terms of entities and relations, drawing on structure-mapping theory (used in the study of analogies). I then use this representation to present a typology of operations that could relate two pieces of academic work. I illustrate the various types of relationships with examples from medicine, physics, psychology, history and philosophy of science, machine learning, education, and neuroscience. The resulting typology not only gives insights into the relationships that might exist between static publications, but also the rich process whereby an ongoing research project evolves through interactions with the research literature.

Introduction

An important part of the research process is literature search: identifying prior work that is of relevance to the present idea being investigated. In many cases, this is an activity that a researcher may defer until writing up the results of the project, in which case, it is primarily an activity one does because one “has to” rather than an activity that can substantially change the course of the research. In some cases, whether due to negligence or the difficulty of finding related works, a researcher may never come across the fact that someone had previously tackled the same problem or made a similar discovery, and perhaps only years later (if ever) it may be realized (Merton, 1963 ; Ke et al., 2015 ; Sacks, 2002 ). But at its best, this is an activity that leads to new insights into the research problem, generates new ideas, and alters the course of the research. In fact, in some cases, searching for related work can become the research process itself; through connecting various pieces of research literature alone, one can discover previously undiscovered public knowledge (Swanson, 1986 ).

Despite the importance of prior literature in the research process, there has been little effort, if any, dedicated to developing a typology of related works, that is, a typology of relationships that might exist among different pieces of research literature. (Of course, it is entirely possible that such a typology has been constructed, but I have missed it due to an inadequate search of the literature!) In this paper, I propose such a typology to help us better understand the kinds of prior work that might have bearing on a research project. I first present a form of knowledge representation that can theoretically be used to represent any piece of research literature or research project. I then present a typology of relationships that can connect two pieces of research in terms of operations that can apply to the two representations, thereby resulting in a representation of the relationship. I will demonstrate the various operations and how they might be employed with a variety of examples from different fields, including medicine, physics, psychology, history and philosophy of science, machine learning, education, and neuroscience. The same form of representation applies to both published research literature and research projects or topics, whether nascent or fully-fledged. In fact, some of the relationships discussed below make more sense in the context of research projects (or broader research agendas) that can dynamically evolve as relevant literature is encountered, rather than research papers whose underlying representations are static. As such, I will use the terms publication, literature, project, and topic somewhat interchangeably.

The specific representation I use is borrowed from structure-mapping theory (Gentner, 1983 ; Falkenhainer et al., 1989 ), which was originally developed as a way to structurally represent analogies. Structure-mapping theory is particularly useful here, both because we can use it when discussing abstractions and analogies, and because the underlying representation can also handle other types of relationships among literature. I could have instead used other forms of knowledge representation, such as conceptual graphs (Sowa, 1976 ), entailment meshes (Pask et al., 1975 ; Pask, 1988 ), or category theoretic representations like ologs (Spivak & Kent, 2012 ). There may be relative advantages to each of these, but the representation used here is both simple and powerful enough to clearly demonstrate the typology. The exact choice of representation may need further consideration if one wants to perform inference on the representations or utilize them in information retrieval tools. For now, we are not concerned with how one might construct these representations or even the fidelity with which it is possible (though we revisit these questions in Sect. 6 ). The possibility that research projects could in theory be represented in the way described below is sufficient to formulate the typology.

While the form of representation and typology presented below may not be directly used in information retrieval tools, I contend that they may be useful in guiding the overall direction that research on such tools might take (e.g., what kinds of papers should a tool search for?). Moreover, the typology may provide some clarity to researchers going through the literature search process for a project. Constructing a graphical representation of one’s paper may be a useful exercise, and can possibly illuminate different searches that are needed to find related work. Seeing how the representation of one’s paper changes over time can also be a useful documentation of the research and literature search process. Beyond such potential practical uses of the typology, I believe it can simply be beneficial to understand the various ways in which one product of research may relate to another. If alongside the physical and social worlds, the world of research literature “also qualifies as an endless frontier” (Swanson, 1986 , p. 115), then our efforts to make sense of the former should be accompanied by efforts to make sense of the latter.

Related work on related work

Related work on related work exists in a number of different disciplines. Literature search is central to all research after all! Fittingly, the typology we develop combines research that exists in different, largely isolated, strands.

In the information sciences and medicine, work on “literature-based discovery” (LBD), dating back to Swanson ( 1986 ), is concerned with making new scientific discoveries by establishing novel connections between different pieces of literature. Swanson ( 1986 ) describes literature-based discovery as a form of scientific discovery that takes place in Karl Popper’s world 3—the “world of the products of the human mind” (Popper, 1978 , p. 144)—whereby search functions are likened to scientific theories and the “logic of undiscovered public knowledge” (p. 116) is analogous to the logic of scientific discovery. In doing so, Swanson ( 1986 ) made a contribution to the philosophy of science, though it seems to have not been recognized in the philosophy of science community. A number of different information-retrieval techniques have been proposed to aid in LBD (Smalheiser, 2017 ; Sebastian et al., 2017 ). Some authors have presented categorizations of different types of “undiscovered public knowledge” or different forms of LBD (Davies, 1989 ; Smalheiser, 2017 ). While these categorizations can be useful in aiding researchers who want to perform literature-based discovery, our typology has a somewhat broader scope in that not all related work necessarily results in LBD. LBD is one potential use case of literature search, and its various methods span across the relationships in the typology presented here, as discussed below.

More broadly, in information retrieval, the notion of “relevance” is central, and some researchers have tried to develop theories around what relevance is—typically conceived of as the relationship between an information need and a document (Saracevic, 1975 , 2016 ; Huang & Soergel, 2013 ). Green ( 1995 ) and Huang and Soergel ( 2013 ) pointed out that most discussions of relevance are around “topic matching,” but that this is only one form of relevance. Green and Bean ( 1995 ) then constructed a typology of different notions of relevance, and Huang ( 2009 ) expanded this to a typology consisting of over 200 notions of relevance. Huang ( 2009 ) considers three broad categories of relationships: (1) “What functional role a piece of information plays in the overall structure of a topic,” (2) “How information contributes to users’ reasoning about a topic,” and (3) “How information connects to a topic semantically” (p. 411). As examples of functional roles, an information source might present a solution to a problem, the cause of an effect, etc. As examples of contributing to reasoning, an information source might provide an analogy to the topic or might be used to deduce something about the topic. While this work is very relevant to the present paper, there are two key differences. First, their work is about the broader concept of relevance between information and needs, while this paper focuses on relevance in academic literature. One would expect that many of the kinds of broader relevance typologies would also hold for research publications, but given the particularities of literature search and the role it plays in the broader process of scientific research, it seems worth studying in its own right. Second, these prior typologies largely focus on the variety of semantic relationships between two topics, while the approach we present here views relevance in terms of operations that operate on knowledge representations of topics. In this sense, the typology I present here can express how to relate different research topics in terms of a small number of mathematically precise operations (that are hopefully easy to remember), rather than a plethora of different possible semantic relationships. The two approaches are complementary, but I contend that the approach taken here is more useful for conceptualizing the evolution of a research project over time.

In computer science and artificial intelligence, there has been a recent thread of work on citation recommendation, concerned with identifying relevant citations given a piece of text and possibly other meta-data (e.g., authors, etc.) (Strohman et al., 2007 ; Liang et al., 2011 ; Ren et al., 2014 ; Bhagavatula et al., 2018 ). Interestingly, this work has not really considered automated techniques for LBD, and it does not cite the vast literature on LBD or on relevance. Indeed, most of the work in this area is concerned with topic matching (finding citations that topically overlap). One notable exception is work by Chan et al. ( 2018 ) and Kang et al. ( 2022 ). Chan et al. ( 2018 ) presented a technique that combines crowdsourcing and machine learning to find analogies between different papers. They utilize a “soft” relational schema, a very coarse-grained representation of a research paper; they explicitly avoid using representations like the one described below, because they can be very difficult to construct for many publications. Kang et al. ( 2022 ) built on this work by training deep learning algorithms on the crowdsourced representations of abstracts to be able to automatically detect the “purpose” and “mechanism” of a paper. An analogy in this context is two papers that have a similar underlying purpose but achieve that purpose through a different mechanism. Kang et al. ( 2022 ) used this to prototype an analogical search engine for scientific literature. While their representation may be useful for LBD, I contend that it can only capture certain kinds of relationships between papers, and, as I discuss further below, some of their methods do not appear to actually look for analogies as per the typology we develop below. As such, our typology can potentially be useful in classifying the different kinds of relationships that various existing LBD and citation recommendation methods can uncover, and the kinds of relationships that they cannot.

A representation of a research project

In our representation, a research project or publication P ∈ Π is represented as a set of entities and relations, P = ( E , R ) . An entity conceptually represents any specific topic of relevance to the project, usually expressed as a noun or a noun phrase (e.g., DNA, the civil rights movement, high blood pressure, theorems). Notice that entities can come in different degrees of specificity (e.g., theorems vs. Gödel’s first incompleteness theorem); the important thing is that entities across all topics and publications are represented at the same level of granularity. We allow entities to be hierarchically defined as functions of other entities (e.g., the entity “volume of a cup” can be thought of as the “volume of” function applied to “cup”).

Relations define a relationship between some number of entities, such that the predicate R ( e 1 , e 2 , ⋯ , e n ) indicates that e 1 , e 2 , ..., e n are related as specified by the relation R . Binary relations are perhaps the most common. For example, in the sentence “stress causes high blood pressure”, “causes” is a relation that takes relates two entities (in this case, “stress” and “high blood pressure”). We might represent this as causes (stress, high blood pressure). As an example of a tertiary relation, consider the sentences “ribosomes translate mRNA into sequences of amino acids” and “Arab translators translated Greek texts into Arabic translations”; they could both be said to use the relation x-translates-y-into-z (though if we think the word “translates” has a very different semantic meaning in these two cases, we could suggest there are two different relations at play here). We also allow for unary relations; for example, “blood pressure is high” can be represented as is-high (blood pressure). Unary relations are called attributes in structure-mapping theory and they effectively allow assigning adjectives to entities; for example, high (blood pressure) would mean “high blood pressure.” With slight abuse of notation, I will use unary relations both as relations (e.g., is-high (blood pressure)) and as attributes (e.g., high (blood pressure)). Finally, we allow for higher-order relations, which take relations as input instead of, or in addition to, entities. causes is a higher-order relation because we can say, for example, causes ( provided ( treatment (subjects), New Curriculum), learn-more-than ( treatment (subjects), control (subjects))).

As with Gentner’s ( 1983 ) structure-mapping theory, the classification of relationships between research has more to do with the structure of the representation (i.e., the presence of certain entities and relations) rather than the semantic meaning of the nodes. However, semantics still play an important role in informing whether a particular relationship is sensible or important in a particular situation. That is, someone without a semantic understanding of a given domain can still apply the operators described below in the sense that one can execute 4 + 7 and 4 × 7 , without regard to which operation makes more sense in the given situation. Furthermore, one aspect of semantics is necessary in the application of some of the operators. Namely, there is a general relation, “is a” (or “is an instance of”), which can capture any situation where a particular entity can be categorized as a special case or instance of another entity. Consistent with earlier work on knowledge representation, we will refer to this relation as is-a (Brachman, 1983 ). For example, is-a (Gödel’s first incompleteness theorem, theorem) and is-a (the civil rights movement, historical occurrence). A single entity can be an instance of many entities (e.g., a cat can be considered an animal, a pet, and an Internet phenomenon). The is-a operator is also reflexive (e.g., is-a (cat, cat)). Finally, with slight abuse of notation, we will also have is-a be a higher-order relation that can designate when one relation is an instance of another. For example, is-a ( holds (person, ball), possesses (person, object)), because holding something is a special case of possessing it and a ball is an object. Some of the operators below can only be applied with an understanding of what things are instances of other things; however, when the relationship is more abstract, sometimes even a domain expert will not readily see these connections.

The set of entities and relations that are used in the representation of a research publication will likely not include all entities and relations included in that publication (e.g., all nouns and verbs), but rather they should include the concepts that are focal to that publication. Of course, that is somewhat subjective, but a useful heuristic is to include all entities and relations that are involved in a system of relationships that might be worth providing citations in reference to, as well as any new entities and relations that are being introduced in the paper. For example, in a paper that runs an experiment with seven conditions, the number of seven is probably not an entity that should be included, but in Miller’s ( 1956 ) paper on working memory capacity or a paper on the religious symbolism of the number seven, it likely should be included.

As suggested above, there is no single correct way to represent a research project. In fact, there can be multiple different views of a research project, which induce different representations. Each of these views can be more or less useful depending on how they are to be used. Moreover, even simple relations can be expressed in different ways. For our purposes, there is a relationship between two research projects if there is at least some view of each that permits the relationship. Since we are not concerned with the practical side of how to best represent projects here, we do not worry about how one would go about discovering the “right” views. In practice though, seeing two related papers from the “wrong” viewpoint is one reason why researchers and information retrieval tools might not notice an important relationship.

We can represent these representations graphically using a graph-like structure as shown in Fig.  1 a. Boxes indicate entities, and the text outside of boxes indicate relations. The arrows coming out of a relation point to its arguments in order from left to right. Nested boxes (e.g., “some part of a new thing”) show hierarchically defined entities. For simplicity, we show binary relations as labeled directed edges for asymmetric relations and labeled undirected edges for symmetric relations, as shown in Fig.  1 b.

Examples of how to graphically represent research projects/publications. a An example of a tertiary relation with three entities that would be read as “An old thing can become some part of a new thing through some process.” There are also two unary relations: is-old and is-new . b Two examples of binary relations. The causes relation is asymmetric while the correlated relation is symmetric

This representation could be couched in the language of model-theoretic philosophy of science (Suppes, 1957 , 1960 ), in particular using the partial structures formalism (French, 2000 ; Da Costa & French, 1990 ), which is also often expressed in terms of entities and (partial) relations. Doing so may be appealing since it would connect literature search to an existing framework for discussing scientific theories. The partial structures formalism has also been used in describing analogies and abstractions in science and provides a way to formalize research undergoing change. However, the ideas presented here not only apply to formal scientific theories, but also to non-scientific literature and more nascent representations of scientific topics, and I do not want to associate the typology presented here with a particular interpretation of scientific theories.

A typology of related works

We can now describe the different kinds of relationships that can exist between a research project and prior work. Suppose we have a research project P = ( E P , R P ) and a piece of literature L = ( E L , R L ) . We assume that P is an ongoing project that can potentially change, while L is already published literature and hence static. Below we describe a set of operations that can be used to describe the relationship between P and L . These operations are functions that take the representations of P and L as inputs and output a representation ρ of the relationship between P and L (as defined by the operation). Since we allow for composing these operations in sequence, some of the operations will actually take as input P , L , and our current representation of the relationship between the two ( ρ i ), and will output a modified representation of the relationship ( ρ i + 1 ). Moreover, when applying multiple operations in sequence, we may want to keep track of the ongoing relationship, which we can do by merging multiple relationships (i.e., taking the union of entities and the union of relations in the sequence of relationships). After each operation is applied, we can also potentially modify P 1 , thereby modifying the relationship between P and L as well. The series of operations and modifications reflects the iterative and influential nature of literature search in the research process. The operations, described below, are called intersection, interpretation, expansion, abstraction, reification, analogy, and substitution. Table  1 lists some basic information about the operations, which may be useful when reading the sections below. I do not make any claims that the typology presented here is complete. There might be other operations, or perhaps more useful categorizations of the operations presented here, which can be elucidated upon in future work. In what follows, I will describe each of the operators in words as well as mathematical formalism when needed; readers can safely skip the mathematical formalism and still grasp the key ideas.

A list of the operations in the typology along with their inputs and outputs

For the meaning of the terms used in the Output column, see the appropriate section

Intersection

The first and probably most prevalent operation is intersection , which outputs a subset of entities that are shared by P and L and a subset of relations shared by the two. Specifically, intersection outputs a representation ρ = ( E PL , R PL ) , where E PL ⊆ E P ∩ E L and R PL ⊆ R P ∩ R L . The exact subset depends on what is determined to be relevant between the two representations. A simple special case of this would be when P and L share just a single entity. For example, suppose P and L both have to do with DNA, but one is about DNA to solve computational problems (Adleman, 1994 ) and the other is about DNA vaccines for coronavirus (Callaway, 2020 ). It is unlikely that these publications have other entities in common. In many such cases, publications are not worth citing, and such an intersection would actually not be relevant. A relationship is worth noting when the degree of overlap is large enough; this can be measured by associating some degree of importance to each entity in P and taking the sum (or some non-linear function) of importances across all the entities in ρ .

In some cases, overlap in a single entity may be enough to warrant citation or even to alter the course of a research project. For example, one of the examples that Swanson ( 1986 ) gives for undiscovered public knowledge has to do with a potential research publication on the “all swans are white hypothesis,” a hypothesis that states that all swans are white. This hypothesis could be supported inductively if there was a lack of any documented evidence of black swans. As Swanson ( 1986 ) says:

Suppose for the sake of argument that scientists living in a remote part of the world were to publish, in a local wildlife journal, some observations about a family of black swans living on a nearby lake. We suppose further that the report comes from a half-dozen people who are reliable observers, and that they are unaware that other people in the world think that all swans are white. (p. 109)

As shown in Fig.  2 , the potential all-swans-are-white hypothesis publication ( P ) is represented using three entities and two relations, although it can be interpreted as two entities and the relationship between them (“the all-swans-are-white hypothesis is proved by the fact that there is no evidence of black swans”); on the other hand, the article in the wildlife journal ( L ) only concerns itself with black swans and possibly other topics of local interest. As such, the two articles overlap in only one entity: black swans. It just so happens that the existence of black swans is a critical refutation of the theory (i.e., “evidence of black swans” is a very important entity in P ), and so this single article can change the course of the research project (e.g., the authors publish a refutation of the all-swans-are-white hypothesis rather than a proclamation of it).

Swanson’s ( 1986 ) black swans example as an example of intersection

Notice that the intersection of the two articles was “black swans” not “evidence of black swans.” (The wildlife journal is not trying to present evidence of black swans; it is discussing a piece of wildlife whose existence they never called into question.) The intersection of “black swans” by itself is not necessarily meaningful. Another paper that discusses black swans but provides no evidence for them is of less value to P . How then can we capture the obvious fact that L presents evidence of black swans, even though it is not captured in its representation? The answer lies in the interpretation operation.

Interpretation

An interpretation takes an existing relationship between P and L and adds additional entities and/or relations from P (not included in L ) that can help interpret the current relationship. Namely, if ρ i = ( E ρ i , R ρ i ) is the output of a previous operation, then ρ i + 1 = ( E ρ i ∪ E PS , R ρ i ∪ R PS ) , where E PS ⊆ E P \ E L and R PS ⊆ R P \ R L . (I use PS and LS as subscripts to denote subsets of P and L .) A natural use of interpretation is to apply it after an intersection. For example, in the black swans example above, we can interpret the intersection of P and L as being “evidence of black swans.” Clearly, L does present evidence of black swans, but it was not interpreted that way until it was interpreted in light of P . Notice that if a researcher conducting project P were to construct the representation of L , they might do so according to their interpretation, whereby “evidence of black swans” would appear in L . Therefore, interpretation steps may often be implicit or hidden in the particular view of L that a researcher adopts. In this paper, I try to represent prior work in a way that is faithful to the original authors’ meaning, though we must recognize that views of prior work will always be informed by our worldview.

An expansion takes an existing relationship between P and L and adds additional entities and/or relations from L (not included in P ) to potentially expand the content of P or to bring new insights into the picture. Notice that structurally, the expansion operation is equivalent to the interpretation operation with P and L swapped; however, semantically, the two are often quite different. An expansion will often result in a change in P . As a result, it makes the most sense when P is an ongoing research topic (or a follow-up investigation to published work), rather than a final publication. Once P has changed to P ′ to incorporate the new entities and relations, what was once an expansion between P and L may be viewed as an intersection between P ′ and L . Therefore expansions play developmental roles in the research process, which are often not captured in publications. That is, many research projects may have changed course as a result of particular publications, but the final publication may only refer to the relationship to prior work at the time of publication, rather than the developmental influence of that prior work.

For example, in the related works section above, I acknowledged connections to Chan et al. ( 2018 ); these connections would be viewed as intersections (e.g., both papers have to do with academic literature, analogies, knowledge representation, etc.). However, what I did not state was that reading Chan et al. ( 2018 ) led me to read about structure-mapping theory (Gentner, 1983 ), and the two publications combined (and considered in relation to Swanson ( 1986 )) resulted in the beginnings of this paper. That is, before this paper was even conceived of, the aforementioned prior works resulted in a series of expansions, which turned into the present piece only after many iterations, which involved a series of other operations applied to various publications (some of which are cited, and some of which may not be). This reflects the role of literature search in the messy process that is research. I suspect that researchers rarely document the series of expansions (and other steps) that lead to the final state of a publication.

In fact, at times, some prior work may only play the role of a stepping stone to discovering other, more relevant, prior works. That is, an expansion of P by L 1 may result in an exploration of the new entities in the expansion, which results in discovering L 2 , which intersects with P . At that point, L 1 may no longer really be relevant; that is, the extent of L 1 ’s relevance may be better captured by L 2 .

One broad category of expansions falls under Swanson’s ( 1986 ) second example—“A Missing Link in the Logic of Discovery” or what is often referred to as the ABC model. As Swanson ( 1986 ) originally expressed it:

Suppose the following two reports are published separately and independently, the authors of each report being unaware of the other report: (i) a report that process A causes the result B, and (ii) a separate report that B causes the result C. It follows of course that A leads to, causes, or implies C. That is, the proposition that A causes C objectively exists, at least as a hypothesis. (p. 110)

Swanson gave a specific example of a discovery he made (the first of his several literature-based discoveries in medicine): connecting (a) literature on how fish oil causes a reduction of blood viscosity with (b) literature on how reducing blood viscosity leads to an improvement in symptoms of Raynaud’s syndrome. The intersection of these two literatures is the entity “reduction of blood viscosity.” An expansion adds the causal link to “relief from Raynaud’s syndrome” and that link is then interpreted in light of the connection to “dietary fish oil.” Connecting these two literatures via these steps can result in a change in P as shown in Fig.  3 . Notice that the addition of a new causal relation between dietary fish oil and relief from Raynaud’s syndrome was inferred from this expansion, but had never been experimentally shown or even published about. Two years later, a clinical trial independently confirmed this hypothesis (Swanson & Smalheiser, 1996 ).

Swanson’s ( 1986 ) example of the ABC model as an example of expansion

Literature-based discovery often involves this kind of linking between two “non-interactive literatures,” literatures that are rarely, if ever, cited in the same publications (Swanson & Smalheiser, 1996 ). However, expansion need not always be between two non-interactive literatures. Indeed, researchers may often be unaware of highly relevant work within their own research community (or other interactive literatures) that build upon the concepts they are investigating. Such cases can often be caught by the researchers themselves when conducting a more expansive literature review, or by reviewers during the peer review process, but likely often go undetected.

Abstraction

An abstraction applies if P contains a subset of entities and relations that are instances of entities and relations in L . In other words, we have an abstraction when L contains a more abstract or generalized representation of part of P . An abstraction can still consist of concrete entities and relations as long as they are more general or more abstract than the entities and relations in P (e.g., as suggested above is-a (cat, animal), is-a (cat, Internet phenomenon), and is-a (the civil rights movement, historical occurrence) can all be single entity abstractions).

Describing an abstraction mathematically requires a bit more care than for previous operations since abstractions must be semantically “consistent” across the entities and relations involved. Formally, an abstraction applies if there is a subset of entities and relations in P —say E PS ⊆ E P and R PS ⊆ R P —and a subset of entities and relations in L —say E LS ⊆ E L and R LS ⊆ R L —such that the following four conditions hold:

• For all e ∈ E PS , there exists a e ~ ∈ E LS such that e is an instance of e ~ .
• For all R ∈ R PS , there exists a R ~ ∈ R LS such that R is an instance of R ~ .
• For all R ∈ R PS , if R ( e 1 , e 2 , ⋯ , e n ) , then R ~ ( e ~ 1 , e ~ 2 , ⋯ , e ~ n ) , where R , e 1 , ⋯ , e n are instances of R ~ , e ~ , 1 ⋯ , e ~ n respectively.
• At least some e ≠ e ~ or some R ≠ R ~ .

The last condition is required to make sure the abstraction is not simply mapping identical representations (in which case it would just be an intersection). The resulting representation is ρ = ( E PS ∪ E LS , R PS ∪ R LS ∪ I S - A ) , where I S - A ( e , e ~ ) and I S - A ( R ( e 1 , e 2 , ⋯ , e n ) , R ~ ( e ~ 1 , e ~ 2 , ⋯ , e ~ n ) ) , for all e , e ~ , R , and R ~ as defined in the conditions above.

Abstractions need not be profound. Consider the black swans example again. The way I presented it above was actually a bit disingenuous: black swans are not the only evidence that disproves the all-swans-are-white hypothesis; any non-white swans would. Thus it might be more accurate to replace the “black swans” entity with “non-white swans” in Fig.  2 a. The relationship between P and L then first involves an abstraction (instead of an intersection)—namely is-a (black swans, non-white swans)—followed by an interpretation, as shown in Fig.  4 . This is a rather trivial kind of abstraction, which likely happens all the time when interpreting prior work in the context of current work.

The black swans example revisited. The relationship between P and L is now an interpretation of an abstraction of L . Notice that we used “are” instead of “is a” simply because the entities are expressed in plural

A more substantial form of abstraction is whenever P reports on empirical findings that can be subsumed into an existing theory described by L . For example, if researchers find that students in a collaborative problem-solving activity learned more than students who were working on the activity on their own, then they might see the ICAP hypothesis (Chi & Wylie, 2014 ), which posits that interactive learning is better than constructive learning, as an abstraction.

Finally, perhaps the most interesting (but also rarest) form of abstraction is when a body of research is interpreted or a problem is solved using some abstract formalism or framework that exists in the literature (often in a different field). For example, a notable example in the history of science is the introduction of group theory to quantum mechanics to solve certain problems related to symmetry (French, 2000 ; Scholz, 2006 ). According to French ( 2000 ):

the relationship between mathematics and physics is represented in terms of an embedding of a scientific theory into a mathematical structure. This effectively gives the theory access to ‘surplus’ mathematical structure which can play an essential role in the further development of theory. (p. 104)

This “surplus structure”—a term originally from Redhead ( 1975 )—is represented in our typology by expansion steps that can follow the abstraction. Namely, once a connection is made between L (say group theory) and P (a particular problem in physics), an expansion can be applied to bring new mathematical machinery from L to bear on P . Furthermore, an interpretation of the abstraction of L in light of P might result in new insights that could lead to further developments in L (if we do not consider L to be static literature). As French ( 2000 ) states, “it is important to acknowledge that both group theory and quantum mechanics were in a state of flux at the time they were brought into contact and both subsequently underwent further development” (p. 110).

Reification

A reification is the inverse of an abstraction. That is, a reification has the same definition of an abstraction, except that P and L are exchanged. We can say P is reified by L if L is abstracted by P . A reification can occur when prior work might contain a concrete example of a phenomenon, which one’s present work presents in more abstract or general terms. Reifications will often be used when interpreting prior empirical findings in light of a new theoretical framework. For example, when articulating his theory of the structure of scientific revolutions, Kuhn ( 2012 ) drew on myriad concrete historical examples from the history of physics, astronomy, chemistry, and other fields. These findings are reifications of particular components of Kuhn’s theory (e.g., paradigms, anomalies, paradigm shifts, etc.).

A reification can also make sense when one is in a formative stage of a project where some of the specifics have not yet been determined. For example, consider Tu Youyou’s work on finding a cure for malaria in the 1970s for which she won the Nobel Prize in 2015. The problem that Tu and her team were working on is represented in Fig.  5 a. According to Tu ( 2015 ):

After thoroughly reviewing the traditional Chinese medical literature and folk recipes and interviewing experienced Chinese medical practitioners, I collected over two thousand herbal, animal and mineral prescriptions within three months after initiation of the project.

One of the substances that showed some initial promise was sweet wormwood ( qinghao ), which was shown in the literature to cure intermittent fevers, as shown in Fig.  5 b. Therefore sweet wormwood is a reification of a potential cure for malaria, as shown in Fig.  5 c, and this can be interpreted in the broader research of finding a cure for malaria, as shown in Fig.  5 d. Yu went on to identify artemisinin as an actual cure for malaria, but there was an additional step of literature-based discovery needed first, which we will return to later.

The discovery of sweet wormwood as a cure for malaria as an example of reification

An analogy applies when P and L both have a subset of entities and relations that have a shared abstraction. More formally, using the same notation as above, an analogy applies if there exists some other representation A = ( E A , R A ) (representing an abstraction) and the following four conditions hold 2 :

• For all e ~ ∈ E A , there exists an e ∈ E PS and an e ′ ∈ E LS such that e and e ′ are both instances of e ~ .
• For all R ~ ∈ R A , there exists an R ∈ R PS and an R ′ ∈ R LS such that R and R ′ are both instances of R ~ .
• For all R ~ ∈ R A and for every pair R ∈ R PS and R ′ ∈ R LS such that R and R ′ are both instances of R ~ , if R ~ ( e ~ 1 , e ~ 2 , ⋯ , e ~ n ) then R ( e 1 , e 2 , ⋯ , e n ) and R ′ ( e 1 ′ , e 2 ′ , ⋯ , e n ′ ) , where e i and e i ′ are instances of e ~ i for all i and R and R ′ are instances of R ~ .
• At least some e ≠ e ′ or some R ≠ R ′ .

We say that A N A L O G O U S ( e , e ′ ) if and only if condition 1 holds for e and e ′ and similarly we say that A N A L O G O U S ( R ( e 1 , e 2 , ⋯ , e n ) , R ′ ( e 1 ′ , e 2 ′ , ⋯ , e n ′ ) ) if and only if the conditions 2 and 3 above hold for those entities and relations. The representation that results from an analogy operation is ρ = ( E PS ∪ E LS , R PS ∪ R LS ∪ A N A L O G O U S ) .

Analogies can span from shallow analogies between two instances of a similar phenomenon in the same field to deep analogies across scientific fields that share little apparent relation to one another on the surface. The further removed that P and L are from the abstraction A , the deeper the analogy becomes (and typically, the harder to notice). Concretely identifying the abstraction implicit in an analogy is not necessary, and in some cases, it can actually be difficult to do, but I suggest that doing so may be a useful exercise (and could lead to refining the analogy).

Like expansions, analogies can sometimes result in modifying P by looking at the research project in a whole new light. Like expansions, this also means the way in which an analogy might have helped develop P over time may not always be apparent from the final product. Even if a publication discusses an analogy, it may not always be clear if that analogy was instrumental in developing the idea in the first place or if it was an afterthought that the two ideas were related.

An example of an analogy where the impact of prior work on a research project is actually made explicit is the analogy between Thomas Kuhn’s historical philosophy of science and Jean Piaget’s psychological and epistemological theory of how a child develops knowledge. In The Structure of Scientific Revolutions , Kuhn ( 2012 ) gives us a brief sense of his indebtedness to Piaget:

A footnote encountered by chance led me to the experiments by which Jean Piaget has illuminated both the various worlds of the growing child and the process of transition from one to the next. (p. xi)

The extent of this has recently been clarified by historians examining Kuhn’s other works and archival materials (Galison, 2016 ; Burman, 2020 ). For example, Kuhn ( 1977 , as cited in Burman, 2020) states:

Almost twenty years ago I first discovered, very nearly at the same time, both the intellectual interest of the history of science and the psychological studies of Jean Piaget. Ever since that time the two have interacted closely in my mind and in my work. (p. 21)

So what was the nature of this close interaction? One can draw a clear analogy between the two. At risk of oversimplification, a representation of the analogy between Kuhn’s theory and Piaget’s is shown in Fig.  6 , adapted from a mapping given by MacIsaac ( 1991 ). This is not at all to say that this is the precise analogy that Kuhn drew which led to a refinement of his theory as presented in The Structure of Scientific Revolutions . However, he probably made similar mappings that changed over time as he developed his theory. Similar analogies can also be drawn from Kuhn’s theory to gestalt theory and Bruner and Postman’s ( 1949 ) psychological theory of how people perceive incongruities, both of which Kuhn ( 2012 ) explicitly builds off of. Interestingly enough, the Piagetian analogy, while very influential on the development of Kuhn’s theory, was not retained in the final representation of his book, while the analogies to gestalt theory and Bruner and Postman ( 1949 ) were explicitly an important part of his narrative. Note that the relations in P and L are identical in this case, but this need not be the case in general; in fact, they may only be identical because I constructed them that way, but perhaps if the representations were to be derived independently, the relations would be non-identical, but share a common abstraction.

The representation of the analogy between Kuhn’s The Structure of Scientific Revolutions and Piagetian theory. The analogous relations are shown as dotted lines without labels for ease of reading

To provide a more recent example of analogy, we can consider the relationship between the recent machine learning literature on fairness ( P ) in relation to older literature from the 1960s-1970s on fairness in educational and employment testing ( L ). As Hutchinson and Mitchell ( 2019 ) point out, the two literatures share much in common including many mathematical definitions of fairness. To formalize this, Hutchinson and Mitchell ( 2019 ) explicitly construct an analogy between the two literatures:

Test items (questions) are analogous to model features, and item responses analogous to specific activations of those features. Scoring a test is typically a simple linear model which produces a (possibly weighted) sum of the item scores....Because of this correspondence, much of the math is directly comparable; and many of the underlying ideas in earlier fairness work trivially map on to modern day ML fairness. “History doesn’t repeat itself, but it often rhymes”; and by hearing this rhyme, we hope to gain insight into the future of ML fairness. (p. 49)

Their last sentence suggests that the goal of pointing out the relationship between these two literatures are further steps of expansion and interpretation, or in other words, exploiting the “surplus of structure.” Indeed, the authors surface several definitions from test fairness that had not been proposed in machine learning (i.e., an expansion). Notice that in this case, the underlying abstraction may not be immediately obvious (e.g., what is the abstraction underlying both a test item and a feature?); in fact, in some cases, there may not be a simple word or phrase to describe the abstraction, but the fact that a clear analogy can be drawn indicates that there must be some more abstract underlying representation.

Finally, in my own research, I have found that there is an analogy between debates in education research and the bias-variance tradeoff in machine learning (Doroudi, 2020 ). Here an analogy was determined by directly formulating the abstraction (a generalized version of the bias-variance decomposition theorem). This abstraction has four components that any instance must specify: a target, an approximator, a random mechanism, and a source of randomness; once these components are specified, one can derive other phenomena (e.g., the meaning of bias, variance, etc.). This naturally sounds very abstract, but it is more concrete once instantiated in specific contexts. Table  2 gives an example of the analogy between these concepts in machine learning and debates around pedagogy. Once this analogy is drawn, it may be possible to expand techniques that are developed in machine learning to bear on educational debates (Doroudi, 2020 ). One benefit of making the abstraction concrete is that the same abstraction can be used to draw analogies to other fields as well.

A mapping of the bias-variance decomposition from machine learning ( L ) to its analog in the study of pedagogy ( P ), along with the abstraction that connects the two ( A )

Note that the last two rows are only examples of mechanisms that are often viewed as having high bias and low variance or high variance and low bias. See Doroudi ( 2020 ) for more details

Substitution

The analogy operator as described above can be applied in cases that do not semantically appear to be analogies. For example, consider two papers that use different methods to achieve the same outcome; many of the entities and relations may be the same across the two representations, but the entity (or entities) representing the methods would be different. Colloquially we would probably not say there is an analogy between the two approaches. For this reason, we make a distinction between substitutions and analogies. A substitution operates exactly in the same way as an analogy, but it should be applied when it is more sensible. The analogous relation can be replaced with the substitutes relation for semantic clarity. Therefore, unlike the other operators, the distinction between the analogy and substitution operators is semantic. However, there are typically clear structural differences between the two. In a substitution, typically only one or a few entities and relations will change, and the rest will be identical across P and L . Moreover, a substitution is similar to what Gentner ( 1983 ) terms a literal similarity. Namely, Gentner ( 1983 ) suggests that the difference between a literal similarity and an analogy is typically that a literal similarity will involve a greater number of identical attributes (or unary relations).

Consider the following four scenarios that loosely describe different papers:

• Convolutional neural networks are trained to classify histopathological images of breast tissue as benign or malignant (Spanhol et al., 2016 ).
• Support vector machines are trained to classify histopathological images of breast tissue as benign or malignant (Aswathy & Jagannath, 2021 ).
• Human crowdworkers are trained to classify histopathological images of breast tissue as benign or malignant (Eickhoff, 2014 ).
• Pigeons are trained to classify histopathological images of breast tissue as benign or malignant (Levenson et al., 2015 ).

In cases 1 and 2, it would be a stretch to say that there is an analogy between “convolutional neural networks” and “support vector machines,” which are both machine learning algorithms that can be applied to the same classification tasks. Thus, here is a clear case of substitution. However, with case 4, even though one could argue a pigeon is being substituted for a machine learning algorithm, the idea of training pigeons and the idea of training machine learning algorithms both have long histories and are often used for different purposes. Thus, it seems more natural to say pigeons are analogical to neural networks or support vector machines in these scenarios (with the underlying abstraction being a learning agent). Pigeons and support vector machines have a lot fewer attributes in common than convolutional neural networks and support vector machines. Unlike pigeons, the latter two are both algorithms implemented in computer code that were designed specifically for classification tasks. Pigeons, on the other hand, are animals, fly, eat, and make sounds. Some attributes of pigeons are actually important for the training process but not shared by any standard machine learning algorithms, such as their hunger. While we might say getting hungry is analogous to the “reward seeking” or “loss minimizing” property of machine learning algorithms, there is no literal hunger in those algorithms.

Case 3 is less clear-cut. While human crowdworkers are also significantly different from machine learning algorithms, crowdsourcing is often used for tasks where state-of-the-art machine learning is not good enough or a machine learning engineer might want to compare the performance of their algorithm against crowdworkers. On the other hand, human crowdworkers and pigeons share a lot of similar attributes that are lacking in machine learning algorithms. These ambiguities point out that ultimately the decision of whether an analogy or a substitution applies is in the eyes of the beholder. In other words, the degree of overlap in attributes depends on what attributes are most salient to the researcher. If a crowdworker is seen as an alternative to artificial intelligence and its humanity is not at the forefront, then perhaps a substitution would apply. On the other hand, researchers interested in using pigeons’ visual properties as a substitute for human labelers (Levenson et al., 2015 ) could also see a substitution between crowdworkers and pigeons.

As mentioned earlier, Kang et al.’s ( 2022 ) analogical search engine looks for papers that overlap in terms of purpose with a researchers’ study (as represented in the form of a search query). However, if the purpose is virtually identical, then replacing one mechanism for another may often be a substitution, not an analogy, as seen in cases 1 and 2 above. In some cases, such as using pigeons vs. neural networks to classify images, swapping mechanisms may result in an analogy. On the other hand, when the purpose is only similar (but not identical), there is no guarantee that the purpose-mechanism relationship will be analogical across different papers. 3 Thus, while Kang et al. ( 2022 ) find that their search engine is more likely to identify papers that trigger creative adaptations of the original idea (when compared to a standard keyword-based search engine), it is important to distinguish related work that might result in generating novel ideas and related work that actually has an analogical relationship with the present work.

Returning to Tu’s work on discovering a cure for malaria, she found that wormwood “showed some effects in inhibiting malaria parasites during initial screening, but the result was inconsistent and not reproducible.” Scouring over the relevant literature, she then identified a relevant sentence in Ge Hong’s fifth century A Handbook of Prescriptions for Emergencies : “A handful of Qinghao immersed in two liters of water, wring out the juice and drink it all” (Tu, 2015 ). Tu realized that while herbs are typically boiled, Ge’s recipe did not advocate for boiling it so perhaps the heat killed the active components in the wormwood. This led to a new method for extracting artemisinin from wormwood. To model this we would have to add entities to Fig.  5 that account for the method by which the drug is extracted. In that case, Ge’s method can be seen as a substitution for Tu’s original method. This substitution led to a drastic change in the research direction, eventually resulting in a cure for malaria.

Putting the pieces together

Now that we have seen the various operations that can relate two pieces of research to one another, it is worth discussing how these operations might be used in sequence over the scope of a research project. To do so, I provide a hypothetical example. As a disclaimer, the example is not from an area I have any expertise in; in fact, I encountered the relationships described below in the process of writing this paper (although not in the exact sequence described below). On the one hand, this suggests that the example may be oversimplified; on the other hand, perhaps it gives a somewhat authentic account of a non-expert navigating a new research field.

Suppose we are interested in conducting a literature review related to the question “how are memories stored in synapses?” This research question can be represented as “memories are stored in synapses through some mechanism” as shown at the top of Fig.  7 . Some of the steps described below are also represented in Fig.  7 ; in those cases, I will mention the number of the step in parentheses. Operator names are italicized below. If the reader wants to assess their understanding of the operators (or perhaps assess the degree to which there could be subjectivity in which operators apply), the reader can guess which operator applies for each step of the figure before reading the rest of this section.

Example of literature search as a sequence of operators applied to a research question on how memory is stored in synapses

When embarking on this literature search process, we are likely already aware of some answers to the question. For example, “some mechanism” could be reified by “synaptic plasticity” (Step 1). But synaptic plasticity is quite broad and could be reified further by several more specific forms of plasticity, such as “long-term potentiation” (Step 2) and “long-term depression.” Further literature search might reveal a plethora of other mechanisms such as “protein synthesis,” “epigenetic mechanisms,” or “the standard model of synaptic consolidation.” However, these mechanisms are not necessarily mutually exclusive, perhaps leading to a revision of the question formulation to “memories are stored in synapses through a combination of X, Y, ...” (or some more hierarchical representation). On the other hand, some proposed mechanisms may be competing, like “the standard model of synaptic consolidation” and “multiple trace theory” (i.e., one can be substituted for the other). Moreover, we might realize that the “memory” entity can also be reified into particular kinds of memory, like “episodic memory” or “semantic memory.”

Searching the literature further may reveal that there are recent suggestions that memory is not (only) stored in synapses, but could be stored in sub-cellular materials. This might result in a substitution of certain molecules (e.g., “RNA”) for synapse (Step 3 ′ ). Alternatively, to keep our options open we may apply an abstraction of “synapse,” such as “parts of the brain” (Step 3). “Parts of the brain” can then be reified with many different entities, like “RNA” (Step 4). But it can also be substituted for regions of the brain where memories are stored, like the hippocampus. This may subsequently lead to the realization that rather than just asking how memories are stored, we should also be asking where memories are stored, leading to an expansion of the initial representation.

So far we have primarily considered literature that directly bears on the initial question. But sometimes surprising related works can also be discovered through intersections . For example, once we have established that RNA may be involved in memory, a colleague who is a molecular biologist might point out that there is an intersection with the literature on RNA interference (Step 5). Indeed, Smalheiser et al. ( 2001 ) noticed connections between a series of controversial 1960s studies on RNA-mediated memory transfer and RNAi; Smalheiser was a pioneer of literature-based discovery. We might then posit a relation that was neither present in our initial representation nor in related work: RNAi is potentially involved in the memory storage mechanism (i.e., “some mechanism” in our representation). Although it took over a decade, Smalheiser eventually found evidence to suggest that RNAi could indeed be involved in memory transfer (Smalheiser, 2017 ).

Finally, upon contemplating the initial representation further, the researcher may recognize an analogy to “how is memory stored in computer hardware?” (Step 6) or “how is memory stored in artificial neural networks?” Studying the literature in either of these areas may lead to the addition of new hypothesized mechanisms through an interpretation in light of the analogies. Notice that while in some cases a researcher notices an analogy when examining related literature, in other cases a researcher might think of an analogy, and then search for related literature. The related literature could either be about the analog (e.g., how memory is encoded in artificial neural networks) or about the analogy itself (Langille & Gallistel, 2020 ,e.g., how do theories of memory storage in the human brain relate to theories of memory storage in computer science). In the latter case, we have an intersection applied to the entire analogy .

The typology in practice

In this section, we discuss some important considerations for how the representation and typology could be used in practice. In theory, an understanding of the various ways in which one piece of literature may relate to a research topic can inform directions in information retrieval and citation recommendation. Such systems could potentially represent papers in terms of entities and relations by using named entity recognition (Nadeau & Sekine, 2007 ) and relation extraction (Bach & Badaskar, 2007 ); they can also leverage a growing body of work on using knowledge graphs for information retrieval (Reinanda et al., 2020 ). The typology can then inform the kinds of relationships that such systems can explore and possibly recommend to users. However, we reiterate that there is no single way to represent a paper or single way of applying the operators to identify relationships to prior work. As noted above, the choice of what operators apply and hence which relationships to related works will be noticed depends on the view one takes of one’s work and related work. One way to potentially mitigate this challenge is by having users specify their current view of their work in terms of its representation, or perhaps by allowing them to simultaneously represent their work in multiple ways. Furthermore, recognizing that different researchers and papers will use slightly different terms to refer to identical or very similar entities and relations, search engines could try to treat semantically similar phrases as being identical or provide a pre-selected set of entities and relations that they recommend users use.

However, even if the representations of papers are completely aligned, the task of retrieving good analogies and abstractions may be computationally intractable in the worst case (Wareham et al., 2011 ). Indeed, in automated analogical search, simplifications are made to make finding potential analogies more tractable. For example, the MAC/FAC algorithm—which is rooted in structure-mapping theory—first finds several examples that have the most surface-level overlap in terms of relations and then identifies the analogy 4 that is structurally strongest (Forbus et al., 1995 ). In Kang et al.’s ( 2022 ) analogical search engine, they look for papers that have a similar purpose, where similarity is measured by neural network embeddings rather than looking for a formally analogical structure. Although such algorithms may not be perfect, they could still potentially surface candidate analogies that would be given to a researcher who would ultimately identify when an analogy operator is applicable and useful.

Given the ongoing challenges in automated search, perhaps the typology would be more useful as a conceptual tool for researchers. Huang and Soergel ( 2013 ) found that “teaching users about the different kinds of topical relevance relationships may open their minds and make them better searchers and users of information.” Similarly, perhaps the typology presented here could be used as a tool to familiarize researchers with the different ways in which their research may relate to prior work, and how to use search tools to find such works. As mentioned before, simply representing one’s paper as a network of entities and relations may be a useful exercise to help researchers realize new insights about their research; future experimental studies could confirm whether this is true. Moreover, in discussing the potential value of their analogical search engine, Kang et al. ( 2022 ) mention the importance of “how deeply the human users can reflect on the retrieved analogs...and recognize how different notions of relevance may exist for their own problem context, despite potential dissimilarity on the surface” (p. 125). They suggest that “one approach to explaining relevance might be to surface a small number of core common features between an analog and a problem query” (p. 126). The representation presented here provides a natural way of showing users the potential relevance of related work. For example, when one searches for literature (even using a traditional search engine), representations could be generated on demand for the resulting papers such that they maximally align with the user’s query (at least in terms of number of entities and relations, if not in terms of higher-order relationships). Moreover, if the user specifies multiple research projects, a search engine could potentially represent each paper in terms of the representation that best aligns with each project.

I have tried to make the case that literature search is a complex process that can influence and be influenced by research in a variety of ways. By describing research papers and projects in terms of concrete representations, we can formally articulate how different pieces of research might relate to one another. As discussed in the last section, this could have practical ramifications in terms of how search engines could better support the literature search process or how to design training for researchers to improve the way they approach literature search.

Beyond practical applications, the typology presented here could give us insight into the ways in which literature search might iteratively change the course of a research project as a sequence of operations. Although it goes beyond the scope of this paper, it might be worth briefly considering some of the ways in which a research project might be modified as a result of these operations. One form of modification is simply adding new entities and relations to P as a result of an expansion; we can view this as a natural extension of the expansion operator. Several other forms of modifications can fall under the category of logical inference (i.e., deduction , induction , and abduction ). For example, in the black swans example, evidence of black swans triggers a modus ponens argument that proves the “all-swans-are-white” hypothesis is false, thereby changing P . Similarly, in Swanson’s ABC model, we can discern the presence of a new relation through the transitivity of the causal relation. If the representations are well-specified, one can imagine creating an inference engine that can automatically detect such changes in P after coming into contact with related work.

However, literature search cannot be considered in isolation from the other aspects of scientific discovery. Another form of modification to P might be the result of an experimentation operation, whereby a deduced relation is tested. We saw this both in the case of medical research that confirmed the causal link deduced by Swanson, and Tu’s experimental confirmation that wormwood can cure malaria. Finally, there is the construction operation, whereby a new entity or relation is created. Construction can result from either literature search (e.g., where an interpretation of some finding results in the discovery of a new finding, or where the expansion of an analogy results in an analogous entity that was not previously conceived of) or from research itself (e.g., the discovery of a new molecule or a new experimental finding). A thorough understanding of the processes of inference, experimentation, and construction is beyond the scope of this paper, but they begin to give us a hint as to how literature search is an iterative process that interacts with other aspects of the research process.

As pointed out by Swanson ( 1986 ), world 3 is also a world where scientific discovery takes place, by interacting with world 1 (the physical world) and world 2 (the subjective world of mental states). Philosophy of science should try to understand how these worlds interact in the process of scientific discovery; this paper is a step in that direction.

Acknowledgements

This material is based upon work supported by the National Science Foundation under Grant No. (2033868). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.

Author Contributions

Sole authored publication

Provided by National Science Foundation, Award Number: 2033868

Availability of data and material:

Declarations.

1 This can be formalized using the partial structures formalism mentioned above (Da Costa & French, 1990 ).

2 Gentner ( 1983 ) did not explicitly define an analogy in terms of an abstraction, but I believe it is useful to recognize that there is always implicitly an abstraction present, and in many cases, it might be useful to reason about what that abstraction is. Gentner ( 1983 ) further differentiates between abstractions, analogies, and literal similarities. These are differentiated by how many attributes and relations are shared between the two and the degree of abstractness of the entities (i.e., in an abstraction, entities are more abstract). While this is sensible, we allow for abstractions that are more concrete, so long as the entities in one representation are still instances of the entities in the other.

3 For example, one participant’s research question was how to “Grow plants better by optimizing entry of nanoparticle fertilizers into the plant” (p. 14). One paper identified by analogical search was about identifying plants by applying image analysis techniques to their leaves. It is not clear what the similar purpose is in this case, but regardless, the paper does not obviously share an analogical relationship with the research question. While this paper inspired a novel idea that the researcher thought would be relevant to her project, the relationship is captured by an intersection (through the “plant” entity) and possibly the application of interpretation and expansion operators.

4 Technically it looks for matches in terms of literal similarity to mimic people’s tendencies to find literally similar matches, but the algorithm could be easily modified to search for analogies.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

• Adleman, L. M. (1994). Molecular computation of solutions to combinatorial problems. Science, 266 (5187), 1021–1024. 10.1126/science.7973651 [ PubMed ]
• Aswathy, M., & Jagannath, M. (2021). An SVM approach towards breast cancer classification from H &E-stained histopathology images based on integrated features. Medical & Biological Engineering & Computing, 59 (9), 1773–1783. 10.1007/s11517-021-02403-0 [ PubMed ]
• Bach, N., & Badaskar, S. (2007). A review of relation extraction. Literature Review for Language and Statistics II, 2 , 1–15. https://www.cs.cmu.edu/~nbach/papers/A-survey-on-Relation-Extraction.pdf
• Bhagavatula, C., Feldman, S., Power, R., & Ammar, W. (2018, June). Content-based citation recommendation. In M. Walker, H. Ji, & A. Stent (Eds.), Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) (pp. 238–251). ACL.
• Brachman RJ. What IS-A is and isn’t: An analysis of taxonomic links in semantic networks. Computer. 1983; 16 (10):30–36. doi: 10.1109/MC.1983.1654194. [ CrossRef ] [ Google Scholar ]
• Bruner JS, Postman L. On the perception of incongruity: A paradigm. Journal of Personality. 1949; 18 (2):206–223. doi: 10.1111/j.1467-6494.1949.tb01241.x. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Burman JT. On Kuhn’s case, and Piaget’s: A critical two-sited hauntology (or, on impact without reference) History of the Human Sciences. 2020; 33 (3–4):129–159. doi: 10.1177/0952695120911576. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Callaway, E. (2020). The race for coronavirus vaccines: A graphical guide. Nature, 576–577. 10.1038/d41586-020-01221-y [ PubMed ]
• Chan J, Chang JC, Hope T, Shahaf D, Kittur A. Solvent: A mixed initiative system for finding analogies between research papers. Proceedings of the ACM on Human-Computer Interaction. 2018; 2 :1–21. doi: 10.1145/3274300. [ CrossRef ] [ Google Scholar ]
• Chi MT, Wylie R. The ICAP framework: Linking cognitive engagement to active learning outcomes. Educational Psychologist. 2014; 49 (4):219–243. doi: 10.1080/00461520.2014.965823. [ CrossRef ] [ Google Scholar ]
• Da Costa NC, French S. The model-theoretic approach in the philosophy of science. Philosophy of Science. 1990; 57 (2):248–265. doi: 10.1086/289546. [ CrossRef ] [ Google Scholar ]
• Davies, R. (1989). The creation of new knowledge by information retrieval and classification. Journal of Documentation . 10.1108/eb026846
• Doroudi, S. (2020). The bias-variance tradeoff: How data science can inform educational debates. AERA Open, 6 (4), 2332858420977208. 10.1177/2332858420977208
• Eickhoff, C. (2014). Crowd-powered experts: Helping surgeons interpret breast cancer images. In F. Hopfgartner, G. Kazai, U. Kruschwitz, & M. Meder (Eds.), GamifIR ’14: Proceedings of the First International Workshop on Gamification for Information Retrieval (pp. 53–56). ACM. 10.1145/2594776.2594788 [ CrossRef ]
• Falkenhainer, B., Forbus, K. D., & Gentner, D. (1989). The structure-mapping engine: Algorithm and examples. Artificial Intelligence, 41 (1), 1–63. 10.1016/0004-3702(89)90077-5
• Forbus, K. D., Gentner, D., & Law, K. (1995). MAC/FAC: A model of similarity based retrieval. Cognitive Science, 19 (2), 141–205. 10.1207/s15516709cog1902_1 [ CrossRef ]
• French, S. (2000). The reasonable effectiveness of mathematics: Partial structures and the application of group theory to physics. Synthese, 125 (1), 103–120. 10.1023/a:1005246608001 [ CrossRef ]
• Galison, P. (2016). Practice all the way down. R.J. Richards & L. Daston (Eds.), Kuhn’s ‘Structure of Scientific Revolutions’ at fifty (pp. 42–70). University of Chicago Press
• Gentner, D. (1983). Structure-mapping: A theoretical framework for analogy. Cognitive Science, 7 (2), 155–170. 10.1207/s15516709cog0702_3 [ CrossRef ]
• Green, R. (1995). Topical relevance relationships. I. Why topic matching fails. Journal of the American Society for Information Science, 46 (9), 646–653. 10.1002/(SICI)1097-4571(199510)46:93C646::AID-ASI2%3E3.0.CO;2-1 [ CrossRef ]
• Green, R., & Bean, C. A. (1995). Topical relevance relationships. II. An exploratory study and preliminary typology. Journal of the American Society for Information Science, 46 (9), 654–662. 10.1002/(SICI)1097-4571(199510)46:9%3C654::AID-ASI33E3.0.CO;2-3 [ CrossRef ]
• Huang, X. (2009). Topical relevance, rhetoric, and argumentation: A cross- disciplinary inquiry into patterns of thinking and information structuring. [Doctoroal dissertation, University of Maryland]. Digital Repository at the University of Maryland. https://drum.lib.umd.edu/handle/1903/9577
• Huang, X., & Soergel, D. (2013). Relevance: An improved framework for explicating the notion. Journal of the American Society for Information Science and Technology, 64 (1), 18–35. 10.1002/asi.22811 [ CrossRef ]
• Hutchinson, B., & Mitchell, M. (2019). 50 years of test (un)fairness: Lessons for machine learning. In Proceedings of the conference on fairness, accountability, and transparency (pp. 49–58). ACM. 10.1145/3287560.3287600 [ CrossRef ]
• Kang, H. B., Qian, X., Hope, T., Shahaf, D., Chan, J., & Kittur, A. (2022). Augmenting scientific creativity with an analogical search engine. ACM Transactions on Computer-Human Interaction, 1 , 1–36. 10.1145/3530013 [ CrossRef ]
• Ke, Q., Ferrara, E., Radicchi, F., & Flammini, A. (2015). Defining and identifying sleeping beauties in science. Proceedings of the National Academy of Sciences, 112 (24), 7426–7431. 10.1073/pnas.1424329112 [ PMC free article ] [ PubMed ] [ CrossRef ]
• Kuhn, T. S. (1977). Concepts of cause in the development of physics. In T. S. Kuhn (Ed.), The essential tension (pp. 21–30). University of Chicago Press. (Original work published 1971. 10.7208/9780226217239-003 [ CrossRef ]
• Kuhn, T. S. (2012). The structure of scientific revolutions: 50th anniversary edition . University of Chicago Press. (Original work published 1962)
• Langille, J. J., & Gallistel, C. R. (2020). Locating the engram: Should we look for plastic synapses or information-storing molecules? Neurobiology of Learning and Memory, 169 , 107164. 10.1016/j.nlm.2020.107164 [ PubMed ] [ CrossRef ]
• Levenson, R. M., Krupinski, E. A., Navarro, V. M., & Wasserman, E. A. (2015). Pigeons (columba livia) as trainable observers of pathology and radiology breast cancer images. PLoS ONE, 10 (11), e0141357. 10.1371/journal.pone.0141357 [ PMC free article ] [ PubMed ] [ CrossRef ]
• Liang, Y., Li, Q., & Qian, T. (2011). Finding relevant papers based on citation relations. In International conference on web-age information management (pp. 403–414). 10.1371/journal.pone.0141357
• MacIsaac, D. (1991). The pedagogical implications of parallels between Kuhn’s philosophy of science and Piagets’ model of cognitive development. Retrieved from http://physicsed.buffalostate.edu/danowner/kuhnpiaget/KP1.html
• Merton, R. K. (1963). Resistance to the systematic study of multiple discoveries in science. European Journal of Sociology/Archives Européennes de Sociologie, 4(2) , 237–282. 10.1017/S0003975600000801
• Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63 (2), 81. 10.1037/h0043158 [ PubMed ]
• Nadeau, D., & Sekine, S. (2007). A survey of named entity recognition and classification. Lingvisticae Investigationes, 30 (1), 3–26. 10.1075/bct.19.03nad
• Pask, G. (1988). Learning strategies, teaching strategies, and conceptual or learning style. In R. R. Schmeck (Ed.), Learning strategies and learning styles (pp. 83–100). Springer.
• Pask G, Kallikourdis D, Scott BC. The representation of knowables. International Journal of Man-Machine Studies. 1975; 7 (1):15–134. doi: 10.1016/S0020-7373(75)80003-4. [ CrossRef ] [ Google Scholar ]
• Popper, K. (1978). Three worlds. [Lecture transcript]. The Tanner Lectures on Human Values. https://tannerlectures.utah.edu/_resources/documents/a-to-z/p/popper80.pdf
• Redhead, M. L. (1975). Symmetry in intertheory relations. Synthese, 32, 77–112. https://www.jstor.org/stable/20115097 .
• Reinanda, R., Meij, E., de Rijke, M., et al. (2020). Knowledge graphs: An information retrieval perspective (Vol. 14(4), pp. 289–444). Foundations and Trends in Information Retrieval. 10.1561/1500000063
• Ren, X., Liu, J., Yu, X., Khandelwal, U., Gu, Q., Wang, L., & Han, J. (2014). Cluscite: Effective citation recommendation by information networkbased clustering. In Proceedings of the 20th ACM SIGKDD international conference on knowledge discovery and data mining (pp. 821–830). 10.1145/2623330.2623630
• Sacks, O. (2002). Scotoma: Forgetting and neglect in science. Prematurity in scientific discovery (pp. 70–84). University of California Press. 10.1525/9780520927735-010
• Saracevic, T. (1975). Relevance: A review of and a framework for the thinking on the notion in information science. Journal of the American Society for Information Science, 26 (6), 321–343. 10.1002/asi.4630260604 [ CrossRef ]
• Saracevic, T. (2016). The notion of relevance in information science: Everybody knows what relevance is. But, what is it really ? 10.1007/978-3-031-02302-6 [ CrossRef ]
• Scholz, E. (2006). Introducing groups into quantum theory (1926–1930). Historia Mathematica, 33 (4), 440–490. 10.1016/j.hm.2005.11.007 [ CrossRef ]
• Sebastian, Y., Siew, E.-G., & Orimaye, S. O. (2017). Emerging approaches in literature-based discovery: Techniques and performance review. The Knowledge Engineering Review, 32 , e12. 10.1017/S0269888917000042 [ CrossRef ]
• Smalheiser, N. R. (2017). Rediscovering Don Swanson: The past, present and future of literature-based discovery. Journal of Data and Information Science, 2 (4), 43. . [ PMC free article ] [ PubMed ]
• Smalheiser, N. R., Manev, H., & Costa, E. (2001). RNAi and brain function: Was McConnell on the right track? Trends in Neurosciences, 24 (4), 216–218. 10.1016/s0166-2236(00)01739-2 [ PubMed ] [ CrossRef ]
• Sowa, J. F. (1976). Conceptual graphs for a data base interface. IBM Journal of Research and Development, 20 (4), 336–357. 10.1147/rd.204.0336 [ CrossRef ]
• Spanhol, F. A., Oliveira, L. S., Petitjean, C., & Heutte, L. (2016). Breast cancer histopathological image classification using convolutional neural networks. In 2016 International joint conference on neural networks (IJCNN) (pp. 2560–2567). 10.1109/IJCNN.2016.7727519 [ CrossRef ]
• Spivak, D. I., & Kent, R. E. (2012). Ologs: A categorical framework for knowledge representation. PLoS ONE, 7 (1), e24274. 10.1371/journal.pone.0024274 [ PMC free article ] [ PubMed ] [ CrossRef ]
• Strohman, T., Croft, W. B., & Jensen, D. (2007). Recommending citations for academic papers. In Proceedings of the 30th annual international ACM SIGIR conference on research and development in information retrieval (pp. 705–706). 10.1145/1277741.1277868 [ CrossRef ]
• Suppes, P. (1957). Introduction to logic . D. van Norstrand Co.
• Suppes, P. (1960). A comparison of the meaning and uses of models in mathematics and the empirical sciences. Synthese, 12 , 287–301. 10.1007/BF00485107
• Swanson, D. R. (1986). Undiscovered public knowledge. The Library Quarterly, 56 (2), 103–118 University of Chicago Press. https://www.jstor.org/stable/4307965
• Swanson, D. R., & Smalheiser, N. R. (1996). Undiscovered public knowledge: A ten-year update. In E. Simoudis, J. Han, & U. Fayyad (Eds.), Proceedings of the second international conference on knowledge discovery and data mining (pp. 295–298). https://www.aaai.org/Papers/KDD/1996/KDD96-051.pdf
• Tu, Y. (2015). Tu Youyou—Biographical. NobelPrize.org. Retrieved from https://www.nobelprize.org/prizes/medicine/2015/tu/lecture/
• Wareham, T., Evans, P., & van Rooij, I. (2011). What does (and doesn’t) make analogical problem solving easy? A complexity-theoretic perspective. The Journal of Problem Solving, 3 (2), 30–70. 10.7771/1932-6246.1091

Educational resources and simple solutions for your research journey

How to Write Review of Related Literature (RRL) in Research

A review of related literature (a.k.a RRL in research) is a comprehensive review of the existing literature pertaining to a specific topic or research question. An effective review provides the reader with an organized analysis and synthesis of the existing knowledge about a subject. With the increasing amount of new information being disseminated every day, conducting a review of related literature is becoming more difficult and the purpose of review of related literature is clearer than ever.  

All new knowledge is necessarily based on previously known information, and every new scientific study must be conducted and reported in the context of previous studies. This makes a review of related literature essential for research, and although it may be tedious work at times , most researchers will complete many such reviews of varying depths during their career. So, why exactly is a review of related literature important?    

Table of Contents

Why a review of related literature in research is important  

Before thinking how to do reviews of related literature , it is necessary to understand its importance. Although the purpose of a review of related literature varies depending on the discipline and how it will be used, its importance is never in question. Here are some ways in which a review can be crucial.  

• Identify gaps in the knowledge – This is the primary purpose of a review of related literature (often called RRL in research ). To create new knowledge, you must first determine what knowledge may be missing. This also helps to identify the scope of your study.  
• Avoid duplication of research efforts – Not only will a review of related literature indicate gaps in the existing research, but it will also lead you away from duplicating research that has already been done and thus save precious resources.  
• Provide an overview of disparate and interdisciplinary research areas – Researchers cannot possibly know everything related to their disciplines. Therefore, it is very helpful to have access to a review of related literature already written and published.  
• Highlight researcher’s familiarity with their topic 1  – A strong review of related literature in a study strengthens readers’ confidence in that study and that researcher.

Tips on how to write a review of related literature in research

Given that you will probably need to produce a number of these at some point, here are a few general tips on how to write an effective review of related literature 2 .

• Define your topic, audience, and purpose: You will be spending a lot of time with this review, so choose a topic that is interesting to you. While deciding what to write in a review of related literature , think about who you expect to read the review – researchers in your discipline, other scientists, the general public – and tailor the language to the audience. Also, think about the purpose of your review of related literature .  
• Conduct a comprehensive literature search: While writing your review of related literature , emphasize more recent works but don’t forget to include some older publications as well. Cast a wide net, as you may find some interesting and relevant literature in unexpected databases or library corners. Don’t forget to search for recent conference papers.
• Review the identified articles and take notes: It is a good idea to take notes in a way such that individual items in your notes can be moved around when you organize them. For example, index cards are great tools for this. Write each individual idea on a separate card along with the source. The cards can then be easily grouped and organized.  
• Determine how to organize your review: A review of related literature should not be merely a listing of descriptions. It should be organized by some criterion, such as chronologically or thematically.  
• Be critical and objective: Don’t just report the findings of other studies in your review of related literature . Challenge the methodology, find errors in the analysis, question the conclusions. Use what you find to improve your research. However, do not insert your opinions into the review of related literature. Remain objective and open-minded.  
• Structure your review logically: Guide the reader through the information. The structure will depend on the function of the review of related literature. Creating an outline prior to writing the RRL in research is a good way to ensure the presented information flows well.  

As you read more extensively in your discipline, you will notice that the review of related literature appears in various forms in different places. For example, when you read an article about an experimental study, you will typically see a literature review or a RRL in research , in the introduction that includes brief descriptions of similar studies. In longer research studies and dissertations, especially in the social sciences, the review of related literature will typically be a separate chapter and include more information on methodologies and theory building. In addition, stand-alone review articles will be published that are extremely useful to researchers.  

The review of relevant literature or often abbreviated as, RRL in research , is an important communication tool that can be used in many forms for many purposes. It is a tool that all researchers should befriend.  

• University of North Carolina at Chapel Hill Writing Center. Literature Reviews.  https://writingcenter.unc.edu/tips-and-tools/literature-reviews/  [Accessed September 8, 2022]
• Pautasso M. Ten simple rules for writing a literature review. PLoS Comput Biol. 2013, 9. doi: 10.1371/journal.pcbi.1003149.

Q:  Is research complete without a review of related literature?

A research project is usually considered incomplete without a proper review of related literature. The review of related literature is a crucial component of any research project as it provides context for the research question, identifies gaps in existing literature, and ensures novelty by avoiding duplication. It also helps inform research design and supports arguments, highlights the significance of a study, and demonstrates your knowledge an expertise.

Q: What is difference between RRL and RRS?

The key difference between an RRL and an RRS lies in their focus and scope. An RRL or review of related literature examines a broad range of literature, including theoretical frameworks, concepts, and empirical studies, to establish the context and significance of the research topic. On the other hand, an RRS or review of research studies specifically focuses on analyzing and summarizing previous research studies within a specific research domain to gain insights into methodologies, findings, and gaps in the existing body of knowledge. While there may be some overlap between the two, they serve distinct purposes and cover different aspects of the research process.

Q: Does review of related literature improve accuracy and validity of research?

Yes, a comprehensive review of related literature (RRL) plays a vital role in improving the accuracy and validity of research. It helps authors gain a deeper understanding and offers different perspectives on the research topic. RRL can help you identify research gaps, dictate the selection of appropriate research methodologies, enhance theoretical frameworks, avoid biases and errors, and even provide support for research design and interpretation. By building upon and critically engaging with existing related literature, researchers can ensure their work is rigorous, reliable, and contributes meaningfully to their field of study.

R Discovery is a literature search and research reading platform that accelerates your research discovery journey by keeping you updated on the latest, most relevant scholarly content. With 250M+ research articles sourced from trusted aggregators like CrossRef, Unpaywall, PubMed, PubMed Central, Open Alex and top publishing houses like Springer Nature, JAMA, IOP, Taylor & Francis, NEJM, BMJ, Karger, SAGE, Emerald Publishing and more, R Discovery puts a world of research at your fingertips.  

Try R Discovery Prime FREE for 1 week or upgrade at just US$72 a year to access premium features that let you listen to research on the go, read in your language, collaborate with peers, auto sync with reference managers, and much more. Choose a simpler, smarter way to find and read research – Download the app and start your free 7-day trial today !  

Related Posts

Research in Shorts: R Discovery’s New Feature Helps Academics Assess Relevant Papers in 2mins 

How Does R Discovery’s Interplatform Capability Enhance Research Accessibility 

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base

Methodology

• How to Write a Literature Review | Guide, Examples, & Templates

How to Write a Literature Review | Guide, Examples, & Templates

Published on January 2, 2023 by Shona McCombes . Revised on September 11, 2023.

What is a literature review? A literature review is a survey of scholarly sources on a specific topic. It provides an overview of current knowledge, allowing you to identify relevant theories, methods, and gaps in the existing research that you can later apply to your paper, thesis, or dissertation topic .

There are five key steps to writing a literature review:

• Search for relevant literature
• Evaluate sources
• Identify themes, debates, and gaps
• Outline the structure
• Write your literature review

A good literature review doesn’t just summarize sources—it analyzes, synthesizes , and critically evaluates to give a clear picture of the state of knowledge on the subject.

Instantly correct all language mistakes in your text

Upload your document to correct all your mistakes in minutes

Table of contents

What is the purpose of a literature review, examples of literature reviews, step 1 – search for relevant literature, step 2 – evaluate and select sources, step 3 – identify themes, debates, and gaps, step 4 – outline your literature review’s structure, step 5 – write your literature review, free lecture slides, other interesting articles, frequently asked questions, introduction.

• Quick Run-through
• Step 1 & 2

When you write a thesis , dissertation , or research paper , you will likely have to conduct a literature review to situate your research within existing knowledge. The literature review gives you a chance to:

• Demonstrate your familiarity with the topic and its scholarly context
• Develop a theoretical framework and methodology for your research
• Position your work in relation to other researchers and theorists
• Show how your research addresses a gap or contributes to a debate
• Evaluate the current state of research and demonstrate your knowledge of the scholarly debates around your topic.

Writing literature reviews is a particularly important skill if you want to apply for graduate school or pursue a career in research. We’ve written a step-by-step guide that you can follow below.

Receive feedback on language, structure, and formatting

Professional editors proofread and edit your paper by focusing on:

• Academic style
• Vague sentences
• Style consistency

See an example

Writing literature reviews can be quite challenging! A good starting point could be to look at some examples, depending on what kind of literature review you’d like to write.

• Example literature review #1: “Why Do People Migrate? A Review of the Theoretical Literature” ( Theoretical literature review about the development of economic migration theory from the 1950s to today.)
• Example literature review #2: “Literature review as a research methodology: An overview and guidelines” ( Methodological literature review about interdisciplinary knowledge acquisition and production.)
• Example literature review #3: “The Use of Technology in English Language Learning: A Literature Review” ( Thematic literature review about the effects of technology on language acquisition.)
• Example literature review #4: “Learners’ Listening Comprehension Difficulties in English Language Learning: A Literature Review” ( Chronological literature review about how the concept of listening skills has changed over time.)

You can also check out our templates with literature review examples and sample outlines at the links below.

Download Word doc Download Google doc

Before you begin searching for literature, you need a clearly defined topic .

If you are writing the literature review section of a dissertation or research paper, you will search for literature related to your research problem and questions .

Make a list of keywords

Start by creating a list of keywords related to your research question. Include each of the key concepts or variables you’re interested in, and list any synonyms and related terms. You can add to this list as you discover new keywords in the process of your literature search.

• Social media, Facebook, Instagram, Twitter, Snapchat, TikTok
• Body image, self-perception, self-esteem, mental health
• Generation Z, teenagers, adolescents, youth

Search for relevant sources

Use your keywords to begin searching for sources. Some useful databases to search for journals and articles include:

• Your university’s library catalogue
• Google Scholar
• Project Muse (humanities and social sciences)
• Medline (life sciences and biomedicine)
• EconLit (economics)
• Inspec (physics, engineering and computer science)

You can also use boolean operators to help narrow down your search.

Make sure to read the abstract to find out whether an article is relevant to your question. When you find a useful book or article, you can check the bibliography to find other relevant sources.

You likely won’t be able to read absolutely everything that has been written on your topic, so it will be necessary to evaluate which sources are most relevant to your research question.

For each publication, ask yourself:

• What question or problem is the author addressing?
• What are the key concepts and how are they defined?
• What are the key theories, models, and methods?
• Does the research use established frameworks or take an innovative approach?
• What are the results and conclusions of the study?
• How does the publication relate to other literature in the field? Does it confirm, add to, or challenge established knowledge?
• What are the strengths and weaknesses of the research?

Make sure the sources you use are credible , and make sure you read any landmark studies and major theories in your field of research.

You can use our template to summarize and evaluate sources you’re thinking about using. Click on either button below to download.

Take notes and cite your sources

As you read, you should also begin the writing process. Take notes that you can later incorporate into the text of your literature review.

It is important to keep track of your sources with citations to avoid plagiarism . It can be helpful to make an annotated bibliography , where you compile full citation information and write a paragraph of summary and analysis for each source. This helps you remember what you read and saves time later in the process.

Don't submit your assignments before you do this

The academic proofreading tool has been trained on 1000s of academic texts. Making it the most accurate and reliable proofreading tool for students. Free citation check included.

Try for free

To begin organizing your literature review’s argument and structure, be sure you understand the connections and relationships between the sources you’ve read. Based on your reading and notes, you can look for:

• Trends and patterns (in theory, method or results): do certain approaches become more or less popular over time?
• Themes: what questions or concepts recur across the literature?
• Debates, conflicts and contradictions: where do sources disagree?
• Pivotal publications: are there any influential theories or studies that changed the direction of the field?
• Gaps: what is missing from the literature? Are there weaknesses that need to be addressed?

This step will help you work out the structure of your literature review and (if applicable) show how your own research will contribute to existing knowledge.

• Most research has focused on young women.
• There is an increasing interest in the visual aspects of social media.
• But there is still a lack of robust research on highly visual platforms like Instagram and Snapchat—this is a gap that you could address in your own research.

There are various approaches to organizing the body of a literature review. Depending on the length of your literature review, you can combine several of these strategies (for example, your overall structure might be thematic, but each theme is discussed chronologically).

Chronological

The simplest approach is to trace the development of the topic over time. However, if you choose this strategy, be careful to avoid simply listing and summarizing sources in order.

Try to analyze patterns, turning points and key debates that have shaped the direction of the field. Give your interpretation of how and why certain developments occurred.

If you have found some recurring central themes, you can organize your literature review into subsections that address different aspects of the topic.

For example, if you are reviewing literature about inequalities in migrant health outcomes, key themes might include healthcare policy, language barriers, cultural attitudes, legal status, and economic access.

Methodological

If you draw your sources from different disciplines or fields that use a variety of research methods , you might want to compare the results and conclusions that emerge from different approaches. For example:

• Look at what results have emerged in qualitative versus quantitative research
• Discuss how the topic has been approached by empirical versus theoretical scholarship
• Divide the literature into sociological, historical, and cultural sources

Theoretical

A literature review is often the foundation for a theoretical framework . You can use it to discuss various theories, models, and definitions of key concepts.

You might argue for the relevance of a specific theoretical approach, or combine various theoretical concepts to create a framework for your research.

Like any other academic text , your literature review should have an introduction , a main body, and a conclusion . What you include in each depends on the objective of your literature review.

The introduction should clearly establish the focus and purpose of the literature review.

Depending on the length of your literature review, you might want to divide the body into subsections. You can use a subheading for each theme, time period, or methodological approach.

As you write, you can follow these tips:

• Summarize and synthesize: give an overview of the main points of each source and combine them into a coherent whole
• Analyze and interpret: don’t just paraphrase other researchers — add your own interpretations where possible, discussing the significance of findings in relation to the literature as a whole
• Critically evaluate: mention the strengths and weaknesses of your sources
• Write in well-structured paragraphs: use transition words and topic sentences to draw connections, comparisons and contrasts

In the conclusion, you should summarize the key findings you have taken from the literature and emphasize their significance.

When you’ve finished writing and revising your literature review, don’t forget to proofread thoroughly before submitting. Not a language expert? Check out Scribbr’s professional proofreading services !

This article has been adapted into lecture slides that you can use to teach your students about writing a literature review.

Scribbr slides are free to use, customize, and distribute for educational purposes.

Open Google Slides Download PowerPoint

If you want to know more about the research process , methodology , research bias , or statistics , make sure to check out some of our other articles with explanations and examples.

• Sampling methods
• Simple random sampling
• Stratified sampling
• Cluster sampling
• Likert scales
• Reproducibility

 Statistics

• Null hypothesis
• Statistical power
• Probability distribution
• Effect size
• Poisson distribution

Research bias

• Optimism bias
• Cognitive bias
• Implicit bias
• Hawthorne effect
• Anchoring bias
• Explicit bias

A literature review is a survey of scholarly sources (such as books, journal articles, and theses) related to a specific topic or research question .

It is often written as part of a thesis, dissertation , or research paper , in order to situate your work in relation to existing knowledge.

There are several reasons to conduct a literature review at the beginning of a research project:

• To familiarize yourself with the current state of knowledge on your topic
• To ensure that you’re not just repeating what others have already done
• To identify gaps in knowledge and unresolved problems that your research can address
• To develop your theoretical framework and methodology
• To provide an overview of the key findings and debates on the topic

Writing the literature review shows your reader how your work relates to existing research and what new insights it will contribute.

The literature review usually comes near the beginning of your thesis or dissertation . After the introduction , it grounds your research in a scholarly field and leads directly to your theoretical framework or methodology .

A literature review is a survey of credible sources on a topic, often used in dissertations , theses, and research papers . Literature reviews give an overview of knowledge on a subject, helping you identify relevant theories and methods, as well as gaps in existing research. Literature reviews are set up similarly to other  academic texts , with an introduction , a main body, and a conclusion .

An  annotated bibliography is a list of  source references that has a short description (called an annotation ) for each of the sources. It is often assigned as part of the research process for a  paper .  

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

McCombes, S. (2023, September 11). How to Write a Literature Review | Guide, Examples, & Templates. Scribbr. Retrieved September 3, 2024, from https://www.scribbr.com/dissertation/literature-review/

Is this article helpful?

Shona McCombes

Other students also liked, what is a theoretical framework | guide to organizing, what is a research methodology | steps & tips, how to write a research proposal | examples & templates, "i thought ai proofreading was useless but..".

I've been using Scribbr for years now and I know it's a service that won't disappoint. It does a good job spotting mistakes”

The Software Engineering PhD Guide

How to write a related work ¶.

Standing on the shoulders of giants is a popular phrase amongst academics. This phrase captures the idea that no research stands in isolation and that all work builds upon existing knowledge. The related work section is where you position your research in the context of recent work and demonstrate scholarship. In this article I will describe how to write a related work section with a template for organising notes on papers.

What is related work? ¶

Fig. 4 Overview of how the related work section is grouped and critiqued. ¶

Related work is research that closely relates to your own and depends on the framing of the overall research outlined in the introduction. The outline of how the related work section is assembled is shown in Fig. 4 .

How to identify your related work? ¶

Pick a venue for your research ¶.

The choice of conference or journal influences the related work that reviewers expect to see in your paper. It’s a good idea to check publications from the last 3 years from the selected venue to confirm you have not missed key papers.

Establish a position for your paper ¶

Start by identifying the position of your paper and how you plan to pitch the research findings. Your supervisors and mentors will help with positioning the research. A position describes the way you plan to present your research paper and includes a motivation.

Identify research areas ¶

Identifying research areas by answering the questions: what have others done to solve related problems? How have others used the techniques I am using? Select the three research areas that are closest to your research.

Scan the literature and come up with themes ¶

Look for research in the three research areas that has been published from top ranked venues in the last 3 years. Group the related work into sub-themes.

Writing the related work ¶

Start by outlining the key research areas related to your research and then summarise the research area by listing out the sub-themes with multiple citations for each sub-theme where possible. Then provide a critique of the related work by a) highlighting a gap, b) showing how your research is different, c) describe how you improve existing techniques, or d) showing how your work is complimentary. A common mistake when writing related work is to list out each relevant paper and talk about the individual findings. This should only be done for a few papers that are very similar to yours to provide a deeper critique. Another common problem in the related work section is to describe what the research papers do and provide no analysis in the context of your work. Assume that the reviewers are familiar with the related work and want to know how your work relates to the existing research.

A template for organising your related work ¶

A goal of related work is to demonstrate scholarship; that you understand the field and can contextualise your research. Also demonstrate that you are an active member of the community by building upon prior knowledge and addressing known gaps. In your own words, fill out each column of this template to help organise your related work. The next step is to start to group papers around core themes that emerge.

How to write a methodology?

How to write a thesis?

Francesco Lelli

Related work/literature review/survey paper: a collection of resources.

A scientific literature review (sometimes also called related work or survey paper) is an integral part of:

• Writing scientific papers
• Writing position reports in a non-academic job
• Writing your Bachelor/Master/PhD thesis

Here, you will find a collection of resources that should help you in addressing your scholarly needs.

Not All Publications Are Equal

Yes, quality matters. I am talking about both (i) the quality of a venue/journal and (ii) the quality of a paper published in the particular venue/journal.

In talking about a venue you want to consider impact factor , self-citation ratio and indexing of the venue as some key heuristics for understanding the “prestige” of the venue. If you have trouble in understanding the meaning of these terms, I described these aspects extensively in one of my recent articles that talks about understanding scientific venues . I also presented what white papers are and how you should consider them in your research.

You have to be aware of the quality of a publication per se and independently from where it has been published. It is particularly important for saving time as well as for being able to read works that can actually help you in solving your problem instead of making it more complicated. Over time, every scientist develops his/her heuristics and in this article I described mine . In a nutshell, it is about looking at the citations of the article, its abstract, the venue, and the authors.

Much more can be said about the topic. This is an extensive lecture series from the University of Washington. If you are curious, you can learn some of the dynamics of scientific publishing. The title “Calling Bullshit in the Age of Big Data” gives you a good idea of the content.

Related Work/Literature Review and Active Reading

Do not limit yourself to only passive reading of scientific papers: instead, follow an active reading approach. In particular, you should take into account that scientific articles follow an   IMRaD structure. It stands for   I ntroduction,  M ethods,  Re sults and  D iscussion . In a recent article, I discussed how to take advantage of that structure for reading scientific papers quickly and effectively .

It is also important to always keep in mind the reason why you are doing a related work/literature review and act accordingly. Maybe you are trying to understand a problem or are trying to find the proper methodology for solving a clear problem. Your reading approach should be finetuned for the particular goal, and in this article you can find suggestions for taking advantage of a literature review for your research. Starting from asking yourself “why should I read this paper?” .

Leverage Proper Tools for Organizing Your Work

The more you will read, in particular if you will practice active reading, the more you will need to effectively organize your work. You should start organizing your work early on, when you have not yet accumulated an unmanageable amount of scientific resources. Otherwise, the inertia will cost you an unbearable amount of time.

There are dedicated tools for this task. In this article, I describe how you could organize your references using specific features of Microsoft words . There are several other tools like Mendeley and Citeulike that could help you in reducing the complexity of managing a large amount of resources.

Other Practical Aspects for a Literature Review:

In this video, Javed Vasillis presents a practical approach for conducting a literature review with a focus on the HCI domain.

However, many of the suggestions are valid for every domain of research. In particular, how to use the keywords of scientific papers as well as scientific research engines.

In this video, Shady Attia presents his view on how to conduct a literature review.

In addition, in case you are doing a literature review for non-scientific purposes or for the purpose of conducting an assignment, you may want to watch this video. You will find a quick and effective approach for this task.

However, in case you are not writing a company report or a white paper I would encourage you to take a more formal approach as described in this article and in the other videos. If your goal is to produce a (relatively) quick deliverable for an assignment you may want to consider it.

Reading a good literature review (or related work or survey paper, call it in the way you prefer) can help you in understanding a problem and in providing you with clear ideas on how to solve a particular challenge. Writing an outstanding literature review can help you in positioning yourself as an expert in a field. They key is leveraging the structure of scientific papers, using an active reading approach, as well as using tools that can help you manage the increasing complexity.

This article (Related Work/Literature Review/Survey Paper: a collection of resources) is part of the miniseries on  how to do a good thesis , you can see the full list of posts at the following links:

How to Do a Good Thesis: the Miniseries

Share this:

• Share on Tumblr
• ← European Factory Platform (EFPF): €2.5 million for Industry 4.0 Initiatives
• Can You Use the Internet of Things to Fight Addiction? →

Leave a Reply Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed .

Privacy Overview

Terms and Conditions - Privacy Policy

Instantly share code, notes, and snippets.

ikbelkirasan / writing-a-state-of-the-art-section.md

• Download ZIP
• Star ( 10 ) 10 You must be signed in to star a gist
• Fork ( 1 ) 1 You must be signed in to fork a gist
• Embed Embed this gist in your website.
• Share Copy sharable link for this gist.
• Clone via HTTPS Clone using the web URL.
• Learn more about clone URLs
• Save ikbelkirasan/848f97c4a1aee1fa6277ced7b5be80af to your computer and use it in GitHub Desktop.

Writing the "Related Work" Section of a Paper/thesis

By: Chamin Morikawa ( https://www.linkedin.com/pulse/writing-related-work-section-paperthesis-chamin-morikawa/ )

For most students, writing about what they did on their own is not hard. But writing about others' work - which is what you have to do in the "State of the Art" or "Related Work" section - is quite hard for them. Here are a few guidelines to make this task a bit easier.

Let's lay down our assumptions before continuing. I assume that you want to write a "Related Work" section for a research paper or a thesis that describes your approach to solve some problem. Let's also assume that there are other publications that attempt to solve the same problem, but the solutions in them are not perfect. Finally, let's assume that your approach has some difference when compared to those by others, and some improvement (faster, more accurate, easier to afford, etc.).

The question is, how do you come up with a good Related Work section for this publication?

The Reasons

Let's start by looking at the reasons for having this section in a paper or a thesis. While most of you already know them, a reminder can help us to compose it properly.

The primary reason for detailing the state of the art is to highlight that somebody else had not already tried what you did, when you started your research. In order to do this convincingly, you will have to have done a good survey of related research, make a good summary of them if there is a lot of work, and identify the need for improvement. Doing this allows you to demonstrate the motivation for your approach to solve the given problem, and also point out the difference between your approach and the others.

There are a few secondary reasons for writing this section. If you are writing a Master's or PhD thesis, this section serves as evidence of your research skills. Including a good description of the state of the art in a research paper will allow readers who are not very familiar with your topic to learn more about it (if you want your paper to be recommended by professors to their students, this will definitely help). A third reason, one that many researchers won't mention directly, is to have a place for "citations". Citations in other publications is the most important metric for assessing the value of a research publication. A description on others' research can help sustain this metric, and also give an opportunity to make favors (not that I recommend it, but many researchers are guilty of mutual citations and citation loops that boost their records).

One thing to keep in mind when writing the "Related work" section is that it should be shaped "like a funnel". To be more specific, the content should be broad at the start, and focused at the end.

Start with a very brief introduction of the basic research area that your work belongs to. For example, if your paper is about automatic age estimation using digital photos of faces, you can start by mentioning Automated Face Image Analysis as the basic area. You don't have to get down to Computer Vision; Automated Face Image Analysis is already a large research area. Selecting a couple of survey papers to show the advances of this area should be sufficient.

Now it is time to mention other papers that try to solve the same problem as the one that your paper does. You can organize them by idea, into paragraphs, so that the reader won't feel lost among a mix of research works. It is fine to mention the accuracies, and it is essential to mention the state-of-the art if there is a clear evaluation metric to identify it.

Now you are coming to the end of this section. If you are writing a thesis, or a survey paper, this is a good place to summarize the approaches with their performances, advantages and disadvantages, on a table.Otherwise, you can write a paragraph that summarize same content. In either case, the ending paragraph should be a pointer to you work; you point out the limitations in the existing approaches and then state that you are going to try approach X that is different from what has been tried before.

If you are writing a short paper or a demo, you can actually avoid having a specific section detailing related research. The primary reason for this is that the page count for such papers is smaller. In such a case, you can extend the introduction of the paper mentioning work that is closest to your approach. If your solution to the given research problem is very different from others, you can keep the related work section short; but some reviewers might be unhappy with this.

Concluding remarks

I gave you some guidelines on writing a "Related Work" section, based on my experience as a student, researcher, reviewer, and a teacher. I hope you find them useful. Finally, an unofficial guideline; know the snakes in your jungle! I know of a few reviewers who used to reject papers that did not cite their work, even though their work was so old that they could be cited only in the first couple of paragraphs of the related work section. Your advisor usually knows the culprits for your field.

This week: the arXiv Accessibility Forum

Help | Advanced Search

Computer Science > Human-Computer Interaction

Title: relatedly: scaffolding literature reviews with existing related work sections.

Abstract: Scholars who want to research a scientific topic must take time to read, extract meaning, and identify connections across many papers. As scientific literature grows, this becomes increasingly challenging. Meanwhile, authors summarize prior research in papers' related work sections, though this is scoped to support a single paper. A formative study found that while reading multiple related work paragraphs helps overview a topic, it is hard to navigate overlapping and diverging references and research foci. In this work, we design a system, Relatedly, that scaffolds exploring and reading multiple related work paragraphs on a topic, with features including dynamic re-ranking and highlighting to spotlight unexplored dissimilar information, auto-generated descriptive paragraph headings, and low-lighting of redundant information. From a within-subjects user study (n=15), we found that scholars generate more coherent, insightful, and comprehensive topic outlines using Relatedly compared to a baseline paper list.

Submission history

Access paper:.

• Other Formats

References & Citations

• Google Scholar
• Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

• Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

What is a related work? A typology of relationships in research literature

Affiliation.

• 1 School of Education, University of California, Irvine, 401 E. Peltason Drive, Suite 3200, Irvine, CA 92617 USA.
• PMID: 36643731
• PMCID: PMC9829224
• DOI: 10.1007/s11229-022-03976-5

An important part of research is situating one's work in a body of existing literature, thereby connecting to existing ideas. Despite this, the various kinds of relationships that might exist among academic literature do not appear to have been formally studied. Here I present a graphical representation of academic work in terms of entities and relations, drawing on structure-mapping theory (used in the study of analogies). I then use this representation to present a typology of operations that could relate two pieces of academic work. I illustrate the various types of relationships with examples from medicine, physics, psychology, history and philosophy of science, machine learning, education, and neuroscience. The resulting typology not only gives insights into the relationships that might exist between static publications, but also the rich process whereby an ongoing research project evolves through interactions with the research literature.

Keywords: Abstraction; Analogies; Information retrieval; Literature search; Literature-based discovery; Relevance; Structure-mapping theory.

© The Author(s) 2022.

PubMed Disclaimer

Conflict of interest statement

Conflicts of interest/Competing interestsN/A

Examples of how to graphically…

Examples of how to graphically represent research projects/publications. a An example of a…

Swanson’s (1986) black swans example…

Swanson’s (1986) black swans example as an example of intersection

Swanson’s (1986) example of the…

Swanson’s (1986) example of the ABC model as an example of expansion

The black swans example revisited.…

The black swans example revisited. The relationship between P and L is now…

The discovery of sweet wormwood…

The discovery of sweet wormwood as a cure for malaria as an example…

The representation of the analogy…

The representation of the analogy between Kuhn’s The Structure of Scientific Revolutions and…

Example of literature search as…

Example of literature search as a sequence of operators applied to a research…

Similar articles

• Macromolecular crowding: chemistry and physics meet biology (Ascona, Switzerland, 10-14 June 2012). Foffi G, Pastore A, Piazza F, Temussi PA. Foffi G, et al. Phys Biol. 2013 Aug;10(4):040301. doi: 10.1088/1478-3975/10/4/040301. Epub 2013 Aug 2. Phys Biol. 2013. PMID: 23912807
• Toward a Psychology of Social Change: A Typology of Social Change. de la Sablonnière R. de la Sablonnière R. Front Psychol. 2017 Mar 28;8:397. doi: 10.3389/fpsyg.2017.00397. eCollection 2017. Front Psychol. 2017. PMID: 28400739 Free PMC article.
• Before theory comes theorizing or how to make social science more interesting. Swedberg R. Swedberg R. Br J Sociol. 2016 Mar;67(1):5-22. doi: 10.1111/1468-4446.12184. Epub 2016 Feb 22. Br J Sociol. 2016. PMID: 26901757
• Exploring types of individual unlearning by local health-care managers: an original empirical approach. Coombs CR, Hislop D, Holland J, Bosley SLC, Manful E. Coombs CR, et al. Southampton (UK): NIHR Journals Library; 2013 Jun. Southampton (UK): NIHR Journals Library; 2013 Jun. PMID: 25642528 Free Books & Documents. Review.
• Analogies in medicine: valuable for learning, reasoning, remembering and naming. Pena GP, Andrade-Filho Jde S. Pena GP, et al. Adv Health Sci Educ Theory Pract. 2010 Oct;15(4):609-19. doi: 10.1007/s10459-008-9126-2. Epub 2008 Jun 5. Adv Health Sci Educ Theory Pract. 2010. PMID: 18528776 Review.
• Adleman, L. M. (1994). Molecular computation of solutions to combinatorial problems. Science,266(5187), 1021–1024. 10.1126/science.7973651 - PubMed
• Aswathy, M., & Jagannath, M. (2021). An SVM approach towards breast cancer classification from H &E-stained histopathology images based on integrated features. Medical & Biological Engineering & Computing,59(9), 1773–1783. 10.1007/s11517-021-02403-0 - PubMed
• Bach, N., & Badaskar, S. (2007). A review of relation extraction. Literature Review for Language and Statistics II,2, 1–15. https://www.cs.cmu.edu/~nbach/papers/A-survey-on-Relation-Extraction.pdf
• Bhagavatula, C., Feldman, S., Power, R., & Ammar, W. (2018, June). Content-based citation recommendation. In M. Walker, H. Ji, & A. Stent (Eds.), Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) (pp. 238–251). ACL.
• Brachman RJ. What IS-A is and isn’t: An analysis of taxonomic links in semantic networks. Computer. 1983;16(10):30–36. doi: 10.1109/MC.1983.1654194. - DOI

LinkOut - more resources

Full text sources.

• Europe PubMed Central
• PubMed Central
• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Stack Exchange Network

Stack Exchange network consists of 183 Q&A communities including Stack Overflow , the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Q&A for work

Connect and share knowledge within a single location that is structured and easy to search.

"Related work" or "related works"

Which expression should be used as a section heading of an academic paper: related work or related works ? This is a question that has been bothering me for years, as googling shows that both have a large number of usages.

• expressions

5 Answers 5

What is the context?

If the heading refers to things you produced in the same or relevant subject area, then work is uncountable , and the heading should be Related Work .

If the heading refers to your oeuvre or the output of a fellow artist, then work is countable in this case, and the heading should be Related Works .

• 6 +1 Works tend to refer to an individuals works, whereas work to a subject area. –  Schroedingers Cat Commented Jan 25, 2012 at 9:10
• 1 I've never seen this section named "Related Works" in a paper, ever. But maybe this is depends on the field of the paper. –  bitmask Commented Jan 26, 2012 at 0:34
• @bitmask Precisely; works are "something produced by a writer, painter, musician, or other artist." –  Gnawme Commented Jan 26, 2012 at 0:50

Related Work

However, there are at least three cases where a distinction may need to be made.

• If it is a list of literary items (works), then it could be plural: Related Works .
• If it is a paragraph on other related work done by the author, then singular: Related Work .
• If it is a list of engineering projects, then plural: Related Works .

In general it is singular: Related Work .

See also, some random examples from Wikipedia: Australopithecus afarensis

If would depend on the context of the writing in question. If you want to pluralize work to works it should be in the case of referring to the body of work as multiple discrete items (ie works of art).

In the case of writing --

John had done a great deal of writing for clients in the past on these same topics. His related works included a blog post on Engadget and a magazine article for Wired.

John had done a great deal of related work writing on these same topics in the past.

This gets back to @Gnawme's point about the work being either countable or uncountable. If you can substitute stuff or another amorphous noun, then use work .

It would depend on whether your use of the word "work" in context refers to a countable work or not. If you're talking about books, paintings, or other things that are referred to as "works", then use the plural. If you're talking about uncountable work, like "I've also worked in engineering and medicine", then it's probably uncountable and thus you should use the singular.

The distinction is not necessarily consistent. A list of books you have written would be called "works". But a list of software products you have written would probably be called "work". That is, it is common to say, "The author of 'Economics of Tibetan Agriculture', Mr Jones, has three previous works in this field ..." and then list the titles. But no one says, "The author of 'Tax Prep Pro version 4.2', Mr Smith, has three previous works in this field ...".

According to Google Ngram Viewer , the capitalized forms (as would be typical for section headings) are both used, but Related Work is considerably more common.

Your Answer

Sign up or log in, post as a guest.

Required, but never shown

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy .

Not the answer you're looking for? Browse other questions tagged expressions or ask your own question .

• Featured on Meta
• Bringing clarity to status tag usage on meta sites
• Announcing a change to the data-dump process

Hot Network Questions

• What`s this? (Found among circulating tumor cells)
• Should Euler be credited with Prime Number Theorem?
• Can reinforcement learning rewards be a combination of current and new state?
• how do I fix apt-file? - not working and added extra repos
• Cardinality of connected LOTS
• The question about the existence of an infinite non-trivial controversy
• how did the Apollo 11 know its precise gyroscopic position?
• Direction of centripetal acceleration
• What's the benefit or drawback of being Small?
• Manhattan distance
• How can I play MechWarrior 2?
• What was the typical amount of disk storage for a mainframe installation in the 1980s?
• Why are there both "was" and "wozu" in this sentence: "Auf jeden fall ist es mir vollkommen egal, was Sie denken, wozu Sie ein recht hätten!"?
• How to clean a female disconnect connector
• How does the phrase "a longe" meaning "from far away" make sense syntactically? Shouldn't it be "a longo"?
• Intersection of Frobenius subalgebra objects
• How high does the ocean tide rise every 90 minutes due to the gravitational pull of the space station?
• Gravitational potential energy of a water column
• Why didn't Air Force Ones have camouflage?
• Do US universities invite faculty applicants from outside the US for an interview?
• Are there alternative methods, beyond Bayesian or probabilistic approaches, for modeling the relationship between evidence and hypothesis credibility?
• What does "dare not" mean in a literary context?
• Show that an operator is not compact.
• \ExplSyntaxOn problem with new paragraph

Diet is the missing link to helping you manage your menopause, according to new Zoe study

A plant-based diet could reduce menopausal symptoms by more than 30 per cent

According to new research, conducted by the science and nutrition company Zoe , sticking to a healthy diet can improve menopausal symptoms in as little as six months.

The research – which analysed over 70,000 peri and postmenopausal women – found that adopting a healthy plant-based diet could reduce overall peri and postmenopausal symptoms by 30% and 37% respectively, while also improving sexual symptoms, such as low libido , by up to 19% in perimenopausal women and 29% in postmenopausal women.

Led by Professor Sarah Berry , chief scientist at Zoe , in collaboration with King’s College London and funded by the British Menopause Society, the data highlights the impact nutrition can have on the quality of life for women during this transitional period.

‘There’s so much misinformation around menopause , so it’s no surprise that many women don’t know what they’re experiencing or how to change it. Previously we didn’t ever talk about menopause and as a result, there’s so much we still don’t know, yet we do know that it encompasses a huge portion of women’s lives,’ Prof Sarah Berry said.

According to Zoe, 99.8% and 93% of the 70,000 peri and postmenopausal participants included in this observational study experienced at least one of the 20 most common menopause symptoms.

After following Zoe’s dietary guidance (a healthy plant-based diet with an increased intake of whole grains and legumes), these were the changes they reported:

• Mood changes, anxiety , and depression decreased by 35% in perimenopausal women and 44% in postmenopausal women.
• Issues such as fatigue , weight gain , memory loss , and sleep disturbances were reduced by 32% in perimenopausal women and 38% in postmenopausal women.
• Night sweats, hot flashes , and chills were reduced by 30% in perimenopausal women and 32% in postmenopausal women.

What does this mean?

Menopause significantly affects a woman’s metabolism and gut microbiome , leading to poorer blood sugar responses, reduced sleep quality, and an increased tendency to consume sugary foods. These changes can exacerbate menopausal symptoms, but this new research indicates that diet can help mitigate these effects.

How? Foods that are high in fibre , like whole grains and legumes, not only reduce inflammation and blood sugar spikes but also positively alter the gut microbiome , which can lead to a reduction in symptoms.

What do the experts say?

‘Ten years ago, I was on the verge of quitting because I thought I’d gone mad and didn’t know what was wrong with me,’ said menopause advocate and previous WH cover star Davina McCall .

‘The fact that you can improve your menopause experience through food is incredible. Women are desperate for solutions, and this research is crucial in helping to provide them.’

Prof Berry caveated that while there is no single solution for menopause symptoms, the study demonstrates the potential benefits of a healthy diet. ‘Despite the many inflated claims and “menowashing”, there is no silver bullet when it comes to diet and menopause symptoms, but what our data does show is that following an overall healthy dietary pattern may help to improve symptoms.’

Read more on menopause

• Jane Baxter’s recipes for a better menopause – once you’ve tried these Mexican eggs, you can never go back
• How gut health can support symptoms of the menopause – Celebrity GP Dr Zoe Williams shares her top tips and advice for a healthy gut
• ‘I'm 46 and going through perimenopause – how should I exercise?’

Rochelle Humes reveals the surprising 'superpower' she's harnessed to thrive in business

Cut through the noise and g et practical, expert advice, home workouts, easy nutrition and more direct to your inbox. Sign up to t he WOMEN'S HEALTH NEWSLETTER

With nearly a decade of journalistic experience – in print, online and social – at national newspapers and lifestyle magazines, it’s fair to say Alice has tried it all when it comes to health and fitness. From packing herself off to an extreme Aveduric retreat in Sri Lanka and sweat-testing every new fitness fad to running the London Marathon and completing a 70.3 IronMan, Alice now looks after WH ’s food content. With a ‘food first’ ethos, she is here to help you decipher exactly which foods will support your health, and which macro-counting, pasta-replacing, intermittent-fasting, 13-day cleanse is just, well, a scam. A keen baker and host, her favourite dessert has to be pavlova (with lots of summer berries and whipped cream, of course).

Perimenopause and Menopause

5 things you need to know about blood sugar levels

New drug reduces menopausal osteoporosis approved

'What I learnt at the Goodwood Menopause Retreat'

Your guide to running through menopause

‘Exercise for sanity, not vanity’ during menopause

5 mindfulness exercises to help ease menopause

Kelly Holmes was in 'denial' about perimenopause

6 ways to prevent 'menopause belly'

Does the menopause cause herpes outbreaks?

Halle Berry experienced this perimenopause symptom

How to eat during perimenopause

Susannah Constantine, her gut health + menopause

Generating a related work section for scientific papers: an optimized approach with adopting problem and method information

• Published: 21 July 2022
• Volume 127 , pages 4397–4417, ( 2022 )

Cite this article

• Pengcheng Li 1 ,
• Wei Lu 2 &
• Qikai Cheng   ORCID: orcid.org/0000-0003-3904-8901 2  

1069 Accesses

2 Citations

Explore all metrics

The rapid explosion of scientific publications has made related work writing increasingly laborious. In this paper, we propose a fully automated approach to generate related work sections by leveraging a seq2seq neural network. In particular, the main goal of our work is to improve the abstractive generation of related work by introducing problem and method information, which serve as a pivot to connect the previous works in the related work section and has been ignored by the existing studies. More specifically, we employ a title-generation strategy to automatically obtain problem and method information from given references and add the problem and method information as an additional feature to enhance the generation of related work. To verify the effectiveness and feasibility of our approach, we conduct a comparative experiment on publicly available datasets using several common neural summarizers. The experimental results indicate that the introduction of problem and method information contributes to the better generation of related work and our approach substantially outperforms the informed baseline on ROUGE-1 and ROUGE-L. The case study shows that the problem and method information enables considerable topic coherence between the generated related work section and the original paper.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

Automatic related work section generation: experiments in scientific document abstracting

An Analytical Study on a Benchmark Corpus Constructed for Related Work Generation

The Concept of System for Automated Scientific Literature Reviews Generation

Explore related subjects.

• Artificial Intelligence

https://github.com/smalot/pdfparser .

https://github.com/dwadden/dygiepp .

https://huggingface.co/models .

https://github.com/nlpyang/PreSumm .

https://github.com/abisee/pointer-generator .

https://github.com/zhongxiangboy/Improving-related-work-generation-by-introducing-problem-and-method-information .

Chen, J., & Zhuge, H. (2019). Automatic generation of related work through summarizing citations. Concurrency and Computation: Practice and Experience, 31 (3), e4261.

Article   Google Scholar  

Chen, Y. C., & Bansal, M. (2018). Fast abstractive summarization with reinforce-selected sentence rewriting. In  Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 675–686).

Cheng, S. W., Kuo, C. W., & Kuo, C. H. (2012). Research article titles in applied linguistics. Journal of Academic Language and Learning, 6 (1), A1–A14.

MathSciNet   Google Scholar  

Das, S., & Paik, J. H. (2021). Context-sensitive gender inference of named entities in text. Information Processing & Management, 58 (1), 102423.

Day, R. A. (1996). How to write and publish a scientific paper. General Pharmacology, 6 (27), 1077.

Google Scholar  

Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of deep bidirectional transformers for language understanding. In  Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies , Vol. 1 (Long and Short Papers) (pp. 4171–4186).

Flowerdew, L. (2008). Corpus-based analyses of the problem-solution pattern: A phraseological approach (Vol. 29). John Benjamins Publishing.

Book   Google Scholar  

Gehrmann, S., Deng, Y., & Rush, A. M. (2018). Bottom-up abstractive summarization. In  Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP) (pp. 4098–4109).

Heffernan, K., & Teufel, S. (2018). Identifying problems and solutions in scientific text. Scientometrics, 116 (2), 1367–1382.

Hoang, C. D. V., & Kan, M. Y. (2010). Towards automated related work summarization. In  Proceedings of the 23rd International Conference on Computational Linguistics: Posters  (pp. 427–435).

Hsu, W. T., Lin, C. K., Lee, M. Y., Min, K., Tang, J., & Sun, M. (2018). A unified model for extractive and abstractive summarization using inconsistency loss. In  Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 132–141).

Hu, Y., & Wan, X. (2014, October). Automatic generation of related work sections in scientific papers: An optimization approach. In  Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) (pp. 1624–1633).

Jaidka, K., Khoo, C., & Na, J. C. (2013). Deconstructing human literature reviews–a framework for multi-document summarization. In  Proceedings of the 14th European Workshop on Natural Language Generation  (pp. 125–135).

Jamali, H. R., & Nikzad, M. (2011). Article title type and its relation with the number of downloads and citations. Scientometrics, 88 (2), 653–661.

Ji, D., Tao, P., Fei, H., & Ren, Y. (2020). An end-to-end joint model for evidence information extraction from court record document. Information Processing & Management, 57 (6), 102305.

Khoo, C. S., Na, J. C., & Jaidka, K. (2011). Analysis of the macro-level discourse structure of literature. Online Information Review, 35 (2), 255–271.

Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., & Zettlemoyer, L. (2020). BART: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. In  Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics  (pp. 7871–7880).

Lin, C. Y., & Hovy, E. (2003). Automatic evaluation of summaries using n-gram co-occurrence statistics. In Proceedings of the 2003 Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics (pp. 150–157).

Liu, Y., & Lapata, M. (2019). Text summarization with pretrained encoders. In  Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) (pp. 3721–3731).

Lu, Y., Dong, Y., & Charlin, L. (2020). Multi-XScience: A large-scale dataset for extreme multi-document summarization of scientific articles. In  Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) (pp. 8068–8074).

Luan, Y., Wadden, D., He, L., Shah, A., Ostendorf, M., & Hajishirzi, H. (2019). A general framework for information extraction using dynamic span graphs. In  Proceedings of NAACL-HLT  (pp. 3036–3046).

Ma, S., Zhang, C., & Liu, X. (2020). A review of citation recommendation: From textual content to enriched context. Scientometrics, 122 (3), 1445–1472.

Miao, L., Cao, D., Li, J., & Guan, W. (2020). Multi-modal product title compression. Information Processing & Management, 57 (1), 102123.

Mohammad, S., Dorr, B., Egan, M., Hassan, A., Muthukrishnan, P., Qazvinian, V., & Zajic, D. (2009). Using citations to generate surveys of scientific paradigms. In Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics (pp. 584–592).

Mutlu, B., Sezer, E. A., & Akcayol, M. A. (2020). Candidate sentence selection for extractive text summarization. Information Processing & Management, 57 (6), 102359.

Nasar, Z., Jaffry, S. W., & Malik, M. K. (2018). Information extraction from scientific articles: A survey. Scientometrics, 117 (3), 1931–1990.

Paiva, C. E., Lima, J. P. D. S. N., & Paiva, B. S. R. (2012). Articles with short titles describing the results are cited more often. Clinics, 67 (5), 509–513.

Papineni, K., Roukos, S., Ward, T., & Zhu, W. J. (2002, July). Bleu: A method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (pp. 311–318).

Putra, J. W. G., & Khodra, M. L. (2017). Automatic title generation in scientific articles for authorship assistance: A summarization approach. Journal of ICT Research and Applications, 11 (3), 253–267.

Radford, A., Narasimhan, K., Salimans, T., & Sutskever, I. (2018). Improving language understanding by generative pre-training . University of British Colombia.

Saggion, H., Shvets, A., & Bravo, À. (2020). Automatic related work section generation: Experiments in scientific document abstracting. Scientometrics, 125 (3), 3159–3185.

Scott, M. (2001). Mapping key words to problem and solution. In M. Scott & G. Thompson (Eds.), Patterns of Text: In Honour of Michael Hoey (pp. 109–127). Benjamins.

Chapter   Google Scholar  

See, A., Liu, P. J., & Manning, C. D. (2017). Get to the point: Summarization with pointer-generator networks. In  Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 1073–1083).

Swales, J. M., & Feak, C. B. (2004). Academic writing for graduate students: Essential tasks and skills (Vol. 1). University of Michigan Press.

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., & Polosukhin, I. (2017). Attention is all you need. In  Proceedings of the 31st International Conference on Neural Information Processing Systems  (pp. 6000–6010).

Wang, P., Li, S., Zhou, H., Tang, J., & Wang, T. (2019). ToC-RWG: Explore the combination of topic model and citation information for automatic related work generation. IEEE Access, 8 , 13043–13055.

Wang, Y., Liu, X., & Gao, Z. (2018). Neural related work summarization with a joint context-driven attention mechanism. In  Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP) (pp. 1776–1786).

Widyantoro, D. H., & Amin, I. (2014). Citation sentence identification and classification for related work summarization. In  2014 International Conference on Advanced Computer Science and Information System  (pp. 291–296). IEEE.

Yasunaga, M., Kasai, J., Zhang, R., Fabbri, A. R., Li, I., Friedman, D., & Radev, D. R. (2019). ScisummNet: A large annotated corpus and content-impact models for scientific paper summarization with citation networks. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 33, pp. 7386–7393).

Zaman, F., Shardlow, M., Hassan, S. U., Aljohani, N. R., & Nawaz, R. (2020). HTSS: A novel hybrid text summarisation and simplification architecture. Information Processing & Management, 57 (6), 102351.

Zhang, M., Zhou, G., Yu, W., & Liu, W. (2021). FAR-ASS: Fact-aware reinforced abstractive sentence summarization. Information Processing & Management, 58 (3), 102478.

Download references

Acknowledgements

This work was partially supported by Major Projects of National Social Science Foundation of China (No. 17ZDA292).

Author information

Authors and affiliations.

School of Economics and Management, Hubei University of Technology, Wuhan, Hubei, China

Pengcheng Li

School of Information Management, Wuhan University, Wuhan, Hubei, China

Wei Lu & Qikai Cheng

You can also search for this author in PubMed   Google Scholar

Contributions

PL: Conceptualization, Methodology, Writing—Original Draft. WL: Conceptualization, Methodology, Formal analysis, Supervision. QC: Data Curation, Writing—Review & Editing.

Corresponding author

Correspondence to Qikai Cheng .

Ethics declarations

Conflict of interest.

The authors declare no conflict of interest.

Rights and permissions

Reprints and permissions

About this article

Li, P., Lu, W. & Cheng, Q. Generating a related work section for scientific papers: an optimized approach with adopting problem and method information. Scientometrics 127 , 4397–4417 (2022). https://doi.org/10.1007/s11192-022-04458-8

Download citation

Received : 10 May 2021

Accepted : 23 June 2022

Published : 21 July 2022

Issue Date : August 2022

DOI : https://doi.org/10.1007/s11192-022-04458-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Automatic related work generation
• Scientific summarization
• seq2seq neural network
• Problem and method extraction
• Find a journal
• Publish with us
• Track your research

• Share this —

• Watch Full Episodes
• Read With Jenna
• Inspirational
• Relationships
• TODAY Table
• Newsletters
• Start TODAY
• Shop TODAY Awards
• Citi Concert Series
• Listen All Day

Follow today

More Brands

• On The Show
• TODAY Plaza

Supreme Court rejects bid to restrict access to abortion pill

In a blow for anti-abortion advocates, the  Supreme Court  on Thursday rejected a challenge to the abortion pill mifepristone , meaning the commonly used drug can remain widely available.

The court  found unanimously  that the group of anti-abortion doctors who questioned the Food and Drug Administration’s decisions making it easier to access the pill did not have legal standing to sue. 

President Joe Biden said in a statement that while the ruling means the pill can remain easily accessible, “the fight for reproductive freedom continues” in the aftermath of the Supreme Court’s ruling two years ago that overturned abortion rights landmark Roe v. Wade.

“It does not change the fact that the right for a woman to get the treatment she needs is imperiled if not impossible in many states,” he added.

Justice Brett Kavanaugh, writing for the court, wrote that while plaintiffs have “sincere legal, moral, ideological, and policy objections to elective abortion and to FDA’s relaxed regulation of mifepristone,” that does not mean they have a federal case.

The plaintiffs failed to show they had suffered any injury, meaning that “the federal courts are the wrong forum for addressing the plaintiffs’ concerns about FDA’s actions,” he added.

“The plaintiffs may present their concerns and objections to the president and FDA in the regulatory process or to Congress and the president in the legislative process,” Kavanaugh wrote. “And they may also express their views about abortion and mifepristone to fellow citizens, including in the political and electoral processes.”

The legal challenge was brought by doctors and other medical professionals represented by the conservative Christian legal group Alliance Defending Freedom.

“We are disappointed that the Supreme Court did not reach the merits of the FDA’s lawless removal of commonsense safety standards for abortion drugs,” said Erin Hawley, one of the group’s lawyers. She told reporters she is hopeful the underlying lawsuit can continue because three states — Idaho, Missouri and Kansas — have brought their own claims and have different arguments for standing.

By throwing out the case on such grounds, the court avoided reaching a decision on the legal merits of whether the FDA acted lawfully in lifting various restrictions, including one making the drug obtainable via mail, meaning the same issues could yet return to the court in another case.

Another regulatory decision left in place means women can still obtain the pill within 10 weeks of gestation instead of seven. 

Likewise a decision to allow health care providers other than physicians to dispense the pill will remain in effect.

The court’s decision to roll back abortion rights two years ago led to a wave of new abortion restrictions in conservative states.

Then, the court suggested it was removing itself from the political debate over abortion, but with litigation continuing to rage over abortion access, the justices are continuing to play a pivotal role. 

Abortion rights supporters welcomed the ruling, with Nancy Northup, president of the Center for Reproductive Rights, saying she was relieved at the outcome but angered about the case lingering in the court system so long.

“Thank goodness the Supreme Court rejected this unwarranted attempt to curtail access to medication abortion, but the fact remains that this meritless case should never have gotten this far,” she said in a statement.

Danco Laboratories, manufacturer of Mifeprex, the brand version of mifepristone, praised the ruling too, saying it was good for the drug approval process writ large.

In rejecting the challenge, the court “maintained the stability of the FDA drug approval process, which is based on the agency’s expertise and on which patients, health care providers and the U.S. pharmaceutical industry rely,” company spokeswoman Abigail Long said.

Anti-abortion groups expressed disappointment, saying that the ruling highlighted the importance of this year’s election in which Democrat Biden, who has pledged to defend abortion rights, faces off against Republican Donald Trump, who has the strong backing of conservatives who oppose abortion.

“Joe Biden and the Democrats are hell-bent on forcing abortion on demand any time for any reason, including DIY mail-order abortions, on every state in the country,” Marjorie Dannenfeiser, president of SBA Pro-Life America, said.

If Trump were to win the election, his appointees to the FDA would be a position to impose new restrictions on mifepristone. Biden’s campaign manager, Julie Chavez-Rodriguez, alluded to the possibility in a call with reporters after the ruling. Calling the case “one tactic in a broader, relentless strategy” by anti-abortion activists, Chavez-Rodriguez said if Trump is elected, his advisers and allies would try to ban abortion nationwide “without the help of Congress or the court,” and also restrict access to contraception — a threat, she said, to blue as well as red states.

The mifepristone dispute is not the only abortion case currently before the court. It is also due to decide whether  Idaho’s strict abortion ban  prevents doctors in emergency rooms from performing abortions when a pregnant woman is facing dangerous complications.

Mifepristone is used as part of a two-drug FDA-approved regimen that is now the most common form of abortion in the United States.

Abortion is effectively banned altogether in 14 states, according to the Guttmacher Institute, a research group that backs abortion rights.

The FDA had the backing of the pharmaceutical industry, which has warned that any second-guessing of the approval process by untrained federal judges could  cause chaos and deter innovation.

Last year, Texas-based U.S. District Judge Matthew Kacsmaryk issued a sweeping ruling that completely invalidated the FDA’s approval of the pill, leading to panic among abortion-rights activists that it would be banned nationwide.

The Supreme Court last April put that ruling on hold, meaning the pill remained widely available while litigation continued.

The New Orleans-based 5th U.S. Circuit Court of Appeals in August then narrowed Kacsmaryk’s decision but left in place his conclusion that the FDA’s move to lift restrictions starting in 2016 was unlawful.

Both sides appealed to the Supreme Court. The court in December took up the Biden administration’s appeal in defense of the later FDA decisions, but it opted against hearing the challenge to the original approval of mifepristone in 2000. 

The Supreme Court focused solely on the later FDA action, including the initial 2021 decision that made the drug available by mail, which was finalized last year.

This article first appeared on NBCNews.com .

Lawrence Hurley covers the Supreme Court for NBC News Digital.

Orlando Bloom says he lost 52 pounds in 3 months to play a boxer in ‘The Cut’

Health & wellness.

Person hospitalized for bird flu in Missouri had no contact with animals, CDC says

11 healthiest fast food breakfasts, according to dietitians

Diet & fitness.

Peaches pack these major benefits for heart and gut health, dietitians say

The FDA is investigating metals in tampons. Are tampons safe? Don't panic, experts say

Women's health.

Make the busy back-to-school season easier with a healthy meal plan

A list of ultraprocessed foods most harmful to your health, according to recent research

Mom, 40, and son, 11, diagnosed with cancer months apart due to rare genetic syndrome

What is the healthiest sport? The No. 1 pick, according to science

Olivia Munn shares detailed timeline of breast cancer diagnosis and treatment to help others

Stack Exchange Network

Stack Exchange network consists of 183 Q&A communities including Stack Overflow , the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Q&A for work

Connect and share knowledge within a single location that is structured and easy to search.

Does the related work section need to be strictly about other's work only?

I would like to clarify a writing style issue concerning separating related work from my own work:

Does the related work section need to be strictly about other work only? Or may you also explicitly explain the relevance of the related work to the text at hand?

That is, should you avoid explaining why a given related paper is related to your work and just report on the paper itself, or is it okay to also explain, WHY that paper is relevant while you talk about it? For example

Related work

[...] Someone et al. (2010) report that "X -> Y". Someoneelse (2011) however found evidence for "not X -> Y". It is not entirely obvious where this difference in findings results from. That is why I study "X" in regard to its influence on "Y" in more detail in this thesis. [...]

A version of this statement would also find itself in the introduction and possible the abstract, but imagine there to be more surrounding explanation in the related work section.

Is that permissible writing style? Or is the That is why I study "X" in regard to its influence on "Y" in more detail in this thesis. part out-of-place in the related work section?

• writing-style

2 Answers 2

The objective of related works is to, well, show how other work is related to yours. It’s not only reasonable, but also expected that you say how other works are positioned with respect to yours. When I read the related work section I also want to understand exactly that “why” of relatedness. Doing so explicitly would really help your readers.

Previous work related to this paper

Usually in papers discussion of related work comes after you have introduced the context of the problem and defined some bounds of it, so in many cases it's obvious how and why it's relevant - but if it isn't, then you definitely should explain the relevance. However, it's not the place for in-depth comparison; if your paper proposes some novel method that should be compared with earlier work by Smith et al, then in 'related work' you introduce Smith's work; then you have chapter(s) talking about your method in detail, and only then (IMHO) you can properly compare and contrast the methods.

Your previous work counts as well

It's worth noting that your own previous papers are likely to be in the 'related work' section. In "related work" you describe the state of art in the field before this paper , not before you first became a researcher; and it's plausible that a proper description of the state of art in your subfield includes some of your earlier papers.

You must log in to answer this question.

Not the answer you're looking for browse other questions tagged writing writing-style ..

• Featured on Meta
• Bringing clarity to status tag usage on meta sites
• Announcing a change to the data-dump process

Hot Network Questions

• Do US universities invite faculty applicants from outside the US for an interview?
• Remove an edge from the Hasse diagram of a finite lattice
• Is there a way to prove ownership of church land?
• Should Euler be credited with Prime Number Theorem?
• What prevents random software installation popups from mis-interpreting our consents
• \ExplSyntaxOn problem with new paragraph
• In a tabular with p-column, line spacing after multi-line cell too short with the array package
• How to sum with respect to partitions
• How can I play MechWarrior 2?
• Is "She played good" a grammatically correct sentence?
• What would be a good weapon to use with size changing spell
• An instructor is being added to co-teach a course for questionable reasons, against the course author's wishes—what can be done?
• Confusion about time dilation
• What is the optimal number of function evaluations?
• What are the most commonly used markdown tags when doing online role playing chats?
• Wien's displacement law
• Does the average income in the US drop by $9,500 if you exclude the ten richest Americans?
• A seven letter *
• Simulate Minecraft Redstone | Where to start?
• Replacing jockey wheels on Shimano Deore rear derailleur
• Is the 2024 Ukrainian invasion of the Kursk region the first time since WW2 Russia was invaded?
• Is this host and 'parasite' interaction feasible?
• Topos notions coming from topology and uniqueness of generalizations
• What does "dare not" mean in a literary context?