03 · What You Need to Know
Emerging topics create a moving-target retrieval problem
Terminology instability is part of emergence
A research topic can become visible before a stable vocabulary develops around it. Researchers may use multiple provisional labels, borrow language from established fields, or emphasize different aspects of the phenomenon.
Research on emerging-topic identification illustrates this problem directly. One study notes that researchers use varied expressions such as emerging topics, emerging trends, research fronts, and other terminology for overlapping forms of scientific emergence, with no single dominant paradigm organizing the area.
The same instability can occur around the topic you are trying to retrieve. Searching one currently fashionable label may capture only one terminological branch of a literature that has not yet agreed what to call itself.
Define the phenomenon before chasing its names
Write a short operational description of what the topic includes. What characteristics must be present? What neighboring phenomena are outside scope? What would make a study clearly relevant even if it never uses your preferred label?
This conceptual anchor is particularly important in an emerging field because the vocabulary may change faster than the underlying research question.
Stable concept definition
Your working account of the phenomenon that determines what evidence belongs within the search scope.
Evolving search vocabulary
The changing collection of words, phrases, acronyms, and indexing terms through which that phenomenon appears in databases.
The first may itself require refinement as knowledge develops, but it should not be casually rewritten every time a new buzzword appears. Academia produces enough of those without assistance.
Start with seed records rather than trying to predict every term
Identify several records that clearly concern the emerging phenomenon. These can come from preliminary searches, expert recommendations, citation trails, conference literature, preprints, known research groups, or other appropriate sources.
Inspect their titles, abstracts, author keywords, terminology, references, and database indexing. Cochrane's technical guidance recommends identifying free-text terms from a sample of relevant references and explicitly recognizes that concepts can be expressed through different words.
This approach is more defensible than trying to brainstorm every possible term before looking at the literature.
Harvest several kinds of terminology
An emerging topic may appear through more than direct synonyms. Organize what you find according to its role.
| Term type |
Why it matters |
Main caution |
| Current labels |
Capture terminology gaining traction in recent literature |
May not appear in early studies |
| Early or provisional labels |
Retrieve literature published before terminology stabilized |
May have been abandoned for conceptual reasons |
| Descriptive phrases |
Capture papers that describe the phenomenon without naming it |
Can be difficult to search precisely |
| Acronyms |
May become common quickly in emerging fields |
Can have unrelated meanings |
| Broader terminology |
May capture records indexed or described before the new concept was differentiated |
Can produce substantial noise |
| Disciplinary variants |
Capture parallel literatures developing under different intellectual traditions |
May represent overlapping rather than identical concepts |
| Controlled vocabulary |
Provides standardized indexing when a suitable heading exists |
May lag behind the emerging concept |
If the topic does not yet have a stable label at all, the broader strategy for searching a concept with no agreed name becomes particularly relevant.
Expect early and later literature to use different vocabulary
Emerging fields often develop terminology in phases. Early papers may rely on descriptions or borrow terminology from an adjacent field. A memorable label may then appear and spread rapidly. Later research may differentiate subtypes and develop more technical vocabulary.
A current search therefore needs temporal breadth as well as lexical breadth. If you use only the terminology dominant today, you may systematically underretrieve the papers that helped create the field before that terminology existed.
Research on topic evolution has documented that scientific topics change not only in popularity but also through shifts in associated vocabulary, including migration of words as topics evolve.
This makes terminology change over time a methodological issue rather than merely an interesting history lesson.
Controlled vocabulary may lag behind an emerging field
Subject headings remain valuable, but new topics may not immediately receive dedicated indexing terms. A broader heading may be the closest available option, or the relevant heading may be introduced only after a body of literature already exists.
There is another practical lag: even when an appropriate subject heading exists, very recent database records may not yet be fully indexed. Cochrane therefore recommends combining controlled vocabulary with free-text searching and explicitly notes that current records may not yet have complete indexing.
Watch Out
Do not conclude that an emerging topic has little literature merely because a database has no dedicated subject heading for it. The literature can precede the indexing vocabulary used to organize it.
Free-text searching carries more weight when indexing is immature
For a new topic, titles, abstracts, and author keywords may contain terminology long before controlled vocabulary catches up. Free-text searching is therefore particularly important.
Cochrane's current guidance recommends both text words and subject headings and calls for a wide range of free-text terminology, including synonyms, acronyms, spelling variants, truncation, and proximity approaches where appropriate.
That principle is useful well beyond intervention reviews: emerging terminology usually demands close attention to the actual language appearing in records.
Search descriptive phrases when the noun has not settled
Sometimes the most reliable representation of an emerging topic is not its name but a description of what happens.
Suppose researchers disagree about what to call a new form of technology-assisted behavior. Their labels may differ, but several descriptions could repeatedly combine the same underlying actions, actors, or technologies.
Phrase and proximity searching can help retrieve these records. Instead of requiring one exact expression, a proximity query can require conceptually important words to occur near each other while allowing differences in wording and word order.
Database syntax varies, so these formulations should be tested and translated rather than copied unchanged between platforms.
Follow disciplinary branches before they disappear into separate vocabularies
Emerging topics frequently develop simultaneously in more than one field. Computer science may emphasize the technology, psychology the behavior, education the learning process, and sociology the institutional consequences.
Those literatures can become increasingly difficult to connect if each develops its own terminology.
Sample relevant papers across the fields contributing to the topic and determine whether different disciplines are naming the concept differently. Do not assume that the vocabulary of the field in which you first encountered the topic is the vocabulary of the entire literature.
Track terminology rather than merely collecting it
A useful terminology log can record each candidate term, where you found it, what period or discipline uses it, how it relates to the target concept, and whether it contributes relevant retrieval.
This turns a growing synonym list into an auditable search-development record.
| Term |
Source |
Relationship |
Period or field |
Search decision |
| Candidate label A |
Seed article |
Direct label |
Early literature |
Retain |
| Candidate phrase B |
Recent review |
Near-synonym |
Current literature |
Retain |
| Candidate term C |
Adjacent discipline |
Related construct |
Field-specific |
Test before inclusion |
This record becomes particularly valuable when the search is updated months or years later and the terminology has moved on again.
Test new terms for marginal retrieval value
Do not add every newly encountered expression automatically. Search candidate terms and inspect the records they contribute beyond the existing strategy.
Recent systematic-search guidance recommends calibrating search strings against preliminary results and checking whether the vocabulary represents the diverse language of the review topic.
A new term may recover a previously invisible branch of the literature. It may also be a fashionable label applied mostly to records you already retrieve. Both findings are useful.
Validate against known relevant records from different periods
Create a benchmark set that includes early and recent papers if possible. If your strategy retrieves the newest literature but consistently misses foundational studies, contemporary terminology may be dominating the search.
Conversely, if the strategy retrieves early conceptual work but misses recent records, new terminology may have emerged since your initial development.
Known-item testing cannot prove completeness, but it can reveal obvious temporal weaknesses in the vocabulary.
An emerging-topic search is not finished forever
Search strategies for mature concepts also require updating, but emerging topics deserve particular attention because vocabulary can change rapidly.
When updating the review or rerunning a saved search, do more than change the date range. Inspect recent records for new terminology, acronyms, subject headings, product names, and distinctions that did not exist during the original search.
PRESS provides a useful quality-control framework by asking whether subject headings, free-text synonyms, spelling variants, acronyms, truncation, and other search elements are appropriate and complete.
04 · A Practical Example
Building a search while a field is still naming itself
Hypothetical Example
A new educational technology acquires several competing labels
Imagine that researchers begin studying AI systems that generate individualized simulated classmates for collaborative learning. No established term exists initially.
Early papers describe the systems as artificial peer simulations. Another research group uses synthetic learning partners. A later conference popularizes generative peer agents, while some computer-science papers describe technically similar systems using agent terminology without mentioning education in the name.
Define Describe the essential phenomenon independently of any one label: AI-generated interactive agents designed to function as peer participants in learning activities.
Seed Identify clearly relevant early and recent papers through targeted searching, citations, conferences, and expert knowledge.
Harvest Extract labels, descriptive phrases, acronyms, author keywords, and available subject headings from those records.
Map Record when and where each expression is used and distinguish direct labels from neighboring concepts.
Test Determine which terms contribute unique relevant records and whether descriptive proximity searches recover papers that use no established label.
Update When rerunning the search, inspect recent literature for terminology introduced since the original strategy was developed.
If one label eventually becomes dominant, it should not erase the earlier terminology from a search covering the field's full history. Today's standard term may be tomorrow's excellent keyword, but it was useless for authors who had not invented it yet.
