Most students search PubMed the same way they'd search Google — type a question in plain English, hope for the best. PubMed isn't built for that. It's built for Boolean search: a more structured way of combining terms that, once you understand it, finds dramatically more relevant papers than a plain-English question ever will.
1. Use OR for synonyms and alternative wordings. A single concept is often described several different ways across the literature — an abbreviation, a full name, alternate spelling. Searching only one misses papers using the others.
2. Use AND to combine distinct concepts. This narrows your results to papers touching on both ideas, not just one.
3. Use quotation marks for exact phrases. Without quotes, PubMed may treat multi-word terms loosely. Quotes tell it to search for that exact phrase.
Searching only the exact words in your essay question is the single biggest reason students miss good papers. A paper studying "Cyp1a2-null mice" is directly relevant to a question about "CYP1A2 knockout" — but won't show up if you only search the word "knockout," since that specific paper never uses it. Building in synonyms up front is what actually surfaces papers like this.
Adding [tiab] after a term (e.g. "gene knockout"[tiab]) tells PubMed to search only the title and abstract, which is usually more precise than searching the entire record.
A well-built search string is often the difference between finding 3 relevant papers and finding 30 — most of the work is just making sure you're not accidentally excluding the ones that use different words for the same idea.