Lesson 02: Reading Strategy¶

Reading a research paper is not like reading a textbook or an article. Papers are written to communicate new results to specialists, not to teach from first principles. A systematic approach helps you extract what you need, but "what you need" is often just the gist, and that is completely fine.

---

Why reading strategy matters¶

Most papers are dense, and most of the detail in a paper is not what you need on a first read. Reading start-to-finish wastes time and makes it easy to lose the thread.

A structured approach gives you:

• A clear picture of what the paper claims and whether it is relevant to you, quickly.
• A way to decide how deeply to read before committing hours to a single paper.
• A framework for notes that are actually useful later.

---

Step 1: The decision filter (2–5 minutes)¶

Before committing any real time to a paper, run it through a quick filter. Most papers you encounter will not be relevant enough to read further, and that is fine. The goal here is to find out fast.

Work through these four steps in order, stopping as soon as the paper is clearly not relevant:

1. Title: is this in the right area at all?
2. Abstract: what does it claim? Does that claim matter to your work?
3. Figures: flip through them quickly. The main results are almost always in the figures. Do the results look relevant, credible, and interesting?
4. Conclusions: what do the authors say they found, and what caveats do they flag?

If the paper is still interesting after all four steps, it is worth reading properly. If it loses you at any step, stop. Move on and come back only if a reason to resurfaces.

---

Step 2: Reading the paper¶

Once a paper has passed the filter, read it in two passes.

Pass 1: Understand the results (30–60 minutes)¶

Goal: understand what was found and how.

• Read the introduction to establish context and the specific question being addressed.
• Return to the figures carefully: axes, captions, error bars, what each panel shows.
• Read the results section alongside the figures.
• Skim the methods to understand how the results were produced; you do not need to reproduce them yet.
• Note any references you do not recognise but that seem important.

After this pass, aim to be able to say what the paper claims and roughly how. If the paper is outside your current background, even "they measured X and found it larger than expected" is a perfectly valid takeaway. That is the gist, and the gist is enough for most papers.

Pass 2: Full detail (several hours)¶

Goal: understand the paper deeply enough to use or build on it.

This pass is only for papers that are directly relevant to your work.

• Read every section carefully, including methods and appendices.
• Work through equations by hand if needed.
• Follow up key references.
• Try to reproduce a result or a figure if you have the tools.

Most papers will only ever need the decision filter and Pass 1. Reserve Pass 2 for the handful of papers that are central to your project.

---

Reading the abstract in detail¶

The abstract is a highly compressed summary of the whole paper. A well-written abstract contains four things:

1. Context: what is the broader problem?
2. Gap or question: what specific question does this paper address?
3. Method or approach: how did the authors address it?
4. Main result: what did they find?

When you read an abstract, try to identify all four parts. If you can, you have understood what the paper is about before reading a word of the body.

---

Dealing with unfamiliar concepts¶

You will regularly encounter terms, methods, and notation you have not seen before. This is expected, not a sign that you are falling behind.

Physics subfields are narrow enough that even a postdoc working in a neighbouring area will often struggle to follow the technical details of a paper outside their speciality. The difference between a beginner and an expert is not that the expert understands everything on a first read. It is that the expert is faster at identifying what they can skip and what they need to look up.

A few strategies when you hit something unfamiliar:

• Note it and continue. Often the meaning becomes clear from context a few sentences later.
• Check a textbook or review article. For standard terms, a quick lookup is faster than asking.
• Ask your supervisor. If the same concept appears repeatedly and you cannot place it, ask. This is exactly what supervisors are for, and they will not think less of you for not knowing.

Some papers will remain largely opaque no matter how carefully you read them. That is not failure. It means the paper requires background you have not yet built. Come back to it in a few months; you may be surprised how much more you can follow.

---

Taking notes while reading¶

Notes you write while reading are far more useful than highlighted text. Highlighting tells you a sentence felt important; a note tells you why.

Suggested notes to take:

• One-sentence summary of the paper's main claim.
• Key result: the most important number, figure, or conclusion.
• Method in brief: what technique or approach did they use?
• Questions: what did you not understand, or what would you want to follow up?
• Relevance: why does this paper matter to your current work?

Even a few sentences per paper adds up to an invaluable record over months of reading. Your reference manager (covered in Lesson 03) is a good place to keep these notes attached to each paper.

---

Reading efficiently over time¶

Reading speed improves with practice. A paper that takes you two hours today may take twenty minutes in six months, because you will recognise the standard approaches, the common notation, and the field's open questions.

A few habits that help:

• Read a little every day rather than in occasional large sessions.
• Follow your curiosity. Papers that interest you are easier to read.
• Discuss papers with colleagues or in journal club (see the Journal Club track). Explaining a paper to someone else is one of the best ways to test your understanding.