How to use this site

Every chapter page follows the same structure, so you always know what to expect:

1. Overview — the chapter’s key concepts, distilled, with figures from the book

2. Hands-on tutorial — compact MATLAB/Python code examples for the main ideas (for chapters where code applies; some early chapters are conceptual and use figures and questions instead)

3. Thought questions — open-ended discussion questions, great for classes and journal clubs

4. Quiz yourself — factual comprehension questions; click any question to reveal the answer

Chapter pages vs. labs — where code actually runs¶

Each code chapter has two levels:

Each code chapter page has a “🖥️ Ways to run this chapter’s code” box near the top, linking to its lab and cloud options — and a quick-run shortcut card in the bottom-right corner on large screens.

▶ Running a lab in your browser (recommended — zero setup)¶

Python labs run directly in your browser via WebAssembly (JupyterLite/Pyodide). Nothing to install, no account needed:

1. Open a lab page and click the ⏻ power icon at the top right of the notebook content (labeled “Click for interactive code”). On small windows the icon can be hidden — widen the browser window if you don’t see it.

2. Wait for the kernel to start — the first start takes ~30–60 seconds while Python downloads into your browser (it’s cached afterward, so later starts are fast).

3. Run cells top to bottom, starting with the first code cell. The first cell sets up imports (and for some labs installs nilearn via %pip). Later cells depend on earlier ones — if you jump ahead you’ll get NameErrors. If things get confused, use the restart button and run again from the top.

Once the kernel is live, each cell gets its own ▶ run button and a run-all control appears in the notebook toolbar.

Cells are fully editable. Once the kernel is running, click into any code cell, change parameters or code, and re-run it — this is the best way to build intuition (What happens with a smaller sample? A different threshold?). Your edits live only in your browser tab; reloading the page restores the original.

If a cell hangs or you want to start over: there is no per-cell stop button, but the ↺ restart control in the notebook toolbar interrupts everything and gives you a fresh kernel — then run again from the top. Note that the first run of a lab can look “hung” for up to a minute while Python and packages download into your browser; the page may even be briefly unresponsive. That’s normal — it’s cached and fast afterward.

☁️ Other ways to run¶

MATLAB and Python tabs¶

Most exercises are provided in both languages using tabbed panels:

% MATLAB version (CANlab tools)
obj = load_image_set('emotionreg');
plot(obj);

# Python version (nilearn)
from nilearn import plotting
plotting.plot_stat_map(img)

The tabs are synchronized: choose MATLAB or Python once, and every tabbed example on the site switches to your choice.

Quiz answers¶

Quiz questions hide their answers until you click:

Try it! Click the question above.