Self-study
One person, a browser, and public data. No cohort, no schedule.
For: Anyone working through the material on their own time, at any career stage.
Modes of Use
The same material, three situations: working alone, running a hosted session, or inside a research programme.
Why there is a mode axis at all
The tracks answer what you are learning. They do not answer how you are using it, and that turns out to change almost everything about what the material needs to be.
A person working alone at midnight needs the answer to a self-check to be available and needs to be told not to open it yet. A facilitator standing in front of eleven people needs a run of show, a rubric visible before the activity starts, and a list of misconceptions to fish for in the debrief. A trainee inside a lab needs neither: they need work whose output somebody downstream actually uses, and a reviewer who will tell them it is not good enough.
Those are three different products built from one body of material. Treating them as one surface is why sites like this one usually end up serving none of them well: the units read as both self-study and lecture notes, and the facilitator material sits orphaned from the content it supports.
Mode is independent of track. Any track can be worked in any available mode, and each track page states what it looks like in each.
Hosted workshop
A facilitator runs a session. Timing, materials, and rubric are already written.
For: Instructors, mentors, and program staff running a taught session, a lab meeting, or a summer-program block.
Research intensive
A longer, contributory programme that draws on this material. Not built yet.
For: Trainees working inside a lab or programme on real data, producing work others will rely on.
Self-study
AvailableOne person, a browser, and public data. No cohort, no schedule.
You set the pace, choose the order, and mark your own work. Every unit is written to be readable without a lecturer standing next to it: worked examples run to completion, self-checks are answerable before the answer is revealed, and each lab ends in an artifact you can look at and judge.
How it works
- Pick a track, or ignore the tracks entirely and follow a question through the core reference.
- Read a unit, attempt its self-checks with the answers closed, then do the lab.
- Compare your lab artifact against the three-level rubric printed in the unit.
- Use the dictionary the moment a term stops you. Most apparent difficulty in this field is vocabulary.
- Who it is for
- Anyone working through the material on their own time, at any career stage.
- What you need
- A browser. Public volumes open in Neuroglancer with no account. A few labs suggest Python, but each states a no-code alternative.
- Assessment
- Self-scored against the rubric in each unit. Nothing is recorded, submitted, or graded.
What this mode does not give you
- You cannot calibrate alone. Self-scored perceptual judgement drifts, and it drifts confidently. The single highest-value thing you can add is one other person to compare calls with.
- There is no deadline, and the units that matter most (05-07) are the slowest and least immediately rewarding. Most people who stall, stall there.
- Nothing here confers a credential. What it produces is a set of artifacts you can show someone.
What to do about that
- Work the perceptual drills with one partner, even asynchronously. Two people scoring the same patches beats three rounds of self-review.
- Post a lab artifact somewhere it will be read. A brief written for an audience is a different document from one written for yourself.
- Use the structured self-checks as tests, not as reading. Opening the answer first converts a retrieval opportunity into re-reading, which feels productive and is not.
Hosted workshop
AvailableA facilitator runs a session. Timing, materials, and rubric are already written.
Every module has a session kit: run of show with timings, prep checklist, the activity and its handouts, the misconceptions to surface in debrief, an assessment rubric, an exit prompt, and a page on what to do when the session goes wrong. The kits are generated from the module pages, so a session and its source material cannot drift apart.
How it works
- Read the facilitator guide once. It covers the assessment stance the kits assume.
- Pick a module, open its session kit, and work the prep checklist the day before.
- Run the session from the run-of-show. Learner worksheets and slide decks are linked from the kit.
- Debrief against the listed misconceptions rather than asking whether there are questions.
- Who it is for
- Instructors, mentors, and program staff running a taught session, a lab meeting, or a summer-program block.
- What you need
- A room, a projector or shared screen, and verified data access. The kits assume no software installation by learners.
- Assessment
- A three-level rubric per activity plus an exit prompt. Grading is of the reasoning, not the answer.
What this mode does not give you
- The kits assume a facilitator who has done the activity themselves. They are not a script for someone encountering the material for the first time in the room.
- Timings are for a group that can already navigate a volume. A first session with absolute beginners runs long.
- Sessions are single-module. Sequencing several into a coherent programme is still the facilitator's job; the track pages are the closest thing to a syllabus.
What to do about that
- Do the activity yourself first, including the part where you are unsure, and narrate that uncertainty in the room.
- Show learners the rubric before they start, not after. It changes what they produce.
- Name one hidden-curriculum norm per session out loud. The kits nominate a candidate for each.
Research intensive
In development — not builtA longer, contributory programme that draws on this material. Not built yet.
The intended third mode is an intensive, research-and-contribution focused programme that uses this site as its reference layer rather than its syllabus: a cohort works on live data, contributes proofreading or analysis that is actually used, and is assessed on work that leaves the classroom. It does not exist yet. This entry is here so the gap is visible rather than implied.
- Who it would be for
- Trainees working inside a lab or programme on real data, producing work others will rely on.
Why it is listed here anyway
- Nothing on this site currently delivers this mode. Any page that reads as though it does is a page to fix.
- It needs three things the self-study and workshop modes do not: a calibration gate that a lab would trust, real tasks whose output is used, and a reviewer who signs off. None are in place.
What it would take
A calibration gate
A perceptual assessment on native-resolution data with labels derived from released segmentations, reporting accuracy within a learner's own high-confidence calls rather than raw accuracy. Units 05-07 argue calibration matters more than accuracy; the gate has to measure what the curriculum claims to value.
Real tasks with real consumers
Proofreading or analysis work that someone downstream actually uses. Practice tasks whose output is discarded do not produce the thing this mode exists to produce.
Structured review
A named reviewer, a rubric, and a return path. The worksheets already carry peer-review sections; what is missing is the exchange.
Choosing
If you are unsure, you are almost certainly in self-study, and the question that matters more is which track. Pick that on the tracks page, or ignore the tracks and follow a question through the core.