How a different exam for every student is marked fairly, what playing games for credit does and does not require of you, and the rest of the rules.
Policies and support#
Everyone sits a different exam. Here is how that is fair.#
Assessment 3 has you pick the course you stalled on and sit an exam we set on the quarter of it you chose. Your exam is therefore harder or easier than the person next to you, and pretending otherwise would be dishonest. We considered four ways to handle it.
Everyone sits the same exam, on one subject the course nominates. Rejected, because it destroys the premise. The subject has to be one you have already failed to get through, and a stall is personal — a subject chosen centrally is merely unfamiliar to most of the room, which is a much easier problem and not the one this course is about.
Scale the scores by subject difficulty. Rejected. It requires a difficulty model across arbitrary subjects that we do not have and could not defend if challenged, and an indefensible model is worse than an acknowledged limitation.
Mark the raw score. Rejected outright. It pays you to pick something easy, and by week 10 you would all have found that out.
What we do: the exam is a criterion, not a score. It carries 40% and it is marked against the baseline you wrote in week 10 — the question is whether genuine movement happened, not how many marks it was worth. Above that bar the remaining marks go to whether the mechanism caused the movement and whether you can evidence it, which is the thing this course actually taught.
The consequence is worth stating plainly: you cannot improve your mark by picking an easy subject, and you cannot lose marks by picking a brutal one. You can lose marks by picking something you have not actually stalled on, because the evidence of the stall is marked too — and “I never got round to it” is not a stall, it is a subject you have not started.
Playing games for credit#
Two of the twelve weeks are spent playing, and this raises real questions that a normal course does not have to answer.
You never have to buy anything. The School licenses copies for both Labs and the machines are provided. If you would rather use your own account and your own save, that is fine and slightly better, but it is not expected and it earns nothing extra.
You do not have to be good. Nothing in this course is marked on how you played. Assessment 1 and 2 are marked on the mechanism you built afterwards. Arriving at Lab 1 having never queued a match is an ordinary starting point and there is a beginners’ queue for exactly that.
If you cannot play, tell us in week 1. Photosensitivity, motion sickness, RSI, a hand injury, or a considered objection to the content all count, and none of them requires a diagnosis. The observations the Labs produce can be gathered by watching someone else play with the note-taking protocol running, or from recorded play with the same protocol. This is a genuine substitute, not a consolation one — the notes are the deliverable and the notes are what gets used.
Time spent playing is coursework time, and it is inside the ordinary 130 hours, not on top of them. If a Lab is running you over your budget for the week, stop playing and come to the tutorial with fewer matches. Eight is a minimum for a reason and six with good notes beats twelve with none.
What this course is not#
It is a course about reward structures. It is not counselling, and the staff teaching it are not qualified to provide any.
Some of the material — learned helplessness, the evening that ends with nothing to point at, the difference between resting and not-working — sits close to things that can be symptoms rather than habits. If working through it starts to feel like more than a study problem, that is worth taking to someone who is trained for it. Ask any of the teaching team and we will point you at the right service; nothing you say to us about it affects your marks or goes into your record.
Nothing in this course is designed to make you work more hours. Several parts of it are designed to make the hours you already work produce something, which is a different goal and occasionally the opposite one.
Assessment mechanics#
Late work. The two demonstrations happen live in the lecture slot and cannot be handed in late; if you cannot attend, arrange an alternative slot beforehand. Assessment 3’s written material is accepted up to five working days late at 5% per working day, and the exam is fixed.
Extensions are granted readily and must be asked for before the due date. The demonstrations can move to another week. Assessment 3’s run cannot be extended past the end of semester, so an extension there moves the written account, not the thirty hours.
Marks and feedback are returned within two weeks. Both demonstrations get verbal feedback on the day.
Using AI#
Expected, and used throughout — the built route for both mechanisms explicitly invites it, and building a working tool in an afternoon is a normal way to do Assessment 1 or 2.
Two limits, and they are about what is being marked rather than about tools.
You must be able to drive what you submit. In a demonstration you will be asked to change something small in front of us. Not being able to is the same problem whether the thing was generated, copied from a friend, or written by you in a hurry a month ago.
The logs must be real. Everything else in this course can be assisted; your own record of your own sittings cannot be, because a fabricated log makes the entire submission meaningless rather than merely dishonest. The failures in a real log are the most valuable thing in it, and a run with no bad fortnight in it is not a run we believe.
Academic integrity#
The mechanism you build is yours, and you may borrow freely from published designs — the reference ladder read in week 5 exists to be copied from. Say where a design came from.
Collaboration on mechanisms is encouraged and collaboration on logs is not, for the reason above.
Getting help#
Email any of the teaching team. Questions about a brief are better asked in the lecture, where everyone hears the answer.