Add a rule to the ruleset, entitled Learning:

Each Rat has an amount of Learning, which is tracked in the GNDT and is an integer that defaults to zero.

Add a rule to the ruleset, entitled Hypotheses:

A Hypothesis is a prediction about the behaviour of Rats in the coming week. A Hypothesis must be specific and measurable - it must contain one or more behaviours or actions that all Rats or a specific subset of Rats can carry out, and the number of times (or range of numbers of times) that each behaviour is Hypothesised to occur.

As a weekly action, any Rat can make a Hypothesis by sending a PM to the Head Scientist detailing their Hypothesis and by making a story post to the blog announcing that a Hypothesis has been formed (but not detailing its specifics). The Head Scientist should, as soon as possible thereafter, respond to that Rat with a Specificity Score for their Hypothesis. The Specificity Score of a Hypothesis should reflect how specific it is and how likely it is to come about, and should be an integer between 0 and 10.

As soon as possible thereafter, but not sooner than 7 days after the posting of the relevant story post, a Rat who formed a Hypothesis may Evaluate it by making a further story post with the detailed terms of their Hypothesis and the results of their Evaluation. A Hypothesis may be Evaluated as Successful if the terms of that Hypothesis have been met completely and in their entirety. Otherwise, it is Unsuccessful. If a Hypothesis is Evaluated to be Successful, the Hypothesising Rat may once increase their Learning by the Specificity Score of their Hypothesis.