Category: Blog
-
Bolt Performance Calibration — A Total Score Can Hide a Changing Skill Profile — and Subscores Can Be Too Noisy to Rescue It
The same total score can be built from different strengths and weaknesses, yet narrow subscores can be less reliable than the total. Bolt calibrates both before anyone turns a profile into a learner label.
-
MindOS Learning Manual: Intertextual-Integration State | Reading Four Sources Is Not the Same as Building One Coherent Understanding
Intertextual integration is the learner operation of selecting, comparing, corroborating and synthesising information across multiple documents into one coherent model. This MindOS manual separates source handling from actual cross-source understanding.
-
Bolt Performance Calibration — A Teacher Can Rank Students Correctly and Still Misjudge Their Actual Level
Teacher judgement has more than one kind of accuracy. A teacher can identify who is stronger than whom while still overestimating or underestimating the class’s absolute performance level.
-
MindOS Learning Manual: Relevance-Filtering State | The Most Interesting Detail on the Page May Be the Least Useful Thing to Study
Interesting but irrelevant details can consume attention, increase extraneous load and weaken learning. This MindOS manual shows how learners identify the study goal, filter for relevance, preserve useful context and resist seductive details without flattening the material.
-
Bolt Performance Calibration — A Report-Card Grade Is Not Automatically a Pure Achievement Score
A report-card grade may combine achievement with effort, behaviour, homework, participation and local grading rules. Bolt calibrates what the grade can and cannot justify before anyone treats it as a pure mastery score.
-
MindOS Learning Manual: Refutation State | Correcting a Misconception Is Not the Same as Telling the Right Answer
A misconception is not always repaired by presenting the correct fact. Refutation State makes the old model visible, explains why it fails, supplies a better model, and tests whether the learner can use the correction after delay and in new cases.
