A branching scenario is an interactive video structure in which the learner's decisions at choice points determine which video segment plays next, creating personalised paths through the content rather than a single linear sequence. The underlying data model is a directed graph: each node is a video clip or scene, and each edge is a choice that leads from one node to the next; the player resolves the graph at run time based on stored learner choices. Branching scenarios are particularly effective for soft-skills training — sales conversations, medical consultations, conflict resolution — where the consequence of a wrong choice can be shown immediately in the next scene rather than explained abstractly. From an xAPI perspective, each choice is emitted as a statement with a result object, so the full decision path is reconstructable in an LRS for debriefing or analytics. The main production cost is exponential: a scenario with three choices at each of four nodes requires up to eighty-one unique paths, meaning video assets multiply quickly unless designers deliberately merge branches back to shared nodes. This exponential growth makes early graph design — deciding where branches converge — one of the most important and cost-controlling decisions in a branching project. Adaptive learning systems can extend branching by selecting the next branch automatically based on past performance, blurring the line between author-defined branching and algorithm-driven personalisation.
