Comprehensible Input

Input hypothesis: we acquire language by understanding messages slightly above our level ("i+1").

Acquisition-learning distinction: subconscious acquisition, not conscious rule learning, is what produces fluency.

Monitor hypothesis: consciously learned rules only edit output; they never generate spontaneous speech.

Natural order: structures are acquired in a predictable order regardless of teaching sequence.

Affective filter: anxiety, self-doubt, and boredom block input from being acquired; a relaxed, confident learner acquires more. Krashen also legitimized the "silent period" before speaking. A frequently quoted corollary: "talking is not practicing" — output is the result of acquisition, not the cause of it.

Level-matched input: vocabulary economy rules keep nearly all non-target words high-frequency at Beginner level — i+1 in practice: mostly known language plus a small set of new structures.

The comprehension scaffold requires any new or difficult word to be clarified through context, contrast, or rephrasing before the story moves on — comprehensible input by construction, with translations available on demand rather than needed in the moment.

Affective filter management: practice happens alone in a safe space, questions are answered to a warm AI storyteller rather than a judging teacher, and a modelled answer always follows so the learner is never left hanging.

The hypotheses are influential but contested — critics (McLaughlin, Gass et al.) call them hard to test, and newer neuroscience emphasizes interaction and embodiment alongside input (which, notably, points back toward TPR-style methods — which Spokn also uses). Where Spokn deliberately goes beyond a strict reading of Krashen: it prompts early speaking (30–40 spoken responses per story) — closer to Blaine Ray's confidence-building rationale than to a strict silent-period reading of Krashen. The honest framing is input-first with confidence-building output, not "pure Krashen."