The Interaction Hypothesis (Long, Gass) claims that interaction provides learners with opportunities to negotiate meaning and to obtain negative feedback:
It is proposed that the environmental contributions to acquisition are mediated by selective attention and the learner's developing L2 processing capacity, and that these resources are brought together most usefully, although not exclusively; during negotiation for meaning. Negative feedback obtained during negotiation work or elsewhere may be facilitative of L2 development […] and essential for learning certain specifiable L1-L2 contrasts.
(Long, 1996)
The study of feedback in second language can be dated to Hatch (1978) and has produced an impressive body of work. Research in this framework examines the use of confirmation checks, comprehension checks, clarification requests, reformulations and recasts in interaction between learners, and with proficient speakers.
However, particular contributions in a given interaction may be ambiguous: a teacher's recast, or reformulation of a learner's utterance may function as approval or a correction, and a learner's uptake, or reaction to feedback, may not be immediately identifiable, nor correspond to what was intended. As Gass admits:
It is difficult to demonstrate acquisition as a result of interaction. Learners do not always recognize feedback as intended, or immediately incorporate corrections in interaction if they do. It is possible that acquisitional effects of conversational interaction may be limited to low-level phenomena such as pronunciation or basic lexical meaning.
(Gass 2003)
Activities