We presented three studies documenting that 5- to 6-year-old English-speaking children and adults are indeed both sensitive to and tolerant of violations of informativeness, and that this holds with scalar and non-scalar expressions to the same extent. We argue that this hitherto ignored tendency towards pragmatic tolerance is a potentially significant factor in previous studies that concluded that young children lack some important aspect of pragmatic competence. We do not deny that other factors proposed in the literature also influence whether participants reject

buy SCH727965 underinformative utterances. Processing demands (Pouscoulous et al., 2007), the presentation of a specific context against which utterances are evaluated (Guasti et al., 2005) and drawing attention to being informative

(Papafragou & Musolino, 2003) have been suggested as relevant considerations for children (and the first two for adults as well). Indeed, we would suggest that some of these factors may interact with pragmatic tolerance, e.g. when in a given Anti-diabetic Compound Library task it is particularly important to be informative. In this case we might expect participants to treat pragmatic violations as gravely as logical ones. This could include cases of explicit intervention, in which children are trained to correct underinformative descriptions (Papafragou & Musolino, 2003, experiment 2; Guasti et al., 2005, experiment 2) or cases where the question asked highlights a certain contrast, for example if Mr. Caveman were asked ‘Did the mouse pick up all the carrots?’ instead of ‘What did the mouse pick up? Turning to the relation between the sensitivity to informativeness and actual implicature derivation, we believe that it is possible to disentangle whether participants are competent with one or the other, PDK4 but not in judgement tasks or sentence-to-picture-matching paradigms. Implicature derivation can be tapped by paradigms that involve the participant operating

on a situation to make it match their interpretation of the critical utterances, rather than evaluating whether the utterances are an adequate description of the given situation. This holds because utterances can be characterised as underinformative only if they are presumed to be describing an existing situation. We are currently exploring this avenue based on the action-based paradigm developed by Pouscoulous et al. (2007, experiment 3). We do not claim that children’s mastery of informativeness and implicature derivation must develop in tandem. As the former is a prerequisite for the latter, the latter is likely to be psycholinguistically more demanding.