Our research on statistical language learning shows that infants, young children, and adults can compute, online and with remarkable speed, how consistently sounds co-occur, how frequently words occur in similar contexts, and the like, and can utilize these statistics to find candidate words in a speech stream, discover grammatical categories, and acquire simple syntactic structure in miniature languages. However, statistical learning is not merely learning the patterns presented in the input. When their input is inconsistent, children sharpen these statistics and produce a more systematic language than the one to which they are exposed. When input languages inconsistently violate tendencies that are widespread in human languages, learners shift these languages to be more aligned with language universals, and children do so much more than adults. These processes explain why children acquire language (and other patterns) more effectively than adults, and also may explain how systematic language structures emerge in communities where usages are varied and inconsistent. Most especially, they suggest that usage-based learning approaches must account for differences between adults and children in how usage properties are acquired, and must also account for substantial changes made by adult and child learners in how input usage properties are represented during learning.

While acknowledging that they have many points of theoretical agreement, the discussion deliberately focuses on areas of incompatibility between integrational and ‘distributed’ approaches to mind, cognition, and language in the light of Roy Harris’ claim that the notion of a ‘distributed mind’ comprises a category mistake. Harris’ position is based on his affirmation of a ‘vulgar concept of mind’ which contrasts sharply with certain accounts of mind originating from within cognitive science. The tension between lay and scientific understandings of mind and language forms a key point of discord between the two approaches. I discuss Harris’ argument that the category mistake inherent in the notion of ‘distributed mind’ can be dissolved by replacing it with the notion of an ‘integrated’ or ‘integrating’ mind. I then consider the derived notion of ‘distributed language’ and its theoretical and explanatory value from an integrational perspective. Finally, I conclude by arguing that although they share many important insights, the differences of perspective which exist between the integrational and distributed approaches are such as likely to prevent the emergence of anything resembling a unified movement.

The present study was concerned with the question whether comprehension is based on mental simulation processes beyond the word level. In Experiment 1, participants were presented with coherent sentence pairs, consisting of a context sentence and a target sentence. Target sentences ended with a word referring to an entity with a typical location in the upper vertical space (e.g., bird in There she sees a bird). Context sentences either supported the target entity’s typical location or not (Anna looks to the sky vs. Anna looks to the ground, respectively). Participants responded to the final word of the sentence pair by pressing an up- or a down-key. The results showed a main effect of response direction (faster up compared to down responses) as well as an interaction between context location and response direction. In Experiment 2, participants were presented with incoherent sentence pairs with the same context sentences and different target sentences (whereby the target word was kept identical), but in an incoherent manner (target sentence: On the poster one sees a bird). Here, the results showed a main effect of response direction but no interaction. The same result was obtained in Experiment 3, in which participants were presented with word pairs consisting of an up- or down-context word (e.g., sky vs. ground) and an up-target word (e.g., bird). Overall, the results provide evidence for the view that comprehension involves simulation processes at the word level as well as simulation processes at the sentence or discourse level.