Exposure and input are often used interchangeably in the study of child bilingualism to describe the learner's interaction with her environment. Carroll (Carroll), in her thought-provoking keynote article, challenges the equation between amount of exposure and amount of input necessary for acquiring a language, and in particular a second language. Carroll, whose diverse research borrows from both generative and psychological models of learning, brings these two perspectives into her exploration of the role of exposure and input in language acquisition, calling for a clearer definition of each. Carroll argues that the input to the language acquisition mechanism is the abstract construct necessary in order to solve the learning problem, and it is not a function of the amount of exposure: very little input is necessary in order to solve a learning problem, e.g., identifying word boundaries, or acquiring a syntactic structure. While scrutinizing the equation often drawn between exposure and input, Carroll further argues that the amount of exposure further misses the quality of this exposure as well as the societal context in which this exposure occurs.

Carroll's proposed dissociation between exposure and input can further be used to disentangle emergent bilingualism from Specific Language Impairment (SLI). Within the generative framework, SLI is considered to be impairment in the innate linguistic representation. In the context of bilingualism, the term representation is used for intra-language – the representation of language which is influenced by exposure to more than one language. Yet, no one would propose that the bilingual representation is an impaired representation. Any attempt to disentangle bilingualism from SLI necessarily has to address the dissociation between exposure and input. Children with SLI are expected to show difficulties in identifying the abstract constructs relevant for acquiring the language due to impaired representation even with extended exposure. By contrast, bilingual children are expected to be able to identify the relevant input despite their limited exposure. Competition between the two languages might yield a difference from monolinguals due to variation in the exposure, or, to be more specific, due to intra-language effects.

Carroll's dissociation between exposure and input can suggest that the input necessary for acquiring syntactic features of a language is independent of the amount of exposure, and the ability to detect the relevant input is affected by the child's inner abilities (i.e., SLI), rather than environmental features (i.e., structures in the other language). This proposal corroborates recent findings which show that sentence repetition (SRep) tasks that challenge the syntactic abilities of the child are the most sensitive and specific tool for screening SLI in bilingual children and are less susceptible than lexical measures to the impact of length of exposure (LoE) and age of onset (AoO) (Thordardottir & Brandeker, Reference Thordardottir and Brandeker2013). Compared to other aspects of language such as morphology or the lexicon, syntactic complexity is least affected by LoE, but rather depends on the child's ability to identify the input relevant for acquiring a syntactic structure. This does not mean that bilingual children use syntactic structures flawlessly in their L2, but it means that the acquisition of syntax is much faster than that of other linguistic domains which are more sensitive to exposure.

Studies of exposure variables in bilingual acquisition show indeed that simultaneous bilingual children score within monolingual norms on SRep regardless of the relative amount of exposure to each language (Thordardottir & Brandeker, Reference Thordardottir and Brandeker2013). Yet, Chiat, Armon-Lotem, Marinis, Polisenska, Roy and Seeff-Gabriel (Reference Chiat, Armon-Lotem, Marinis, Polisenska, Roy, Seeff-Gabriel and Gathercole2013) found AoO and LoE effects for some sequential bilingual groups (Russian–Hebrew) but not others (Turkish–English and Russian–German). This ties up well with Carroll's observation that time related measures of exposure interact with cultural difference such as measures of identity and parental motivation and support. Chiat et al. (Reference Chiat, Armon-Lotem, Marinis, Polisenska, Roy, Seeff-Gabriel and Gathercole2013) further suggest that syntactic complexity does not affect bilinguals’ ability in SRep, but vocabulary does; and recommend using sentences that are syntactically age-adequate, but with a more restricted vocabulary.

A series of such tasks have been developed under the umbrella of LITMUS–SRep (Marinis & Armon-Lotem, Reference Marinis, Armon-Lotem, Armon-Lotem, de Jong and Meir2015). This tool targets syntactic complexity without tasking lexical access, under the assumption that the former is less sensitive to exposure in its broad sense than the latter. In a study of Arabic–French six-year-old children (Abboud, Tuller, Henry, Prévost & Saad, Reference Abboud, Tuller, Henry, Prévost and Saad2013) in Lebanon and France, it was found that SRep-French distinguished bilinguals with TLD from those with SLI in both settings, despite major sociolinguistic differences. Likewise, a study of Russian–Hebrew bilinguals (Armon-Lotem & Meir, in press) showed that SRep-Hebrew reached 91% of total diagnostic accuracy. Carroll's keynote provides a theoretical framework for understanding why LITMUS-SRep tasks can help disentangle bilingualism and SLI.