Nick C. Ellis
2002 Studies in Second Language Acquisition  
This article shows how language processing is intimately tuned to input frequency. Examples are given of frequency effects in the processing of phonology, phonotactics, reading, spelling, lexis, morphosyntax, formulaic language, language comprehension, grammaticality, sentence production, and syntax. The implications of these effects for the representations and developmental sequence of SLA are discussed. Usage-based theories hold that the acquisition of language is exemplar based. It is the
more » ... cemeal learning of many thousands of constructions and the frequency-biased abstraction of regularities within them. Determinants of pattern productivity include the power law of practice, cue competition and constraint satisfaction, connectionist learning, and effects of type and token frequency. The regularities of language emerge from experience as categories and prototypical patterns. The typical route of emergence of constructions is from formula, through low-scope pattern, to construction. Frequency plays a large part in explaining sociolinguistic variation and language change. Learners' sensitivity to frequency in all these domains has implications for theories of implicit and explicit learning and their interactions. The review concludes by considering the history of frequency as an explanatory concept in theoretical and applied linguistics, its 40 years of exile, and its necessary reinstatement as a bridging variable that binds the different schools of language acquisition research. 145 in the lexicon isolated from "core" grammatical information; rather, it is relevant at all stages of lexical, syntactic, and discourse comprehension. Comprehenders tend to perceive the most probable syntactic and semantic analyses of a new utterance on the basis of frequencies of previously perceived utterance analyses. Language users tend to produce the most probable utterance for a given meaning on the basis of frequencies of utterance representations. These accounts readily contribute to explanations of sociolinguistic variation. The review discusses how these effects can be simulated using mathematical or computational models. The effects of frequency in input are modulated by the need to simultaneously satisfy the constraints of all other constructions that are represented in the learner's system. The interactions of input and existing representation can be described as Bayesian interactions in a rich network of interacting associations and connections, some competing, others mutually reinforcing as a result of the many redundancies of language and representation. Recent work shows that in syntax, as in phonology, the productivity of pattern depends on type frequency of the construction. The implications for theories of SLA are described: a developmental sequencefrom formula, through low-scope pattern, to construction-is proposed as a useful starting point to investigate the emergence of constructions and the ways in which type and token frequency affect the productivity of patterns. The review finishes with consideration of the consequences of exemplar views of language for theories of implicit and explicit learning. To the extent that language processing is based on frequency and probabilistic knowledge, language learning is implicit learning. This does NOT deny the importance of noticing (Schmidt, 1993) in the initial registration of a pattern-recognition unit. NOR does it deny a role for explicit instruction. Language acquisition can be speeded by explicit instruction. The last 20 years of empirical investigations into the effectiveness of L2 instruction demonstrate that focused L2 instruction results in large target-oriented gains, that explicit types of instruction are more effective than implicit types, and that the effectiveness of L2 instruction is durable. An outline is given of the mechanisms by which explicit knowledge affects implicit learning. FREQUENCY LEARNING Humans are sensitive to the frequencies of events in their experience. Ask them to make explicit judgments from memory about the relative frequency with which things happen and they are typically pretty good at it. College students, for example, can accurately estimate the frequency with which words occur in a list (Hasher & Chromiak, 1977) and can also estimate, with a high correlation to actual counts, the frequency of English words (Shapiro, 1969) , of single letters (Attneave, 1953) , and even of pairs of letters (Underwood, 1971) . There are few individual differences in these abilities; children are just about as good as young adults, and this ability remains robust in the face of
doi:10.1017/s0272263102002024 fatcat:homsmm7odjhxri66jcxz72uwva