I have often been struck by how highly fluent second language (L2) speakers of English can make errors in, say, possessive determiner gender agreement (e.g., Chinese, French, or Russian speakers saying “his” instead of “her”) without being disturbed at all by what they have said. To me, as a first language speaker of English, the error is extremely jarring and can disrupt understanding. For the L2 speaker, the error has much less impact. By contrast, an error in lexical reference (e.g., saying “boy's” instead of “girl's”) is generally experienced as jarring and potentially disruptive, even by L2 speakers. Why, then, do L2 speakers perceive errors in linking grammatical form to meaning so differently than errors in linking lexical units to meaning? Does this difference pose a challenge for L2 instruction and, if so, how should the challenge be met?

For a little over a decade now, VanPatten and his colleagues have systematically investigated this problem—namely, how to organize learning experiences to help learners make the appropriate connections between grammatical form and meaning. To accomplish this, VanPatten and his colleagues have advocated a method called Processing Instruction (PI). The edited volume reviewed here represents an important milestone in this program of work and will become a useful resource for SLA scholars interested in the problems of relating form and meaning in language.

The volume contains 17 chapters, with a healthy mix of theory, empirical research, and constructive commentaries, divided into five parts, plus links between parts. The volume lays out the principles that have guided research in PI, addresses questions raised in the past by critics, offers new empirical research, and presents some very thoughtful commentaries by leading scholars who are not necessarily part of the PI research community. The volume succeeds admirably in doing all this and thereby provides the reader with a full and balanced overview of the field.

The Processing Instruction referred to in the title is the name given to a pedagogical approach based on a particular model of L2 input processing. Here, input processing refers specifically to “the initial process by which learners connect grammatical forms with their meanings as well as how they interpret the roles of nouns in relationship to verbs” (p. 5). This process of making form-meaning connections is believed to reflect a set of perceptual biases that characterize all language learners. The problem is that these perceptual biases can sometimes get in the way of noticing the relationship between grammatical form and meaning. In this volume, VanPatten summarizes these biases in terms of nine perceptual principles (two major principles—Primacy of Meaning and the First Noun Principle—with six and three sub-principles, respectively). The goal of PI, therefore, is to focus attention on form-meaning connections by forcing learners to avoid using their natural processing inclinations. To accomplish this, PI typically involves activities that include both explicit instruction about the target form-meaning connection to be learned and exposure to structured input designed to deactivate the learner's usual perceptual biases (e.g., presenting examples in which the learner must focus on the form to extract the correct meaning).

A major focus of the reported research—in keeping with earlier studies—is to determine whether PI is more successful than traditional, non-PI methods, and whether both the explicit instruction and the structured input components of PI are equally necessary. New data are presented from learners of French, Spanish, and Italian regarding a variety of grammatical constructions (negatives, causatives, copula, subjunctives, future tense). The general finding is that PI is more successful than a control, non-PI instructional method and that the explicit instruction component of PI may not be necessary for PI to work, whereas the structured input component is necessary. Such findings generally replicate past results. As with previously published papers, however, these results continue to generate lively discussion about how to define so-called traditional instruction in the context of such research, about the extent to which this research is really about explicit versus implicit learning, whether the tests of learning are appropriate from an ecological point of view, and about whether the learner's systemic knowledge of the target language is changed by PI and, if so, how. The authors of the empirical chapters (Benati, Cheng, Farley, Fernandez, Sanz, VanPatten, Wong) often reply to past critics on these issues. The authors of the commentary chapters (Carroll, Collentine, Doughty, Harrington, Lee, Lightbown), among whom are some past critics, take these issues up again. Although it is unlikely that many of these controversies will be fully resolved by this volume, newcomers to the debate will find that the positions are well-stated, and some earlier misunderstandings over terminology and methodological issues are clarified.

As a cognitive psychologist working in SLA, I found this book to be informative, interesting, and thought-provoking. Some of the contributors make the valid and timely point that research into PI should now depart from its initial, somewhat narrow focus (e.g., the comparison of PI with one particular form of traditional teaching; the focus on explicit versus implicit learning) and further test the nature and limits of different learning approaches for promoting skill with form-meaning associations in the target language. To the ideas suggested in the volume, one could also add more basic cognitive questions. What conditions will promote automatic form-meaning processing, as opposed to generating metalinguistic knowledge? What are the mechanisms responsible for there being a special role for meaningful communication in form-meaning learning (i.e., can we go beyond just demonstrating that meaningful communication is crucial)? How do skills with form-meaning association fare in especially challenging environments (e.g., noisy contexts; situations placing heavy demands on attention and memory) and why? What relation is there, if any, between the skills targeted by PI research and oral fluency? Are there insights to be gained from looking at eye-movement measures, reaction time performance, and other more finely grained cognitive and perceptual measures?

Overall, this volume presents a well-rounded account of where PI research is today and provides some thoughtful observations on its accomplishments, limits, and future directions.