We discuss two inconsistencies between Levelt et al.'s model and existing theory about, and data from, picture-word experiments. These inconsistencies show that the empirical support for weaver++ is much weaker than Levelt et al. suggest. We close with a discussion of the proposed explanation for phonological coactivation of near synonyms of a target word.

Speakers retrieve words to use them in sentences. Errors in incorporating words into sentential frames are revealing with respect to the lexical units as well as the lexical retrieval mechanism; hence they constrain theories of lexical access. We present a reanalysis of a corpus of spontaneously occurring lexical exchange errors that highlights the contact points between lexical and sentential processes.

A highly constrained model of phonological encoding, which is minimally affected by the activation levels of alternative morphemes and phonemes, is proposed. In particular, weaver++ predicts only facilitatory effects on word production of the prior activation of form related words. However, inhibitory effects of form priming that suggest that a more strongly competitive system is required exist in the literature.

We discuss two key assumptions of Levelt et al.'s model of lexical retrieval: (1) the nondecompositional character of concepts and (2) lemmas as purely syntactic representations. These assumptions fail to capture the broader role of lemmas, which we propose as that of lexical–semantic representations binding (compositional) semantics with phonology (or orthography).

The commentaries provide a multitude of perspectives on the theory of lexical access presented in our target article. We respond, on the one hand, to criticisms that concern the embeddings of our model in the larger theoretical frameworks of human performance and of a speaker's multiword sentence and discourse generation. These embeddings, we argue, are either already there or naturally forgeable. On the other hand, we reply to a host of theory-internal issues concerning the abstract properties of our feedforward spreading activation model, which functions without the usual cascading, feedback, and inhibitory connections. These issues also concern the concrete stratification in terms of lexical concepts, syntactic lemmas, and morphophonology. Our response stresses the parsimony of our modeling in the light of its substantial empirical coverage. We elaborate its usefulness for neuroimaging and aphasiology and suggest further cross-linguistic extensions of the model.

This target article discusses the verbal working memory system used in sentence comprehension. We review the concept of working memory as a short-duration system in which small amounts of information are simultaneously stored and manipulated in the service of accomplishing a task. We summarize the argument that syntactic processing in sentence comprehension requires such a storage and computational system. We then ask whether the working memory system used in syntactic processing is the same as that used in verbally mediated tasks that involve conscious controlled processing. Evidence is brought to bear from various sources: the relationship between individual differences in working memory and individual differences in the efficiency of syntactic processing; the effect of concurrent verbal memory load on syntactic processing; and syntactic processing in patients with poor short-term memory, patients with poor working memory, and patients with aphasia. Experimental results from these normal subjects and patients with various brain lesions converge on the conclusion that there is a specialization in the verbal working memory system for assigning the syntactic structure of a sentence and using that structure in determining sentence meaning that is separate from the working memory system underlying the use of sentence meaning to accomplish other functions. We present a theory of the divisions of the verbal working memory system and suggestions regarding its neural basis.

We argue that if a different definition of sentence complexity is adopted and processing capacity is assessed in a way that is consistent with that definition, then the Caplan & Waters distinction between interpretive versus postinterpretive processing is unnecessary insofar that it applies to the thematic role assignment in relative-clause sentences.

A distinction between interpretive processing and post-interpretive processing calls for a consideration of interface relations in systems of verbal memory. Syntactic movement of a phrase and the cognitive system of thought/mind interact. Systems of declarative memory and procedural memory interact.

Normals display selective deficits in morphology and syntax under adverse processing conditions. Digit loads do not impair processing of passives and object relatives but do impair processing of grammatical morphemes. Perceptual degradation and temporal compression selectively impair several aspects of grammar, including passives and object relatives. Hence we replicate Caplan & Waters's specific findings but reach opposite conclusions, based on wider evidence.

We agree with Caplan & Waters that there are problems with the single-resource theory of sentence comprehension. However, we challenge their dual-resource alternative on theoretical and empirical grounds and point to a more coherent solution that abandons the notion of working memory resources.

A separation between interpretive and post-interpretive processes is central to Caplan & Waters's theory of language comprehension. This commentary raises some issues that are intended to help sharpen the distinction.

This commentary considers how far one can go in making inferences about functional modularity or segregation, based on the sorts of analyses used by Caplan & Waters in relation to the underlying neuronal infrastructure. Specifically, an attempt is made to relate the “functionalist” approach adopted in the target article to “neuroreductionist” perspectives on the same issue.

We discuss several issues raised by Caplan & Waters's distinction between interpretative and post-interpretative processes in sentence comprehension, including the nature and properties of the two systems, problems with measuring their respective capacities, and the relationship between the hypothesized separate-language-interpretation-resource (SLIR) and the general verbal working memory system that supports post-interpretive processing.

Individuals may differ in the general-attention executive component or in the subordinate domain-specific “slave” components of working memory. Tasks requiring sustained memory representations across attention shifts are reliable, valid indices of executive abilities. Measures emphasizing specific processing skills may increase reliability within restricted samples but will not reflect the attention component responsible for the broad predictive validity of span tasks.

Caplan & Waters's neuropsychological evidence for two types of verbal working memory rests entirely on a very restricted definition of “syntactic complexity,” one in terms of word order. This opens the possibility that the dissociation they observe relates to the differential use of word-order strategies rather than to the structure of verbal working memory.

In challenging current conceptions of the role of working memory in sentence processing, Caplan & Waters consider studies comparing young and older adults on sentence processing. This commentary raises two challenges to Caplan & Waters's conclusions: first, working memory tasks appear to be age invariant. Second, the production of complex syntactic constructions appears not to be age invariant.

Even though there is ample evidence from the sentence- comprehension literature for specialized working memory systems in normal and patient populations, some open questions remain. One of them is an explanation for a missing “post-interpretive” processing deficit in a variety of accuracy-judgment tasks in an aphasic patient with a severe verbal working memory problem.

Scalp-recorded event-related potential (ERP) measures of reading and listening have been proved more sensitive to the time course of syntactic processing than the chronometric and behavioral data described by Caplan & Waters. ERP studies using sentences containing relative clauses indicate that there are individual differences in syntactic processing that appear at the earliest theoretically relevant time points and are attributable to working memory operations.

A promising approach to more refined models consistent with the Caplan & Waters hypothesis is based on similarity-based interference, a general principle that applies across working memory domains. This may explain both the fine details of syntactic working memory phenomena and the gross fractionation for which Caplan & Waters have found evidence. Detailed models of syntactic processing that embody similarity-based interference fare well cross-linguistically.

Although the working memory capacity involved in syntactic processing may be separate from the capacity involved in word list recall, other aspects of initial sentence interpretation appear to depend on some of the same capacities tapped by span tasks. Specifically, there appears to a capacity for lexical–semantic retention involved in both sentence comprehension and span measures.