Traditional mechanistic accounts of language processing derive almost entirely from the study of monologue. Yet, the most natural and basic form of language use is dialogue. As a result, these accounts may only offer limited theories of the mechanisms that underlie language processing in general. We propose a mechanistic account of dialogue, the interactive alignment account, and use it to derive a number of predictions about basic language processes. The account assumes that, in dialogue, the linguistic representations employed by the interlocutors become aligned at many levels, as a result of a largely automatic process. This process greatly simplifies production and comprehension in dialogue. After considering the evidence for the interactive alignment model, we concentrate on three aspects of processing that follow from it. It makes use of a simple interactive inference mechanism, enables the development of local dialogue routines that greatly simplify language processing, and explains the origins of self-monitoring in production. We consider the need for a grammatical framework that is designed to deal with language in dialogue rather than monologue, and discuss a range of implications of the account.

Pickering & Garrod (P&G) claim that the automatic mechanisms that underlie language processing in dialogue are absent in monologue. We disagree with this claim, and argue that dialogue simply provides a different context in which the same basic processes operate.

I argue that alignment of linguistic representations and situation models in dialogue are qualitatively distinct. By virtue of the isomorphy between interlocutors' linguistic representations, interlocutors align their linguistic representations fully. However, evidence about situation models is indirect and mediated through language, with the result that alignment of situation models is only partial.

Pickering & Garrod's (P&G's) call to study language processing in dialogue context is an appealing one. Their interactive alignment model is ambitious, aiming to explain the converging behavior of dialogue partners via both intra- and interpersonal priming. However, they ignore the flexible, partner-specific processing demonstrated by some recent dialogue studies. We discuss implications of these data.

We agree with Pickering & Garrod's (P&G's) proposal that dialogue is an important empirical and theoretical test bed for models of language processing. However, we offer two cautionary notes. First, the enterprise will require explicit computational models. Second, such models will need to incorporate both joint and separate speaker and hearer commitments in ways that go beyond priming and alignment.

Pickering & Garrod (P&G) argue that contemporary models of language use are inadequate. This has resulted largely because of an experimental focus on monologue rather than dialogue. I agree with the need for increased experimentation that focuses on the interplay between production and comprehension. However, I have some concerns about the Interactive Alignment model that the authors propose.

Pickering & Garrod (P&G) describe a mechanism by which the situation models of dialog participants become progressively aligned via priming at different levels. This commentary attempts to characterize how alignment and routinization can be extended into the language acquisition domain by establishing links between alignment and joint attention, and between routinization and grammatical construction learning.

Pickering & Garrod's (P&G's) mechanistic theory of dialogue is a major advance for psycholinguistics. But the commitment to representational parity in production and comprehension is problematic. Recent research suggests that speakers frequently produce a structure that listeners find ungrammatical and have trouble understanding. If the grammars of the two systems are different, then the assumption of representational parity must be relaxed.

Pickering & Garrod's (P&G's) theory of dialogue production cannot completely explain recent data showing that when interactants in referential communication tasks have different views of a physical space, they accommodate their language to their partner's view rather than mimicking their partner's expressions. Instead, these data are consistent with the hypothesis that interactants are taking the perspective of their conversational partners.

Pickering & Garrod's (P&G's) claim that conversationalists do not explicitly keep track of their interlocuters' information states is important. Nonetheless, via alignment, they seem to create a virtually symmetrical view of the information states of speaker and addressee – a key component of their accounts of collaborative utterances and of self-monitoring. As I show, there is significant evidence for intrinsic contextual misalignment between conversationalists that can persist across turns.

Pickering & Garrod (P&G) argue that language processing in dialogue is in principle easier than in monologue. Although dialogue situations may provide more opportunities for facilitative priming, those priming mechanisms are also available in monologue situations. In both cases, the interactive alignment model calls strict modular accounts of language processing into serious question.

Pickering & Garrod (P&G) deserve appreciation for their cogent argument that dialogue merits greater scientific consideration. Current models make little contact with behaviors of dialogue, motivating the interactive alignment theory. However, the theory is not truly “mechanistic.” A full account requires both representations and processes bringing those representations into harmony. We suggest that Grossberg's (1980) adaptive resonance theory may naturally conform to the principles of dialogue.

The interactive alignment model treats within- and between-individual co-ordination as essentially equivalent. It is argued that this leads to a conservative account of alignment that presupposes high levels of linguistic and conceptual co-ordination. Data from the maze task are used to argue that this approach is not sufficient to account for important co-ordination phenomena.

Pickering & Garrod's (P&G's) interactive alignment model explains the existence of alignment between speakers via an automatic priming mechanism. We propose that it may be preferable to explain alignment through processes of memory retrieval. Our discussion highlights how memory retrieval can produce the same results as the priming mechanism and presents data that favor the memory-based view.

This commentary argues that dialogue alignment can be explained if parsing-directed grammar formalisms are adopted. With syntax defined as monotonic growth of semantic representations as each word is parsed, alignment between interlocutors is shown to be expected. Hence, grammars can be evaluated according to relative success in characterizing dialogue phenomena.

Pickering & Garrod's (P&G's) mechanistic theory of dialogue attempts to detail the psychological processes involved in communication that are lacking in Clark's theory. By relying on automatic priming and alignment processes, however, the theory falters when it comes to explaining much of dialogic interaction. We argue for the inclusion of less automatic, though not completely conscious and deliberate, processes to explain such phenomena.

Pickering & Garrod (P&G) suggest that communicators synchronize their processing at a number of linguistic levels. Whereas their explanation suggests that representations are being compared across individuals, there must be some representation of all conversation participants in each participant's head. At the level of the situation model, it is important to maintain separate representations for each participant. At other levels, it seems less crucial to have a separate representation for each participant. This analysis suggests that different mechanisms may synchronize representations at different linguistic levels.

Dialogue requires ability beyond the production and comprehension of word strings. The interactive alignment account is good as far as it goes, but it must be embedded in a broader model encompassing alignment of paralinguistic representations.

Pickering & Garrod's (P&G's) interactive alignment account corresponds directly with the account Skinner (1957) gave in his book Verbal Behavior. This correspondence becomes evident when “properties of verbal stimuli” substitutes for “channels of alignment.” Skinner's account appears to have the dual advantages of requiring fewer basic terms and integrating the field of verbal behavior with the whole field of human behavior.

I share the authors' stance on the dialogic or interactional character of language. The authors, however, have left actual interaction out of their conception of dialogue. I sketch a number of organizations of practices of talking and understanding that supply the basic arena for talk-in-interaction. It is by reference to these that mechanisms for speech production and understanding need to be understood