Top-down feedback does not benefit speech recognition; on the contrary, it can hinder it. No experimental data imply that feedback loops are required for speech recognition. Feedback is accordingly unnecessary and spoken word recognition is modular. To defend this thesis, we analyse lexical involvement in phonemic decision making. TRACE (McClelland & Elman 1986), a model with feedback from the lexicon to prelexical processes, is unable to account for all the available data on phonemic decision making. The modular Race model (Cutler (...) & Norris 1979) is likewise challenged by some recent results, however. We therefore present a new modular model of phonemic decision making, the Merge model. In Merge, information flows from prelexical processes to the lexicon without feedback. Because phonemic decisions are based on the merging of prelexical and lexical information, Merge correctly predicts lexical involvement in phonemic decisions in both words and nonwords. Computer simulations show how Merge is able to account for the data through a process of competition between lexical hypotheses. We discuss the issue of feedback in other areas of language processing and conclude that modular models are particularly well suited to the problems and constraints of speech recognition. Key Words: computational modeling; feedback; lexical processing; modularity; phonemic decisions; reading; speech recognition; word recognition. (shrink)

Although researchers studying human speech recognition (HSR) and automatic speech recognition (ASR) share a common interest in how information processing systems (human or machine) recognize spoken language, there is little communication between the two disciplines. We suggest that this lack of communication follows largely from the fact that research in these related fields has focused on the mechanics of how speech can be recognized. In Marr's (1982) terms, emphasis has been on the algorithmic and implementational levels rather than on the (...) computational level. In this article, we provide a computational-level analysis of the task of speech recognition, which reveals the close parallels between research concerned with HSR and ASR. We illustrate this relation by presenting a new computational model of human spoken-word recognition, built using techniques from the field of ASR that, in contrast to current existing models of HSR, recognizes words from real speech input. (shrink)

Although researchers studying human speech recognition (HSR) and automatic speech recognition (ASR) share a common interest in how information processing systems (human or machine) recognize spoken language, there is little communication between the two disciplines. We suggest that this lack of communication follows largely from the fact that research in these related fields has focused on the mechanics of how speech can be recognized. In Marr's (1982) terms, emphasis has been on the algorithmic and implementational levels rather than on the (...) computational level. In this article, we provide a computational-level analysis of the task of speech recognition, which reveals the close parallels between research concerned with HSR and ASR. We illustrate this relation by presenting a new computational model of human spoken-word recognition, built using techniques from the field of ASR that, in contrast to current existing models of HSR, recognizes words from real speech input. (shrink)

Language production and comprehension are intimately interrelated; and models of production and comprehension should, we argue, be constrained by common architectural guidelines. Levelt et al.'s target article adopts as guiding principle Ockham's razor: the best model of production is the simplest one. We recommend adoption of the same principle in comprehension, with consequent simplification of some well-known types of models.

Comparisons of indirect measures with direct measures can help elucidate the relationship between nonconscious and conscious perception. We report three experiments on masked word priming in which we observed a negative correlation between prime discriminability and priming , i.e. where priming decreased with increasing prime visibility. While such observations are rare , they may indicate a conflict between conscious and nonconscious processing when primes are shown close to the subjective visibility threshold for the priming-relevant information. For instance, such a conflict (...) could occur between nonconscious processing of a prime’s meaning and conscious perception of prime letters. Theoretical accounts that discuss similar conflicts assume that the conflict is resolved either by automatically prioritising conscious processes or by discounting the estimated confusion caused by a prime-target pair . In both cases, priming is predicted to decrease when prime visibility increases from below threshold to perithreshold levels. Therefore, we suggest that negative priming-d’ relationships are most likely observed when the d’ measure assesses prime visibility at a level of representation that is below the level of representation at which priming arises, in terms of a putative hierarchy of word processing. (shrink)

The central thesis of our target article is that feedback is never necessary in spoken word recognition. In this response we begin by clarifying some terminological issues that have led to a number of misunderstandings. We provide some new arguments that the feedforward model Merge is indeed more parsimonious than the interactive alternatives, and that it provides a more convincing account of the data than alternative models. Finally, we extend the arguments to deal with new issues raised by the commentators (...) such as infant speech perception and neural architecture. (shrink)

We agree with Frost that the variety of orthographies in the world's languages complicates the task of Frost suggests that orthographic processing must therefore differ between orthographies. We suggest that the same basic orthographic processes are applied to all languages. Where languages differ is in what the reader must do with the results of orthographic processing.

Rather than worrying about Bayesian Fundamentalists, I suggest that our real concern should be with Mechanistic Fundamentalists; that is, those who believe that concrete, but frequently untestable mechanisms, should be at the heart of all cognitive theories.