Academic journal article Canadian Journal of Experimental Psychology

Neighbourhood Density Effects in Reading Aloud: New Insights from Simulations with the DRC Model

Academic journal article Canadian Journal of Experimental Psychology

Neighbourhood Density Effects in Reading Aloud: New Insights from Simulations with the DRC Model

Article excerpt

Abstract A series of nine simulations with the Dual Route Cascaded (DRC) model (Coltheart, Rastle, Perry, Langdon, & Ziegler, 2001) investigated neighbourhood density (N) effects in nonword and word naming. Two main finding emerged from this work. First, when naming nonwords there are two loci for the effect of N in the model, contrary to Coltheart et al.'s single locus explanation of what the model is doing. The early N effect involves interactive activation between the orthographic lexicon and the letter units such that high N facilitates letter identification, which in turn affects the nonlexical route. The late N effect arises from activation in the orthographic lexicon that feeds forward to the phonological lexicon and primes phonemes in the phoneme system. Second, when naming words the presence/absence of an effect of N on the Letter Units through feedback from the lexical level depends on the parameter settings. Implications and suggestions for future directions are made.

A recent trend is for theories to be elaborated computationally rather than verbally. Indeed, computational models are sometimes described as having two advantages over "verbal models" (Coltheart, Rastle, Perry, Langdon, & Ziegler, 2001; Jacobs & Grainger, 1994). The first is that implementing a computational model requires completeness; no assumptions can remain unspecified. The second is that the adequacy of the implemented model can be tested by comparing simulation results with human data. Under conditions where the model does not successfully simulate human performance, attempts can then be made to reformulate the model. A third advantage (capitalized on in the present work) is that at least some computational models can be dissected to assess how a factor affects performance.

As computational models become more complicated, with an increasing number of pathways and parameters, one has to ask how transparent they are. This issue is not new and has been raised in the context of models trained with learning algorithms (e.g., the Parallel Distributed Processing model of Plaut, McClelland, Seidenberg, & Patterson, 1996). As Coltheart et al. (2001) point out:

It is difficult to discover the functional architecture of a trained network for a network of realistic size - that is, to discover how the trained network has been structured by the learning algorithm so as to perform the task it has learned. (p. 205)

This problem is reduced (but not eliminated) in a localist model to the extent that the functional and structural architecture are identical (e.g., Coltheart et al., 2001; Grainger & Jacobs, 1996; McClelland & Rumelhart, 1981). However, such a relationship does not necessarily make transparent how an effect arises. For example, although layers in a localist network have specific functions, it is not always clear how two or more layers in a network will (or will not) interact to produce a particular effect (e.g., see Besner & Roberts, 2002, for a finding that many cognitive psychologists see as counterintuitive). Thus, until a careful analysis is carried out, claims concerning how an effect arises in a computational model should be treated with caution.

The Effect of Neighbourhood Density

The present paper explores, in a particular computational model, how lexical knowledge affects naming of both nonwords and words. The factor indexing lexical knowledge, Neighbourhood Density (N), which is defined as the number of words that can be created from a stimulus by replacing a single letter at a time (Coltheart, Davelaar, Jonasson, & Besner, 1977). Higher levels of N are associated with slower decisions for nonwords and faster decisions for words in the context of lexical decision (Andrews, 1989; Coltheart et al., 1977; Sears, Hino, & Lupker, 1995) and faster naming for both nonwords and words (Andrews, 1989; McCann & Besner, 1987; Peereman & Content, 1995; Sears et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.