Visual word recognition across orthographies: On the interaction between context and degradation.

Borowsky, Ron; Besner, Derek

doi:10.1037/0278-7393.17.2.272

Cited by 40 publications

(86 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The effect of introducing some form of degradation of the visual form of words, however, does not produce a consistent influence on performance. Visual degradation achieved through contrast reduction has been shown to be additive with word frequency (Borowsky & Besner, 1991, 1993, suggesting that frequency-sensitive processes are rather late in the chain of processing components leading to word recognition--at least late enough to miss all the action of early perceptual processes that establish a clear signal against variable noise background. Interestingly, in the same experiments that show additivity between contrast and frequency, there is clear evidence for much larger semantic priming effects under conditions of low visual quality, in line with the argument that slowing causes an increase of effects riding on RT.…”

Section: Theoretical Relevance For Priming Studiesmentioning

confidence: 99%

A linear mixed model analysis of masked repetition priming

2010

View full text Add to dashboard Cite

Section: Theoretical Relevance For Priming Studiesmentioning

confidence: 99%

A linear mixed model analysis of masked repetition priming

2010

View full text Add to dashboard Cite

“…This semantic priming effect sometimes interacts with the effect of stimulus quality. That is, a reduction in stimulus quality slows the processing of unrelated targets more than it slows the processing of related targets (e.g., Becker & Killion, 1977;Borowsky & Besner, 1991Meyer, Schvaneveldt, & Ruddy, 1975). There are also conditions under which these same two factors produce additive effects on lexical decision RT that are within the same range of RTs that showed an interaction in some of the former studies (Borowsky & Besner, 1991Brown & Besner, 2002;Stolz & Neely, 1995).…”

Section: The Joint Effects Of Semantic Priming and Stimulus Qualitymentioning

confidence: 99%

Parallel distributed processing and lexical-semantic effects in visual word recognition: Are a few stages necessary?

Borowsky

Besner

2006

Psychological Review

Self Cite

View full text Add to dashboard Cite

(2000) presented a parallel distributed processing model that purports to simulate human lexical decision performance. This model (and D. C. Plaut, 1995) offers a single mechanism account of the pattern of factor effects on reaction time (RT) between semantic priming, word frequency, and stimulus quality without requiring a stages-of-processing account of additive effects. Three problems are discussed. First, no evidence is provided that this model can discriminate between words and nonwords with the same orthographic structure and still produce the pattern of factor effects on RT it currently claims to produce. Second, the level of representation used by the model to make a lexical decision is inconsistent with what is known about how skilled readers with damage to their semantic system make word/nonword discriminations. Finally, there are a number of results that are difficult to reconcile with the single mechanism account. The authors' preference is to retain the stages-of-processing account.Keywords: stages-of-processing, parallel distributed processing (PDP), lexical-semantic priming, word frequency, stimulus quality OverviewThe idea that one can make inferences about the organization and nature of underlying mental processes by examining the pattern of factor effects on the time to respond in various tasks has a long and distinguished history in cognitive psychology. In particular, the observation that two or more factors have additive effects on reaction time (RT) has traditionally been interpreted as evidence that these different factors affect separate, serially organized stages of processing (Sternberg, 1969(Sternberg, , 1998; see also McClelland, 1979 for how cascaded processing can also yield additive effects of two factors on RT and Roberts & Sternberg, 1993 for limitations to this claim). The present commentary considers an account proposed by . They argue that a single mechanism embedded in an implemented parallel distributed processing (PDP) model with a sigmoid activation function correctly simulates the results of multifactor RT experiments in which the effects of some factors interact (e.g., A and B; B and C) whereas other factors have additive effects (e.g., A and C), without the need to attribute the effects of the additive factors (A and C) to separate stages. The analysis of Plaut and Booth's computational model provided here suggests that this single mechanism account fails in several ways. We therefore continue to prefer the traditional separable stages-of-processing view of how additive effects of factors on RT are best understood.The model is also potentially important for another reason. In earlier work, Seidenberg and McClelland (1989) claimed that their implemented PDP model, which was trained using all the monosyllabic words in English, produced lexical decision accuracy on a par with that seen by skilled human participants. Unfortunately, closer examination of the Seidenberg and McClelland model showed that its lexical decision accuracy was, in a particular context, actually qu...

show abstract

“…In word-identification tasks such as naming aloud or lexical decision, semantic priming of target words is enhanced if the targets are presented in a visually degraded form such as low contrast (e.g., Becker, 1979;Borowsky & Besner, 1991Stanovich & West, 1983). The typical form of this interaction is overadditive, whereby the influence of semantic priming is greater when performance is slowed by a lowcontrast stimulus.…”

mentioning

confidence: 99%

Modulation of additive and interactive effects in lexical decision by trial history.

Masson

Kliegl

2013

Journal of Experimental Psychology: Learning, Memory, and Cogni

View full text Add to dashboard Cite

Additive and interactive effects of word frequency, stimulus quality, and semantic priming have been used to test theoretical claims about the cognitive architecture of word-reading processes. Additive effects among these factors have been taken as evidence for discrete-stage models of word reading. We present evidence from linear mixed-model analyses applied to 2 lexical decision experiments indicating that apparent additive effects can be the product of aggregating over-and underadditive interaction effects that are modulated by recent trial history, particularly the lexical status and stimulus quality of the previous trial's target. Even a simple practice effect expressed as improved response speed across trials was powerfully modulated by the nature of the previous target item. These results suggest that additivity and interaction between factors may reflect trial-to-trial variation in stimulus representations and decision processes rather than fundamental differences in processing architecture.Keywords: additive and interactive effects, effects of trial history, lexical decision, linear mixed models Many formal models of word-reading processes assume the existence of separate processing modules that are responsible for computing different types of information (e.g., deriving orthography from visual input, computing phonology from orthographic patterns, selecting semantic information that corresponds to a presented word). A fundamental issue in the design of such models is the manner in which information is shared between the constituent modules. It is commonly assumed that input between processing modules is cascaded, so that partial information moves between modules even before a module has completed its operations (e.g., McClelland, 1979), and most current models of visual word identification include feedback from higher (e.g., lexical) to lower (e.g., letter) levels of processing (e.g., Coltheart, Rastle, Perry, Langdon, & Ziegler, 2001;Perry, Ziegler, & Zorzi, 2007;Plaut & Booth, 2000).These architectural features play an important role in allowing computational models to simulate some fundamentally important behavioral results involving word-identification tasks and, in particular, interactions between independent variables that influence the speed with which words are identified. In word-identification tasks such as naming aloud or lexical decision, semantic priming of target words is enhanced if the targets are presented in a visually degraded form such as low contrast (e.g., Becker, 1979;Borowsky & Besner, 1991Stanovich & West, 1983). The typical form of this interaction is overadditive, whereby the influence of semantic priming is greater when performance is slowed by a lowcontrast stimulus. In general, an overadditive interaction is one in which the simultaneous influence of two independent variables is larger than what would be expected by a simple summation of their individual effects. For example, unrelated primes and low stimulus quality both slow responding relative to related primes and high...

show abstract

Visual word recognition across orthographies: On the interaction between context and degradation.

Cited by 40 publications

References 17 publications

A linear mixed model analysis of masked repetition priming

A linear mixed model analysis of masked repetition priming

Parallel distributed processing and lexical-semantic effects in visual word recognition: Are a few stages necessary?

Modulation of additive and interactive effects in lexical decision by trial history.

Contact Info

Product

Resources

About