Low-frequency Exception Words Research Articles

The consequences of progressive phonological impairment for reading aloud

The ‘primary systems’ view of reading disorders proposes that there are no neural regions devoted exclusively to reading, and therefore that acquired dyslexias should reliably co-occur with deficits in more general underlying capacities. This perspective predicted that surface dyslexia, a selective deficit in reading aloud ‘exception’ words (those with atypical spelling-sound characteristics), should be a consistent feature of semantic dementia, a progressive disorder of conceptual knowledge, and just such a pattern has been observed in previous research. In a similar vein, one might expect the gradual deterioration of phonological processing seen in the nonfluent forms of progressive aphasia to be accompanied by phonological dyslexia, a selective deficit in reading of unfamiliar letter strings, i.e., nonwords. The present study, reporting a case-series consideration of reading-aloud data from 16 progressive nonfluent aphasic patients, revealed a pattern in which both low-frequency exception word and nonword reading were comparably compromised. The severity of the reading disorder was predicted by scores on the expressive language task of picture naming but not the receptive task of spoken word-to-picture matching. Our hypothesis that a phonological deficit underpins diminished performance for both naming and reading was supported by the finding that reading-aloud performance was predicted specifically by the rate of phonological errors in picture naming. Moreover, the strength of this relationship was similar for low-frequency exception words and nonwords, suggesting that reading deficits for these two types of items in this disorder shared a common cause: a progressive impairment of phonological processing.

Localizing the Frequency × Regularity word reading interaction in the cerebral cortex

The aim of this work is to combine behavioural and functional magnetic resonance imaging (fMRI) data to advance our knowledge of where the Frequency × Regularity interaction on word naming is located in the cerebral cortex. Participants named high and low frequency, regular and exception words in a behavioural lab (Experiment 1) and during an fMRI study (Experiment 2). We used the Additive Factors Method (AFM) to localize the expected overadditive Frequency × Regularity interaction both temporally, through word naming reaction times (whereby low frequency exceptions produce the longest reaction times), and spatially on the cortex, through hemodynamic response measures from fMRI (whereby low frequency exceptions produce the highest activation intensities). Activation maps revealed significant activation for low frequency exception words in the supplementary motor association cortex (SMA). We interpret the SMA activation as increased articulatory preparation, given previous demonstrations of the SMA's involvement in motor programming. Hemodynamic time courses were extracted from four regions of interest: the middle temporal gyri, SMA, insula and the inferior frontal gyri. Importantly, hemodynamic intensities within the SMA displayed an overadditive interaction pattern parallel to that found with naming reaction times. Thus, we provide an application of the AFM to fMRI intensity measures and evidence that the SMA is a potential cortical source of the Frequency × Regularity interaction during a basic naming paradigm. While the AFM has traditionally been used to localize factors in time we provide evidence that the AFM is useful in understanding how variables influence one another in the brain.

SD-squared revisited: Reply to Coltheart, Tree, and Saunders (2010).

The connectionist triangle model of reading aloud proposes that semantic activation of phonology is particularly important for correct pronunciation of low-frequency exception words. Our consideration of this issue (Woollams, Lambon Ralph, Plaut, & Patterson, 2007) reported computational simulations demonstrating that reduction and disruption of this semantic activation resulted in the marked deficit in low-frequency exception word reading that is characteristic of surface dyslexia. We then presented 100 observations of reading aloud from 51 patients with semantic dementia (SD) demonstrating a universal decline into surface dyslexia, a phenomenon we termed "SD-squared." Coltheart, Tree, and Saunders have more recently provided a simulation of the SD-squared data within the dual route cascaded (DRC) model, achieved by varying the amount of damage to components of the lexical and nonlexical pathways. Although they suggested that these simulations provide a closer fit to the SD patients' reading data than our own, we demonstrate here that this is not the case. Moreover, we argue that the connectionist triangle model account has substantially greater explanatory and predictive power than the DRC account.

Pathway control in visual word processing: Converging evidence from recognition memory

The extent to which readers can exert strategic control over oral reading processes is a matter of debate. According to the pathway control hypothesis, the relative contributions of the lexical and nonlexical pathways can be modulated by the characteristics of the context stimuli being read, but an alternative time criterion model is also a viable explanation of past results. In Experiment 1, subjects named high- and low-frequency regular words in the context of either low-frequency exception words (e.g., pint) or nonwords (e.g., flirp). Frequency effects (faster pronunciation latencies for high-frequency words) were attenuated in the nonword context, consistent with the notion that nonwords emphasize the characteristics of the frequency-insensitive nonlexical pathway. Importantly, we also assessed memory for targets, and a similar attenuation of the frequency effect in recognition memory was observed in the nonword condition. Converging evidence was obtained in a second experiment in which a variable that was more sensitive to the nonlexical pathway (orthographic neighborhood size) was manipulated. The results indicated that both speeded pronunciation performance and memory performance were relatively attenuated in the low-frequency exception word context in comparision with the nonword context. The opposing influences of list context type for word frequency and orthographic neighborhood size effects in speeded pronunciation and memory performance provide strong support for the pathway control model, as opposed to the time criterion model.

Imageability x phonology interactions during lexical access

Although many studies have demonstrated the effects of imageability and phonological neighborhood size, few have examined if these factors interact. Strain, Patterson, and Seidenberg (1995) explained an imageability effect in naming low-frequency exception words (only) as being due to a slowing of orthographic-to-phonological mapping for these words, which allowed semantics to have an effect. Tyler, Voice, and Moss (2000) showed an interaction between imageability and phonological cohort size in word repetition. Westbury and Buchanan (2006) found an interaction between imageability and phonology using an auditory false memory paradigm that measured the false recognition rate for phonological associates of semantically primed words. They explained the finding in terms of a greater reliance of abstract than concrete words on phonological representations. In this paper we test three related hypotheses: that the imageability x phonology interaction should be modulated by modality; that measures of phonological processing fluency should predict the size of the interaction; and that concrete and abstract words should show a systematic difference in number of phonological neighbours. We find support for all three hypotheses, suggesting that the interaction between imageability and phonology reflects a difference in the representation of abstract and concrete words in the lexicon.

The neural basis of surface dyslexia in semantic dementia

Semantic dementia (SD) is a neurodegenerative disease characterized by atrophy of anterior temporal regions and progressive loss of semantic memory. SD patients often present with surface dyslexia, a relatively selective impairment in reading low-frequency words with exceptional or atypical spelling-to-sound correspondences. Exception words are typically 'over-regularized' in SD and pronounced as they are spelled (e.g. 'sew' is pronounced as 'sue'). This suggests that in the absence of sufficient item-specific knowledge, exception words are read by relying mainly on subword processes for regular mapping of orthography to phonology. In this study, we investigated the functional anatomy of surface dyslexia in SD using functional magnetic resonance imaging (fMRI) and studied its relationship to structural damage with voxel-based morphometry (VBM). Five SD patients and nine healthy age-matched controls were scanned while they read regular words, exception words and pseudowords in an event-related design. Vocal responses were recorded and revealed that all patients were impaired in reading low-frequency exception words, and made frequent over-regularization errors. Consistent with prior studies, fMRI data revealed that both groups activated a similar basic network of bilateral occipital, motor and premotor regions for reading single words. VBM showed that these regions were not significantly atrophied in SD. In control subjects, a region in the left intraparietal sulcus was activated for reading pseudowords and low-frequency regular words but not exception words, suggesting a role for this area in subword mapping from orthographic to phonological representations. In SD patients only, this inferior parietal region, which was not atrophied, was also activated by reading low-frequency exception words, especially on trials where over-regularization errors occurred. These results suggest that the left intraparietal sulcus is involved in subword reading processes that are differentially recruited in SD when word-specific information is lost. This loss is likely related to degeneration of the anterior temporal lobe, which was severely atrophied in SD. Consistent with this, left mid-fusiform and superior temporal regions that showed reading-related activations in controls were not activated in SD. Taken together, these results suggest that the left inferior parietal region subserves subword orthographic-to-phonological processes that are recruited for exception word reading when retrieval of exceptional, item-specific word forms is impaired by degeneration of the anterior temporal lobe.

Switch costs when reading aloud words and nonwords: Evidence for shifting route emphasis?

Reynolds and Besner (2005) examined contextual control over the use of lexical and nonlexical routes by requiring participants to alternate between reading pairs of low-frequency exception words and pairs of nonwords. Their main finding was that latencies for both words (e.g., wad) and nonwords (e.g.,flad) were slower when the immediately preceding trial involved the opposite item type rather than the same item type (a switch cost). The authors interpreted this result as evidence that under certain circumstances, readers have the ability to shift emphasis between their lexical and nonlexical routes. The present research shows that these results can be replicated using Reynolds and Besner's items; however, the switch cost for words, but not for nonwords, disappears when more easily named nonwords are used. This result suggests that Reynolds and Besner's results were likely due to something other than shifting route emphasis.

SD-squared: On the association between semantic dementia and surface dyslexia.

Within the connectionist triangle model of reading aloud, interaction between semantic and phonological representations occurs for all words but is particularly important for correct pronunciation of lower frequency exception words. This framework therefore predicts that (a) semantic dementia, which compromises semantic knowledge, should be accompanied by surface dyslexia, a frequency-modulated deficit in exception word reading, and (b) there should be a significant relationship between the severity of semantic degradation and the severity of surface dyslexia. The authors evaluated these claims with reference to 100 observations of reading data from 51 cases of semantic dementia. Surface dyslexia was rampant, and a simple composite semantic measure accounted for half of the variance in low-frequency exception word reading. Although in 3 cases initial testing revealed a moderate semantic impairment but normal exception word reading, all of these became surface dyslexic as their semantic knowledge deteriorated further. The connectionist account attributes such cases to premorbid individual variation in semantic reliance for accurate exception word reading. These results provide a striking demonstration of the association between semantic dementia and surface dyslexia, a phenomenon that the authors have dubbed SD-squared.

Commonalities and Differences in Word Identification Skills Among Learners of English as a Second Language

This study examined word identification skills among Chinese and Korean college students learning to read English as a second language in a naming experiment and an auditory category judgment task. Both groups demonstrated faster and more accurate naming performance on high‐frequency words than low‐frequency words and on regular words than exception words. Moreover, the difference in naming accuracies between regular and exception words was more pronounced for low‐frequency words than for high‐frequency words. A first language (L1) effect was observed: The Korean students were overall more accurate in naming all categories of words, showed a higher percentage of regularization errors in naming low‐frequency exception words, and were more accurate and faster in auditory meaning retrieval. These findings suggest that there are both common and unique processes in learning to read English for students from different L1 backgrounds.

Modulation of regularity and lexicality effects in reading aloud

We examined the question of whether the sizes of the regularity and lexicality effects in naming can be modulated as a function of filler type (nonwords or low-frequency exception words). The lexicality effect was larger in the exception word filler condition than in the nonword filler condition, but the size of the regularity effect was essentially unaffected by filler type. This pattern is at odds with what is generally assumed to be the predictions from dual-route theories of reading aloud. An attempt was next made to determine whether the dual-route cascaded model of Coltheart, Rastle, Perry, Langdon, and Ziegler (2001) could possibly simulate this pattern when changes were introduced to each of the three parameters that affect the contribution of the nonlexical route. We discuss the implications of these results for the idea that reliance on the lexical and nonlexical routes is under strategic control.

Priming and attentional control of lexical and sublexical pathways in naming: a reevaluation.

The authors report 3 naming experiments using J. D. Zevin and D. A. Balota's (2000) multiple prime manipulation. They used 2 sets of nonword primes (fast and slow) and low-frequency exception word primes to separate the effects of prime speed from those of prime type. The size of the regularity effect was unaffected by prime type. Relative to the low-frequency exception word prime condition, the frequency effect was reduced in the fast, but not in the slow, nonword prime condition. Lexicality effect size was reduced in both nonword prime conditions, a result consistent with the lexical checking strategy described by S. J. Lupker, P. Brown, and L. Colombo (1997). The authors suggest that these results are better explained in terms of S. J. Lupker et al.'s time-criterion account than J. D. Zevin and D. A. Balota's pathway control hypothesis.

Effects of filler type in naming: change in time criterion or attentional control of pathways?

We report two naming experiments examining the effects of filler type on the size of regularity and frequency effects. Low-frequency exception words were used as one filler type in both experiments. Their effects were contrasted with the effects of nonword fillers (Experiment 1) and low-frequency regular word fillers (Experiment 2). In both experiments, the size of the regularity effect was unaffected by the filler type manipulation. In contrast, the frequency effect interacted with filler type such that relative to the low-frequency exception filler environment, the size of the frequency effect was reduced in the environment of low frequency regular word fillers, but not in the environment of nonword fillers. These results appear to be better explained in terms of Lupker, Brown, and Colombo's (1997) flexible time-criterion framework than in terms of a pathway control hypothesis (e.g., Zevin & Balota, 2000).

A similarity in word-recognition procedures among second language readers with different first language backgrounds

This study investigated word recognition among fluent readers of English as a second language (ESL). Specifically, the study examined whether ESL readers' first language (L1) affects the procedures underlying second language word recognition, with respect to the effects of word frequency and regularity on word recognition. The results revealed a similarity in word-recognition procedures between fluent ESL readers with various L1 backgrounds (i.e., Chinese, Japanese, and Persian). In processing high-frequency words, all the ESL groups recognized exception words as quickly as regular words; low-frequency exception words, on the contrary, took longer to recognize than low-frequency regular words.

Reply to Strain, Patterson, and Seidenberg (2002).

E. Strain, K. E. Patterson, and M. S. Seidenberg (2002) reported an effect of imageability and a RegularityImageability interaction in a regression analysis of naming latencies to 120 words. One of their items (couth) was named correctly by just 5 of their 24 participants, and its reaction time was an outlier on their distribution. When that single item is removed, the significant predictors are age of acquisition (AoA), word frequency, regularity, and length. Analyses of the combined data from J. Monaghan and A. W. Ellis's (2002) Experiments 1-3 indicate that AoA predicts naming latencies for exception words but not consistent words. E. Strain et al.'s other points are considered in the light of these observations. Monaghan and Ellis (2002) sought to test some predictions and clarify some assertions concerning the impact of spelling-sound consistency on word naming. Some of Monaghan and Ellis's observations were of relevance to a study by Strain, Patterson, and Seidenberg (1995), who suggested that consistency, or regularity (we shall regard the two as interchangeable for present purposes), interacts with imageability such that the reading of low-frequency exception words is affected by imageability more than is the reading of low-frequency regular words. By the end of their own investigations, Monaghan and Ellis were less sanguine. Monaghan and Ellis took a factorial approach to exploring interactions, ma- nipulating spelling-sound consistency plus one other factor in each experiment (word frequency in Experiment 1, age of acquisition (AoA) in Experiment 2, and imageability in Experiment 3). Con- sistency interacted with frequency and AoA, but no interaction between consistency and imageability comparable to that claimed by Strain et al. (1995) was found. Strain, Patterson, and Seidenberg (2002) disputed the effect of AoA and its interaction with consis- tency in Monaghan and Ellis's Experiment 2. In addition, they queried the reality of the consistency effect in Monaghan and Ellis's Experiment 3 and hence the relevance of that experiment to the issue of whether consistency interacts with imageability. Strain et al. (2002) reported a new experiment in which naming response times (RTs) to 120 words were analyzed using simulta- neous multiple regression. The predictor variables were imageabil- ity, AoA, word frequency, positional bigram frequency, ortho- graphic neighborhood size (Coltheart's N), letter length, spelling- sound regularity (coded as exception words � 0 and regular words � 1), and an interactive variable generated by multiplying each word's regularity by its (transformed) imageability value. Finally, there were 10 variables taken from Treiman, Mullenix, Bijeljac-Babic, and Richmond-Welty (1995) that represented prop- erties of the initial phonemes of the spoken words. Strain et al.

Priming and attentional control of lexical and sublexical pathways during naming.

A modified priming task was used to investigate whether skilled readers are able to adjust the degree to which lexical and sublexical information contribute to naming. On each trial, participants named 5 low-frequency exception word primes or 5 nonword primes before a target. The low-frequency exception word primes should have produced a greater dependence on lexical information, whereas the nonword primes should have produced a greater dependence on sublexical information. Across 4 experiments, the effects of lexicality, regularity, frequency, and imageability were all modulated in predictable ways on the basis of the notion that the primes directed attention to specific processing pathways. It is argued that these results are consistent with an attentional control hypothesis.

Priming and attentional control of lexical and sublexical pathways during naming.

A modified priming task was used to investigate whether skilled readers are able to adjust the degree to which lexical and sublexical information contribute to naming. On each trial, participants named 5 low-frequency exception word primes or 5 nonword primes before a target. The low-frequency exception word primes should have produced a greater dependence on lexical information, whereas the nonword primes should have produced a greater dependence on sublexical information. Across 4 experiments, the effects of lexicality, regularity, frequency, and imageability were all modulated in predictable ways on the basis of the notion that the primes directed attention to specific processing pathways. It is argued that these results are consistent with an attentional control hypothesis.

Spelling-sound regularity effects on eye fixations in reading.

An interaction of word frequency and word regularity has typically been observed in naming and lexical decision experiments in which, in addition to an overall effect of word frequency, responses to low-frequency exception words are slower than those to low-frequency regular words, while no such difference occurs with high-frequency words. The only eye movement study to examine this effect in reading (Inhoff & Topolski, 1994) reported only transient effects of regularity. In the present experiment, we examined the frequency x regularity interaction using different stimuli than those of Inhoff and Topolski and also varied the parafoveal preview of the target word prior to fixation. When the preview was valid, the frequency x regularity interaction appeared. However, with an invalid preview, the effect of regularity disappeared. The results suggest that the activation of phonological codes is a very early component of reading.

Serial processing in reading aloud: No challenge for a parallel model.

K. Rastle and M. Coltheart (1999) challenged parallel models of reading by showing that the cost of irregularity in low-frequency exception words was modulated by the position of the irregularity in the word. This position-of-irregularity effect was taken as strong evidence of serial processing in reading. This article refutes Rastle and Coltheart's theoretical conclusions in 3 ways: First, a parallel model, the connectionist dual process model (M. Zorzi, G. Houghton, & B. Butterworth, 1998b), produces a position-of-irregularity effect. Second, the supposed serial effect can be reduced to a position-specific grapheme-phoneme consistency effect. Third, the position-of-irregularity effect vanishes when the experimental data are reanalyzed using grapheme-phoneme consistency as the covariate. This demonstration has broader implications for studies aiming at adjudicating between models: Strong inferences should be avoided until the computational models are actually tested.

Word reading in Alzheimer's disease: cross-sectional and longitudinal analyses of response time and accuracy data

Using a list of high- and low-frequency regular and exception words, we measured the reading performance of three groups of subjects: patients with mild dementia of the Alzheimer's type (DAT) at presentation (Stage 1) and 2–3 years later (Stage 2); moderately severe DAT patients; and control subjects. In the longitudinally studied patients, there was a dramatic increase in response times (RTs) from Stage 1 to Stage 2, but little change in either error rate or pattern. By contrast, a comparison of the Stage 2 with the Moderate patients revealed similar RTs but a significant increase in error rate for the Moderate group, particularly on low-frequency exception words. These two results for word naming were almost exactly mirrored by effects in picture naming. We conclude that correct word reading, especially for less common words with atypical spelling-sound correspondences, is compromised in DAT only when the disease produces a significant deterioration of semantic memory. These results are relevant to current theories about normal and impaired reading processes.

Semantic anomia without surface dyslexia

A computational model of oral reading developed by Plaut et al. (1996) proposes that reading aloud low-frequency exception words (e.g. yacht) relies upon a reading pathway that maps semantic representations directly onto output phonology. One prediction of this model is that, if the semantic reading pathway is damaged following brain damage, a pattern of surface dyslexic reading will be observed, characterized by frequent regularization errors when reading aloud low-frequency exception words (e.g. shoe → ‘show’). We report the oral reading of a semantic anomic patient (B.P.) who makes frequent semantic intrusion errors on confrontation naming tasks as well as semantic errors on word-picture and picture-picture matching tasks. Contrary to the prediction of the Plaut et al. (1996) model, B.P.'s oral reading of exception words, including his oral reading of picture names that are exception words, is intact. An important additional feature of B.P.'s language profile is that his oral reading of non-words is poor, suggesting that there is impairment to a phonological reading pathway. We argue that B.P.'s preserved oral reading reflects the operation of a lexical, non-semantic reading pathway.