On the role of competing word units in visual word recognition: the neighborhood frequency effect

Current models of word recognition generally assume that word units orthographically similar to a stimulus word are involved in the visual recognition of this word. We refer to this set of orthographically similar words as an orthographic neighborhood. Two experiments are presented that investigate the ways in which the composition of this neighborhood can affect word recognition. The data indicate that the presence in the neighborhood of at least one unit of higher frequency than the stimulus word itself results in interference in stimulus word processing. Lexical decision latencies (Experiment 1) and gaze durations (Experiment 2) to words with one neighbor of higher frequency were significantly longer than to words without a more frequent neighbor. This neighborhood frequency effect is discussed in terms of the different types of candidate selection process postulated by contemporary models of visual word recognition.     

Neighborhood effects in spelling in adults

The participants were asked to spell aloud words for which there were either many orthographically similar words (a dense neighborhood) or few orthographically similar words (a sparse neighborhood). Words with a dense neighborhood were spelled faster and more accurately than were words with a sparse neighborhood. This finding is consistent with the hypothesis of Rapp, Epstein, and Tainturier (2002), that the cognitive spelling system has an interactive architecture incorporating feedback between individual graphemes and orthographic lexeme representations.     

A test of parafoveal‐on‐foveal effects with pairs of orthographically related words

One of the main controversies in the field of eye movements in reading concerns the question of whether the processing of two adjacent words in reading occurs in sequence, or in parallel. To distinguish between these views, the present experiment tested the presence of parafoveal‐on‐foveal effects with pairs of orthographically related words (or neighbours that differed by a single letter) in a controlled but reading‐like situation. Results revealed that fixation times on a foveal target word were shorter when the target was accompanied by an orthographically similar parafoveal word than when the parafoveal word was dissimilar. Furthermore, the size of the effect tended to vary with both the relative frequency of target and parafoveal words, and the position of the critical letter. These results were interpreted in the framework of a pure parallel processing hypothesis, where the processing of adjacent words is only limited by visual acuity, and the respective lexical properties of the foveal and parafoveal words.     

Guessing Strategies, Aging, and Bias Effects in Perceptual Identification

In the typical single-stimulus perceptual identification task, accuracy is improved by prior study of test words, a repetition priming benefit. There is also a cost, inasmuch as previously studied words are likely to be produced (incorrectly) as responses if the test word is orthographically similar but not identical to a studied word. In two-alternative forced-choice perceptual identification, a test word is flashed and followed by two alternatives, one of which is the correct response. When the two alternatives are orthographically similar, test words identical to previously studied items are identified more accurately than new words (a benefit) but tests words orthographically similar to studied words are identified less accurately than new words (a cost). Ratcliff and McKoon (in press) argue that these are bias effects that arise in the decision stage of word identification. We report five experiments that examined the alternative hypothesis that these bias effects arise from postperceptual guessing strategies. In single-stimulus perceptual identification, repetition priming benefits were equally great for young and older adults who claimed to use deliberate guessing strategies and those who did not (Experiment 1). In contrast, only groups of young and older people who claimed to deliberately guess studied words in a two-alternative forced-choice task (Experiments 2 and 5) showed reliable benefits and costs. Costs and benefits were abolished in the two-alternative forced-choice task when a very long study list was used, presumably because the increased retrieval burden made the use of deliberate guessing strategies less attractive (Experiment 3). Under conditions similar to those of Experiment 3, repetition priming was observed in single-stimulus perceptual identification (Experiment 4). These results are consistent with the view that costs and benefits in the forced-choice perceptual identification task arise from deliberate guessing strategies but that those in the single-stimulus task do not. The possibility that the observed relationship between strategy reports and priming effects reflects erroneous postexperimental assessments of strategies by participants is also considered.     

When do primes go bad? A corpus of orthographically related primes that inhibit fragment completion

The memory block effect (MBE) occurs when orthographically similar words inhibit retrieval. Previous studies have published 55 different stimuli that produce the MBE in word fragment completion. This small number of stimuli constrains experimental designs, presents serious obstacles for using neuroimaging to elucidate neural substrates of blocking, and raises concern that the MBE is limited to a particular group of words. A pool of 315 stimulus words was tested in a traditional MBE paradigm, and the results demonstrated that the MBE generalizes to other stimuli. This study also expands the number of stimuli that produce the MBE because 185 new stimuli produced blocking effects. As a result, the current list of 240 MBE stimuli can be used for word fragment research including cognitive neuroscience investigations of retrieval inhibition. A table of MBE stimuli is available in an archived appendix that can be downloaded from www.psychonomic.org/archive.     

Shared neighborhood effects in masked orthographic priming

Several studies have found effects of orthographically related masked nonword primes on lexical decisions to target words. These effects have been explained by the neighborhood characteristics of the target word (Forster, 1987), but the neighborhood characteristics of the prime in combination with the target are also found to be important (Hinton, Liversedge, & Underwood, 1998). In this study, we present a new account of masked form-priming effects based on the shared neighborhood of prime and target. Shared neighbors are words that are activated by both prime and target. According to the interactive activation model (McClelland & Rumelhart, 1981), shared neighborhood determines the size of priming effects. This prediction was tested and confirmed in a masked priming experiment that manipulated the shared neighborhood density of complete primes.     

Distributional analyses in auditory lexical decision: Neighborhood density and word-frequency effects

In the present article, the effects of phonological neighborhood density and word frequency in spoken word recognition were examined using distributional analyses of response latencies in auditory lexical decision. A density × frequency interaction was observed in mean latencies; frequency effects were larger for low-density words than for high-density words. Distributional analyses further revealed that for low-density words, frequency effects were reflected in both distributional shifting and skewing, whereas for high-density words, frequency effects were purely mediated by distributional skewing. The results suggest that word frequency plays a role in early auditory word recognition only when there is relatively little competition between similar-sounding words, and that frequency effects in high-density words reflect postlexical checking.     

Masked inhibitory priming in english: evidence for lexical inhibition

Predictions derived from the interactive activation (IA) model were tested in 3 experiments using the masked priming technique in the lexical decision task. Experiment 1 showed a strong effect of prime lexicality: Classifications of target words were facilitated by orthographically related nonword primes (relative to unrelated nonword primes) but were inhibited by orthographically related word primes (relative to unrelated word primes). Experiment 2 confirmed IA's prediction that inhibitory priming effects are greater when the prime and target share a neighbor. Experiment 3 showed a minimal effect of target word neighborhood size (N) on inhibitory priming but a trend toward greater inhibition when nonword foils were high-N than when they were low-N. Simulations of 3 different versions of the IA model showed that the best fit to the data is produced when lexical inhibition is selective and when masking leads to reset of letter activities.     

Differential cognitive processing of Kanji and Kana words: do orthographic and semantic codes function in parallel in word matching task.

Relative engagements of the orthographic and semantic codes in Kanji and Hiragana word recognition were investigated. In Exp. 1, subjects judged whether the pairs of Kanji words (prime and target) presented sequentially were physically identical to each other in the word condition. In the sentence condition, subjects decided whether the target word was valid for the prime sentence presented in advance. The results showed that the response times to the target swords orthographically similar (to the prime) were significantly slower than to semantically related target words in the word condition and that this was also the case in the sentence condition. In Exp. 2, subjects judged whether the target word written in Hiragana was physically identical to the prime word in the word condition. In the sentence condition, subjects decided if the target word was valid for the previously presented prime sentence. Analysis indicated that response times to orthographically similar words were slower than to semantically related words in the word condition but not in the sentence condition wherein the response times to the semantically and orthographically similar words were largely the same. Based on these results, differential contributions of orthographic and semantic codes in cognitive processing of Japanese Kanji and Hiragana words was discussed.     

Phonology as the source of syllable frequency effects in visual word recognition: Evidence from French

In order to investigate whether syllable frequency effects in visual word recognition can be attributed to phonologically or orthographically defined syllables, we designed one experiment that allowed six critical comparisons. Whereas only a weak effect was obtained when both orthographic and phonological syllable frequency were conjointly manipulated in Comparison 1, robust effects for phonological and null effects for orthographic syllable frequency were found in Comparisons 2 and 3. Comparisons 4 and 5 showed that the syllable frequency effect does not result from a confound with the frequency of letter or phoneme clusters at the beginning of words. The syllable frequency effect was shown to diminish with increasing word frequency in Comparison 6. These results suggest that visually presented polysyllabic words are parsed into phonologically defined syllables during visual word recognition. Materials and links may be accessed at www.psychonomic.org/archive.     

Orthographic neighborhood effects in perceptual identification and semantic categorization tasks: a test of the multiple read-out model

How should a word's orthographic neighborhood affect perceptual identification and semantic categorization, both of which require a word to be uniquely identified? According to the multiple read-out model (Grainger & Jacobs, 1996), inhibitory neighborhood frequency effects should be observed in these types of tasks, and facilitatory neighborhood size effects should not be. In Experiments 1 and 2 (perceptual identification), these effects were examined as a function of stimulus visibility (i.e., high vs. low visibility) to provide as full a test as possible of the model's predictions. In the high-visibility conditions, words with large neighborhoods were reported less accurately than words with small neighborhoods, but there was no effect of neighborhood frequency (i.e., whether the word had a higher frequency neighbor). In the low-visibility conditions, low-frequency words with large neighborhoods and low-frequency words with higher frequency neighbors showed superior identification performance. In the semantic categorization task (Experiment 3), words with large neighborhoods were responded to more rapidly than words with small neighborhoods, but there was no effect of neighborhood frequency. These results are inconsistent with two of the basic premises of the multiple read-out model--namely, that facilitatory neighborhood size effects are due to a variable response criterion (the sigma criterion), rather than to lexical selection processes, and that the lexical selection processes themselves produce an inhibitory neighborhood frequency effect (via the M criterion). Instead, the present results, in conjunction with previous findings, suggest that large neighborhoods (and perhaps higher frequency neighbors) do aid lexical selection.     

Neighborhood frequency does affect performance in the Reicher task: encoding or decision?

Three experiments using the Reicher task show that performance on low-frequency words is disrupted if the incorrect alternative forms a higher frequency word. This neighborhood frequency effect occurs for both energy and pattern masks and for different sets of items. When the upcoming word is primed and its accessibility is enhanced, the neighborhood frequency effect is eliminated. Experiments 4a and 4b tested the neighborhood frequency effect using a same-different task and a signal-detection analysis. Neighborhood frequency affected the decision criterion but not the sensitivity of the perceptual system. Experiment 5 showed that many words with a frequency in the range from 1 to 5 per million are not recognized out of context. These results are consistent with the hypothesis that frequency effects in the Reicher task are caused by a bias in the decision system and can be simulated with the stochastic activation-verification model.     

中文阅读中的字形与语音加工

该文报告两个探讨字形及语音加要于中文阅读中所起之作用的实验。基本实验作业是让被移动鼠标以阅读计算机屏幕逐字呈现之短文。计算机记录文中个别单字中呈现时间作为反映阅读理解即时加工的因变量。在每篇文章中段都有一关键字,用来观察在此字位置出现原字、同音字、形似字或无关字对阅读所造成的影响。结果发现,在阅读初期,同音字、形似字及无关字所造成的干扰效应,无论在效果大小及出现位置上并无差异。别外,干扰效应在同音     

Naming regular and exception words: Further examination of the effect of phonological dissension among lexical neighbours

Abstract

It is generally assumed that visual word recognition is accompanied by the activation of lexical representations corresponding to words orthographically similar to the target (neighbours). With regard to the pronunciation of their constituent units, these words can either converge with or diverge from the target pronunciation. The role of the frequency of the divergent pronunciations in print-to-sound conversion was examined in a naming experiment in which subjects pronounced regular and exception words. The results showed that naming latencies for exception words were affected by the orthographic similarity of the target with frequent phonologically divergent words (enemies). In a similar vein, regular words which include the letters G or C (whose pronunciations are contextually determined) and which are orthographically similar to words favouring an incorrect pronunciation of the letter took longer to pronounce than regular controls. A delayed naming experiment indicated that these differences were not attributable to the articulatory characteristics of the items. Finally, it also appeared that enemy frequency influenced naming latencies but not regularisation rates and regularisation latencies. The results are discussed within the framework of current dual-route and parallel distributed processing models of reading.     

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.     

A compilation of 800 word neighborhoods by frequency

A number of inconsistencies are evident in the literature examining word-neighborhood size and frequency effects. One reason for the inconsistency may be that there are no standardized materials and criteria used in the different studies. Each experimenter has devised his or her word neighborhoods using different criteria for neighborhood size and frequency. The purpose of the present study was to develop a standardized set of word neighborhoods. Eight hundred orthographic neighborhoods were constructed with 4- and 5-letter words. The word lists were devised relative to the key elements that have been identified in the literature: (1) target-word frequency, (2) number of words in the neighborhood, (3) number of words higher in frequency than the target word, (4) number of letter positions contributing to the neighborhood, and (5) summation of the frequency of all neighbors (providing a standard metric for high- vs. low-frequency neighborhoods).     

Evidence for right hemisphere phonology in a backward masking task

The extent to which orthographic and phonological processes are available during the initial moments of word recognition within each hemisphere is under specified, particularly for the right hemisphere. Few studies have investigated whether each hemisphere uses orthography and phonology under constraints that restrict the viewing time of words and reduce overt phonological demands. The current study used backward masking in the divided visual field paradigm to explore hemisphere differences in the availability of orthographic and phonological word recognition processes. A 20 ms and 60 ms SOA were used to track the time course of how these processes develop during pre-lexical moments of word recognition. Nonword masks varied in similarity to the target words such that there were four types: orthographically and phonologically similar, orthographically but not phonologically similar, phonologically but not orthographically similar and unrelated. The results showed the left hemisphere has access to both orthography and phonology early in the word recognition process. With more time to process the stimulus, the left hemisphere is able to use phonology which benefits word recognition to a larger extent than orthography. The right hemisphere also demonstrates access to both orthography and phonology in the initial moments of word recognition, however, orthographic similarity improves word recognition to a greater extent than phonological similarity.     

The dynamics of lexical competition during spoken word recognition

The sounds that make up spoken words are heard in a series and must be mapped rapidly onto words in memory because their elements, unlike those of visual words, cannot simultaneously exist or persist in time. Although theories agree that the dynamics of spoken word recognition are important, they differ in how they treat the nature of the competitor set-precisely which words are activated as an auditory word form unfolds in real time. This study used eye tracking to measure the impact over time of word frequency and 2 partially overlapping competitor set definitions: onset density and neighborhood density. Time course measures revealed early and continuous effects of frequency (facilitatory) and on set based similarity (inhibitory). Neighborhood density appears to have early facilitatory effects and late inhibitory effects. The late inhibitory effects are due to differences in the temporal distribution of similarity within neighborhoods. The early facilitatory effects are due to subphonemic cues that inform the listener about word length before the entire word is heard. The results support a new conception of lexical competition neighborhoods in which recognition occurs against a background of activated competitors that changes over time based on fine-grained goodness-of-fit and competition dynamics.     

Bias effects in word fragment completion in young and older adults

Young and older adults were tested on a word fragment completion task in which correct solutions were studied words, words orthographically similar to studied words, or new words. In Experiments 1 and 2, the standard production version of the word fragment completion task was used; older adults had reduced benefits of prior exposure to target words and slightly decreased costs. However, costs and benefits did not differ across age in a forced-choice version of the task (Experiment 3). At a behavioral level, the results are contrary to predictions that age differences in word fragment completion priming effects will be greater when there is a strong competitor for the correct solution and that age differences in both costs and benefits will be smaller for identification than for production tasks. Theoretical implications of these findings are considered.     

Orthographic similarity: The case of “reversed anagrams”

How orthographically similar are words such as paws and swap, flow and wolf, or live and evil? According to the letter position coding schemes used in models of visual word recognition, these reversed anagrams are considered to be less similar than words that share letters in the same absolute or relative positions (such as home and hose or plan and lane). Therefore, reversed anagrams should not produce the standard orthographic similarity effects found using substitution neighbors (e.g., home, hose). Simulations using the spatial coding model (Davis, Psychological Review 117, 713-758, 2010), for example, predict an inhibitory masked-priming effect for substitution neighbor word pairs but a null effect for reversed anagrams. Nevertheless, we obtained significant inhibitory priming using both stimulus types (Experiment 1). We also demonstrated that robust repetition blindness can be obtained for reversed anagrams (Experiment 2). Reversed anagrams therefore provide a new test for models of visual word recognition and orthographic similarity.