The role of lateral masking and orthographic structure in letter and word recognition.

The present experiments evaluated the contribution of orthographic structure and lateral masking in the perception of letter, word, and nonword test displays. Performance was tested in a backward recognition masking experiment in which a masking stimulus followed the test display after a variable blank interstimulus interval. In agreement with previous findings across different experiments, words were recognized better than single letterd at short interstimulus intervals, but the opposite was the case at long intervals. Performance on the nonwords resembled performance on letters at short masking intervals and performance on words at long masking intervals. The quantitative results were described by a processing model that incorporates the effects of lateral masking and orthographic structure in the dynamic processing of letter strings. Lateral masking tends to lower the potential perceptibility of letters whereas orthographic structure can reduce the uncertainty of the candidate letters in the letter sequence. The present model predicts that the quantitative contribution of each of these processes to performance is critically dependent upon the processing time available before the onset of the masking stimulus.     

Identification and pronunciation effects in a verbal reaction time task for words,pseudowords, and letters

The question addressed in this investigation was whether faster reading and pronunciation of words than orthographically regular pseudowords is due to faster identification or to faster programming and execution of the motor response. In Experiment I, three different response conditions (naming, threechoice signaled responding, and one-choice signaled responding) were employed to separate the identification and articulation processes in a verbal reaction time task. It was found that, for all intents and purposes, single, isolated letters are processed as if they were very short words. Words are read and pronounced 72 msec faster than pseudowords. Words are also pronounced 30 msec faster than pseudowords even if the reader has longer than 1 sec to identify the stimulus (three-choice condition) or to both identify the stimulus and preprogram the response (one-choice condition). The data indicate that words are identified about 52 msec faster and articulated about 30 msec faster than pseudowords. Since the number of response alternatives (one or three) does not interact with stimulus type (letter, word, or pseudoword) in the signaled response control condition, the 30-msec difference is due to response execution and not to differential response programming. Response programming takes in the neighborhood of 236 msec. Experiment 2 investigated the effect of local orthographic context upon the identification of the first letter of a string of letters. No difference was found in identifying the initial letter of words and pseudowords, but the initial letter of these orthographically regular letter strings was identified and named 10 msec faster than the initial letter of orthographically anomalous strings of letters (anagrams). The data from the two experiments are supportive of theories of reading that assume (1) that the letters of visually presented words are processed simultaneously, in parallel, (2) that there is a relatively direct access and retrieval of the phonological memory codes for the names of words, and (3) that orthographically regular pseudowords having no representation in the phonological lexicon undergo a grapheme-to-phoneme transformation that takes longer to finish than the direct spelling-to-sound process used for words.     

Processing of repeated letters in search and matching tasks

Several published experiments have used a variant of the matching or “same”-“different” paradigm in which the subject indicates whether or not all of the items in a visual array are the same. Contrary to the results of most paradigms in which subjects process multiple items, reaction time (RT) in this matching task is independent of the number of items in the display (N). One possible explanation of this independence holds that subjects respond to some overall property of the display, such as its general symmetry, and not to the individual items. I tested this hypothesis with a visual search task in which subjects searched a circular display of N=2, 4, or 6 letters for a specified target letter under three stimulus conditions. In the varied condition, the nontarget letters were all different from each other. In the repeated condition, the nontargets were repetitions of the same letter, but a single discrepant letter was always present in the display. The uniform condition also used repeated nontargets, but the negative (target-absent) displays contained N repetitions of a single nontarget item. I found that RT increased at a rate of 27 msec/item for the varied condition, but at only 13 and 8 msec/item in the repeated and uniform conditions, respectively. The slopes in the latter conditions were significantly lower than the slope for the varied condition, and they also differed significantly from each other. Hence, there was some advantage attributable to properties of the overall patterns associated with positive and negative responses, suggesting that at least some responses were based on these properties. There was a much larger advantage, however, attributable to the use of repeated nontargets, regardless of whether the overall patterns differed. This latter observation implies that the flat RT functions observed in matching studies must be interpreted primarily in terms of responses to individual items rather than to the pattern as a whole.     

Automatic letter priming in an alphabetic decision task

An alphabetic decision task was used to study effects of form priming on letter recognition at very short prime durations (20 to 80 msec). The task required subjects to decide whether a stimulus was a letter or a nonletter. Experiment 1 showed clear facilitatory effects of primes being either physically or nominally identical to the targets, with a stable advantage for the former. Experiment 2 demonstrated that uppercase letters are classified more rapidly as letters (vs. non-letters) when they are preceded by a briefly exposed, forward- and backward-masked, visually similar uppercase letter than when they are preceded by a visually dissimilar uppercase letter. Finally, Experiment 3 demonstrated that nominally identical and visually similar primes facilitate processing more than do nominally identical, visually dissimilar primes. The alphabetic decision task proved to produce sensitive and stable priming effects at the feature, letter, and response-choice level. The present results on letter-letter priming thus constitute a solid data base against which to evaluate other priming effects, such as word-letter priming. The results are discussed in light of current activation models of letter and word recognition and are compared with data simulated by the interactive activation model (McClelland & Rumelhart, 1981).     

Do transposed-letter similarity effects occur at a prelexical phonological level?

Nonwords created by transposing two letters (e.g., RELOVUTION) are very effective at activating the lexical representation of their base words (Perea & Lupker, 2004). In the present study, we examined whether the nature of transposed-letter (TL) similarity effects was purely orthographic or whether it could also have a phonological component. Specifically, we examined transposed-letter similarity effects for nonwords created by transposing two nonadjacent letters (e.g., relovución-REVOLUCION) in a masked form priming experiment using the lexical decision task (Experiment 1). The controls were (a) a pseudohomophone of the transposed-letter prime (relobución-REVOLUCION; note that B and V are pronounced as /b/ in Spanish) or (b) an orthographic control (relodución-REVOLUCION). Results showed a similar advantage of the TL nonword condition over the phonological and the orthographic control conditions. Experiment 2 showed a masked phonological priming effect when the letter positions in the prime were in the right order. In a third experiment, using a single-presentation lexical decision task, TL nonwords produced longer latencies than the orthographic and phonological controls, whereas there was only a small phonological effect restricted to the error data. These results suggest that TL similarity effects are orthographic--rather than phonological--in nature.     

Temporal integration of vibrotactile patterns

Three experiments were conducted to measure the temporal integration of vibrotactile patterns presented to the fingertip. In Experiment 1, letters were divided in half and the time between the onsets of the first half of the letter and second half of the letter, stimulus onset asynchrony (SOA), was varied. The recognizability of the letters declined as the SOA was increased from 9 to 100 msec. In Experiment 2, the time between two patterns constituting a masking stimulus was varied and the stimulus effectiveness in interfering with letter recognition was determined. The amount of masking increased as the SOA increased from 9 to 50 msec. In Experiment 3, the SOA between a letter and its complement (the portions of the tactile array not activated by the letter) was varied. Increasing SOA from 9 to approximately 50 msec led to increasingly accurate letter recognition. The results of the three experiments suggest that the skin is capable of complete temporal integration over a time period of less than 10 msec, and that the temporal integration function becomes asymptotic in 50 to 100 msec. The results also suggest that the onset of a vibrotactile pattern is critical for generating contours. The implications of the results for modes of generating tactile patterns and for temporal masking functions are discussed.     

Friends in Low-Entropy Places: Orthographic Neighbor Effects on Visual Word Identification Differ Across Letter Positions

Visual word recognition is facilitated by the presence of orthographic neighbors that mismatch the target word by a single letter substitution. However, researchers typically do not consider where neighbors mismatch the target. In light of evidence that some letter positions are more informative than others, we investigate whether the influence of orthographic neighbors differs across letter positions. To do so, we quantify the number of enemies at each letter position (how many neighbors mismatch the target word at that position). Analyses of reaction time data from a visual word naming task indicate that the influence of enemies differs across letter positions, with the negative impacts of enemies being most pronounced at letter positions where readers have low prior uncertainty about which letters they will encounter (i.e., positions with low entropy). To understand the computational mechanisms that give rise to such positional entropy effects, we introduce a new computational model, VOISeR (Visual Orthographic Input Serial Reader), which receives orthographic inputs in parallel and produces an over-time sequence of phonemes as output. VOISeR produces a similar pattern of results as in the human data, suggesting that positional entropy effects may emerge even when letters are not sampled serially. Finally, we demonstrate that these effects also emerge in human subjects' data from a lexical decision task, illustrating the generalizability of positional entropy effects across visual word recognition paradigms. Taken together, such work suggests that research into orthographic neighbor effects in visual word recognition should also consider differences between letter positions.     

Letters and words in word identification

Studies by Barron and Henderson (1977) and Johnson (1975) provide evidence that whole words may be the unit of identification in word perception, rather than single letters. Johnson found that words were matched faster than a letter to the first letter in a word. Barron and Henderson found faster matching times for words than for legal non-word items in a letter-matching task. These findings support the interpretation that words are identified before individual letters. If so, a word-frequency effect should be expected. Experiments 1 and 2 tested for word vs. first-letter-in-word differences, as well as for a word-frequency effect in simultaneous and delayed visual matching tasks. In the simultaneous task, first letters in words were matched faster than words. In the delayed task, there was no difference between matching words or matching the first letters in words. With both tasks there was a word-frequency effect for word matches but not for first-letter-in-word matches. In Experiment 3, first-letter matching time was unrelated to word frequency or lexical status, although it did vary with orthographic legality. These results, on the whole, are consistent with a race model in which identifications take place simultaneously at word, letter-cluster, and letter levels, rather than a sequential model in which the whole word is identified before the component letters.     

Task- and location-switching effects on visual attention

In two experiments, we examined the effects of task and location switching on the accuracy of reporting target characters in an attentional blink (AB) paradigm. Single-character streams were presented at a rate of 100 msec per character in Experiment 1, and successive pairs of characters on either side of fixation were presented in Experiment 2. On each trial, two targets appeared that were either white letters or black digits embedded in a stream of black letter distractors, and they were separated by between zero and five items in the stream (lags 1-6). Experiment 1 showed that report of the first target was least accurate if it immediately preceded the second target and if the two targets were either both letters or both digits (task repetition cost). Report of the second target was least accurate if one or two distractors intervened between the two targets (the U-shaped AB lag effect) and if one target was a letter and the other a digit (task switch cost). Experiment 2 added location uncertainty as a factor and showed similar effects as Experiment 1, with one exception. Lag 1 sparing (the preserved accuracy in reporting the second of two targets if the second immediately follows the first) was completely eliminated when the task required attention switching across locations. Two-way additive effects were found between task switching and location switching in the AB paradigm. These results suggests separate loci for their attentional effects. It is likely that the AB deficit is due mainly to central memory limitations, whereas location-switching costs occur at early visual levels. Task-switching costs occur at an intermediate visual level, since the present task switch involved encoding differences without changes in stimulus-response mapping rules (i.e., the task was character identification for both letters and digits).     

Visual search under conditions of very rapid sequential Input rates

The purposes of the present experiment were to provide information on rate of information processing in visual perception and to determine the degree to which the “sequential blanking” effect found by Mayzner. Tresselt, and Cohen (1966) constituted a limitation on rapid sequential input rates. A 10-channel tachisto scope was employed that permitted controlled durations of each of the 10 channels and the 9 inter channel intervals. The S’s task was one of visual search or detection in which he searched for a target letter among noise letters. A temporal interval forced-choice procedure was used. In addition to varying the rate at which letters were sequentially presented, various irregular temporal spatial orders of presentation of the letter sequences were employed and the direction and orientation of the display in the visual field was varied as was also the spacing between adjacent stimuli. No evidence of “sequential blanking” was found either in terms of the detection criterion or in the Ss’ phenomenal reports. Detection performance was as good at a rate of 2 1/2 msec per letter as it was at a rate of 30 msec per letter.     

Do transposed-letter similarity effects occur at a prelexical phonological level?

Nonwords created by transposing two letters (e.g., RELOVUTION) are very effective at activating the lexical representation of their base words (Perea & Lupker, 2004). In the present study, we examined whether the nature of transposed-letter (TL) similarity effects was purely orthographic or whether it could also have a phonological component. Specifically, we examined transposed-letter similarity effects for nonwords created by transposing two nonadjacent letters (e.g., relovución–REVOLUCIÓN) in a masked form priming experiment using the lexical decision task (Experiment 1). The controls were (a) a pseudohomophone of the transposed-letter prime (relobución–REVOLUCIÓN; note that B and V are pronounced as /b/ in Spanish) or (b) an orthographic control (relodución–REVOLUCIÓN). Results showed a similar advantage of the TL nonword condition over the phonological and the orthographic control conditions. Experiment 2 showed a masked phonological priming effect when the letter positions in the prime were in the right order. In a third experiment, using a single-presentation lexical decision task, TL nonwords produced longer latencies than the orthographic and phonological controls, whereas there was only a small phonological effect restricted to the error data. These results suggest that TL similarity effects are orthographic—rather than phonological—in nature.     

Modes of vibrotactile pattern generation

Two modes of generating vibrotactile patterns, static and scanned, were examined. In the static mode, all elements making up the pattern were turned on and off simultaneously. In the scanned mode, the pattern to be identified was moved across the tactile array. The patterns were letters of the alphabet presented to the fingertips by means of the Optacon, a reading aid for the blind. The results of the first experiment showed that the performance on a letter recognition task decreased as pattern duration decreased and that at all durations below 200-msec performance in the static mode was better than in the scanned mode. Good letter recognition was possible at durations of 4 msec in the static mode. The results of the second experiment showed that letter recognition in the static mode was highly dependent on the perceived intensity of the letters presented, whereas intensity changes in the scanned mode had little effect on letter recognition. The results of the third experiment showed that both modes of pattern presentation produced similar results in the presence of masking stimuli. The implications of the results for cutaneous pattern perception and information transmission and visual letter recognition are discussed.     

Asynchrony of lateral onset as a factor in difference in visual field.

An experiment in matching judgments was designed to examine a role of perceptual process in apparent asymmetry. Recognition of Hirakana letters (Japanese letters) was required. The experimental condition in which stimuli were presented to the left visual field first and to the right visual field second produced more errors for all stimulus intervals (0 to 60 msec.) than experimental conditions where stimuli were presented to the right visual field first and to the left one second. Especially, superiority of the latter condition was marked with the longest stimulus interval employed. These results indicate superiority of the left hemisphere function for recognizing Hirakana letters and suggest that not only memory but also perceptual process contributes to this laterality effect.     

Scanning,holistic encoding,and the word-superiority effect

The role of holistic encoding induced by scanning in the superior identifiability target letters in orthographic strings (words) enjoy over those in strings of unrelated letters was examined in two experiments. Both the set of potential targets and the critical position in the letter arrays were predesignated. One group in each experiment was induced to scan each array from left to right in a dot-counting task. A dot could precede the array to its left and/or appear with the array to its right. Only this group displayed the word superiority. For the second group in the first experiment, any dot(s) present always appeared with the array. The second group in the second experiment was shown no dots but, rather, performed the letter-identification task only. The absence of a word-superiority effect in other studies is related to holistic encoding stemruing from scanning requirements, and the implications for positional uncertainty explanations of the phenomenon are discussed.     

Levels of representation in word processing

Abstract

Four experiments are described examining the effects of words frequency, orthographic structure, and letter spacing on a range of tasks designed to tap different levels of representation in word processing. In Experiment 1, the task was lexical decision. Effects of both frequency (high-frequency words were recognized faster than low-frequency words) and orthographic regularity (illegal non-words were rejected faster than legal non-words) were found. In Experiment 2 subjects had to detect a rotated letter within the letter strings. Effects of orthographic structure emerged, with a marked disadvantage for illegal non-words with respect to the other types of string. No difference was found among high-frequency words, lowfrequency words and legal non-words. In Experiment 3, subjects had to detect a letter elevated above the horizontal plan with respect to the rest of the string. Effects of both spatial arrangement of letters and number of letters were found (spaced strings were responded to less accurately than non-spaced strings and seven-letter extra-spaced strings were responded to slower than the other strings). Neither lexical nor orthograpic variables affected this task. In Experiment 4 subjects had to detect the presence of a bold segment contained in one of the letters in the strings. Performance was unaffected by both lexical and spatial variables. The pattern of results is discussed with reference to a multi-stage model of word recognition in which lexical and spatial variables affect processing at different stages. At a feature map level, in which features are extracted from the discontinuities of light intensities, processing is independent of both spatial and lexical factors. At a letter-shape map level, in which spatial relationships between features are coded, spacing between letters affects encoding. At a graphemic map level, in which letter identities and their relative positions within strings are coded, orthographic variables have an effect. Lexicality and frequency affect only subsequent stages of processing, when stored lexical information is retrieved (e.g. for lexical decision).     

Reduced orthographic learning in dyslexic adult readers: Evidence from patterns of letter search

Visual letter search performance was investigated in a group of dyslexic adult readers using a task that required detection of a cued letter target embedded within a random five-letter string. Compared to a group of skilled readers, dyslexic readers were significantly slower at correctly identifying targets located in the first and second string position, illustrating significantly reduced leftward facilitation than is typically observed. Furthermore, compared to skilled readers, dyslexic readers showed reduced sensitivity to positional letter frequency. They failed to exhibit significantly faster response times to correctly detect target letters appearing in the most, compared to least, frequent letter position within five-letter words, and response times correlated with positional letter frequency only for the initial, and not the final, letter position. These results are compatible with the SERIOL (sequential encoding regulated by inputs to oscillations within letter units) model of orthographic processing proposed by Whitney and Cornelissen (2005). Furthermore, they suggest that dyslexic readers are less efficient than skilled readers at learning to extract statistical regularities from orthographic input.     

Reduced orthographic learning in dyslexic adult readers: evidence from patterns of letter search

Visual letter search performance was investigated in a group of dyslexic adult readers using a task that required detection of a cued letter target embedded within a random five-letter string. Compared to a group of skilled readers, dyslexic readers were significantly slower at correctly identifying targets located in the first and second string position, illustrating significantly reduced leftward facilitation than is typically observed. Furthermore, compared to skilled readers, dyslexic readers showed reduced sensitivity to positional letter frequency. They failed to exhibit significantly faster response times to correctly detect target letters appearing in the most, compared to least, frequent letter position within five-letter words, and response times correlated with positional letter frequency only for the initial, and not the final, letter position. These results are compatible with the SERIOL (sequential encoding regulated by inputs to oscillations within letter units) model of orthographic processing proposed by Whitney and Cornelissen (2005). Furthermore, they suggest that dyslexic readers are less efficient than skilled readers at learning to extract statistical regularities from orthographic input.     

Evidence for an addition to the reading scan hypothesis

Letters were tachistoscopically presented in pairs to the left or right of fixation for 40 msec. The pair consisted of a target letter (B, D, P, or T) and a noise letter (all other letters of the alphabet except I), with the noise letter appearing either to the inside or the outside of the stimulus letter. S’s task was to report verbally the target letter as fast as possible. Reaction times and number of errors indicated that there was more interference when the noise letter occurred further from the fixation point than the target letter for both the left and right visual field. Since a reading scan cannot account for the results in the right visual field, an outside-toward-fixation scan is proposed. An analysis of the different kinds of errors indicated that the outside to fixation scan is more primitive than the reading scan in that it produces less detailed information about the letters than does the reading scan.     

Priming of abstract letter representations may be universal: The case of Arabic

Recent research on the Roman alphabet has demonstrated that the magnitudes of masked repetition priming are equivalent for letter pairs that have similar visual features across cases (e.g., c-C) and for letter pairs with dissimilar features (e.g., g-G). Here, we examined whether priming of abstract letter representations occurs in an orthographic system, Arabic, in which the letters show an intricate number of contextual forms. Arabic does not have a lowercase/uppercase distinction, but the letters exhibit different forms that depend on their position (initial, medial, final, or isolated) and their connectivity. Importantly, some letters look quite different across positions (e.g., (symbol in text) and (symbol in text), which correspond to the letter 'ayn), whereas others look very similar (e.g. (symbol in text), and (symbol in text), which correspond to the letter fā'). We employed a masked priming same-different task, in which native speakers of Arabic decided whether a target letter was the same as or different from a reference letter presented in a different position (middle vs. isolated). The results showed masked repetition priming effects of the same magnitude for letter pairs with similar and with dissimilar visual features across letter positions. These data support the view that priming of abstract letter representations is a universal phenomenon.