Whole-word units are used before orthographic knowledge in perceptual development

A visual search task for target letters in multiletter displays was used to investigate information-processing differences between college students and presecond-grade children (mean age = 7 years, 4 months). The stimulus displays consisted of single words, pronounceable pseudowords, and unpronounceable nonwords varying in length from three to five letters. The mean response times for indicating whether or not a target letter occurred in the display increased with the number of display letters for both groups, although there were apparent differences between groups in the rate of search and type of search strategy used. Pre-second-grade children responded faster to word displays than to pseudoword and nonword displays, indicating that familiar letter strings could be processed faster than unfamiliar strings regardless of whether or not the latter were consistent with rules of English orthography. In contrast, college students processed words and pseudowords about equally well, and both resulted in faster responses than nonwords. As reading skills develop, children apparently come to process familiar words differently from other letter strings. Only after a significant sightword vocabulary is established do children seem to recognize the regularities of standard English orthography and make use of this knowledge to facilitate perceptual processes.     

The development of visual information processing skills related to reading

Literate adults can use their familiarity with specific words and their knowledge of English orthography to facilitate word recognition processes. The development of word superiority effects in visual perception was investigated in the present study using a search task with kindergarten (5.7 years old), second (8.0 years old), and fourth grade children (10.0 years old), and college students. The search task consisted of the visual presentation of a target letter followed by a three-, four-, or five-letter display. The target letter was included in the display on half the trials, and the displays were common words, orthographically regular pseudowords, and irregular nonwords. Although response times decreased with age, the three oldest groups showed similar effects for the size and structure of the displays. That is, response times increased linearly with the number of display letters, and responses were faster for word and pseudoword displays than for nonwords. The data for the kindergarten children showed evidence for the use of a different search strategy, and they did not respond differentially to the three types of displays. The results are discussed in terms of the implications for developmental models of visual search and word superiority effects in visual perception.     

Types and tokens in visual letter perception

Five experiments demonstrate that in briefly presented displays, subjects have difficulty distinguishing repeated instances of a letter or digit (multiple tokens of the same type). When subjects were asked to estimate the numerosity of a display, reports were lower for displays containing repeated letters, for example, DDDD, than for displays containing distinct letters, for example, NRVT. This homogeneity effect depends on the common visual form of adjacent letters. A distinct homogeneity effect, one that depends on the repetition of abstract letter identities, was also found: When subjects were asked to report the number of As and Es in a display, performance was poorer on displays containing two instances of a target letter, one appearing in uppercase and the other in lowercase, than on displays containing one of each target letter. This effect must be due to the repetition of identities, because visual form is not repeated in these mixed-case displays. Further experiments showed that this effect was not influenced by the context surrounding the target letters, and that it can be tied to limitations in attentional processing. The results are interpreted in terms of a model in which parallel encoding processes are capable of automatically analyzing information from several regions of the visual field simultaneously, but fail to accurately encode location information. The resulting representation is thus insufficient to distinguish one token from another because two tokens of a given type differ only in location. However, with serial attentional processing multiple tokens can be kept distinct, pointing to yet another limit on the ability to process visual information in parallel.     

Whole and part comparisons of words and nonwords

Under conditions of sequential presentation, two words are matched more quickly than are a single letter and the first letter of a word. An exception to this whole-word advantage was reported in 1980 by Umansky and Chambers, who used word pairs as stimuli, and asked subjects to compare the entire words or the words’ first letters. Experiment 1 showed that the stimulus lists used by Umansky and Chambers may not have constrained subjects to process the displays differently for wholistic and component comparisons. In those studies, the two words were identical onsame trials for both wholistic and first-letter comparisons, so that first-letter decisions could have been based on wholistic information. In the present study, lists were constructed so that first-letter decisions could not be determined correctly by wholistic information (e.g., BLAME/BEACH), and the whole-word advantage was replicated. Experiment 2 tested whether wholistic comparisons are generally superior to component comparisons. For consonant strings, first-letter comparisons were made more quickly than were whole-string comparisons. These results are interpreted as support for hierarchical models of visual word processing.     

Measuring the size of mental images

Mental images seem to have a size; the experimental problem was to map that image size onto a scale of physical measurement. To this end, two experiments were conducted to measure the size of mental images in degrees of visual angle. In Experiment 1, college students employed light pointers to indicate the horizontal extent of projected mental images of words (the letter string, not the referent). Imagined words covered about 1.0 degress of visual angle per letter. In Experiment 2, a more objective eye-movement response was used to measure the visual angle size of imagined letter strings. Visual angle of eye movement was found to increase regularly as the letter distance between the fixation point and a probed letter position increased. Each letter occupied about 2.5 degrees of visual angle for the four-letter strings in the control/default size condition. Possible relations between eye movements and images are discussed.     

Response latency in visual search with redundancy in the visual display

Two experiments were run to investigate the effects of redundant display items upon response latency in a visual search task. In the first study, Ss searched five-letter displays for a predesignated critical letter. Both critical and noncritical letters could be repeated in the displays. Mean response latency decreased markedly with increasing redundancy in the critical letter and was affected to a lesser extent by redundancy in the noncritical letters. In the second study, Ss were required to detect the presence of redundant letters in displays of from two to five letters, first with no information as to what letter might be repeated, then with knowledge of which letter would be repeated if the display contained a redundant letter. Response latencies in the former case were much slower than in the latter. The implications of these findings for current views of visual information processing were discussed.     

Coordinate frame for symmetry detection and object recognition

Can subjects voluntarily set an internal coordinate frame in such a way as to facilitate the detection of symmetry about an arbitrary axis? If so, is this internal coordinate frame the same as that involved in determining perceived top and bottom in object recognition and shape perception? Subjects were required to determine whether dot patterns were symmetric. Cuing the subjects in advance about the orientation of the axis of symmetry produced a substantial speedup in performance (Experiments 1 and 3) and an increase in accuracy with brief displays (Experiment 2). The effects appeared roughly additive, with an overall advantage for vertical symmetry; thus, the vertical axis effect is not due to a tendency to prepare for the vertical axis. The cuing advantage was found to depend upon the subject's knowing in advance the spatial location as well as orientation of the frame of reference (Experiment 4). The fifth experiment provided evidence that the frame of reference responsible for these effects is the same as the one that determines shape perception: Subjects viewed displays containing a letter (at an unpredictable orientation) and a dot pattern, rapidly naming the letter and then determining whether the dots were symmetric about a prespecified axis. When the top-bottom axis of the letter was oriented the same way as the axis of symmetry for the dots, symmetry judgments were significantly more accurate. Thus, the results suggest a single frame of reference for both types of judgment. The General Discussion proposes a theory of how visual symmetry may be computed, which might account for these phenomena and also characterize their relation to "mental rotation" effects.     

Sources of visual field interference in children and adults

Subjects at three grade levels, third, sixth, and college (8, 11, and 20 years of age, respectively), were asked to identify briefly presented words accompanied by either words or letter strings in the same or opposite hemifield. Results indicated two sources of visual field interference, one relating to the presence of alphanumeric information elsewhere in the visual field and the second concerning the meaningfulness (lexical identity) of that information. The former showed a decrease in importance with age, and the latter an increase. Results were discussed in terms of the development of selective attention and hemispheric interaction.     

Part-whole relationships in the processing of small visual patterns

Subjects seem to react to a word faster than they react to a letter within a word. One interpretation is that words are processed holistically; another is that all visual stimuli are processed in terms of components, but that more stimulus information is available for use when the targets are words than when they are letters within words. The results of three experiments indicate that the word or pattern-level advantage occurs even when the stimulus information in the two situations is equated, but if the perceptual arrays cannot be unitized (e.g., consonant sequences), a pattern-level advantage does not occur. In addition, the experiments provide substantial evidence to indicate that if letter arrays cannot be unitized, then they are processed on a componentby-component basis, rather than holistically. Finally, the appropriate definition ofholistic processing is considered.     

Positional dyslexia

Position-specific errors in word reading are usually associated with neglect or visual extinction on the same side as the reading problem. In this study, two patients with left-hemisphere lesions showed visual extinction on the right but reading difficulty on the left side of words and pseudowords. Further study of one patient revealed that he also had problems reading the beginning of words presented tachistoscopically or in vertical orientation. In addition, the positional difficulty was apparent when he named the letters in words. The pattern of results indicates that the positional dyslexia in these patients was not likely attributable to general deficits in visual perception or attention but may have reflected a disorder at a later stage of letter processing.     

Similarity-related channel interactions in visual processing

Lateral interference between letters in the visual field is a joint function of their similarity and physical separation. Data needed to evaluate hypothesis about the processes implicated in these effects were obtained from two experiments in which the task was identification of a target letter always presented in the center of a three-letter display. Variation of target-flanker similarity, the primary variable, was combined factorially with spacing of target and flanker, location of the display in the visual field, delay of patterned postmasks, and exposure duration. The effect of target-flanker similarity on target identification yielded a nonmonotonic function with a minimum at an intermediate degree of similarity. Data from same-different judgments regarding target-flanker similarity indicate that some information about similarity is available even at levels of visibility that do not permit identification of individual letters. All of the findings could be accommodated by a model assuming that only variables determining visibility--exposure duration, mask properties, location in the visual field, separation of letters--influence extraction of featural information pertaining to letter identification. In contrast, visual similarity influences performance by way of sometimes subtle effects on subjects' criteria or response biases and by effects on the encoding and retention of information regarding relative positions of characters in the visual field. The varying effects of similarity reported in the literature on letter identification all appear to be interpretable in these terms.     

The interfering effect of word perception on letter identification

Subjects identified a single lowercase letter in a visual display by pressing one of two buttons. Two letters were assigned to each response. Groups received one of three context conditions: word, nonword, or single-letter displays. In words and nonwords, the flanking letters adjacent to the target varied as to whether they were response compatible or incompatible with the target. Single letters produced faster response latencies than either multi-letter condition, and words yielded slower latencies than did nonwords. Items Containing an incompatible-response flanking letter produced longer latencies than items containing a compatible flanking letter. Subgroups of subjects with different characteristic processing patterns were identified with a separate test. These subgroups were differentially affected by the context conditions in the letter-identification task. A greater subgroup difference was found in nonwords than in words.     

Does familiarity affect transfer from an iconic to a short-term memory?

Six Ss discriminated seven-letter nonsense words from comparison words. Target and comparison words differed on randomly selected trials by one randomly chosen letter. Target words were displayed for 50, 55, 60, 70, 90, or 200 msec, and were preceded and followed by a masking field. In one condition the Ss were familiarized with the comparison words, and in another they were not. Discrimination was better for familiar words at all display durations. There was an interaction between familiarity and the letter position effect. For unfamiliar words the typical bow-shaped position effect occurred. For familiar words no marked position effect occurred. An identification condition using unfamiliar words found no interaction between letter position and display duration. The results are interpreted as evidence that familiarity removes a letter position effect that depends upon serial transfer from a nonmaskable mediating visual representation that is constructed from a maskable representation by nonserial processes.     

The effects of forced serial processing on identification of words and random letter strings

Skilled readers identified words and random letter strings displayed one letter at a time on the screen of a computer-controlled oscilloscope. Letters were either displayed in a single position, or painted left-to-right in adjacent positions. Adjacent displays were either unmasked, masked by a letter-like pattern following each letter (backward masking) or masked by a pattern preceding each letter (forward masking). Unmasked words were identified with a high degree of accuracy, and accuracy was independent of the exposure duration of the letters over the range 50–200 msec. Masked adjacent word displays and one-position word displays were identified much less accurately, and accuracy was highly dependent on letter exposure duration. In contrast to these results, identification of random strings, or of letters within random strings, was almost unaffected by the presence of the mask and was dependent on letter exposure duration under all display conditions. The results are interpreted as evidence that (a) masking forces subjects to process serial displays one letter at a time, and (b) words, or large segments of words, are habitually processed in parallel, while random strings are processed as a series of individual letters or small chunks.     

Conflicting strategies and hemispheric suppression in a lexical decision task

The research tests the prediction of the inhibitory-interaction hypothesis that experience with a task accentuates the functional imbalance between the hemispheres. Right-handed males who were experienced readers were presented a letter string to the centre visual field for lexical decision. The string was or was not accompanied by a blinking light to the left or right visual field. It was predicted that asymmetry would be greatest for strings that spelled words, less for strings that were orthographically correct (pseudowords) and least for strings that were orthographically incorrect (nonwords) because efficient adult readers have more experience with letter strings that do than do not spell a word, and have more experience with orthographically correct than incorrect letter strings. The analysis of response times supported the prediction. Moreover, in the nonword and in early trials of the pseudoword conditions, response times were faster when one or other hemisphere was distracted than when both were engaged suggesting the hemispheres use strategies that conflict when suppression has not been accentuated by practice. As well, as the trials progressed in the pseudoword condition, asymmetry reversed before increasing suggesting that the hemispheres reduce conflict by competing for and then strengthening suppression.     

Word superiority over isolated letters: The neglected case of forward masking

Previous research shows that when briefly presented alphabetic stimuli are followed by pattern masks, letters in words are reported more accurately than are isolated letters (the “Word-Letter Phenomenon,” or WLP); however, when these masks are replaced by blank fields, the WLP disappears. These findings have led to the popular notion that the WLP reflects selective masking of ongoing stimulus processing and so critically depends on the use of poststimulus masks. Here we report three experiments which re-examine the role of masking in the WLP by contrasting the effects of postmasked displays with the effects of premasked displays in which words and isolated letters werepreceded by a pattern mask and followedby a completely blank field. Despite the critical role generally assigned to poststimulus pattern masks, similar WLPs were obtained with pre- and postmasked displays. Implications for theories of word and letter recognition are discussed.     

Oral report of words and word approximations presented to the left or right visual field

Subjects reported letter strings forming words, pronounceable high approximations to words, and unpronounceable low approximations to words presented tachistoscopically to the left or right visual field (LVF, RVF). (a) For number of strings totally correct, the same RVF superiority was obtained with high approximations as with words, the field difference with low approximations being negligible. (b) In contrast, for letter scores from partially correct strings, RVF superiority did not vary with string type. Finding (a) is interpreted to indicate that the left hemisphere is differentially specialized for processing words as units and that requiring oral report makes pronounceable strings processable as word-like units. Finding (b) suggests that the left hemisphere is not specialized for processing subword fragments.     

Semantic predictability eliminates the transposed-letter effect

Semantic predictability facilitates word recognition during language processing. One possible explanation for this facilitation is that highly specific predictions generated online during language processing preactivate some features of upcoming words. To explore whether, how, and when these predictions affect visual word recognition, in the two experiments reported here we investigated the influence of semantic predictability on transposed-letter priming. In order to do so, a paradigm that combines self-paced word-by-word reading with masked priming was developed. Transposed-letter priming occurred in nonconstraining contexts but not in constraining contexts, indicating that readers use context to make predictions about both letter identity and position in upcoming words, and that these predictions have an early influence on visual word recognition.