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.     

The repeated letter effect in the misspellings

The study consisted of a comparison of the spontaneous misspellings of a dysgraphic (L. H. Oswald) with the ease of perception and recall of similar misspellings by normal Ss. The main results were as follows: (1) Oswald frequently misspelled repeated letters in words (e.g., ELDERY). Similarly, normal Ss found repeated-letter misspellings more difficult to perceive and recall than nonrepeated-letter misspellings. (2) Oswald frequently misspelled repeated letters which were close together as in ANLYZE (for ANALYZE), but he rarely misspelled widely separated repeated letters as in MISSPELLNG. Similarly, normal Ss found repeated-letter misspellings hardest to perceive and recall when the repeated letters were close together, and very easy when they were widely separated. (3) Oswald misspelled the second of two repeated letters (as in ELDERY) more frequently than the first (as in EDERLY). Similarly, normal Ss found repeated-letter misspellings easier to perceive and recall when the first of the repeated letters was misspelled, as in EDERLY, than when the second was misspelled. (4) Oswald sometimes repeated a letter on his own, as in HABITITUATED—but more frequently he dropped a repeated letter as in ELDERY. However, repeated-letter misspellings involving deleted letters were easier for the normal Ss to recall than those involving deletions. Evidence suggesting that repeated phonemes induce analogous errors in normal speech production was discussed, and implications for general models of the serial order of speech behavior were pointed out.     

Repetition blindness is orientation blind

In identifying rapid sequences of three letters, subjects were worse at identifying the first and third letters when they were the same than when they were different, indicating repetition blindness (RB). This effect occurred regardless of the angular orientations of the letters, but was more pronounced when the orientations of the repeated letters were different than when they were the same. In a second experiment, RB was also evident when the first and third letters were lowercase bs or ds, presented upright or inverted, even though they are differently named when inverted (q and p, respectively). Conversely, a third experiment showed that RB occurred when the letters had the same names but were repeated in different case. These results suggest that the early extraction of letter shape is independent of its orientation and left-right sense, and that RB can occur at the levels of both shape and name.     

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.     

Identification of typographically transformed words: instance-based skill acquisition

Two hypotheses concerning the development of skill at identifying typographically transformed words were tested. One claim is that a general skill independent of specific training instances is involved, and the other is that skill is based on memory for the analysis of specific instances encountered during training. Contrary to the general skill view, a series of experiments demonstrated that transfer of word identification skill was highly specific and occurred only when training and test instances shared common letters printed in the same case (i.e., uppercase or lowercase). Transfer of skill also depended on the visual patterns formed by adjacent letters and word shape. Presentation of a word in training and test phases significantly improved test phase identification of that word even when a unique visual pattern was used. These results are compatible with an instance-based view of word identification skill in which it is assumed that subjects develop skilled analysis of the visual and conceptual characteristics of specific words, and that this skill can be used to identify repeatedly presented words as well as predictable sets of novel words.     

Orthographic repetition blindness

Repetition blindness (RB) is the failure to report the second occurrence of a repeated word, when words are sequentially and briefly displayed (Kanwisher, 1987). RBis also observed for non-identical words, such as home, dome . Explanations for non-identity RB assume that similarity at the level of the whole word causes the secondword to be suppressed ('similarity inhibition'). Three experiments demonstrate that RB is robust for diverse types of orthographic relatedness, including critical words that share only their first initial letter, their last two letters, first three letters, middle three letters, beginning and final letters, three alternating letters, and three non-aligned letters (as in chance hand ). The theoretical construct of similarity inhibition may be able to account for these data, although one mechanism previously proposed in the literature, neighbourhood inhibition, is probably not a useful way to explain the data on RB for words sharing only one or two letters. Weintroduce an alternative explanation for orthographic RB: Only the repeated letters are suppressed, and amount of RB depends on howeasily the perceiver can reconstruct the target word from the non-suppressed letters.     

Two bs or not two Bs? A signal detection theory analysis of repetition blindness in a counting task

Stimulus repetition usually benefits performance. A notable exception is repetition blindness (RB), in which subjects fail to report a repeated stimulus in a rapid serial visual presentation. Theories differ in attributing RB to either perceptual encoding or memory retrieval and to impaired discrimination versus response bias. In the present study, subjects judged whether one or two letters were imbedded in sequences of digits. Unlike previous studies, false guesses of two unrepeated letters were distinguished from false guesses of two repeated letters. When repeated- and unrepeated-letter trials were randomly intermixed (Experiment 1), RB was entirely attributable to response bias. However, when they were separately blocked (Experiments 2 and 3), RB was manifested in discriminability (d'). The results support perceptual-encoding accounts of RB but indicate that effects on discriminability depend on subjects' processing strategies.     

Taking a bright view of negative priming in the light of dim stimuli: further evidence for memory confusion during episodic retrieval.

A same-different letter-matching task was used to examine the effects of stimulus intensity on negative priming, which is poorer performance when target letters have been presented as distractor letters on the immediately preceding trial. In Experiment 1, stimulus intensity was manipulated between-participants, whereas in Experiment 2, it varied randomly from trial-to-trial within-participants. In Experiment 1, negative priming was equivalent for both stimulus intensities. In Experiment 2, negative priming effects were larger for repeated intensity stimuli than for nonrepeated intensity stimuli, when stimulus intensity was dim. Furthermore, for repeated intensity stimuli, negative priming effects were enhanced when the overt response required to the stimulus was repeated from prime to probe trial. These results are consistent with the hypothesis that negative priming may be due to memory confusion, rather than to inhibition of the distractor stimuli.     

Sex differences in the reading process revealed by a letter-detection task

Four experiments used a task in which males and females had to work through a passage and circle instances of a target letter. Coltheart, Hull and Slater (1975) have previously used this task to show that females have greater difficulty than males in detecting target letters, especially letters in the and other silent letters. The four experiments failed to replicate Coltheart et al.'s findings and, in fact, frequently found significant sex differences in the opposite direction. This discrepancy in results could not be accounted for in terms of procedural differences between the two studies. It was concluded that Coltheart et al.'s (1975) results with the letter detection task may not be reliable.     

Inducing the familiarity effect

The task required subjects to judge whether members of six-letter consonant string pairs were physically identical. Some (repeated) strings were presented on each of the 16 training sessions, while other (nonrepeated) strings were presented only during a single session. An advantage in matching time of repeated strings over nonrepeated strings was obtained which increased in a roughly linear fashion during training and which persisted for at least 7 weeks after the last training session. Detailed analyses of thedifferent responses and the results of several transfer tasks suggest that perceptual processing is indeed facilitated by acquired information about which letters usually follow others and about the positions in which these sequences of letters are likely to occur, but provide no evidence that this facilitation derives from the formation of visual features which are larger than single letters.     

Memory for attended and unattended visual stimuli

Subjects were presented with lists of 'compound letters,' letters whose overall shapes were described by repeated use of replicates of other, smaller letters. In Experiment 1 subjects were asked to attend to either the overall letter or the smaller, constituent letter. At the end of list presentation, recall of all letters was required, but a postlist cue determined whether the attended or unattended letters were to be reported first. The results for four-item lists accorded with those of Martin (1978, 1980): order of report had a larger effect upon retention of attended letters than upon retention of unattended letters. The findings for three-item lists did not agree with Martin, however: first, the interaction of attention and order was weak; second, sharp recency for unattended letters was not found.

Experiment 2 required that subjects recall either in temporal pairs or by letter size. The results strongly suggest that the present paradigm does not constitute an analogue to dichotic listening. In particular, there is little evidence for a role for sensory retention of compound letters at time of recall.     

Loss of vision during the retinal stabilization of letters

Upper- and lower-case letters of the alphabet were stabilized until loss of vision occurred. Loss of straight-line visibility was the most frequently reported perceptual event. Occasionally, features of letters separated spatially before their loss of visibility. In both instances, loss of visibility often resulted in the perception of a less complex letter. Confidence ratings for each loss of letter visibility indicated that participants were quite certain about perceived fragmentations. In a control experiment, participants were asked to guess how letters would fragment during stabilization. Again, loss of line visibility was the most frequently reported event. However, spatial separation of features was rarely predicted and complex letters were not predicted to fragment into simpler letter forms. Furthermore, the confidence in predicted fragmentation was quite low. These results are consistent with the view that losses of visibility during retinal stabilization constitute a distinct perceptual experience. Fragmentations appear to be determined by the availability of less complex letter forms and by the loss of subletter information, consisting of letter features and information specifying spatial configurations.     

Levels of processing in speech production

When subjects repeated short alphabetic sequences at experimenter-controlled rates while retaining consonant trigrams, subsequent recall was unrelated to speech rate if single letters or a long sequence was repeated, but recall was inversely related to speech rate for sequences three or five letters in length. The latter effect was predicted by a model in which sequences were organized as single response units in a central phonetic processor but independently executed by means of mechanisms not requiring the services of the central processor. Quantitative predictions were generated from this model by estimating organization time for the sequences and assuming that the probability of forgetting would be equal to the organization time/execution time ratio. All obtained recall probabilities were as predicted for sequences three of five letters in length with both experimenter- and subject-controlled speech rates. Repetition of a nine-letter sequence from the end of the alphabet produced forgetting patterns clearly out of accord with any organization time/execution time ratio, but a similar sequence from the beginning of the alphabet produced evidence of being a marginal case with unstable organization.     

Processing of distractors inside and outside the attentional focus in a priming procedure

A priming procedure was used to study the processing of distractors located either inside (between the location of two targets) or outside (peripherally to the locations of the targets) the focus of attention. The stimuli were five-letter arrays, and participants had to decide whether two marked target letters were the same or different. In Experiments 1 and 2, positive priming was obtained both when targets and in-distractors in primes repeated as targets in probes; negative priming was found when out-distractor primes repeated as targets in probes. In Experiment 3, we also manipulated the match in letter case from primes to probes. In-distractors produced reliable positive priming, irrespective of whether the letters matched in case. In contrast out-distractors produced negative priming but only when the letters had the same case in primes and probes. These results are attributed to a spatial attention process operating (in this case) on low-level visual features, and an object-based selection process that enables more abstract information to be processed for selected stimuli.     

Retrograde Amnesia after ECT. An Attempt to Demonstrate Reminder-effects in Humans

In a study of the retrograde amnesic effect of bilateral ECT, twelve patients were subjects in a repeated measures design. Both verbal and nonverbal materials were used and retention was tested with different methods: Yes-no recognition, forced-choice recognition and recall cued by initial letters. The absolute amount of retrograde amnesia was small; there seemed to be no difference between verbal and nonverbal materials, and cueing with initial letters was not found to be an effective method of reducing amnesia.     

Interference effect caused by repetition in visual recognition of letters: Examination of the orthographic difference between two same letters1

Abstract: The present study examined the effect of phonological identity between two letters on the visual recognition of the letters. Participants were required to identify the two same or different letters that were successively presented for a short duration. In order to manipulate the phonological identity of the two letters, the orthography of the Kana letters was varied. In half of the trials, the first and second letters were presented in Hiragana (the same‐orthography condition). In the other half, the first letter was presented in Katakana, and the second letter in Hiragana (the different‐orthography condition). The results revealed that the identification performance for the second letter was reduced in the same‐orthography condition when the two letters were the same, compared with when they were different. In contrast to this, in the different‐orthography condition, the identification performance of the second repeated letter was marginally superior to that of the nonrepeated letter. Considering the present findings together with those of the author's previous study (Kuwana, 2004), it is suggested that the interference effect caused by repetition could result from the reduction in the availability of visual pattern information stored in long‐term memory.     

Grammatical class effects in relation to normal and aphasic sentence processing

Agrammatic, Broca's aphasic patients, Wernicke's aphasic patients, and neurologically intact control subjects were asked to detect target letters in prose passages and in a scrambled word passage. The targets were embedded, in some instances, in content words (open-class vocabulary items), and in other instances, in function words (closed-class vocabulary items). With respect to the prose passages, both the control subjects and Wernicke's aphasic patients were more apt to notice target letters when they appeared in the open-class items than when in closed-class items; by contrast, the agrammatic Broca's patients showed no vocabulary class detection difference. The Wernicke's patients were not entirely normal, however: Whereas the normal subjects showed a much smaller vocabulary class effect for letter detection in the scrambled condition, the Wernicke's maintained the pattern they had shown in the prose condition. These and other findings obtained on the letter cancellation task are discussed in relation to lexical access mechanisms geared to sentence parsing.     

Detection of intraword and interword letter repetition: A test of the word unitization hypothesis

Do words, as familiar units or gestalts, tend to swallow up and conceal their letter components (Pillsbury, 1897)? Letters typically are detected faster and more accurately in words than in nonwords (i.e., scrambled collections of letters), and in more frequent words than in less frequent words. However, a word advantage at encoding, where the representation of the string is formed, might compensate for, and thus mask, a word disadvantage at decoding and comparison, where the component letters of the representation are accessed and compared with the target letter. To better reveal any such word disadvantage, a task was used in this study that increased the amount of letter processing. Subjects judged whether a letter was repeated within a six-letter word or a nonword (Experiment 1; intraword letter repetition) or was repeated between two adjacent unrelated six-letter words or nonwords (Experiment 2; interword letter repetition). Contrary to Pillsbury's word unitization hypothesis, both types of letter repetition (intraword and interword) were detected faster and just as accurately with words as with nonwords. In Experiment 2, however, interword letter repetition was detected less accurately on common words (but not on rare words or third-order pseudowords) than on the corresponding nonwords. Thus, although the familiar word does not deny access to its own component letters, it does make their comparison with letters from other words more difficult.     

Object and action picture naming in English and Greek

When two orthographically similar words are displayed using rapid serial visual presentation (RSVP), the repeated letters in the second critical word (W2) are not detected, leading to a deficit in reporting this word known as repetition blindness (RB). In Turkish, letters containing diacritic markings (e.g., y, ö) are considered separate letters, yet are visually highly similar to their non-diacritic analogues (s,o). Two experiments used the phenomenon of RB to investigate whether diacritic letters are represented as more similar to their non-diacritic analogues than are two unrelated letters. In Experiment 1, substantially more RB was found for words differing in just a diacritic (i y im-isim) compared to orthographic neighbours (words differing in a visually non-similar letter, such as ilim-isim). In Experiment 2, the amount of RB for identical words (isim-isim) was comparable to words that differed by a single diacritic marking (i y im-isim). We conclude that diacritic letters are mentally represented as variants of their non-diacritic analogue. Letter / word recognition researchers may be interested in pursuing these findings using standard techniques such as backward masking and orthographic priming.     

Repetition blindness and illusory conjunctions: errors in binding visual types with visual tokens.

Repetition blindness (Kanwisher, 1986, 1987) has been defined as the failure to detect or recall repetitions of words presented in rapid serial visual presentation (RSVP). The experiments presented here suggest that repetition blindness (RB) is a more general visual phenomenon, and examine its relationship to feature integration theory (Treisman & Gelade, 1980). Experiment 1 shows RB for letters distributed through space, time, or both. Experiment 2 demonstrates RB for repeated colors in RSVP lists. In Experiments 3 and 4, RB was found for repeated letters and colors in spatial arrays. Experiment 5 provides evidence that the mental representations of discrete objects (called "visual tokens" here) that are necessary to detect visual repetitions (Kanwisher, 1987) are the same as the "object files" (Kahneman & Treisman, 1984) in which visual features are conjoined. In Experiment 6, repetition blindness for the second occurrence of a repeated letter resulted only when the first occurrence was attended to. The overall results suggest that a general dissociation between types and tokens in visual information processing can account for both repetition blindness and illusory conjunctions.