To what extent do we use wholistic visual cues when reading?
Researchers have proposed that individuals use the outline shape of a word, meaning wholistic visual cues, in visual word recognition in reading. Various methods have been used to investigate reading in individuals with differing reading abilities. The reading abilities of young children and individuals with dyslexia have been studied to determine how this deviates from normally reading adults. The majority of research found the use of wholistic visual cues is most beneficial to individuals with dyslexia (Perea & Panadero, 2013) and those in the early stages of learning to read (Ehri, 1995). However, disparity between the research conclusions exist, proposing that wholistic visual cues are used in conjunction with other recognition processes.
The holistically biased hybrid model (Allen, Wallace & Weber, 1995) and the process model (Besner & Johnston, 1989) both provide theories for word recognition. The holistically biased hybrid model uses either an addressed or an assembled pathway to process words, with the addressed pathway being dominant in normal reading (Allen et al., 1995). This model accounts for both word frequency and provides an explanation for reading behaviours when presented with a mixed case paradigm, supporting the use of wholistic visual cues in reading (Allen et al., 1995). Alternatively, the process model recognises words using either a familiarity assessment, letter analysis or multi-letter identification (Besner & Johnston, 1989) and rejects the use of wholistic visual cues. The interactive activation model uses similar analytical methods as the process model (McClelland and Rumelhart, 1981), whereby numerous variables of the word are processed for recognition. The different uses of visual cues was determined by Paap, Newsome and Noel (1984) who concluded that wholistic visual cues are used in the initial stages of word recognition and this is then followed by abstract letter identification. Evidence for both of these models have been found and provide evidence for and against the use of wholistic visual cues using a variety of research methods.
A method used to research the importance of word shape in reading is that of the cloze test, whereby participants must anticipate the upcoming word. Haber, Haber and Furlin (1983) tested adult reading using cues including word length, envelope shape or providing the following word. They found that participants used the shape envelope to recognise the target word. The cue provided word length information which decreased the number of alternative applicable words. However, the wholistic visual cues did not provide semantic cues, suggesting that other methods of word recognition are required in conjunction with wholistic visual cues. Mirman and Magnuson (2008) discovered that words which are semantically similar increase reaction times in visual word recognition. These pieces of research support the interactive activation model because it requires the processing of several aspects of the word in parallel, including semantics (McClelland & Rumelhart, 1983). Fisher and Murray (1987) replicated Haber et al.’s (1983) research using children as their participants. No age difference was discovered in the use of wholistic visual cues between the ages of 10 to 13 years old. However, Johnston, Anderson and Duncan (1991) determined that at 8 years old salient external features improved reading accuracy which was not seen at the age of ten. This research therefore suggests that wholistic visual cues may be used more in reading behaviours when improving reading ability.
Naming tasks have been particularly important in researching developing reading abilities. Webb, Beech, Mayall and Andrews (2006) studied the effects of concealing either the inner or outer sections of words for children. The outer sections of a word elicited more accurate reading ability, whereas inner visual information of words had no influence on the individual’s reading behaviours even when accounting for frequency effects. Increased performance when presented with the outer sections of words as opposed to the inner sections was also evidenced by Beech and Mayall (2005). This concluded that individuals process word shape envelopes in visual word recognition, suggesting that wholistic visual cues may play a role in the initial stages of the interactive activation model (Webb et al., 2006). It is also possible to explain the findings using Gestalt theory, proposing that individuals form connections between the outer sections of words in order to make up for the missing word sections (Beech & Mayall, 2005). The Gestalt theory provides evidence for the use of wholistic visual cues and it is evident when researching the effects of presenting other sections of words.
This was then analysed further by presenting only the upper section of words to normally reading adults. Perea, Comesana and Soares (2012) determined an improved reaction time when upper sections of words were presented. However, this did not apply to pseudo-words, implying that the word must be known in order to have the desired effect. The decreased reaction time indicates that upper sections of words have more salient cues than lower sections of words, meaning they are more representative of the target word. These pieces of research therefore emphasise the importance of wholistic visual cues in visual word recognition, however research by Pelli, Farell and Moore (2003) contrasted these conclusions. The researchers determined that individuals focus on minor cues in words and collate these for an overall representation of the word, contrasting the evidence of using wholistic visual cues when reading. Further information other than word shape envelopes must therefore be known in order to accurately process the given word.
An alternative way in which the use of wholistic visual cues has been researched is that of using a mixed case paradigm, presenting an array of uppercase and lowercase letters to participants (Coltheart & Freeman, 1974). Presenting words in a mixed case format increased reaction times as well as reducing the recognition of the word, providing evidence for the importance of wholistic visual cues and contradicting the interactive activation model (Coltheart & Freeman, 1974). However, the mixed case paradigm had no influence on the identification of individual letters in this case, supporting the interactive activation model and opposing the use of wholistic visual cues. Besner and Johnston (1989) also found pseudo-words in a mixed case format to be detrimental to reading ability. This therefore demonstrates the need for the word shape envelope and rejects the interactive activation model due to the reduced reading ability. Allen et al. (1995) used a lexical decision task to compare the reading of lower and mixed cases when given a time constraint on processing. The experiments concluded an increased reaction time for pseudo-words in a mixed case format, indicating the importance of wholistic visual cues in word recognition. Participants struggled to accept or reject pseudo-words presented for 400ms, representing the processing limitations during short exposure periods. This research is in line with Allen et al.’s (1995) holistically biased hybrid model but contrasts the analytical models, including the process model. The research carried out by Allen et al. (1995) indicates that for the successful completion of lexical decision tasks wholistic visual cues are necessary, meaning that reading methods may adapt to the task’s requirements. This provides an explanation for the great degree of variation seen in the research into the use of wholistic visual cues and shows the high level of validity of this method due to the replicability of the findings.
Further research has led to the suggestion that wholistic visual cues may only be used in circumstances where normal reading behaviour is inhibited. This was recently demonstrated by Perea and Panadero (2013) using a lexical decision task to analyse reading behaviours for adults, children and children with developmental dyslexia. There was no effect on the reaction times for word recognition when pseudo-words were presented as having the same shape as real words for adults and children. However, those with developmental dyslexia were found to be affected by the word shape of pseudo-words (Perea & Panadero, 2013), reflecting how wholistic visual cues are used to a different extent. This indicates the use of more analytical processing methods in normal reading. Lavidor (2011) also found word shape envelopes to be beneficial to individuals with dyslexia. These pieces of research suggest that children and those with dyslexia fixate on particularly salient cues in order to reliably process words (Ehri & Wilce, 1985). This again dismisses the word shape hypothesis and provides further evidence for the importance of the interactive activation model (McClelland & Rumelhart, 1981) during normal reading due to the use of feature analysis. These studies highlight the requirement for top-down processing in normal reading behaviours compared to dyslexia, as the interactive activation model (McClelland & Rumelhart, 1981) requires the processing of multiple factors in word recognition.
Cognitive processing, including top-down processing, is a vital part of visual word recognition. Research carried out by Yates (2013) provides further evidence with clustering effects that word shape alone has limited influence on normal reading behaviour. If a set of words only differ by a phoneme then word recognition requires a higher level of activation and is processed more slowly (Yates, 2013). This infers that the use of a word’s shape envelope is influenced by cognitive processing ability. Cognitive processing can again be seen by analysing parafoveal vision. When monitoring eye movements McConkie and Zola (1979) discovered normally reading adults do not detect any changes to manipulations of word shape in parafoveal vision. Contrasting evidence by Haber, Haber, Furlin, Paap, Newsome and Noel (1984) determined that when proofreading, participants remained unaware of changes to words unless the word shape envelope was manipulated. An alternative explanation for the lack of evidence for the word shape hypothesis may be that the use of wholistic cues becoming automated (Webb et al., 2006). Research has proven the importance of the word shape envelope for young readers and dyslexics, however it has not been consistently reported in adults. If word shape is processed in a more automated manner the individual’s cognitive load would be significantly reduced (Webb et al., 2006). This reduction would allow other, more efficient, reading behaviours to occur. These pieces of research led to the conclusion that wholistic visual cues are important in visual word recognition, however as an individual matures their reading behaviours may be modified.
Previous research has found that individual’s reading methods change as their reading ability improves. Children initially use wholistic visual cues and then potentially develop more effective reading methods (Perea & Panadero, 2013). Research by Seymour and Eldre (1986) determined that in order to read children have to be specifically taught to read each of these words, meaning they are unable to determine the phonological information of a word and as such rely on wholistic visual cues (Webb et al., 2006). This was also supported was Ehri (1995) who described stages in children’s visual word recognition. When children learn to read they engage in sight word reading or logographic reading, meaning that the word is read through memory retrieval. As children’s reading ability develops they learn the relationship of phonemes and graphemes and are thus able to apply this to more complex words in the consolidated alphabetic stage (Ehri, 1995). Research using children is therefore of great advantage when focusing on adult reading. The conclusion has been reached that as children develop their reading ability they refine their use of wholistic visual cues. Thus as visual word recognition becomes more refined individuals use other recognition strategies in parallel with wholistic visual cues.
Research into the use of wholistic visual cues has shown that adults do not necessarily use wholistic visual cues. However, evidence has shown that children and individuals with developmental dyslexia use these cues to a great extent. This difference may occur due to the processing of wholistic visual cues becoming more automated as reading develops. Alternatively, the varying use of wholistic cues may be explained by the demands of the task, as shown by lexical decision tasks (Allen et al., 1995). For this reason wholistic visual cues provide a more accurate account of visual word recognition when taken in conjunction with analytical models, such as the interactive activation model and the process model.
Allen, P. A., Wallace, B., & Weber, T. A. (1995). Influence of case type, word frequency, and exposure duration on visual word recognition. Journal of Experimental Psychology: Human Perception and Performance, 21(4), 914-934.
Beech, J. R., & Mayall, K. A. (2005). The word shape hypothesis re-examined: Evidence for an external feature advantage in visual word recognition. Journal of Research in Reading, 28(3), 302-319.
Besner & Johnston (1989) Beech &M
Coltheart & Freeman (1974) Allen
Ehri, L. C. (1995). Phases of development in learning to read words by sight. Journal of Research in Reading, 18(2), 116-125.
Ehri & Wilce (1985) Lavidor
Fisher & Murray (1987) Webb
Haber, L. R., Haber, R. N., & Furlin, K. R. (1983). Word length and word shape as sources of information in reading. Reading Research Quarterly, 18(2), 165-189.
Haber, Haber, Furlin, Paap, Newsome & Noel (1984) Beech & Mayall
Johnston, Anderson & Duncan (1991) Beech & M
Lavidor, M. (2011). Whole-word shape effect in dyslexia. Journal of Research in Reading, 34(4), 443-454.
McClelland and Rumelhart (1981)
McConkie, G. W., & Zola, D. (1979). Is visual information integrated across successive fixations in reading? Perception and Psychophysics, 25(3), 221-224.
Mirman & Magnuson (2008) yates
Seymoure & Eldre (1986) Webb
Paap, K. R., Newsome, S. L., & Noel, R. W. (1984). Word shape’s in poor shape for the race to the lexicon. Journal of Experimental Psychology: Human Perception and Performance, 10(3), 413-428.
Pelli, D. G., Farell, B., & Moore, D. C. (2003). The remarkable inefficiency of word recognition. Nature, 423, 752-756.
Perea, M., Comesana, M., & Soares, A. P. (2012). Does the advantage of the upper part of words occur at the lexical level? Memory and Cognition, 40, 1257-1265.
Perea, M., & Panadero, V. (2013). Does viotin activate violin more than viocin? On the use of visual cues during visual-word recognition. Experimental Psychology, 61(1), 23-29.
Webb, T. M., Beech, J. R., Mayall, K. M., & Andrews, A. S. (2006). It’s what’s on the outside that matters: An advantage for external features in children’s word recognition. Journal of Experimental Child Psychology, 94, 163-181.
Yates, M. (2013). How the clustering of phonological neighbours affects visual word recognition. Journal of Experimental Psychology: Learning, Memory and Cognition, 39(5), 1649-1656.
Cite This Work
To export a reference to this article please select a referencing style below: