Study on Orthographic Markers at the Beginnings of Words 

Abstract

The disyllabic verbs which have iambic stress and nouns which have trochaic stress in context of Australian English are phonologically categorized. The current study examines (1) whether monolingual native English speakers (NS) use orthographic markers at the beginnings of words to assign lexical stress? (2) If so, can the grammatical context in which words are embedded reduce NS’ reliance on orthographic markers? (3) Do non-native bilingual speakers of English (NNS) differ from NS in their demonstrated sensitivity to orthographic markers of lexical stress and/or grammatical context.  Investigating 71 NS and 50 NNS by conducting the Boston Naming Test and stress assignment tasks both in context and isolation revealed possible expectations that orthography indicates grammatical category at the onset of disyllabic words. It was found that NS were sensitive to the manifestation of the signals found in non-words when they were asked to assign stress. The reliability of orthography when allocating stress patterns decreases when it comes to the grammatical context. This study highlights that English orthography incorporates signals for lexical stress in addition to grammatical category; it will also review and critique prior studies. 

Study on Orthographic Markers at the Beginnings of Words 

Previous studies focusing on phonological acquisition have disputed the concept of lexical stress assignment within the use of both non-words and words.  The way in which orthographic markers and the grammatical categories such as verbs and nouns are positioned is imperative to finding out whether the word whether it is a non-word, or a word sounds like a a verb or a noun. These have been an influential variable of lexical stress in speech production of both NS and NNS of Australian English.

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

The stress pattern that is found in the English language depicts the disyllabic verbs and nouns. Patterns such as trochaic which is used for nouns and has stress found on the first syllable and iambic which is used commonly for verbs which is where the stress is found on the second syllable are the most noticeable and common patterns found in the English language (Arciuli & Cupples, 2007).  Past researchers have come across an issue in regard to English speakers and whether they would allocate a non-words’ grammatical category (eg. Verb or noun) via their patterns of stress.

There have been suggestions namely by Arciui and Cupples (2007) that there is an association between Orthographic indicators and grammatical category assignment. Kempt et al (2009) exemplifies this association when he stated that proficient English readers were more likely to treat pseudo words that had noun like endings as nouns as opposed to words with verb-like or control endings being treated as their respective grammatical categories. While this study was unable to illustrate why stronger reading ability, sensitivity to orthographic cues and grammatical category shared a correlation they further solidified the likelihood of such a correlation, as a result of their own findings and prior research. Due to the selection of participants for this study consisting of predominately female university students who were in their twenties there is an implicit bias, despite attempts being made to achieve a diverse sample size/pool. Therefore, the results are likely to be somewhat biased because of a lack of variation in age and gender (Kemp et al,.2009).

Previous studies highlight that individuals who are sensitive to orthographic cues to the grammatical category as well as lexical stress which is presented at the start of a disyllabic non-word are the individuals who are deemed as skilled NS (Arciuli & Cupples, 2007). An example would be when Aruicili and Cupples (2007) explored this phenomenon by examining individuals to find out whether they would be sensitive to the orthographic markers or not in relation to allocating their grammatical category. The individuals who were involved in this study were asked to complete a stress assignment task and a sentence construction task, when fifty percent took the stress assessment task the other half took the sentence construction task at the same time. Both the tasks were comprised of non-words that were selected from the dictionary analysis that was completed prior to completing the tasks the non-words used the beginnings that are characteristically used for nouns and verbs (Arciuli & Cupples, 2007). It was found that the non-words that were used as nouns had a higher chance to be allocated trochaic stress, similarly iambic stress was assigned when dealing with non-words that were used as verbs (Arciuli et al., 2009).  It was also suggested through the results that the individuals that took part in the construction task had the tendency to allocate non-words that had a noun like beginning as a noun, in addition to that non-words that had a verb like beginning would be categorized as a verb (Arciuli & Cupples, 2007). Yet, within normal reading the distinction of orthography being a cue for stress pattern grammatical category is still unknown.

Arciuli and Cupples (2007) and Davis and Kelly (1997), illustrated that NNS were typically faster and generally more accurate at identifying words that had stress patterns that followed similar patterning of grammatical category rather than words that had a different uncharacteristic pattern.

 It was found by Davis and Kelly (1997) that both NS and NNS had the ability to assess the stress patterns of pseudo words as well as disyllabic verbs and nouns. NNS illustrated more precise results than NS when they were required to create a sentence based on both iambic and trochaic pseudo words or real words. This suggested that NNS are able to allocate the grammatical category via their stress patterns. There was a gap in this study as the stimuli was acoustically presented thus no words printed out, this limitation was corrected in this study as the stimuli was printed out.

In light of previous research, the research question was proposed: Firstly, do native, monolingual, speakers of English (NS) use orthographic markers at the beginnings of words to assign lexical stress? Secondly, if so, can the grammatical context in which words are embedded reduce NS’ reliance on orthographic markers? Lastly, do non-native, bilingual, speakers of English (NNS) differ from NS in their demonstrated sensitivity to orthographic markers of lexical stress and/or grammatical context?

 It was hypothesised that :

 If NS use orthographic markers to assign lexical stress, they will:

Assign first syllable stress (the typical nominal pattern) more often to non-words with “noun-like” orthographic beginnings than to non-words with “verb-like” orthographic beginnings, when presented in isolation.

If NS use grammatical context to reduce reliance on orthographic markers to lexical stress, their preference for assigning first syllable stress to non-words with “noun-like” beginnings more often than to non-words with “verb like beginnings” will be reduced when non-words are presented in grammatical contexts that conflict with their orthography (i.e., “noun-like” non-words used as verbs, and “verb-like” non-words used as nouns).

 If NNS resemble NS in their sensitivity to orthographic markers and grammatical context, they will also: 

Assign first syllable stress more often to “noun-like” non-words than to “verb-like” non-words presented in isolation.

Show a reduced preference for assigning first syllable stress to non-words with “noun-like” than “verb-like” beginnings when non-words are presented in grammatical contexts that conflict with their orthography.

Method

Participants

In this study the participants comprised of 71 NS and 50 NNS all these participants were volunteers who were undergraduate student that attended Macquarie University.

Tasks and Materials

The tasks that were involved in this study was the Boston Naming Test (BNT) which was used to evaluate English vocabulary knowledge; secondly a stress assignment in isolation was conducted so the participants can choose if they prefer to put stress on either first or second syllable (appendix); Lastly a stress assignment within context, the purpose of the third task is similar to the second task, it is also conducted to see if a circumstantial difference would impact participants stress preference.

The materials that were used in this study were chosen from Arciuli and Cupple’s (2007) study. The materials can be found in the appendix.

Procedure

 All the students that were available and present in the tutorials were asked to participate in the experiment as the experiment took place as a normal class activity.  There were three tasks that the students had to participate in; the Boston Naming Task and the stress assignment in both isolation and in context.  For the BNT the students were requested to write the names of sixty images that were shown in a PowerPoint slide (appendix). For the non-word task in isolation, there was three columns of twenty-four no-words, twelve noun like and twelve verb like that were given to the students that participated in the study; they were asked to choose which part of the word they wanted to emphasize and example would be custew for the beginning and custew for the end. Similarly, for the last task non-word in context they were given twenty-four sentences with different non-words within each sentence again twelve noun-like and twelve verb-like. The participants were then asked to underline which section of the non-word they decided to put more emphasis on. The participants were specifically asked to not read the non-words that were targeted out loud but to instead just think about how they were stressed.

Results

The results in this study signified four key findings. The Vocabulary results from the BNT illustrated that NS’ mean (51.9) did better than the NNS (45.4) which was expected (table 1). Nevertheless, the SD=9.0 illustrating that the vocabulary of NNS does vary.  Suggesting that NNS have a general English vocabulary while some NNS may not know a lot of English words.

Table 1

Mean BNT scores (and SDs) as a function of speaker group

Group

Native speakers (n = 71)

51.9 (4.5)

Non-native speakers (n = 50)

45.4 (2.0)

Group Difference

6.5

t-value and significance level

t(119) = 5.35, p < .001

Secondly, it was shown that the grammatical context did influence their choice of stress pattern. NS did show an inclination to take orthography into account when completing the stress assignment; this supports hypothesis. Which is NS having the tendency to stress the initial syllable of noun like non-words (6.69 out of 12) than the verb like words (3.39 out of 12). For noun like non-words, first syllable stress reduced from 6.69 in isolation to 5.75. This illustrates the tendency of NS to select the first syllable stress pattern for verb like non-words in context (4.39) rather than in isolation (3.39).

 Thirdly, the stress patters found in NS were identical to NNS. This illustrates that NNS like NS favor stressing the first syllable of noun like non-words (6.34) than verb like non-words (2.68), showing NNS take orthography in to mind when determining the stress patterns. In addition, the number of first syllable stress reduced from 6.34 in isolation to 4.84; highlighting that grammatical context does play a role in changing the stress patterns. Furthermore, NNS illustrated a preference when it came to put stress on the first syllable for noun like non-words as there was a difference of 3.66 in regard to no context; 0.58 for context this was higher than NS’s preference as there was a difference of 3.30 not in context; 0.36 for in context.

Table 2

Stress assignment as a function of orthography, context, and speaker group (number of first syllable stress out of 12)

Native Speakers

Non-native Speakers

Orthography

No context

Context

No context

Context

Noun-like

6.69

5.75

6.34

4.84

Verb-like

3.39

4.39

2.68

4.26

Difference

3.30

1.36

3.66

0.58

Finally, it was discovered that there are three predominant elements which influence a speakers’ choice: Grammatical context and orthography and the way in which they interact the most (significant effect is the interaction). There was no significant main effect of language background and interactions. This indicates the same pattern of sensitivity in NS as NNS (Table3).

  Table 3

Analysis of variance (ANOVA): Summary of main findings

Orthography

F(1,119) = 132.66 (p < .001)

Context

F(1,119) = 0.03 (p = .865)

Orthography x context

F(1,119) = 38.5 (p < .001)

Discussion

The present study was set out to investigate if orthography and grammatical category are possible cues to lexical stress. All 3 hypotheses were supported, as orthography did have a tendency to affect the stress patterns for non-words for both NS and NNS.  NS typically decide to choose trochaic stress when it comes to non-words that have noun like features and similarly they like to lean towards iambic stress for non-words that have verb like features.  The sensitivity of stress markers were the same when it came to both NS and NNS. In addition to that NNS did illustrate a larger difference when it came to either stressing the initial syllable in noun like non-words and verb like non-words in both ‘isolation and in context’ rather than NS.

The present study results were consistent with and also extended upon Arciuli and Cupples (2007) as both studies found that Native English Speakers used the orthography and grammatical category to allocate the stress patterns of non-words. As this study followed the guidline of Arculi and Cupples (2007) it was able to further the research of how Non-native bilingual speakers of English have tendencies to allocate the lexical stress of non-words as there are not many studies that have performed such an experiment.

A strength of this study is that it is easily replicable increasing its reliability. However, a limitation could include the use of a convenience sample. All the students who participated in this study were psycholinguistic students. As the unit all the students are enrolled in is a 200-level unit they would have had to complete a 100-level unit beforehand as a prerequisite which were linguistics or psychology, this could alter the results as the participants could predict what the study is about.

In conclusion, this study has extended on previous research by demonstrating that orthography and grammatical category have been found to be cues to lexical stress also that English native speakers use orthographic markers at the beginnings of words to assign lexical stress.  This is a very important contribution, and future research should extend on this work by using NNS with different levels of English taking into account English fluency.

References

Arciuli, Joanne, Cupples, Linda, & Macquarie University. Department of Linguistics. (2007). Would You Rather ’embert a Cudsert’ or ‘cudsert an Embert’? How Spelling Patterns at the Beginning of English Disyllables Can Cue Grammatical Category.

Arciuli, J., & Monaghan, P. (2009). Probabilistic Cues to Grammatical Category in English Orthography and Their Influence During Reading. Scientific Studies of Reading, 13(1), 73-93.

Cupples,L. (2016. Introduction to Psychololinguistics class experiment, 2016[Power Point slides]. Retieved from Ling214, Macquarie University, Sydney, NSW, Australia.

Davis, S. M., & Kelly, M. H. (1997). Knowledge of the English noun–verb stress difference by native and nonnative speakers. Journal of Memory and Language, 36(3), 445-460.

Kemp, N., Nilsson, J., & Arciuli, J. (2009). Noun or verb? Adult readers’ sensitivity to spelling cues to grammatical category in word endings. Reading and Writing, 22(6), 661.