The role of systematicity between orthography and morphology in reading and writing

Lead Research Organisation: Royal Holloway, University of London
Department Name: Psychology

Abstract

English is a language that is difficult to read and to spell. Words that are written similarly often have different sounds (compare bush and lush). Further, words that sound similar are often represented differently in writing (compare field and feel). Yet, English is the most widely spoken language in the world. Children and adults learning to read and write in English (in their mother tongue or foreign language) must cope with this complexity. Cross-linguistic studies on literacy acquisition have consistently shown that English-speaking children make many errors in reading and spelling and lag behind their peers who learn to read in other European languages (Seymour et al., 2003).

However, English does have important regularities between spelling and sound. Substantial research effort has been undertaken in recent years to understand these regularities, and to teach learners to capitalise on them to support their literacy development. This research has had powerful implications for literacy teaching, and has motivated the implementation of systematic phonics instruction in UK schools (Rose, 2006).

This project focuses on a different kind of regularity that has received relatively little research attention -- that between orthography and morphology. It turns out that one can predict aspects of a word's spelling if its grammatical class is known (noun, adjective). For example, if one hears an adjective that ends in the sound "ick", then one can almost be sure that the correct spelling is -ic (e.g. cosmic). Conversely, if this word is not an adjective, then it is likely to have a different spelling (for example, -ik, beatnik). Such regularity was described decades ago, but more recent evidence has shown that it is ubiquitous in English (Berg & Aronoff, 2015). Interestingly, systematicity between orthography and grammatical class is unique to spelling; the spoken forms of words do not carry this meaningful information.

Is it possible that humans unconsciously make use of this morphological information encrypted in English spelling? We will conduct a series of five experimental investigations in skilled adult readers to find out whether this is so, using classic behavioural, as well as cutting-edge eye-tracking and word learning methods. Our experiments will ask whether participants show enhanced reading of words whose spelling accords with their grammatical class. We will also determine whether participants produce different spellings for spoken novel words depending on their morpho-syntactic role in a sentence. Finally, we will ask whether participants show superior learning of novel words that are spelled consistently with their morpho-syntactic function.

The finding that skilled readers are sensitive to regularities between spelling and morphology would lead to an intriguing question: is explicit emphasis of these regularities beneficial for learners of English? If so, in the longer term, we expect that highlighting the regularities between spelling and morphology could enhance literacy instruction programmes, and yield positive outcomes for children and adults learning to read and spell English.

This work will be guided by a psychologist with expertise in the processes that underpin reading (Prof. Rastle) and a linguist with expertise in morphology (Prof. Aronoff, Stony Brook), both of whom are international leaders in their fields. Thus, we propose an international collaboration between the disciplines of psychology and linguistics, which should give rise to novel research questions and surprising theoretical developments. We expect this work to have a strong immediate impact on several disciplines within academia. Further, we have designed a strong programme of engagement with non-academic stakeholders to ensure that this work's potential for enhancing methods for literacy instruction is realised.

Planned Impact

The proposed work has the potential to influence the current state of affairs both within and outside academia.

On the academic side, the alliance between psychologists and linguists who strive to understand how morphology is represented in the English orthography and in the human brain will have a double effect: first, this work will become known to a wider community of scholars including psychologists, linguists and psycholinguists; and second, joint perspectives may foster innovation in both fields. The dissemination strategy covers a wide scientific audience and ensures that the researchers in domains adjacent to cognitive psychology, for example, developmental psychology and second-language acquisition, will benefit from this work. It is especially important to embrace these fields, because in the future, our proposed line of research may be followed up by intervention studies of reading and spelling acquisition, and our work is expected to have implications for enhancing methods for reading and spelling instruction (see Pathways to Impact).

On the non-academic side, this work is expected to inform education through the enhancement of methods for teaching reading, spelling, and vocabulary in English. This goal is, beyond question, long-term, for two reasons. First, realistic goals should be set that can be accomplished within the duration of this grant. Second, we embrace the tenets of evidence-based practice, which means that, prior to any practical applications, interventions following from this project's findings will need to be designed and evaluated. Nonetheless, we keep these distant goals in mind, and suggest pathways for our work to have a positive influence on a wide range of non-academic users already in the short-term.

First, we intend to disseminate the outcomes of this research widely in general and more specialised outlets, targeting primary and secondary school teachers and professionals involved in teaching adults to read and write in English (as a native or a foreign language). Raising awareness of the role of morphology in reading and spelling among teachers and educators results in significant gains in spelling among their students (Hurry et al., 2005). If the hypotheses proposed are upheld, then these specialist teacher audiences will benefit from knowing that regularity between the spelling of morphemes and their morpho-syntactic function is common in English, and this regularity facilitates reading, spelling and vocabulary acquisition in adult native speakers. In addition, we will make the list of regularities between orthography and morphology freely available online on the project website, so that the teachers can use it in their practice.

Second, we shall focus on building the necessary foundation to facilitate future research and intervention work. In particular, we intend to build a network of professionals who are interested in evidence-based methods to support literacy acquisition, consisting of teachers, educators, policy makers, school governors, professionals involved in developing children's reading books, and other practitioners. I will join two relevant communities (Forum for Research in Language and Literacy, ResearchED) and attend their meetings on a regular basis and interact with target groups via social media. The formed links will be strengthened by organising a workshop on the role of morphology in reading and spelling (for both academic and non-academic audiences) toward the end of the grant. A test bed for evaluating novel interventions will be in place by the end of this fellowship, which will enable a smooth transition of this project to the next phase and secure maximal impact in the future.

Publications

10 25 50
 
Description The study of English spelling has focussed almost exclusively on letters and sounds, the inconsistency of correspondence between these, and difficulties that this inconsistency poses for L1 and L2 learners of English. For instance, different spellings exist for the same final sound sequence in words bonus, atlas, and nervous, and this variability has been traditionally viewed as a complication or even a defect of English spelling by linguists and psychologists alike. Our study shifts the perspective onto the relationship between letters and meanings. We were interested to see if any cognitive advantages are gained through the proliferation of spellings in English. To test this idea, we first conducted a large-scale analysis of English suffixation and demonstrated that even when spelling-to-sound consistency is low, spelling-to-meaning consistency tends to be high. We designed and described tools for measuring the amount of unique spelling-to-meaning information that is carried by suffix spellings as opposed to suffix pronunciations that can be used in future research. Further, in a series of psychological experiments (novel word classification, spelling, and natural sentence reading with eye tracking), we showed that adult skilled readers are sensitive to meaningful information that is carried by English suffix spellings and exploit it rapidly when they are engaged in reading and writing. Importantly, we demonstrate that the degree of readers' sensitivity mirrors the strength of spelling-to-meaning regularities in the writing system. This finding indicates that this meaningful information supports our ability to read and to write, and it can be learnt by most people from experience with language. Our findings have been published in a high-impact interdisciplinary journal "Cognition". We are currently working on a follow-up study investigating the mechanics of the acquisition and development of spelling-to-meaning representations using a artificial language learning approach.

One objective of the Future Leaders award was to provide opportunities for the PI to develop general meta-academic skills, such as dissemination of findings and non-academic communication. The PI presented the outcomes of this award at a number of international and national conferences (Psychonomic meetings in Amsterdam, New Orleans; European Society for Cognitive Psychology in Potsdam; Experimental Psychology Society meeting in Reading; Statistical Learning conference in Bilbao; Morphological Processing conference in Trieste). The PI was invited as a keynote speaker to a specialised workshop on orthography and morphology organised by Dr Kristian Berg in Delmenhorst, Germany. Further, the PI has become a member of the Forum for Research in Language and Literacy (FRiLL) that brings together academics and practitioners interested in questions that surround literacy and literacy acquisition. One presentation was delivered at their annual conference. Further, Dr Ulicheva co-presented with Prof Rastle at the South East Research Network for Schools Research and Practice. This was an opportunity to raise awareness about the importance of morphology in the English writing system and to exchange knowledge with educators and teachers. On the academic side, the PI has developed a collaboration with Prof Mark Aronoff following a research visit to Stony Brook University in the USA early in the fellowship. Further, she benefited from visiting Dr Marco Marelli to learn advanced computational techniques for corpus analysis (i.e. distributional semantics). Dr Ulicheva has submitted several funding bids, one of which was successful (Marie Curie Individual fellowship).
Exploitation Route Spelling-to-meaning regularities that we have described may have important implications for how spelling is taught in the classroom. Our study demonstrates that spellings that appear inconsistent on surface are in fact predictable when word meaning is taken into account. For instance, adjectives cannot be spelled with -as or -us at the end: the -ous spelling must be used instead. Importantly, our experiments show that readers are capable of learning and applying these spelling-to-meaning dependencies when they perceive and produce written text. We have formulated a set of most frequent spelling rules that can be used in educational settings. While further research needs to test the efficacy of using these rules for spelling instruction, raising awareness of these regularities among teachers may already have a positive effect on students' reading and spelling outcomes - this is the impact path we are currently pursuing.
Sectors Education

URL https://doi.org/10.1016/j.cognition.2018.09.013
 
Description In this project, we identified and described a level of regularity that presents English spelling in a new light. We have shown that components of words can be spelled differently depending on word meaning, and these dependencies between spellings and meanings are ubiquitous in English. Sporadic evidence on this has been reported in linguistic literature, but our work is the first successful attempt to describe this regularity on a larger scale - as an important property of the English writing system as a whole. Importantly, we have shown that skilled readers of English capitalize on systematic relationships between spelling and meaning, which opens numerous possibilities for translating our research findings into practice and improving the way in which English spelling is currently taught to children and second-language learners. We have explored several impact paths. Firstly, Prof Rastle and Dr Ulicheva presented at the South East Research Network for Schools https://www.royalholloway.ac.uk/research-and-teaching/departments-and-schools/psychology/research/serns/ This meeting brings together teachers, educators, practitioners, and researchers. We covered general issues surrounding the learning of morphology, and shared novel findings arising from our project. In particular, a list of meaningful regularities was constructed and distributed to all attendees. This raises awareness of morphological regularities that are encoded in English spelling. We are excited about the possibility that teachers will draw on this information in their teaching practice. Further, our findings are influencing major works on reading, that are being discussed widely in the education community. For example, this research was cited in Castles, Rastle, & Nation (2018), in a well respected outlet for sharing psychological research with general audience (over 55,000 downloads), which in turn has just been featured in an impact publication for teachers, Deans for Impact. Castles, A., Rastle, K., & Nation, K. (2018). Ending the Reading Wars: Reading Acquisition From Novice to Expert. Psychological Science in the Public Interest, 19(1), 5-51. https://doi.org/10.1177/1529100618772271
First Year Of Impact 2018
Sector Education
Impact Types Policy & public services

 
Description Marie Curie European Individual Fellowship
Amount € 183,455 (EUR)
Funding ID 747987 
Organisation European Research Council (ERC) 
Sector Public
Country European Union (EU)
Start 10/2018 
End 09/2020
 
Title Behavioral data from judgement, spelling, and eye-tracking experiments 
Description These experiments tested skilled readers' sensitivity to meaningful information contained in word endings. Associated data can be used by other researchers to investigate similar research questions. The database is publicly available. 
Type Of Material Database/Collection of data 
Year Produced 2017 
Provided To Others? Yes  
Impact Our paper "Skilled readers' sensitivity to meaningful regularities in English writing" was published in Cognition. 
URL https://osf.io/hac5j/
 
Description Dr Marco Marelli 
Organisation University of Milano-Bicocca
Country Italy 
Sector Academic/University 
PI Contribution Contributing the data, research questions, and performing analyses
Collaborator Contribution Expertise in distributional semantics, naive discriminative learning and other modelling approaches
Impact No outputs yet. This collaboration is multi-disciplinary (computational linguistics and psychology).
Start Year 2018
 
Description Mark Aronoff 
Organisation Stony Brook University
Department Linguistics Department
Country United States 
Sector Academic/University 
PI Contribution 1. Expertise in computational linguistics and psychology 2. Research visit to Stony Brook ( Dr Anastasia Ulicheva and Prof Kathy Rastle) 3. Two talks given at Stony Brook by Dr Anastasia Ulicheva and Prof Kathy Rastle to promote our cognitive psychology work among linguists
Collaborator Contribution 1. Intellectual input, expertise in theoretical linguistics
Impact Abstracts jointly co-authored by Ulicheva, Aronoff & Rastle were submitted to several conferences. A paper is in preparation. This collaboration is multi-disciplinary, between the disciplines of linguistics and psychology.
Start Year 2016
 
Description SERNS research and practice meeting 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Professional Practitioners
Results and Impact Dr Ulicheva co-presented with Prof Rastle. We shared general issues surrounding learning to read and learning of morphology, as well as relevant results arising from this project with teachers, educators, and practitioners. We also formulated and distributed a pamphlet with 40 spelling rules that could be used by teachers in their practice.
Year(s) Of Engagement Activity 2018