In search of cognitive factors in functional illiteracy
Insufficient reading ability in adulthood has consistently been reported in large-scale studies worldwide (e.g. Grotlüschen et al., 2019; OECD, 2016). According to a recent survey carried out in Germany, 12.1% of the working-age population is able to process written sentences but not continuous texts, even if they are brief (LEO 2018, Grotlüschen et al., 2019). Notably, more than half of the adults with low literacy skills in this survey spoke German as their native language. Thus, possible insufficient competence in the testing language due to a migratory background, may only explain part of these results. When adults with low literacy skills present poor functional use of printed information, they are referred to as functional illiterates. It has been suggested that this definition be operationalized by relating it to one’s ability to subtract meaning from reading and to express oneself in writing at a reasonable pace (Egloff et al., 2011).
Alongside social-cultural and/or emotional disadvantages, cognitive deficits have also been suggested to underlie functional illiteracy (Egeloff et al., 2011; Eme, 2011). The question which cognitive factors are associated with functional illiteracy is the focus of this symposium, and is common to its three presentations. Notably, while there is an extensive research on the cognitive factors of children with reading difficulties, studies exploring this subject in adults with low literacy skills are scarcer.
The first two presentations of this symposium address the relations between functional illiteracy and dyslexia –a developmental learning disability explained by cognitive deficits. The first presentation of Vágvölgyi and her colleagues, explores this subject by carrying out a systematic review of scientific quantitative studies, which include the two types of populations. While taking into account the extensive research on dyslexia, and the more rare research on functional illiteracy, the question at the center of this review is what knowledge is transferable from research on dyslexia to that of functional illiterate adults. The results of this review point to difficulties in a direct transfer based on the existing literature, as the studies differ, among others, in the inclusion criteria of the two populations. More research is needed then in order to understand to which extent functional illiteracy and dyslexia share common characteristics. This is the question at the center of the second presentation of Bar-Kochva and her colleagues. In their study of 54 functional illiterate adults, it was examined whether the diagnostic criteria of dyslexia apply to them. The vast majority of the sample (above 80%) presented the same deficits in the basic reading skills (word reading and decoding), which are the main deficits defining dyslexia. In addition, about 50% of the sample presented a deficit in the main reading-related cognitive skill thought to underlie dyslexia (phonological awareness). The third presentation of Bulajić and his colleagues extends the examination of the cognitive factors of functional illiterates to more general cognitive skills: memory systems and IQ, and examines the relations of these skills with the actual results achieved by these adults in a basic adult literacy program. In this study of 63 adults, the optimal model explaining progress in this program included two predictors: verbal intelligence and complex working memory. Considering the significant heterogeneity among functional illiterates (Vágvölgyi, 2018), a better understanding of this population is required both from a theoretical and a practical perspective. The results of the three presentations add in this respect by shedding some light on the cognitive profiles of functional illiterate adults and on the cognitive factors, which play a role in their ability to acquire literacy skills in adulthood.
Beiträge des Symposiums
Learning from the past: What knowledge is transferable from the research into dyslexia to that of low literate adults?
Despite having average intelligence and adequate basic education, children diagnosed with developmental dyslexia still encounter considerable difficulties in the full acquisition of their reading and writing skills (WHO, 2015). Although their basic skills might improve with the support of different intervention programs, some difficulties persist into adulthood (e.g. Law et al., 2015).
Moreover, although being a well-known phenomenon for decades, developmental dyslexia may remain undiagnosed or untreated in childhood. Thus, certain studies assume that dyslexia can be one of the causes of becoming functionally illiterate (i.e. low literate) in adulthood (Greenberg et al., 1997). Recent surveys have shown that the high prevalence of developmental dyslexia among low literate adults in Germany is higher (7-15 %, Grotlüschen & Solga, 2019; Fickler-Stang, 2011) than in the general population (2.1%; Fischbach et al., 2013). Furthermore, low literate adults performed similarly in decoding, word recognition, and text reading tasks, as would be expected from dyslexic readers, thus indicating common difficulties of both groups (Bar-Kochva et al., under review). Although, there is a number of data suggesting commonalities between dyslexia and low literacy, the available evidence has not been systematically reviewed.
We aim to investigate in our ongoing project, whether the research findings regarding risk and protective factors, cognitive processing profile, prevention and intervention on developmental dyslexia are transferable to low literate adults.
In order to address the research question, a systematic literature review is being conducted. We searched 9 electronic databases to identify all the available evidence with original quantitative data, written in either English or German, including both low literate and dyslexic samples. The screening of the identified publications, the data extraction and the quality evaluation of the fitting studies were carried out by two independent coders. The quality of the included studies was assessed using the Quality Assessment Tool for Reviewing Studies with Diverse Design (QATSDD; Sirriyeh et al., 2012).
Our search identified 6.973 publications, of which 5 met the inclusion criteria. Low literate adults are mostly investigated from an educational, while dyslexic individuals from a psychological point of view. The two disciplines have in general different research focus and are using different methodology, which can explain the low number of studies including both samples. Furthermore, the fitting studies point to the diversity of the topic: a variance has been found between the 5 studies in critical categories, such as in the applied criteria and definition for functional illiteracy, in the testing methods and in the quality of reporting. This variance indicates that more direct comparisons have to be made in order to understand to which extent functional illiteracy and dyslexia share common characteristics. Although these preliminary results impose a difficulty on comparing the two populations, some similarities and differences between the two concepts will be presented.
Do functional illiteracy in adulthood and dyslexia share common characteristics? An examination of readers of a transparent orthography
Functional illiteracy has been suggested to be caused by multiple factors (Rüsseler, Boltzmann, & Grosche, 2019). One of these is dyslexia, defined by difficulties in accurate and/or fluent word recognition and by poor spelling and decoding skills (Lyon et al., 2003). Deficits in phonological awareness and rapid-naming have also been shown to be related to dyslexia (Wolf & Bowers, 1999). While the definition of functional illiteracy puts an emphasis on insufficient reading comprehension (Egloff et al., 2011), reading comprehension is not considered a primary deficit in dyslexia (Lyon et al., 2003). Studies examining more basic skills of reading (e.g. word-reading and decoding) in functional illiterate adults were carried out mainly with readers of English (e.g. Greenberg et al., 1997; Mellard et al., 2010). However, due to its opaque spelling-sound relations, the English orthography has been shown to impose special demands of processing –and particularly on reading accuracy (Share, 2008).
This study set out to examine the relations between functional illiteracy and dyslexia among adults who read the more transparent German orthography. The study adds to the few previous studies on functional illiteracy among readers of German (e.g. Grosche, & Grünke, 2011) a separate analysis of reading accuracy and fluency. This study aims at contributing to the understanding of the cognitive factors involved in functional illiteracy in adults who struggle with reading a transparent orthography.
Three questions were addressed: (1) Do functional illiterate adults reading the German orthography present the same difficulties associated with dyslexia in a transparent orthography, i.e. in fluency in decoding and in word recognition, in addition to deficits in phonological awareness and rapid-naming? (2) Is accuracy in reading affected in these adults despite reading a transparent orthography? (3) To which extent the deficits associated with dyslexia explain the variance in reading comprehension?
The sample included 54 participated recruited from vocational trainings and literacy classes (ages 16-63, 33 men). These participants presented a reading comprehension level below the level of 6th graders. Participants were born in Germany, had a normal non-verbal IQ score, and did not have a diagnosed hearing deficit, uncorrected sight problems or an attention and hyperactivity disorder. Standardizes tests were administered, including: reading comprehension, sentence reading, word reading and decoding (pseudoword reading), phonological awareness and rapid-naming.
The first and second questions were examined by comparing performance of participants to the tests' norms. Participants showed on average significant deficits in measures of reading fluency (decoding, word recognition, and text reading), as well as in phonological awareness (but not in RAN). Frequency analyses confirmed that these averages represent the vast majority of the sample in the case of reading fluency, as 83%-89% of the sample presented insufficient performance in these measures. A smaller proportion of the sample –approximately 50%, showed an additional weakness in phonological awareness, while a deficit in rapid-naming was observed in only 25% of the participants.
The examination of the second question indicated deficits also in reading accuracy (when compared to the norms). In addition, an advantage in this measure was obtained in reading of single words compared to decoding and text reading.
The third question was examined using hierarchical regression. Measures of reading fluency and accuracy explained a considerable amount of variance (13%, p < .001 and 20% p < .001, respectively) in reading comprehension beyond age, IQ and phonological awareness.
Together, these results indicate common difficulties of functional illiterate adults and dyslexic readers. Despite reading a transparent orthography, difficulties in both accuracy and fluency in the basic skills of reading appear to impose an obstacle on the ability of functional illiterates to comprehend texts.
Which cognitive capacities best predict learning success of adults who are functionally illiterate?
Previous research investigating the relationship between cognitive capacities and success in formal learning focused mainly on children (Alloway, 2006; Alloway, 2009; Gathercole et al., 2004a; Gathercole, 2004b) and university students with and without learning difficulties (Alloway & Gregory, 2013; Gropper & Tannock, 2009; Ishak et al., 2012; McKenzie & Schweitzer, 2001). Some studies showed that working memory (WM) predicts academic success as strong, or even stronger than intelligence in children with or without learning difficulties (Alloway, 2009; Alloway & Alloway, 2010). The contribution of IQ and WM to explaining the variance of learning success has further been found to decreases with age and educational level (Abraham et al. 2012). However, when it comes to university students with learning difficulties such as dyslexia, IQ and WM were found to be strong predictors of learning success (Alloway & Gregory, 2013). Not much is known about the cognitive profile of functional illiterate adults who have little educational experience. The aim of the present study was to investigate the role(s) played by basic cognitive capacities (short-term memory [STM], WM and IQ) in learning success in this population.
In this study, we examined whether and to which extent the cognitive capacities of STM, WM, general and verbal intelligence predict learning success in functional illiterate adults who attend an adult basic education program (ABE). An additional question raised was which of these capacities make up for a larger individual contribution to the variance in learning success.
Sixty-three functionally illiterate adults (Age: 17-28, M=21.8, SD=3.19; Sex: M=32, F=31) participated in the study. All were students of six Serbian ABE institutions, attending a 3-year intensive functional adult basic education formal programme. Participants had little to no completed grades of elementary school in childhood. Cognitive capacities (predictor variables) were measured by different STM and WM tests/tasks: Digit forward, Digit backward (VITI – WAIS III Serbian standardization), Faces I, Family pictures I (WMS III), and two STM/WM tasks created for the purpose of this study (KI-A1, KI-A5). In addition, a nonverbal test, Raven’s SPM was administered as a measure of general intelligence, while the verbal intelligence assessment included the Vocabulary and Comprehension tests from the VITI.
Learning success (criterion variable) was represented by the overall ABE average grade, i.e. learning success in various subjects (grade point average GPA), including Serbian language (literacy) and Mathematics. In addition, average grades in following subjects were taken as learning success measures: Serbian language, Mathematics, English language, Digital literacy, Applied natural sciences. These were used as an ecologically more relevant alternative to the literacy and basic skills testing.
Correlation analyses and backward elimination regression analyses were conducted to determine the relationship between cognitive capacities and learning success. All cognitive measures were significantly related to GPA and subject grades, except for recognition of faces STM (Faces I). Results of the regression analyses suggest that the optimal model of predicting ABE students’ GPA includes two predictors: verbal intelligence and complex working memory (CWM). These two conjointly explain 34% of the GPA’s variance, while there is a greater contribution of verbal intelligence (18.2%) than CWM (6.7%) to learning success variance. The same two predictors explain even more variance of learning success for average grades in subject of Serbian language (literacy): 37.6% (verbal intelligence 15.6%, CWM 11%).
These findings express the importance of verbal intelligence for learning success, and indicate that WM is a critical predictor of learning success in functional illiterate adults. Furthermore, it indicates potential similarities of the pattern of cognitive predictors of literacy skills of ABE students and university students with learning difficulties (Alloway & Gregory, 2013).