Research on reading has traditionally focused on how children learn to map letters and syllables onto sounds, which is critical for understanding the mental processes involved in the very early stages of reading development. However, little is known about the processes that are at play at later stages of reading development, when children learn to read more complex words that consist of smaller meaningful units (e.g., teacher:  teach + er = someone who teaches), called “morphemes”. Morphologically complex words form the vast majority of words in most of the world’s languages. The MORPHEME project investigates how and when children learn to decompose such words into their constituent units in an innovative, longitudinal, interdisciplinary, and cross-linguistic fashion.

Illlustration Lesende Maus
© Max-Planck-Institut für Bildungsforschung / Max Planck Institute for Human Development

Significance and innovation of the MORPHEME project

Skilled readers can rapidly and effortlessly decompose morphologically complex words into smaller meaningful units, known as morphemes. Recognizing morphemes during reading is critical for mapping word forms onto meanings, because the stems preserve their meaning in the derived forms (e.g., the meaning of teach does not change in teacher or teaching), while the affixes modify the meanings of the stems in a highly consistent manner (e.g., the affix –er always refers to someone who performs a certain action). Hence, the rapid identification of morphemes promotes efficient reading and further facilitates the recognition of unfamiliar and new words (e.g., unfriend). The aim of the MORPHEME project is to investigate the mental processes involved in morphological decomposition during language processing and to determine how and when during reading development, morphological knowledge becomes incorporated into the reading system. Further, this project investigates these research questions longitudinally and cross-linguistically, by comparing the French and German languages, which differ considerably in terms of their morphological productivity.

The ANR-DFG funded project MORPHEME is a collaboration between researchers from the Research Group REaD at the Max Planck Institute for Human Development (MPIB) and the Laboratoire de Psychologie Cognitive at Aix-Marseille University (CNRS). Both German and French partners are experts in research on reading and use a combination of methodologies from experimental cognitive psychology, computer science, linguistics, and education to investigating literacy development.

Aims of the MORPHEME project

Aim 1: To determine how and when during reading development children acquire the ability to decompose morphologically complex words into their constituent morphemes.

Aim 2: To investigate how the degree of morphological productivity of a language may influence the acquisition of morphological knowledge.

Aim 3: To test if the empirical findings on morphological decomposition during word identification generalize to natural reading.

Aim 4: To use the empirical data to further develop and test computational models of reading aloud that provide an explicit account of how morphological decomposition occurs in the skilled and developing reading systems.

How MORPHEME will be conducted

We will investigate how and when during reading acquisition morphology comes into play using a variety of tasks that tap word identification processes (e.g., aural and visual lexical decision, reading aloud, masked priming). Such tasks will be administered to German and French children in a longitudinal fashion (Grades 2 to 4). Moreover, we will investigate whether the observed effects generalize to a more natural reading situation such as sentence reading. The use of eye-tracking techniques in this task to record eye movements during reading will further allow us to uncover potential differences in the processing of prefixed and suffixed words, which is critical for delineating the nature of morphological decomposition during reading development. The reading aloud data will also be used to evaluate and further develop the French and German versions of the Connectionist Dual Process computational model of reading aloud. These computer simulations will allow us to gain a better understanding of the processes underpinning morphological decomposition in the skilled and developing reading systems and to provide an explicit account of the processes involved in learning to read the vast majority of words that comprise our vocabulary.

Applied implications of the MORPHEME project

Previous research on reading has focused on the importance of a phonics-based approach to learning to read. However, in languages such as French and German, children are thought to move beyond phonological decoding early on during reading development. The MORPHEME project seeks to investigate the processes that are at play at later stages of reading development, when French and German children learn to read morphologically complex words. Thus, the findings from this research will offer the potential to develop new evidence-based programs of teaching instruction in both France and Germany. These findings will be of most interest to cognitive psychologists interested in investigating how children learn to read, though we also envisage interest from clinicians studying language impairment and from professionals working in the area of literacy education and policy. Our aim is to contribute to a wider conversation between academics and practitioners in the pursuit of developing a sound and effective literacy strategy.


For questions or comments concerning MORPHEME please contact
Betty Mousikou

mousikou [at] mpib-berlin [dot] mpg [dot] de