In recent studies we have been investigating mirror writing by typical 4- to 6-year-old children. The term is used because the characters – numbers and capital letters – are reversed, yet are correct when looked at in a mirror. In the case of standard writing, the mirror must be placed to the right or to the left, perpendicular to the horizontal plane of writing. There can also be vertical mirror writing, which look correct if we look at them in a mirror placed below or above.
At first one might think that children, who often sit face-to-face in kindergarten, reverse the characters because they see them on the sheet of the child who faces them. But this is not so because such an origin of reversal would lead children to double mirror writing – reversed both horizontally and vertically (see Figure 1).
The origin of the horizontal mirror writings – particularly impressive when the letters are in cursive and thus attached, see the writing of Joséphine in Figure 2 – has long remained a mystery. But it can be slightly disconcerting as well because children spontaneously produce writings they have never seen before and certainly didn’t learn. The American linguist Noam Chomsky has essentially used such an argument – that children make sentences they have never heard (nor read, of course) – to support the notion that language is innate. On the contrary, we will see here how the horizontal mirror writing of the characters is explained by the culture, within the constraints imposed by the cerebral processing.
Mirror writing entered the scientific literature with an article of the German neurologist Alfred Buchwald in 1878 (in German, mirror writing is called Spiegelschrift), but over the following 125 years the explanations for the phenomenon were not only insufficient but also often wrong. One of the main reasons for the explanations’ failure is that they often involved a “culprit” – writing with the left hand. For a long time, this dominant discourse was supported by the observation of left-handed children writing reverse characters, their names or even whole words and sentences. Thus throughout the 20th century, scientific journals have published mirror writing almost exclusively produced by left-handed children. Even today, left-handedness is often the favourite explanation of teachers when children produce mirror writing.
Cerebral and behavioural components
The explanation we find for the phenomenon of mirror writing of characters works on two successive levels, the first cerebral and the second behavioural. The cerebral level was long been limited to the simplistic 1925 theory by Samuel Orton that one of the cerebral hemispheres (usually the left) would correctly represent the letters while the other would represent them in mirror. More recently, however, it has been shown that the brain eliminates orientation (left or right) when storing images, a process called symmetrisation or mirror generalization. This mirror-generalization process, which can be very useful – for example, to recognise a face by both its left and right profiles – leads children aged 5 to know, from memory, the shape of the characters, but not their left/right orientation. Given the features of the process – horizontal mirror in the visual modality – it is important to note that the initial implicit learning of the form of the characters by the children is mainly visual, and that the children produce almost exclusively horizontal mirror writing.
At the behavioural level, when children write the characters from memory, they must give them an orientation. In countries whose primary languages are written in Latin characters – written from left to right – children most often point them toward the right. This leads them to reverse mainly the left-oriented characters: J, Z, 1, 2, 3, 7, and 9 will be considerably more reversed than other characters (see Figure 3). However, when spatial constraints cause them to write from right to left, the children instead reverse the right-oriented letters (see the E, N, and C of MAXENCE in Figure 2). This suggests that children usually orient characters in the direction of their writing.
Since this explanation has nothing to do with the handedness of the children, it can be predicted that right-handed children will reverse characters almost often as left-handed children, and that – left-handed or right-handed – children will primarily reverse left-oriented characters in Western culture. This prediction was confirmed in a 2016 study, as was another subtle prediction: children who reverse left-oriented characters the most are also those who reverse the right-oriented characters the least. This because they strictly orient the characters in the direction of writing.
Our explanation is supported by an analysis of tens of thousands writing samples from more than a thousand children that was published in different journals, notably in the Journal of Educational Psychology. The theory remains relatively unknown, perhaps due to its recency, and some parents continue to wonder if mirror writing by their children might be the precursor of a disorder such as dyslexia. And some paediatricians or occupational therapists today still have no answer other that suggest thwarted left-handedness or a bad lateralization in the child, neither of which are supported by our research on typical developing children.