The Pen and the Keyboard

AFor about fifteen years after portable computers reached the lecture hall, nobody thought the choice of writing instrument worth an argument. Typing was quicker, the resulting file was legible and searchable, and students who could not spell were rescued by the machine. The mood changed after 2014, when a widely reported experiment found that undergraduates who wrote by hand answered conceptual questions about a lecture more accurately a week later than those who had typed, despite having recorded far fewer words. Within four years the finding had been cited in more than a thousand papers and had reached places that academic work seldom reaches: teacher-training manuals, newspaper columns and, eventually, the regulations of individual departments. By 2019 roughly a fifth of the faculties surveyed at one large European university restricted laptops in seminars, and a handful forbade them outright. What travelled, however, was the headline rather than the design. Very few of those who repeated the claim had noticed that the original participants received no instruction of any kind in how to take notes, so that the comparison was between two untrained habits rather than between two technologies.

BThe strongest evidence that the hand contributes something of its own comes not from lecture halls but from children who are still learning their letters. Naledi Ravelli, a developmental psychologist at the Fennimore Institute of Reading Research, has spent a decade comparing groups of five-year-olds taught the alphabet in different ways. Children who trace each character with a finger or a pen produce, on average, 22 per cent more correct identifications a fortnight later than children who select the same characters from a keyboard, and the gap survives when the two groups spend identical time on the task. Ravelli's explanation is that drawing a letter builds a record of the movements required to make it, so that recognising the shape afterwards can draw on two sources instead of one. In support she points to brain-imaging work published in 2017, in which regions associated with planning movement were active while children merely looked at characters they had previously written. The effect, she is careful to add, has been demonstrated only in the years when the alphabet is still being acquired.

CCamille Duflot, an experimental psychologist at the Brant Laboratory for Learning Technology, accepts the findings from early literacy and disputes their extension to adults. A student at a keyboard, she maintains, can transcribe a lecture almost word for word, and transcription requires no decisions; a student with a pen falls behind within a minute and must therefore choose what matters. On this account the pen is not the cause of anything, but a device that imposes a useful limit, and any other constraint should work as well. Dictation, a word limit, or an instruction to write nothing that already appears on the slide should, by her reasoning, produce much the same result. Her own trial, conducted in 2022, gave sixty undergraduates four weeks of instruction in summarising while they typed, and the advantage previously held by the handwriting group disappeared entirely on the conceptual questions. A second group, allowed to type without such instruction, recorded 40 per cent more words than the trained group and scored below it on every measure except word-for-word recall. Duflot's conclusion is blunt: what schools are banning is a symptom, and what they ought to be teaching is selection.

DNeither position emerges from the exchange intact. Ravelli's account rests on participants under seven, and she concedes that nothing in her data explains why an eighteen-year-old should benefit from a pen, since by that age the movements are automatic. Duflot's training study, for its part, ran for four weeks at a single institution, tested factual and conceptual recall but nothing that had to be applied elsewhere, and lost eleven of its sixty participants before the final session. Ravelli is sceptical that four weeks of coaching would survive an examination term, and observes that the trained students were volunteers, who may already have been the more deliberate note-takers. Both accounts, moreover, have been tested almost entirely on material that participants were told in advance would be examined, which is not how most learning outside a classroom proceeds. What the two researchers agree on is easier to state than either of their theories. Writing instrument is a poor variable, because it stands in for several others at once: the speed of the medium, the training of the user and the purpose for which the notes are made.

EThe practical advice that follows is narrower than either side of the public debate would like. Where the aim is to understand an argument, a slower medium helps, and it helps mainly because it forces a decision that a faster one allows the student to postpone. Where the aim is to capture a definition, a formula or a list of dates exactly as given, the faster medium is plainly better, and insisting on a pen serves nobody. Bans, which take no account of that distinction, are therefore a blunt instrument, though they are easy to enforce and easy to defend to parents. Universities that have relaxed their rules since 2021 report no measurable change in examination performance, although nobody claims that the comparison is a controlled one. In the early years the case for the pen rests on quite different ground and looks considerably firmer, which is why both researchers would keep handwriting instruction in primary schools while treating the seminar ban as unfinished business. The question that remains is not which instrument is superior, but what a student has been taught to do with either.