This website is intended for healthcare professionals

Scrabble-ing with dementia

Posted in Special Feature on 2nd Aug 2019

AJ Larner, Cognitive Function Clinic, Walton Centre for Neurology and Neurosurgery, Liverpool, L9 7LJ, UK.
Correspondence to: Email:
To cite: Larner AJ. ACNR 2019;18(4):25
Published online: 2/8/19


In a previous article, it was suggested that board games, puzzles, and quizzes might be characterised as neuropsychological tests, probing different domains of cognitive function.1 One of the board games mentioned, Scrabble, was deemed to tap essentially linguistic skills.  In recent years I have had the opportunity to witness first-hand the changes in Scrabble skills in a previously competent player who developed dementia, namely my mother.

The game

For those unfamiliar with the game, the aim of Scrabble (originally marketed as “Criss-Crosswords”) is to make words using one hundred letter tiles on a 15×15 board.  Individual letters are assigned different values according to the frequency of their use (e.g. in the English version a, e, i, o and u all score 1, whereas q and z score 10).  Each player has a hand of seven tiles, picked at random, with which to construct words, which may run either horizontally or vertically on the board, building upon letters already in place on the board.  Scores for each turn are the sum of the individual letter scores, and these may be augmented by placing tiles on squares labelled as double or triple letter or double or triple word scores.  The player with the highest aggregate score once all the letters are played wins the game.  For those unfamiliar and/or curious, the full rules may be consulted.2

Although purchased initially for the entertainment of their children, my mother and father continued to play Scrabble frequently, often nightly, after we had passed on to other interests, and indeed after we had left home.

The patient

My mother developed cognitive decline in her mid-eighties, having previously been in good health aside from occasional migraine (with perimenopausal onset3) and late life hypertension.  An elective hip replacement in her eighties was certainly associated with change in cognitive and functional abilities.  She was diagnosed with Alzheimer’s disease in her early nineties, based on clinical assessment and cognitive testing (ACE-III 49/100) by an old age psychiatrist. She is treated with donepezil. My sister is her named carer and provides resident assistance for all activities of daily living.

Now 94, my mother continues to play Scrabble on a daily basis.  Although she never initiates the plan to play, she generally expresses willingness if it is suggested, often commenting that it keeps the mind active and provides an opportunity to learn new words.  She perseveres at each game, never expressing any wish to stop playing once a game is underway, although on occasion a game has been terminated because she is in pain (particularly from the previous hip replacement).

Observed changes in Scrabble skills

My mother is sometimes slow in making words, and certainly her playing vocabulary is impoverished in comparison with her past abilities, now with a predilection for short (usually three- or four letter) words.  Nevertheless, she can on occasion “see” words and play immediately (pre-processing?), such that usual turn taking may be overridden in her enthusiasm to play a word, requiring a reminder that “It’s not your go yet!”.  When it is her go, the previously formulated word may have been forgotten.

Difficult letters, such as q, z, and j, may be marginalised, literally and physically, and apparently ignored, sometimes for the whole duration of a game.  The two blank tiles, which can be used to represent any letter, are often a source of confusion, and may be turned upside down.  Nevertheless, she can sometimes surprise us by working out unfamiliar words (e.g. from an opening hand of letters, d,d,e,j,r,s,u, she eventually played all seven as “judders”).

On occasion she may play letters in the wrong direction (i.e. backwards), suggesting visuospatial problems, or may attempt to play a word for which she does not have all the letters.  She may make a word with the tiles in her hand which cannot be played onto the board for lack of a suitable letter/space to which it can be joined.

Playing strategy has also changed in other ways.  Whereas in earlier years the opportunity to play on, for example, a triple word score would not be overlooked, this is no longer the case.  She does not calculate or keep her own score.  Generally her completed game scores are between 100-150, occasionally 200, whereas in her heyday she regularly scored around 300.

The option to replace some or all of a hand of letters in exchange for foregoing a turn of play, which she frequently used in the past if stuck with a handful of vowels, is never now used.  The use of a dictionary to check word spelling is eschewed, indeed those availing themselves of this facility, entirely within the rules of the game, may be accused of cheating!

In summary, I suggest that my mother’s current standard of Scrabble play reflects a general decline in her linguistic skills, perhaps some loss of memory for words, occasional visuospatial errors, and change in executive function as reflected in her less competitive strategy.

Another phenomenon, which seems very curious, occurs on occasion.  Despite her experience of playing Scrabble over many decades, my mother will sometimes state “I don’t think I’ve played this game before”.  With encouragement and example she may pick it up again, but on other occasions this is not the case.  She may report that she just cannot see how she could play the letters in her hand, sometimes laying them on top of letters already on the board.  I wonder if this is a form of “closing in”, a term used to describe copying drawings close to or superimposed upon the original, seen in some patients with Alzheimer’s disease, and variously interpreted as “constructional apraxia” or a visuospatial deficit.  This fluctuation in the ability to play occurs particularly (but not invariably) in the afternoons, and I think might reflect waning of attentional resources, manifest as an executive deficit of not understanding how the game is played.


My previous judgment that Scrabble is essentially a test of linguistic skills1 has been shown, in part from observation in the change in my mother’s play, as far too simplistic.  Evidently, Scrabble requires the allocation of attentional resources, intact verbal and visual memory, visuospatial skills, and executive function, as well as linguistic abilities, for its successful performance.


  1. Larner AJ. The neuropsychology of board games, puzzles and quizzes. Adv Clin Neurosci & Rehabil 2009;9(5):42.
  2. (accessed 22/04/19).
  3. Larner AJ. Familial migraine without aura with perimenopausal onset. Int J Clin Pract 2010;64:128-9.
Download this Article