A method for the computer-assisted learning of orthography, the method includes executing the following steps by a data processing system: retrieving (11) a main set of words from a data storage; retrieving (12) an error data set associated with said main set of words from the data storage; repeatedly executing the following steps: selecting (13) a word to prompt the user with, by computing, for each word from the error data set, a statistic measure related to the probability of an error occurring in the word, and selecting the word which has the maximum value of the statistic measure; prompting (14) the user with the word; accepting (15) a user input specifying a sequence of symbols; comparing (16) the user input with the word; updating (17, 18) and storing the error data set.