User Interaction with Word Prediction

Abstract

Word prediction systems can reduce the number of keystrokes required to form a message in a letter-based AAC system. It has been questioned, however, whether such savings translate into an enhanced communication rate due to the additional overhead (e.g., shifting of focus and repeated scanning of a prediction list) required in using such a system. Our hypothesis is that word prediction has high potential for enhancing AAC communication rate, but the amount is dependent in a complex way on the accuracy of the predictions. Due to significant user interface variations in AAC systems and the potential bias of prior word prediction experience on existing devices, this hypothesis is difficult to verify. We present a study of two different word prediction methods compared against letter-by-letter entry at simulated AAC communication rates. We find that word prediction systems can in fact speed communication rate (an advanced system gave a 58.6% improvement), and that a more accurate word prediction system can raise the communication rate higher than is explained by the additional accuracy of the system alone due to better utilization (93.6% utilization for advanced versus 78.2% for basic).

How do People Type on Mobile Devices?

“…Usefulness is therefore determined in a complex interplay of many factors, including the efficiency of the used text input method, the experience of the user, the accuracy of the prediction and other factors. Accordingly, results reported in the literature have been mixed [1,18,34]. In particular, for typing on mobile keyboards, a recent study showed that the use of word prediction methods can be detrimental to performance [29].…”

A Cost-Benefit Study of Text Entry Suggestion Interaction

“…Prediction and error correction systems aim to reduce the number of actions required from a user to enter their desired text [4,7,8,24,31,32]. Augmentative and Alternative Communication (AAC) research has investigated prediction systems to improve input speed for users with physical impairments, but has found that the benefits of suggestions are not always clear [8,31]. Koester and Levine [13][14][15] suggested that the cognitive and motor costs of using suggestion interfaces may outweigh their benefits, and their evaluations of several AAC How Fast Can You Type on Your Phone?…”

The Effects of Predictive Features of Mobile Keyboards on Text Entry Speed and Errors

“…Our results show that typing a word is faster than picking that word from prediction panel, with a difference of an average of 2.09 seconds per word. Based on our data, participants took on average of .97 seconds before choosing a word from the prediction panel, which includes the (non-negligible) overhead of switching attention, as explored in previous work, e.g., [13,51]. A potential reason behind this relatively long time is that users need to read the three entries within the prediction panel to make a choice which one to select, and then carry out that selection.…”

“…The effectiveness of word prediction as a feature is unclear for mainstream mobile text entry [37], considering also that switching the attention from the keyboard and the typed text to the prediction panel can cause delays [51]. Many factors play a role in the effectiveness of the use of predictive features [26], including the efficiency of text entry method and the experience of the user [37].…”