NavTouch: Making Touch Screens Accessible to Blind Users

Touch screens have showed to be a successful and enthusiastic way of human computer interaction. Due to their fast learning curve, novice users benefit most from the directness of touch screen displays. The ability to directly touch and manipulate data on the screen without using any intermediary devices is a very strong appeal. However, these devices face several interaction issues, once again augmented in text input scenarios. While they also restrict the interaction performed by full capable users, blind individuals are disabled to interact as no feedback is offered. The problem in this scenario is even more drastic than when a physical keypad is available as the keys give the user the required cues to select targets (although obligating to memorize the associations).




Although pointing or selecting may be impossible, performing a gesture is not. We present an approach similar to NavTap (NavTouch) that uses the user’s capacity to perform a directional gesture and through it navigate in the alphabet (similarly to the keypad based approach). Once again, the user is not forced to memorize or guess any location in the screen as the interaction is limited to directional strokes.

Special actions are linked to the screen corners as those are easily identified. After performing a gesture, if the user keeps pressing the screen, the navigation will continue automatically in last direction. The bottom right corner of the screen erases the last character entered and the bottom left corner of the screen enters a space or other special characters. In contrast to keypad, where the user has to find the right key to press, with these gestures that extra cognitive load does not exist.



NavTouch outperforms NavTap because of the additional effort in finding the appropriate directional key with NavTap. Indeed, we found that users are able to quickly navigate in all four directions with NavTouch as gestures can start at almost any point on the screen with no extra associated load. Furthermore, users are able to write sentences faster with navigational approaches, improving their performance across sessions. Overall, experimental results show that navigational approaches are far easier to learn and users are able to improve performance without further training. Moreover, NavTouch was more effective than NavTap because of the more fluid mapping of gestures to actions.

See the video of a blind user testing the system...

Credits:
Hugo Nicolau
Paulo Lagoá
Tiago Guerreiro
Joaquim Jorge