The effect of word prediction settings (frequency of use) on text input speed in persons with cervical spinal cord injury: a prospective study.

Purpose: To determine whether activation of the frequency of use and automatic learning parameters of word prediction software has an impact on text input speed. Methods: Forty-five participants with cervical spinal cord injury between C4 and C8 Asia A or B accepted to participate to this study. Participants were separated in two groups: a high lesion group for participants with lesion level is at or above C5 Asia AIS A or B and a low lesion group for participants with lesion is between C6 and C8 Asia AIS A or B. A single evaluation session was carried out for each participant. Text input speed was evaluated during three copying tasks: without word prediction software (WITHOUT condition) with automatic learning of words and frequency of use deactivated (NOT_ACTIV condition) with automatic learning of words and frequency of use activated (ACTIV condition) Results: Text input speed was significantly higher in the WITHOUT than the NOT_ACTIV (p<0.001) or ACTIV conditions (p = 0.02) for participants with low lesions. Text input speed was significantly higher in the ACTIV than in the NOT_ACTIV (p = 0.002) or WITHOUT (p<0.001) conditions for participants with high lesions. Conclusions: Use of word prediction software with the activation of frequency of use and automatic learning increased text input speed in participants with high-level tetraplegia. For participants with low-level tetraplegia, the use of word prediction software with frequency of use and automatic learning activated only decreased the number of errors.