App based word retraining for people with Semantic Dementia

South West PeninsulaNeurological
Start Date: 1 Dec 2015 End Date: 3 Jun 2017

Reclaiming words in everyday life: App based word retraining for people with Semantic Dementia


Language skills are central to our everyday living, but can be significantly impacted by dementia. In particular, patients with Semantic Dementia (SD) experience marked difficulties with naming and word comprehension. For these people, impairments in language skills are contrasted by relative strengths in episodic memory, visuospatial skills, basic attention and mental tracking. While this suggests that these people would make good candidates for cognitive intervention, unfortunately, to date, very little has been offered to assist these people.

Encouragingly, recent evidence has emerged from series of single case experimental design studies to show that word retraining programs run on patients’ home computers can deliver meaningful and lasting benefits. By engaging in a simple word practice, SD patients with various levels of semantic impairment demonstrated significant improvements in their word retrieval. Importantly, these improvements occurred within a few weeks, and additional benefits to word comprehension were also observed.  Moreover, the improvements were retained, even by those people with severe semantic impairments if regular but less intense practice was continued.

Despite the success of these studies, no user-friendly, freely available version of this program is currently accessible to patients with SD. Optimally, this program should be simple, portable and flexible in design, to allow families to modify training as needed (reducing the time required by clinicians to construct and monitor such therapies). App-based programs have become increasingly popular within the general public and certain patient populations. Such programs provide an opportunity for a flexible, simple, accessible service delivery option, but require research to confirm their efficacy and feasibility within the SD population.

Project aims

This study aims to:

  1. Develop an app-based version of the word retraining program used in previous research.
  2. Acquire preliminary data regarding usability of the program within the Semantic Dementia population.

Project activity

Part 1 – Initial program development

The original program developed for the internet, as described in Savage et al (see Related publications), will be re-coded into an app based program, initially for android capability. The training program will be based on the simple “Look, Listen, Repeat” procedure (i.e. look at the picture, listen to the word, repeat the word aloud). At the conclusion of each training session, participants will complete a quiz of their learning. The program will also include a naming assessment component, where participants are tested on their ability to name each item, one at a time. App developers will construct the code, creating an interface which mirrors what has previously been used in the research studies. The program will draw upon the device’s capabilities for camera and voice recorder to allow custom built lists of words to practice. 

Part 2 – Preliminary testing and program revision

Once the initial coding is complete, this early version of the program will be made available to a small number of participants to test the features and evaluate the usability of the program. Up to four participants with a clinical diagnosis of SD will be invited to use the program. Participants will be recruited through the neurology clinic run by Prof Zeman, but the project will be run within participants’ homes. 

Participants will be loaned an android device for the duration of the testing, and participants and family members will be trained on how to use the program. Three individually tailored lists of target words will be set up for each participant. Firstly, the app will be used to assess naming ability for each target word prior to training to establish baseline performance. Following the baseline period, participants will use the app to train their first word list for three weeks. Training will then be applied to a second list for three weeks. During this 6-week period the third list will remain untreated as a control. During the training period, participants will complete regular naming assessments using the app. Follow-up assessment will also be undertaken at 4 weeks after initial training is complete to assess naming performance and to interview participants regarding their experience using the app.  

At the conclusion of the study, participants and their families will be invited to complete a short questionnaire to measure overall satisfaction with the app regarding the usability and functionality of the program. Participants will be given the opportunity to provide suggestions for improvements and rate how useful they found the program. Following this feedback, any necessary modifications and adjustments will be made prior to finalising the app.

Anticipated outputs

The main output of the study is the word retraining app itself. Preliminary usability data will also be gained to inform any future modifications required to the app prior to running a main trial.

Next steps

If patient feedback and usability results are positive, the next step will be to submit a grant application to the Research for Patient Benefit scheme to fund a full trial.

Dr Sharon Savage