etherhall School are exploring how eye tracking systems can be used to support and provide accurate assessments of learners with disabilities and creating support materials for staff wanting to use eye tracker technology with learners. You can read the initial project post here.
The project is being led by Erica Dennies and Helen Robinson. The school has provided the following update:
In December 2014 we trailed a range of different programmes to use with the eye tracking technology. Having hands on experience of the range software available allowed us to make informed choices about which were most appropriate to use for the project. We’ve invested in a programme called Tobii Gaze Viewer. This is an assessment tool which records the eye tracking data from any application which can then be saved as a single image or a video. It uses heat maps and gaze plots which we can then use to assess the pupil’s ability to use the eye tracker and their cognitive understanding. For example we can assess understanding of language comprehension, prepositions, colours or size by analysing where the pupil is looking on the screen. It can also record the audio track of people talking in the room so we can record the level of verbal prompting staff gave to pupils. It will provide us with exactly the data we need to be able to use eye tracking technology as an assessment tool. It is however expensive at £617 for a single user licence.
The other software packages we trialled and purchased gave us a range of graded activities to use with eye tracking technology. This provides us with pre-made activities to use with pupils.
Originally we anticipated having 2 tablets of different sizes, the larger 18” and a smaller 10”. The smaller screen would allow us to work more closely with a small number of pupils who are not easily positioned in front of a screen; when positioned over a physiotherapy wedge for example. The smaller visual field would ensure ease of view with less screen to scan and less head movement required.
However, escalating costs of software was making it difficult to keep within our budget. To keep in budget, we reduced the size of the screen view on the 18” screen and put a black backdrop behind the images for those pupils who needed a smaller image. So we purchased just the 18” screen. For most pupils however, a full image on an18” screen provides an excellent sized image.
Our final order of items required for the project were as follows;
- SB18 18” classroom touchscreen PC running Windows 8 which comes with additional amplification, switch connections and mounting plate.
- Tobii PC Eye to Eye tracking camera
- Floor stand
- Gaze Viewer assessment software
- Look to learn software
- Scenes and Sounds software
- Sensory Eye FX
The Tobii camera is portable and can be used on any PC or laptop. It is highly accurate and quick to set up. We have found it very easy to use and reliable. The software we purchased is designed for the earliest level of computer access allowing all our pupils to engage with the technology at the simplest level. We will be able to comment further on their value when we have completed the project.
We also bought a day’s training from Smartbox who supplied all the hardware and software. This took place in January and provided us with detailed knowledge of how to set up and use the Tobii eye gaze and all the activities available on the software and how the assessment software worked.
We chose a group of seven pupils for the cohort, assuming that some pupils would be absent from school or non-co-operative. We could therefore guarantee a cohort of four in the final project.
To gather baseline information about our cohort of pupils, we devised and distributed a simple, open question questionnaire to send to parents, teachers and other professionals who work with the child. We asked questions about what the child likes to watch, did they recognise photos of themselves, favourite activities, favourite sounds etc. We also asked if the respondent thought that the child could understand drawings, symbols, photos, real objects, and questions about their ability to track objects and see them in different positions.
This information gave us a view as to what understanding different people held about the child’s visual skills and their cognitive level with regard to taking meaning from two dimensional images. It also gave us information about what would interest the child and be more likely to hold their attention. We received all the questionnaires back from professionals and a few from parents, collecting sufficient detail to inform our strategies and develop some of our own activities to use with individual children with the eye gaze.
We videoed each child in their normal classroom situation so we could record their usual levels of attention and comprehension. We could observe if they were comprehending the activities they were participating in and what form of representational images they were gathering information from. This formed our baseline.
We were now ready to start using the eye gaze with individual pupils and recording their levels of interaction with the screen using the Gaze Viewer software. Each pupil had an initial session with the eye gaze participating in activities which we thought they would like and which would engage them in using the technology. We also videoed the pupil during the session so we could link their physical movements with their heat maps. The maps show where the pupil had looked on the screen and for how long. The areas where pupils had looked the longest show a red blob, less frequent gazes are shown in green So, if the heat map was not showing any evidence of the child looking at the screen, we could check the video to see in the child had looked away, or closed their eyes.
At the end of the session we had evidence of the heat maps showing where the pupil had looked on the screen and for how long, the gaze plot of the order in which they looked at images on the screen, the audio track of what was said during the session and video of the pupil’s face whilst they were participating in the session.
Our initial learning from this first round of sessions is:
- Gaze Viewer is an excellent tool for analysing if the child is looking at the screen and what they are looking at.
- Activities on the purchased software have a lot of detail and can be confusing or just not interesting for the child. So we had to make some simple activities using our own single images produced using Powerpoint.
- We found it was worth taking time to collect evidence from others who knew the young people well about their interests – we designed activities in response to this and these did hold our pupils attention.
- The eye gaze does not need to be calibrated for that specific child for it to pick up the child’s gaze and provide informative data.
Now the pupils are happy to engage with the eye gaze and we have seen them with the technology, we are now going to run a second set of sessions with specific activities to test the pupil’s cognition and check if our understanding of their understanding of specific concepts is accurate.