Human-Centered Systems Design and Unconscious Bias

5 minute read


As part of the TAPAS Marie Curie ETN, I had the opportunity along with my research peers in the program to listen to a series of talks on unconscious bias and human-centred systems designs. While the talks were all from a different perspective, there were certainly common lessons that can be learned from these talks. In this article, I’m going to summarise these lessons and reflect a bit on the user-centred design part.

The challenge of all public events nowadays is the COVID-19 crisis, where speakers have to present through an online service. The organisers decided to have pre-recorded talks instead, which was nice for the listeners, but it was probably very difficult for the presenters.

But why should we care about human-centred system design? Making software and hardware accessible to everyone is not only an ethical agenda that one should follow out of purely altruistic reasons. It is also the financial interest of companies who want to serve the widest user base possible with their products.

In the TAPAS program, this is especially relevant, as the users of our applications will be people with cancer, dementia or children. These are all vulnerable groups who might not be able to interact with the products in the same way as a typical user.

Getting to know the user

If there is a way to summarise all the talks, then it would be this sentence: include the user as early as possible. According to Dr. Hazel Roddam there is no wrong way to do this: make a simple focus group study, have an interview with your user and get to know the user needs.

In the very early stage of product development, you might not even know who is your user. Rebecca Bright’s talk emphasised that you a preliminary concept of the user is needed before the first contact. She calls this a lo-fi persona and describes how user interaction enables Therapy Box to make a hi-fi persona later in the process. This hi-fi persona can be then used for product development decisions.

Engaging with the users

It seemed clear from these talks that depending on the size of the company or users that you have, you might have to resort to different solutions engaging with your users.

Apart from the focus group approach mentioned above, Lorenzo Desideri from AT Bologna showed that often a more hand-tailored approach is needed. According to his talk, 30% of assistive technologies are abandoned and he believes this is mainly because we are not conscious of contextual problems. He brought the example of wheelchairs: these are not great assistive products in countries where step-free access is not provided on public transport or to most buildings. In his presentation, he showed how they work together with users even in the early research stage to adapt their solutions.

Larger projects often have to resort to metrics-based solutions and user activity tracking in applications: Rebecca Bright explains how their team looks for anomalies in clickthrough, sign up, user engagement data to find potential problems in their software. Similar to AT Bologna, she also advocated personal interaction with the users adding that the greatest help is coming from the “complainers” – people who understand the product enough to provide detailed and constructive feedback.

Proxy persons are another way to engage with the user if we cannot get information directly from the users according to Hazel. These proxy persons are often family members and caretakers who can provide useful information about the user that the users are often unable to do. Lorenzo saw the family members and caretakers role not only as information providers but also as people who can provide help in using the assistive technologies.

Thinking about the user

I really liked most of the talks: most of them were very thought-provoking. As researchers, we are very focused on our particular research question, which is often a metric to optimise and we often forget the user in the loop.

Getting in contact with our users is important. A personal contact as in Lorenzo’s situation is an ideal situation, as you can make the solution completely custom. However, general solutions often have the advantage is affordability in my experience.

Also, it is increasingly more regulated how we can get information about our users through the GDPR and various other data protection laws. Thus, I feel often times a researcher can feel discouraged to get to know its user base, simply due to the lack of understanding of the legal process.

If you wish to know more about the speakers, you can find out more about them in the links below (these are only the speakers who talked about human-centred design):