The Oculus Go Wasn’t Designed for Black Hair
In January 2020, I ran an in-person experiment for my thesis at the MIT Media Lab. The project, called Allo-I, is a virtual reality human interaction study that centers the lives and experiences of Black women — specifically, Kenyan women living in informal settlements in Nairobi. As far as I know, it’s the first research of its kind, and I wanted to understand how the participants interact with novel technologies like virtual reality (VR) and the content within them and tested to see whether the technology can result in more imaginative thinking about the future.
The study had 220 participants, half of whom participated in a VR experience. For the study, I used the Oculus Go device by Facebook. I chose that device because it was easy to use and lightweight. I can use the device anywhere and pre-downloaded all my content so that Wi-Fi wouldn’t be an issue. From an accessibility perspective, the Go felt like the right decision. But that was before I realized that Black hair was going to be a major issue for the completion of my work.
Over the course of running the study, it became overwhelmingly apparent that the Oculus Go (and, I would imagine, any other high-definition VR headset for that matter) was not designed with Black women in mind. The texture, size, and styling of Black hair was not suitable for the device, and the strap attachment piece kept pulling out at the joints from extensive stretching.
I had four headsets for the study that participants used 25 times on average, and each headset “broke” approximately 10 times during the study. This means that the headset strap snapped nearly half the times that a participant tried to use it.
It’s not that African women have uniquely big heads. It’s that they have uniquely big hair that includes but is not limited to braids, twists, locs, head wraps, and hijabs. Again, not everyone wearing braids or a hijab couldn’t put on the headset. But depending on the texture of the hair or how it was tied (or how it was hidden), a good number of participants either couldn’t wear the headset or struggled significantly to put it on. While the headset never permanently broke since the joints could be reattached, the reattaching takes significant effort and shows a hole in the design system.
Zoom’s Virtual Background Feature Isn’t Built for Black Faces
A scientist warns that bias in facial recognition software could lead to false arrests, lost job opportunities
The Oculus Go system is designed to be a light headset that can be used by anyone in the comfort of their home. Its design has been structurally sound for most people, except for one specific demographic. The straps go on the side of the headset and around the head. A little space has been created that serves a double purpose: extra support and a space for a little ponytail. Because headsets are heavy, a third strap is added from the middle of the headset to the back for balance.
In my study, the straps were often extended beyond their stretching limits, pulling off the strap connector from the socket on one or both sides of the headset. By the time I was halfway through the study, I’d learned my lesson and removed the middle strap that goes over the center of the head and meets the second strap at the back. The strap connector could be reattached, and in cases where the whole strap system was not functional, I asked the participants to hold the device to their face and ignore the straps, which, for obvious reasons, may have affected the feeling of immersion and embodiment, therefore affecting the study. While I couldn’t take photographs of the participants, I’ve included some images of example hairstyles to demonstrate how and why the headsets kept failing.
The stretchy and versatile strap system was not stretchy and versatile enough for Black hair, a subject matter that is as personal as it is political for Black women. Even if I had experienced the hardware issue myself, I was not prepared for how significant it would be when running my study. When I’ve struggled to put on a headset over a high bun of braids, I temporarily take down the braids. I’m in no position to do or ask that of another woman even if she is a participant in my study. I personally know how time-consuming it can be to get braids or head wraps tied (and what lies beneath a headscarf may be extra personal). Further, a good number of participants didn’t just have big hair, they had hijabs as well, and these, more than any other updo, cannot be removed.
Suddenly, I’d stumbled upon a design challenge I couldn’t resist. It made me ask the question how can I design a VR strap system that is versatile enough for all manners of Black hair?
Before diving into the design challenge, I want to borrow from feminist philosopher Sandra Harding’s notion of “strong objectivity.” This is the idea that even “neutral” research is biased and that there is value in beginning research from the strong viewpoint of the lives of women. Feminist objectivity is important when iterating on a design as it directly confronts the design of “neutrality.” In this case, not only is this work coming from feminist objectivity but particularly Black feminist objectivity.
The design challenge
Design a VR strap system that is versatile enough for all kinds of Black hair and can be easily used on the current Oculus Go device.
The new strap system involved three additional parts that joined together to form one dynamic system: (1) the main attachment piece, (2) a rotational attachment piece, and (3) a screw cap. The parts were designed using Rhino and Unity, which are 3D modeling software, and 3D-printed with thermoplastic polyurethane (TPU).
The attachment piece
The main attachment piece is a short part that bisects, allowing for two separate strap systems. The rectangular cutout allows for one strap, and the circular cutout allows for an additional piece, the rotational part. Two incomplete cylindrical formats were excavated from the short edge of the piece. These are the teeth of the piece, allowing it to be clipped on to the Oculus Go’s existing latch for the straps.
The rotational piece and screw cap
Because the VR headset is heavy, most strap systems are secured by a strap that is attached from the center of the headset and goes down the middle of the head until it reaches the secure strap at the back of the head. However, as seen earlier, the central strap was not useful. The rotational piece gives way for a second strap that can be secured over any part of the head. This piece enters the circular cutout from the main attachment piece and is secured by a screw cap.
The pieces attach to form a device that you can easily attach to the Oculus Go.
The three images side by side show the movement of the rotational piece and its flexibility to sit at any angle.
For the straps themselves, I chose to go with kanga, a cotton material that is popularly worn and used in East, West, and Central Africa. Kangas are usually colorful and patterned and can be used for clothing (like a sarong) or as a scarf or head wrap. I chose this material because of its cultural significance in East Africa and with the hope that its inclusion makes the technology more relatable and approachable.
Due to the coronavirus pandemic, I have been unable to carry out the user-testing sessions, although testing on myself was successful (see below). That said, this is only a low-fidelity prototype, and you can expect to see a high-fidelity prototype developed in the coming months.
This design is not meant to be an authoritative design but rather a provocation and a start for new ways of thinking about accessible design. While this design is specifically seen as an add-on to the Oculus Go system, I hope it inspires other product designers to consider how systems might be unwittingly excluding Black women from using their product comfortably.