Exploring alternative interactions through touchscreen based IVIS interfaces

A Research Through Design approach to automotive HCI design
Project description
This case study was conducted as part of my master’s thesis project at the IT University of Copenhagen with Andreas Dehn & Thor O’Hagan Petersen. For this project we sought to explore alternative forms of touch screen interaction with touch-based IVIS interfaces (in-vehicle infotainment system). The motivation for our research stemmed from frustrations with the user experience of digital interfaces in modern electric vehicles. During the project's six-month run-time, we conducted 17 user tests, 3 expert interviews and an auto-ethnographic field study across 4 design phases and 3 design iterations. For the purpose of the project we constructed a car simulator to test and evaluate our prototypes in a simulated "real" context.
Methods
Autoethnography, Expert Interviews, Brainstorming, User Testing, Sketching, Prototyping
Role
UX Designer, UX Researcher, UI Designer, Interaction Designer
Year
2023
Duration
6 months
Read full thesis report

Problem statement

One of the leading causes of car accidents globally is driver-distraction. In recent years, the car manufacturing industry has pivoted towards digitization of infotainment systems for completing secondary and tertiary tasks opposed to the primary task of driving. These systems rely heavily on visually appealing touchscreen interfaces, which often prove problematic as they are more visually demanding, which in turn can cause cognitive fatigue ultimately adding to the cause of driver-distraction.

Context

Background

Hyphothesis

By introducing alternative forms of interaction to the current status quo of the automotive manufacturing of touchscreens with emphasis on reducing visual and cognitive load on the driver, we might be able to reduce driver-distractions while simultaneously providing a better User Experience, when interacting with touchscreen based IVIS interfaces to perform secondary or tertiary tasks.

The why

With the growing industry tendency for touchscreen based IVIS interfaces, drivers are more prone to distractions than ever before. While these interfaces are visually appealing, the vast majority requires more visual attention and cognitive capacity to operate than traditional tangible interfaces in older car models. Meanwhile, our initial research found that these interfaces more often than not also provides a poor user experience.

Research / Exploration

Preliminary phase

Autoethnographic field study
To better grasp the current industry practice beyond the academic literature as well as to experience and document the different forms of interactions related to IVIS design, we conducted an Autoethnographic Field Study. Here, one team member would test drive a vehicle with IVIS, while performing a set of predefined tasks with the IVIS and then evaluate the subjective user experience after the drive had concluded following a predefined framework for comparison.
Figure 1: Autoethnographic Field Study results
Figure 2: Driving Simulator Construction
Building a Driving simulator
To properly evaluate designs in a form of realistic context, we decided to construct a simulated driving experience. This allowed us reliably and ethically conduct User Testing of our later prototypes without having to jeopardize the safety of our test participants. Research also states that the basic cognitive psychology is similar, as the emotional reactions of drivers in simulators is comparable to real driving and can therefore also be used for emotional research.
Expert interviews
To further explore our field to inform our understanding as well as get professional feedback on our driving simulator and discuss our testing setup, we conducted three semi-structured expert interviews with one design director and two academic scholars all whom have practical experience working with or conducting research in the field of automotive HMI.
We asked questions about
  • Their background, experience and concrete area of expertise.
  • Their work in relation to the industry and perspective on the current situation of automotive HMI and HCI.
  • Our testing framework, drive simulator and testing approach.
Figure 3: IVIS interface behind wheel

INSights

INSIGHTS

The dominant form of touchscreen IVIS interaction is point-touch based, which is experienced as demanding and distracting.
Multimodal interactions are present in most tested vehicles, but are largely underutilized often resulting in a bad user experience and distraction.
Industry design practice is driven by many factors; most of which are not user-driven or entirely removes the user from the equation of driving.
Automotive designers and HMI designers are distinctly separate units. As a result, HMI designers often work in a black box in regards to hardware capabilities.
The validity of our driving simulator, testing framework, and testing approach was deemed sufficient for our situation prior to testing by the experts.
Your interaction as a driver should be multimodal, because interactions should always be contextually bound.

Ideation

Concept

brainstorming
For our first iteration, we synthesized our collective findings from the preliminary phase through a Reverse Brainstorming excercise. By doing this, we were able to focus on the negative aspects unveiled in our initial research and counter these findings with positive solutions afterwards. We then did an adapted KJ Brainstorming excercise to identify the top priorities and ideas from the Reverse Brainstorming excercise. This resulted in the establishment of the following design guidelines for the first and following iterations:

1) Decrease accuracy needed from user input.
2) Reduce visual distraction.
3) Increase perceived User Experience in a driving context.
Figure 4: Brainstorming Excercise
Sketching
In the later iterations, our ideation process consisted primarily of different Sketching excercises. with the purpose of determining the overall concept, modality of interaction and interface layout based on our findings from the first iteration. To centralize sketching around the act of working with multi-finger sub-global touch gestures, a car interior was drawn along with paper cutouts of arms using different number of fingers for the interaction.
Figure 5: Sketching & Hand-drawn mock-ups
Material Exploration
For the last iteration, we conducted the same sketching excercises as in the previous iteration for adjustments, while also going through multiple rounds of Material Exploration with the introduction of Sound and Haptics as forms of multimodal feedback. To enhance vibration strength of the haptic feedback, we explored multiple shapes of Silicone.
Figure 6: Silicone shape exploration to enhance haptics

Design

Prototyping touchscreen interactions

Through multiple iterations of prototypes, we explored alternative ways of interacting with prototypes based on our findings from our initial research and ideation phases.

Our goals was to not only explore different forms of interaction comparatively, but also gradually improve on our design over the course of the three iterations.

First iteration

For our first iteration, we choose to build two prototypes inspired by industry-standard touchscreen interfaces; the first emphasizing point-touch interactions, and the latter focused on touch gestures. We chose to design two visually similar prototype variants, as we wanted to isolate the forms of interaction for testing subjective User Experience and usefulness with different interaction modalities comparatively.

Figure 8: Prototypes iteration 1
Second iteration

Our second iteration focused on the concept of multi-finger sub-global touch gestures. Herein by using touch gestures we made interactions more direct, tangible and hedonically pleasing. Additionally, multi-finger interactions are performed using a larger surface area as a touch target to allow for decreased precision and using combinations of multiple fingers to perform different functions within the same area.

Figure 8: Prototypes iteration 2
Final design

For the final iteration, we focused on small prototype adjustments as well as the new addition of multimodal feedback improvement based on our findings from the previous iterations User Testing. These aspects were amplified through a combination of haptic response, audible feedback and visually distinguished sections.

Figure 9: Prototype final iteration

Testing

User testing

Over three iterations we tested 5 different prototypes comparatively with the purpose of testing the subjective user experience of completing different tasks with different interaction and feedback modalities while driving. Over the three iterations we tested on 17 test participants.

We analysed our findings for each iterations testing through Affinity Diagramming to synthesize our insights and derive action pionts and design guidelines for the following iteration.

Figure 10: Affinity Diagramming example
Who we tested on

Over the course of the three iterations, we tested our prototypes with 17 test participants. Important critereia for our recruitment of test participants were, ownership of drivers license, the age segment for people whom are statistically most likely to buy electric or modern vehicles and lastly, their experience with digital IVIS interfaces.

Figure 11: Test participant demografic example
How we tested

To ensure consistency and comparability throughout our user testing we developed a testing framework that we consistently refined throughout the project. For this an important aspect was the counterbalancing of the test, as it allowed us to neutrally compare the subjective User Experience of our prototypes.

Figure 12: Testing framework procedure
Moderated testing

As our testing procedure was quite extensive, we decided to conduct it in a 1:1 moderated format, where a group member would lead the test following a predetermined manuscript along with each individual test participant, while recording each test for later analysis.

Figure 13: Moderated test setup

Results / Outcome

Findings from design research

Our research found that the status quo of touch interaction in modern electric vehicles relies too heavily on outdated or borrowed interaction paradigms from mobile or web-based user experiences.  However, while multimodal interactions do exist in cars today, the potential of these modalities, such as through touch gestures or multimodal feedback remains largely under-utilized.

The results from our user testing indicates that gestural and multi-finger interaction in combination with multimodal feedback can vastly decrease the degree of visual distraction and improve the hedonic quality of the user experience.

Additionally, when implemented correctly, it can help the user more easily visually distinguish functions and improve error correction while driving.In this way, we explored different interaction modalities through adaptation of HCI practices in a traditionally HMI-driven domain. We argue that this was an effective measure of exploring, validating, and generating knowledge about interactional qualities in the context of IVIS interfaces. However, this approach should be seen as an addition to the existing HMI methods and measures of testing rather than a replacement.

Figure 14: AttrakDiff Survey Results & Follow-up Interview Quotes

Thanks for scrolling!

Expect a reply within 24 hours.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Get in touch

Let's work together

The main reason I love UX and design is because I get to meet so many amazing individuals. I love to work with people who are passionate about their ideas. Let's talk more about your next project.