Overview
This was the class project for the Cognitive Ergonomics course at University of Michigan. I was the only User Experience Designer and Researcher in a team of five students tasked with evaluating flaws in the Google Maps interface and using human factors to redesign revolutionary and evolutionary ideas which would support human cognition and ergonomics.
Process
Initiation
Research
Design
Test

Software: Google Documents, Adobe XD, Illustrator, Sketch
Key Role: UX Designer
Team: 2 Mechanical Engineers, 1 Product Manager
Timeline: Sep 2016 - Dec 2016 (4 months)
Initial Research
Studying the navigation market
The navigation app market is increasing rapidly due to two main reasons: the high demand coming from the users as well as the increased technological capabilities in the last decade. 
Required characteristics
Additional Features:
    - Sharing locations with friends and family
    - Saving locations on maps
    - Exploring geographical areas
Areas of improvement
We wanted to create evolutionary and revolutionary designs for Google Maps application to support human cognition.
User research efforts
Heuristic Evaluation
I evaluated Google Maps based upon the three sets of design principles that we learned in class. These sets of design principles are from Shneiderman’s, Nielsen’s, and Wickens’s. The design principles were used as my guide for this process. I conducted this heuristic evaluation, in order to provide a breadth and depth of analysis based on each of our differing perspectives.
Problems identified incude:
Cognitive Task Analysis
According to our heuristic evaluation, the team created and conducted a cognitive task analysis with a variety of users. This led us to getting a better understanding of the specific pain points users encountered based upon groupings of tasks performed.
Personas
I worked with three novice users and five experienced users for this analysis. In order to get a better understanding of our users, I conducted surveys to gather user information about their familiarity with the app, the frequency of use, and the primary tasks they perform while using the app. 
The novice users had little experience with Google Maps, however they were familiar with a smartphone and other apps. The experienced users, on the other hand, were tech savvy, and used Google Maps daily. 
Testing tasks with existing map application
The tasks that were performed by each of the users were grouped into three categories: navigation, searching, and button toggling. 
These tasks are attached below.
The team used think-aloud protocol throughout each of these tasks to help direct users on which tasks to perform and the goals of the tasks they were performing. Our main objective during the think-aloud protocol was to understand user’s thoughts while they were performing the tasks. We encouraged and reminded the users to communicate with us concurrently while performing the task.
After the tasks were completed with each user, we conducted a post-task interview to highlight significant pain points and garner a further understanding of the user’s thought processes. We identified the results of the task for each user and had a retrospective period of reflection following the task performance. Some of the questions I asked were, “Which tasks were most troubling for you? Why? What were you thinking when you were conducting this task?” These interviews were crucial in helping us understand the users’ thought processes and where they faced challenging problems. The results and the problems identified during our CTA are described in the next section, “Problems Identified.”
Problems Identified with current system
The problems identified through the heuristic evaluation and cognitive task analysis fall under two main categories: an inability to bridge the gulf of execution and attention-based issues.
Gulf of execution/attention based issues
Feature based issues
Designing the right map system
Based on the user research, the design goals include:
    - Reduce data availability
    - Eliminate interruptions due to selectivity
    - Reduce information access cost
    - Aid time-sharing with other tasks
I created the evolutionary and revolutionary designs for the Google Maps. I first used a notebook and sketched out ideas for the app. After a few iterations of designs, I created medium fidelity prototypes using Adobe Illustrator and Sketch.
Evolutionary design
1. Attention based issue
Alan is driving to his date and is running a few minutes late. He tries to navigate to an unknown part of town for dinner and is unfamiliar with the roads. At this time he receives a phone call from his mother and all the instructions from the maps have now disappeared right before Alan has to make a crucial turn.
Issues: The salience of incoming phone call is very high. The pre-attentive reference is violated and Alan cannot timeshare between the tasks of navigation and taking the call.
Please see the attached revised design that would allow Alan to timeshare successfully:
2. Mode of transport
Ron wants to navigate to the neighborhood grocery store, he knows that it is close by but wants to drive there. When he uses this destination on the google maps application it defaults to walking as the mode of transportation. Ron needs to cancel out of the current instructions to change his mode to driving to route properly.
Issues: High information access cost. There is violation of the principle of predictive aiding. This design leads to slips and eventually could lead to errors.
The new design recommends that the user have all the modes of transport available on the main home screen such that all information is available easily. There is high observability and low information access cost.
3. Adding Extra Stops on Routes
Ron wants to pick up coffee on the way back from the grocery store. He tries to add an extra stop but it gets added after his primary destination. He wants to reorder the list such that he goes from grocery store to coffee shop and then home. He tries to drag and drop what he perceives to be are icons representing the destination but these locations are not switching on the screen.
Issues: This violates the proximity compatibility principle and the principle of moving part.
The redesign allows the user to easily add multiple stops, move them and remove them altogether from the routing on google maps.
4. Sharing locations
Alan wants to share his location with his friend but he cannot find how this feature works. He drops a pin on the map and can then share the location with his friend but he can't find out how he can share the location of a shop or a restaurant directly.
Issues: There is poor conceptual model and high information access cost to use the sharing feature for google maps. This feature violates the naturalness principle and can cause errors due to bounded rationality.
The redesign allows the user to view the sharing icon on the navigation screen and shows the options of sharing and the route in the background to eliminate any walkarounds.
5. Consistency in search features
Ron travels for his job sometimes, this enables his to try new restaurants and experience different cuisines. Ron wants to explore all the nearby restaurants using google maps but always has a hard time figuring out how the feature works. It is available in two cases, if you are not routing anywhere and through the options on the top right of the screen if you are routing to a particular destination.
Issues: Top-down processing is not allowed and user expectations are not fulfilled.
This feature will be available to user in any screen, this will open up a set options while making sure that the routing feature is not disabled for the user.
Revolutionary designs
Our design encompasses a three-fold approach to address the design violations and the safety concerns.
We propose a redesign that uses the benefits of auditory and visual communication in combination with tactile modality:
    - Auditory (voice activated inputs in addition to the voice-assisted output)
    - Tactile “Haptic Steering Wheel” (use of touch and in particular vibrations)
    - Visual (using a ‘Heads-Up Display’)
Hands-free voice activated interface
Google Maps allows the driver to change or update destination through two options:
    - Physically modify and manipulate the application on their handheld device while driving
    - Physically select the voice activation icon on the Google Maps App to allow speech inputs
Our proposed design allows the user to call the Google Maps app like current Apple phone users call Siri (Apple’s voice assistant), this will be similar to the ‘Hey Siri’ feature available on Apple devices. This feature will allow the driver to continue focusing on the road and call out instructions to Google Maps, without using Google Map’s DMI, as displayed below.
The Google Maps voice assistant will continue to provide outputs like turn instructions and route navigation through the phone’s speaker acting as an auditory cue.
Haptic steering wheel
Our second design introduces a versatile feature which uses the sense of touch through vibrations and force emitted by the Haptic Steering Wheel. The Haptic Steering Wheel is created with twenty miniature motors embedded within the steering wheel that are tuned to emit low (100-150 Hz) and high (200-250 Hz) frequency vibrations to achieve several functionalities including:
    - Navigational Assistance through graded feedback
    - Error prevention using force 
    - Lane assistance
    - Monitoring feedback of external objects (inter-car spacing, proximity concerns) 
The focus for our design is primarily on navigation assistance and error prevention using force . The Google Maps app would use the navigation route as input to the steering wheel to assist the driver in order to navigate in the right directions. The wheel is designed to emit vibrations in clockwise-directions denoting right-turns and counterclockwise directions denoting left-turns. 
The driver still has the entire power and flexibility to override the warning and continue to turn left. This force is exerted only for the initial wrong turn the driver takes and waits for Google Maps to re-route.

Heads-Up Display (HUD)
The third component to our re-design ties together the auditory and visual modes with a delay, and provides visual cues to the driver on a heads-up display. The main objective is to design the interface of Google Maps to conform to the user’s dynamic driving environment. This redesign is to create knowledge in the world over knowledge in the head.
The heads-up display setup for Google Maps will ensure that the user hears the auditory cue, feels the tactile vibrations on the steering wheel, and gets a visual representation of the instructions to provide redundancy gain and facilitate better judgement.
There are several risks associated with bringing this technology and design to fruition including:
    - Distractions from the notifications on the heads-up display
    - Overtrust in the heads-up display and losing cognition over the external changing world
    - Colors being overlaid on objects in the external environment with the same color
    - Google Maps re-routes and updates, this would leave the driver in limbo
Evaluation plan
This final section outlines how we would evaluate our evolutionary and revolutionary design improvements.
Simulation study
We would initiate our design evaluations by conducting a simulation study to determine the effectiveness of our design and evaluate the potential risks resulting from the new cognitive demands of our design. 
We will study the users’ response time and the number of fixations and saccades in eye tracking as key metrics of evaluation. We plan to evaluate our risks effectively, which includes the use of heads-up display leading to user’s inability to see actual obstructions in the road while focusing on the heads-up display icons. This simulation study will then be extended into a real life pilot study which will be used to study human behavior while using Google Maps for navigation.
Pilot driving study
The main focus of our pilot study will be to use our findings from the simulation study to further evaluate our design in real time while users are actually navigating and operating a vehicle. The two areas of focus include comparison of response times for the user across different modality cues and evaluation of the user’s ability to multitask while navigating using Google Maps. Our new design would allow the user to continue correctly on the navigation path while a call comes through during navigation.
Lessons learned
1. Improvements for the future: I recognized several usability problems as participants’ interactions with products change day by day, which made me realize that as researchers we need to enter users’ natural environments to talk with them and understand how they think in order to create technology with enhanced usability.
Back to Top