Major Human Factors Problems of 2024: Tesla Electronic Door Lock Manual Overrides, Apple Vision Pro, Sonos App Update.
Charles L Mauro CHFP
Founder and President of Mauro Usability Science / Neuroscience-based Design Research / IP Expert
Who Would Have Thought??2024 was a year to remember regarding major human factors problems from world-class corporations like Tesla, Apple, and Sonos. These companies are all known for superior product design, market penetration, and advanced consumer-facing technology. Yet they all delivered products to the marketplace with massive human factors oversights. Some cost actual lives, others resulted in market failure and widespread negative customer response.
A Major Change Is Coming: The scope, cost, and impact of human factors oversights by these successful companies suggest that while their leading product development teams focus aggressively on developing the most advanced technology possible, these same teams fail to think systematically about the missing human factors science behind their consumer-facing solutions which impact safety, ease of use, overall interaction quality, and, most importantly, market acceptance.
The Tesla case study is a prime example of a new human factors science problem facing corporate product development executives who believe advanced technology is the primary pathway to success. An entire generation of development executives have thought (mistakenly) that with enough technology, consumers will accept anything. This a fool's errand, as these examples demonstrate. This way of thinking brought the Apple Vison Pro to market in a failed form factor and is how Sonos launched an app update that bricked many users' entire systems. First, the human factors problem that took human lives.
The Tesla Electronic Door Lock Manual Override: In what has now become an unfortunate and terrifying story, four individuals in a Tesla were recently incinerated when they could not override the dysfunctional electronic door lock system using manual overrides accessible from inside their burning car. A detailed analysis of the incident shows that this terrible event was a fundamental human factors research problem.
How does a leading automotive product reach the consumer with human factors problems so glaring that it takes human lives without reason? The overrides were so complex and hidden that gaining egress from inside the burning Tesla was apparently impossible for the four occupants trapped inside the vehicle. The one occupant who did escape did so through a broken window created by a passing motorist. The root cause of this HF problem is both systemic and product-specific.
Too Many Electronic Interfaces and Automated Control Systems: The problem with the Tesla door lock override is another example of the overuse and misuse of electronic interface elements like closures, large screen-based feature access panels, and the all-time electronic interface Autonomous Driving Systems. A production Tesla is festooned with electronic interfaces that are both active and passive in nature. The Tesla product development approach has used as much automation as possible spread across most user interaction touch points. It has been described as a computer on wheels. Taken from the point of industry trends, over-automation is the seminal human factors science problem facing product development teams today.
Use-Case vs. Human Factors Validation:?Did Tesla execute formal emergency human factors entrance and egress testing utilizing properly screened user-profiles and simulated environmental conditions to validate the overrides? In most EV accidents, manual overrides are the only means of rapid egress in a battery fire besides breaking windows, which is impossible without a dedicated glass-breaking device (not provided by Tesla)
Risk Analysis Drives Interface Design Decisions: Even when windows are broken, occupants must be pulled through the broken window unless the manual door release is activated from inside the car. With this knowledge in mind, it is clear that a Failure Mode and Effects Analysis (FMEA) or other formal human factors risk assessment would undoubtedly show this use case as life-threatening and a top priority in mitigating risk for Tesla drivers. Yet these functions were dedicated to battery-powered electronic subsystems, the functionality of which was determined by the status of the Tesla battery. In terms of human factors science, this is a CLASSIC function allocation problem. The discipline of human factors science employs many robust and actionable methodologies focused on these exact types of problems. For example...!
User Situation Awareness (SA): A primary component of this specific human factors problem was the occupants' need for Situation Awareness and near-total lack of interface affordances in the mechanical override system. What does this mean in simple terms?
Situation Awareness (SA) is a formal methodology embedded in human factors science. SA is defined by the user's expertise in a specific task or series of tasks—knowing how to do something complex, accurately, and rapidly. A properly designed system builds SA organically through repeated use and, in some systems, combined with training. However, relying on training to operate core safety features in a complex consumer product is widely understood to be unreliable and never a state to be relied upon.
When A Procedure and Interface Is Not Self-Evident By Design: The mechanical override egress controls in the Tesla contributed minimal support for developing a workable SA profile for automobile users. This left the user with a near complete reliance on prior instructional materials or training AND the self-evident operation of the mechanical override controls in their Tesla. When a user interface design communicates its function by design based on visual inspection, we say that the solution has high affordance. SA and affordance development work together.
Affordance Is Not Optional: The more an object's physical design tells us by visual inspection about its function and purpose, the faster we move on to performing tasks with the object or system. This facilitates cognitive minimization, a core feature of the human operating system. J. James Gibson introduced the concept of affordance more than 7 decades ago. So, Tesla cannot claim this is a new idea. It is interesting to note that affordance has recently generated new interest in cognitive neuroscience because how we navigate the world is determined by how we assess the objects we see in our environment. When we conduct major human factors research studies for clients on these types of life-critical interfaces, we always start by testing the designer's proposed solution for affordance.
Total System HF Failure: The human factors problems with the actual design and layout of the Tesla emergency door lock release system overall run through the entire egress use-case for front and back seat door release. In some Tesla models, the rear door release is buried under plastic mats, and secondary access doors that require a tool to pry open the tiny door covering a release pull. The actual release pull is visually inconsistent with the function of operating the actual release mechanism.
Tesla Owners Post Solutions! None of this entire interaction sequence offers the rear seat occupants ANY affordances or Situation Awareness formation. Should the backseat be occupied by children the release is likely impossible to access and utilize. It is interesting and important to note that even regular consumers who owned Tesla were aware of the defective design of the manual override system. Some Tesla owners even posted YouTube videos suggesting makeshift workarounds for the mechanical overrides in their Tesla.
Details Matter In Affordance Terms: The front seat manual overide control is impropertly grouped with window controls and is neither marked nor does the shape offer the most basic affordance in terms of operation. It does not convey the action "Pull Up To Open Door". Under high-stress conditions users commit errors of omission which means that forget steps in an interaction sequence. Clearly, the two front seat drivers who perished in the recent Tesla fire did not understand the basic operation of the front seat manual door lock release physical design.
Learning Transfer Is a Powerful Measure: The routine electronic door lock employs a different shape coding, operation, and location than the manual override control. This is important because it means that there is virtually no learning transfer between the electronic system and the manual override. In fact, there is negative learning transfer, which is a well-understood serious human factors problem. Learning Transfer is a large field of study in human factors science.
The Solution Set Is Well Understood: The solution to these types of human factors problems is well-understood human factors science based on risk assessment followed by function allocation analysis, control/display compatibility analysis, Interface design, simulation development, user testing, and finally, validation under appropriate safety guidance utilizing a representative group of consumers. Did Tesla undertake even the most basic human factors methodology in the design of this system? The lawyers will determine the answer to that question.
How We Learn Matters...A LOT: Under the Situation Awareness methodology, the entire field of skill acquisition theory provides a framework for how we learn certain types of complex procedures under routine instances and how such skills translate to extreme conditions. The Tesla egress system's problem is that it was likely never tested utilizing professional human factors research methods for the different use cases requiring the necessary skills to operate the mechanical override.
What We Remember and Forget: In my experience conducting human factors testing for this type of system, I found that a component of skill acquisition called "learning decay" is the most important. Even if the driver understood the system when they bought their Tesla, there is no guarantee that they will recall the exact procedures with time. Learning decay is central to situation awareness and designing all critical interfaces.
领英推荐
User Testing Is A Science, Not An Art: A significant aspect of human factors science explicitly deals with simulating complex environmental conditions. There is no need to set a car ablaze with a respondent inside to determine that the human performance of egress procedures is a massive human factors problem. It's done every day. I assure you that when discovery starts on the litigation surrounding this accident, it will be first on the discovery request list.
Root Cause Summary: At the very core of this design failure by Tesla is a likely failure to employ core professional human factors research methodology. The very first step in such a program would have been Risk assessment followed by "Function Allocation Analysis." This is the structured determination of which functions in the egress task structure should be controlled by human skills, rules, and knowledge and which functions should have been allocated to machine control (Tesla electronic subsystems). Thinking that the occupant can manipulate a new physical egress interface modality is equivalent to machine control, especially under high levels of environmental stress. There is no world in which rapid, emergency egress is left to machine control when it is well understood that said machine enters a failure mode, leaving the occupant encased in an inferno.
Why This Human Factors Problem Matters:?By almost any measure, Tesla has pushed the envelope by stuffing loads of advanced digital technology into the automotive form factor. However, this example demonstrates that there is far more to a consumer-facing automotive solution than digital interfaces and EV propulsion. Thinking that technology is the primary feature benefit for a new generation of automotive products is a fool's errand without carefully considering the core human factors science surrounding this new customer experience. The courts (and juries) will likely take a harsh eye toward product development behavior that ignores fundamental human factors validated design solutions.
Tesla Is Not Alone:?From a corporate cultural point of view, this is exactly the same problem that Tesla has repeatedly faced with its autonomous driving systems: a major lack of human factors science. Tesla is not alone in this problem space. Both Apple and Sonos launched products in 2024 that reflected the same problem of too much technology and a failure to address core human factors requirements leading to massive product failure and negative consumer press.
The Apple Vision Pro Failure:?Although the Apple Vision Pro VR headset turned heads with its mind-bending UI at its core, the product was a massive failure due to weight, center of gravity position, and fit on the head/face. These core ergonomic factors show an enormous need for Apple to take seriously the real-world human factors science related to human head/neck biomechanics and facial vascular structures known to impact all head-mounted devices. Yet Apple, much like Tesla, pushed to market a product of inevitable failure based on relatively simple ergonomic variables. Perhaps the simple things are more complex than one might assume. Even the most flashy UI will never overcome the simple problem of VR fatigue and head/neck biomechanics.
Sonos App Update Failure:?It is hard to imagine a recent App upgrade by a major consumer electronics brand that was handled worse than the 2024 App update by Sonos. From the moment the update was launched, message boards, news feeds, and REDDIT overflowed with massive negative comments about the update. The update was so bad that it essentially bricked the entire audio systems of many customers. One of the most successful consumer audio brands in recent history put into action an App update that could have never undergone independent user testing or usability research. Interviews with company executives aimed at addressing the problem only produced more negative responses. It's hard to imagine how the CEO of Sonos responsible for the release still has a job...and a seven-figure paycheck to boot. Where is the SONOS board? They are probably still trying to log into their Sonos accounts.
Founders, Co-Founders, and Development Executives: These three examples demonstrate that creating world-class products involves much more than stuffing the last technology into an attractive form factor. From this point forward, success will require far more attention to human factors performance during development, whether it be comfort, safety, or traditional usability.
If you are a product development executive, founder, or co-founder, these examples drive home the need to balance human factors science and advanced digital technology. Thinking that ever more advanced technology raises all boats is an assumption for times past. The future is not what it used to be. Who would have thought?
President / Founder (1975-Present)
References / Additional Content
Failure Mode and Effects Analysis: https://en.m.wikipedia.org/wiki/Failure_mode_and_effects_analysis
News Description of The Accident: https://www.mirror.co.uk/news/world-news/terrified-friends-burned-death-tesla-34087725
Video of lock operation: https://www.carexpert.com.au/car-news/four-dead-after-being-trapped-inside-burning-tesla
Consumer Work Around Solution: https://www.youtube.com/watch?v=d1gq1hXVYPw
Emergency Response Guide Video https://www.youtube.com/watch?v=O0hVuR7Zmbc
Security hardware and software architect
1 个月Actually, the concept of affordance was introduced by the American psychologist James J. Gibson, not William Gibson. James J. Gibson coined the term "affordance" in his 1966 book, The Senses Considered as Perceptual Systems, and further developed it in his 1979 book, The Ecological Approach to Visual Perception
Leadership and Keynote Speaker and member of the Data Science Research Centre at University of Derby
1 个月An excellent analysis. Many thanks.
Senior Performance and Development Scientist @ Charlotte Hornets
1 个月This is fantastic. Thanks for sharing!
Research, UX Maps and Service Design @ Cognitive Ink | Experience Anatomist & Technology Philosopher @ Adventures In A Designed World
2 个月We had this figured out decades ago! Bit of an unfortunate regression: https://www.christopherroosen.com/blog/2023/6/5/human-factors-sanders-mccormick
UX Architect, Futurist Artist, Inventor, & UX Activist
2 个月Great article, as usual. Special obscure mechanisms for emergency are a recipe for disaster. In the Rhode Island night club fire, 100 people died because, despite well-marked exits, people were inclined to leave by the way they entered. Strong mental models imprinted by immersive experiences win out over verbal or written instructions. In the 737MAX MCAS instances, pilots naturally attempted to use the normal controls to correct the attitude of the airplane. And why not? An emergency is a terrible time to process new information or try to recall something learned, but seldom used. This is why first responders are trained repeatedly to reinforce desired mental models. Similarly, for the sake of accessibility, special tools or features often fail because they defy the main established mental models. I'm convinced more than ever of the importance of a universal design approach. Simple straightforward concepts that address broad use cases, primary to edge case. It's easier to say than to do, but worth the effort. Unfortunately marketers might view it as boring. We already suffer from fragmented solutions. I'm afraid as long as clever sci-fi magic is valued over sound usability and safety, the situation will only worsen.