The car was belting along at 120kph when I forgot how to drive. It was late on a Friday evening as I hit for home after a week on a client site in West Cork. I had the motorway all to myself and decided to take it easy, threw on Jimi Hendrix’s Axis: Bold As Love and shifted into cruise control. The red lights of the car in front were upon me before I knew it. I looked for the brake but my foot didn’t budge. I swerved into the outside lane at the last second, blessed that there was no one else alongside me. Regaining my composure I switched out of cruise and began to fumble with the pedals, my foot-to-pedal co-ordination returning within seconds. On previous journeys I’d noticed a similar lack of facility that lasted no more than a moment when switching back to manual. I had driven on cruise control uninterrupted for the guts of 40 minutes. So what happened?
Similar to anything by Lawrence Lessig, a new Nicholas Carr book invariably leads to my dropping whatever I’m doing and diving straight in. He’s that rare species; a journalist concerned with the nexus of technology and human beings who makes the complex simple and, more importantly, interesting. His latest book, The Glass Cage, is a logical progression from 2009’s The Shallows: How the internet is changing the way we read, think and remember. It deals with how automation is leading to the atrophy of human skill and intelligence in both work and our personal lives.
Carr’s essay in last weekend’s Wall Street Journal Automation Makes Us Dumb opens with the following paragraph:
“Artificial intelligence has arrived. Today’s computers are discerning and sharp. They can sense the environment, untangle knotty problems, make subtle judgments and learn from experience. They don’t think the way we think—they’re still as mindless as toothpicks—but they can replicate many of our most prized intellectual talents. Dazzled by our brilliant new machines, we’ve been rushing to hand them all sorts of sophisticated jobs that we used to do ourselves.”
We are not only automating the routine tasks but the more complex and specialised ones too. Carr cites a Harvard Business School study carried out by Professor James Bright which showed that automation had reduced the lot of factory workers to carrying out drab and mundane work devoid of skill thereby reducing them into being merely “pusher[s] of buttons.” This is what British Aviation researcher Matthew Ebbatson calls “skill fade.” Ebbatson was concerned with the effect that delegating more and more cockpit tasks to computers was actually having on a pilot’s actual flying abilities and skills. One of his studies found that “flying skills decay quite rapidly towards the fringes of ‘tolerable’ performance without relatively frequent practice.”
Two aircraft accidents in early 2009 were attributed to human error or “skill fade” as a result of the unscheduled disengagement from autopilot to manual. These were Continental Connection Flight 3407 from Newark to Buffalo which killed all 49 people on board, and Air France Flight 447 from Rio de Janeiro to Paris which crashed into the Atlantic Ocean on June 1st 2009 killing all 228 passengers and crew.
Due to autopilot pilots now spend more time pushing buttons and monitoring screens instead of actually flying the plane manually. It looks as though the joke that pilots are now glorified bus drivers is more than a little unfair…on the bus drivers. In fact a pilot today spends no more than three minutes holding the actual controls on a typical passenger flight. As a result pilots are making mistakes when the autopilot unexpectedly disengages thereby placing the pilot in a role with which he is now unfamiliar. “We’re forgetting how to fly” were the words of Rory Kay a former top safety official of the Airline Pilots Association. In January 2013 according to Carr, the Federal Aviation Authority became “so concerned…it issued a ‘safety alert’ to airlines urging them to get their pilots to do more manual flying. An overreliance on automation, the agency warned, could put planes and passengers at risk.”
Automation deprives us of the practice we need to build skill in the first place and reinforce what is already there. In effect, it steals our skill, ensuring that we don’t get anywhere near the 10,000 hours of purposeful practice that psychologist Anders Ericsson asserts is the common denominator of elite performers in any pursuit. Ericsson’s studies and findings, that it was practice and not talent that ultimately matters, led to a revolution in how people viewed high performance:
“We deny that these differences [in skill level] are immutable; that is, due to innate talent…Instead we argue that the differences between expert performers and normal adults reflect a life-long persistence of deliberate effort to improve performance.”
Carr agrees with this when he makes the point that “automation turns us from actors into observers. Instead of manipulating the yoke, we watch the screen. That shift may make our lives easier, but it can also inhibit the development of expertise.” He introduces the phenomenon psychologists call the “generation effect” which highlights that human beings are more likely to remember words when they have to actively summon them from memory than if they read them from a page. They are also more likely to remember how to carry out a task if they actively try to practice it and make mistakes as opposed to following the prompts of a youtube video. As a person struggles through a task they fire of intricate mental processes which forge new neural circuits dedicated to the activity. This generates skill… “skill that requires exactly the kind of struggle that modern software seeks to alleviate.”
While modern business will jump at automating any task that leads to cost reduction they often do so at considerable risk. Call centre operators look to reduce risk and increase productivity by automating as much of the interaction with customers as possible. For instance, call centre advisors will often speak to customers using canned scripts which direct the conversation and prompt the call centre advisor to ask particular questions. These interactions lack the warmth and spontaneity of a normal conversation. The advisor is biased towards the information being fed through to him on his computer screen, is unable to anticipate the needs of the customer and neglects to trust his own instincts and creativity when problem solving. Worst of all he doesn’t or rather is unable to listen to what really is at stake on the call. They serve up a service that reeks of giving customers what the provider thinks they should want as opposed to what they really want. A labour saving device according to Carr “doesn’t just provide a substitute for some isolated component of a job or other activity. It alters the character of the entire task, including the roles, attitudes, and skills of people taking part.”
The owners of Zappos recognised the damage that scripted interactions were doing to customer service and looked to restore some measure of individual freedom in jobs usually known for the lack of it. Zappos doesn’t monitor its customer service employees’ call times or require them to use scripts. The reps handle calls the way they want to. Their job is to serve the customer, and how they do that is up to them. The turnover at Zappos is minimal and although it’s still a young company Zappos consistently ranks as one of the best companies for customer service in the US ahead of better known names like Cadillac, BMW, and Apple, and roughly equal to brands like Jaguar and Ritz Carlton.
The call-centre example is merely the tip of the iceberg. There are numerous eyebrow-raising studies available which show the skill atrophying effect of software on the more specialised professions of architecture, medicine, and accounting. I’d wager that a college graduate or a newly qualified accountant has a far superior knowledge of double-entry book keeping than the partners of any of the big four accounting firms. In fact, I was recently an interested observer in a meeting when the CEO and CFO of a respected organisation struggled to recall the formula for calculating Contribution – one of the fundamentals of management accounting.
The “Skill Fade” phenomenon is not limited to the professions. As a diver I’ve had first-hand experience as to how the introduction of dive computers changed how I approached the activity. Prior to their introduction I was rigorous about planning each dive and used dive tables to schedule each decompression stop. As dive computers calculate decompression stops as you go I found I relaxed my preparation and left things to chance. Needless to say, I’ve been lucky and have never come close to having an accident, but what if one day my dive computer fails mid dive? I am no longer practiced in how to use dive tables, and shudder to think about having to use them to make a decompression stop calculation while I’m twenty metres below the surface of the Irish Sea. I am certain that the instincts and skills that I originally developed are no longer sharp enough for me to assess with confidence what depth and for how long I’d have to carry out a decompression stop. The consequences – getting decompression illness i.e. “The Bends” – are potentially fatal.
As a keen triathlete I’m suspicious of the over-reliance of many athletes on heart-rate monitors. If my heart rate monitor is telling me that I can continue to perform at a certain level while my body is telling me otherwise I know whose counsel I will ultimately follow.
Nick Carr has hit another home run with The Glass Cage and I’ll leave the last word to him:
“Whether it’s a pilot on a flight deck, a doctor in an examination room, or an Inuit hunter on an ice floe, knowing demands doing. One of the most remarkable things about us is also one of the easiest to overlook: each time we collide with the real, we deepen our understanding of the world and become more fully a part of it. While we’re wrestling with a difficult task, we may be motivated by an anticipation of the ends of our labour, but it’s the work itself – the means – that make us who we are.”