Reality-based computer simulations allow staff to grow through failure

By Ann Macaulay
|Canadian HR Reporter|Last Updated: 09/08/2003

“Everything that’s wrong with training can be stated in four words:

It’s just like school

,” writes Roger Schank in his book

Virtual Learning

, subtitled

A Revolutionary Approach to Building a Highly Skilled Workforce


Schank’s scorn for traditional teaching methods is evident: “You don’t learn anything in school that’s worth knowing and you don’t retain it. You memorize (something) for an exam and then you forget it like everyone else. Stand-up training has mimicked that. Most companies want to do training, but they don’t want to do it right. If they wanted to do it right, they would actually put people through complicated practice.”

People learn by doing, failing and practicing, Schank believes. “You don’t learn anything you haven’t experienced. The idea that you can (learn) by somebody telling you something is illusory. It has to be learned by doing, but it also has to be repetitive doing. Practice is very important. The key issue in learning is the possibility of failure and having had failures that are so emotionally impactful that you have to spend a fair amount of time considering what went wrong and want to correct the situation. If you don’t go through that, you’ll never learn,” he says.

Schank, currently the chair of Cognitive Arts, a U.S.-based firm that creates training software, is a former professor of computer science and linguistics at Stanford and Yale universities. He made two discoveries: computers didn’t get bored, and the gap between what he learned about learning and what children are taught in school is enormous. He realized that his team at Yale could create educational software to help students learn in what he describes as “the right way.”

Schank subsequently met some executives from Andersen Consulting, who were extremely interested in finding a better, faster and cheaper way to train people. In 1989, Andersen brought Schank’s team to Northwestern University in Illinois and helped fund the Institute for the Learning Sciences (ILS), which invents new learning technologies for students. Schank also created a for-profit company, Learning Sciences Corporation (LSC), which adapts these technologies for use by corporate clients. LSC has tailored reality-based computer simulations for several organizations, including Target discount stores and the Bennigans restaurant chain.

After working with organizations, Schank made three discoveries:

•people hate training;

•people like multimedia technologies; and

•it’s easier to sell to the pocketbook than to the mind — focus on cost and time savings

Schank believes organizations are headed for serious trouble if they don’t find a better way to teach employees how to do their jobs. People who go through a week of training “aren’t learning anything. But they may very well be learning what other guys in the company are like, and what the leaders look like, and the way the facilities are. It isn’t as if they gain nothing from the experience. That would be wrong. They might have a great bonding experience, they might meet other people who will come up later in their business lives. There are all kinds of things they learn. It just isn’t training. The actual training to do their jobs wouldn’t occur there.

“Corporate training is so busy replicating the existing school system that sometimes they actually teach the same subjects. A lot of corporations actually have training in mathematics, they don’t even know why they’re doing it — they’re just doing it.” Computer-simulated teaching, on the other hand, allows employees to experiment, to create situations that are realistic, and to fail in privacy.

Virtual Learning explains how, at Andersen Consulting, the ILS team took the lessons the firm wanted to teach and turned them into a computer simulation. One part of the course is about human resources — the trainee becomes the HR person at a printing and publishing company, with responsibility for hiring and firing, pay increases, counselling and other managerial tasks. The trainee must face a “mind-boggling array of issues, decisions and consequences,” including:

•who to send to a training program (if they make the wrong decision, there are morale problems);

•an unproductive employee who is a nice guy (when he’s fired, the firm faces a wrongful dismissal suit); and

•a supervisor who suddenly retires and the firm doesn’t have a replacement ready.

The trainee is given one situation after another, and as each simulated month goes by, the computer shows the consequences of the decisions. The trainee is reprimanded, commended or fired, depending on how he or she has reacted. People appear on-screen to tell stories related to the decisions that were made. As Schank says, “everyone fails. But it’s a fun, private failure that encourages learning. People who have taken this course routinely compare computer-simulated war stories about how they got fired or what they messed up…you might think they were describing events that really occurred. Perhaps that’s the proof that a simulation was effective and learning took place — participants talk about it as if it were real.”

Schank says his “sense of it is that people who are in the stand-up training business want to stop being in the stand-up training business. The idea that you can put one person in front of 50 people or 1,000 people and have them say something and have that be a meaningful experience is absurd.

“I think HR executives would be wise to eliminate all training completely that does not involve practice,” he says. “They have this illusion that we put people through training because we send them for a week of listening to somebody talk. This is just a waste of time. Don’t do it. They always have some other reason for it — well, it’s good for bonding, and community, it makes people like each other. That’s actually true. But it’s not training.”


10 rules for training

Author and trainer Roger Schank sets out 10 rules to learn by in his book Virtual Learning:

•people remember best what they feel the most;

•dumb employees aren’t born — they’re made;

•deliver training just in time (or when a learner has just failed and really needs help);

•you can fail to learn just about anything;

•learners will teach themselves better than the world’s best trainer of highest-paid motivational speaker;

•memorization without corresponding experience is worthless;

•when a company buys a learning system, it should come with all the options;

•training should open with a bang;

•trainees should be learning from the world’s best; and

•it’s better to train the many rather than the few.


Simulation building

(Excerpted from Roger Schank’s

Virtual Learning: A Revolutionary Approach to Building a Highly Skilled Workforce


An overview of the process to produce a workplace simulation. The following is the process from Learning Science Corporation’s perspective; for in-house trainers or HR professionals, the steps can be easily translated to apply in that setting.

1. Meet with prospective client. Sometimes corporate leaders come to us in crisis, they desperately need to train people in a skill in which the firms are in short supply. Maybe the companies have an expensive new computer system and no one understands how to use it. Perhaps the companies are driven by a competitive need — they need a skill that will help them gain a competitive edge. More often than not, organizations don’t know what skill they want to teach. The managers simply say, “We spend a lot of money on training and we want to improve its effectiveness. Can you help us?”

2. Obtain buy-in. It’s not always easy for companies to make the commitment to developing a computer simulation. Certainly there’s the money issue — multimedia isn’t cheap. But there are also psychological and emotional issues. People have to buy into failure as a learning catalyst; into the idea that trainers can simulate reality on a computer; that trainers have to stop trying to tell people what companies want them to know.

3. Pick a skill. Once the buy-in is obtained, the training team starts working with companies to determine which skill to teach. Usually there are numerous choices and needs. Ultimately, the designers narrow down those choices to what is teachable — the skills in which companies have expertise and experience.

4. Conduct interviews. Next LSC team members interview employees, as much to obtain good stories as to understand what to teach. Asking the right questions is critical. In this step designers are searching for the common failures that can help people master targeted skills.

5. Script (build the simulation). The challenge here is to hook the stories to the failures so that learning occurs and to do so in a realistic manner. If the simulation seems phoney or people don’t have access to stories that “explain” the failures, learning won’t happen. Once all of the alternatives are scripted, it’s simply a matter of putting everything together from a technological standpoint.

Add Comment

  • *
  • *
  • *
  • *