Finding the most effective balance between direct instruction and discovery learning is a tricky business. We just finished the first group computer and Second Life training in our Women in Control study (Drs. John Wiecha and Milagros Rosal, Co-PIs), and I have to confess that, despite all of our careful planning and design, that particular fulcrum turned out to be pretty darned illusive. There were nine women in this first group (three more groups to come) – all of them with type 2 diabetes, new to computers, and definitely new to Second Life. Our job? To get them up and running on their brand new laptops, and comfortable in the virtual world of Second Life. The challenge is made all the more daunting by the fact that the women in our study are not computer literate and most of them have a fourth-grade reading level.
Each subject in our portion of the controlled study receives a new Macbook computer and a 4G wireless device which will allow them to join their diabetes healthcare group sessions, from home, in the virtual world (the control group will participate in face-to-face sessions at Boston Medical Center). They will meet online, as a group, once per week, for the next eight weeks, with nurse educators, to learn about their disease and how to manage it. In order to get them up to speed for the group sessions, we conduct two face-to-face computer training workshops. The first introduces them to their new laptop, shows them how to assemble their modem and connect to the internet, and get familiar with the keyboard, mouse, and interface. The second workshop introduces them to Second Life. In advance of the training, we set up their SL accounts, create and cloth their avatars, and configure their interface (set favorites and preferences). So, when they first fire up Second Life, they arrive in a configured “home” and meet the avatar that bears their name. Lessons abound…
* A passive introduction to SL (project SL up onto a screen, dim the lights, and give them a basic tour) at the end of the first computer training really helped. “Just sit back and relax, you don’t have to do anything right now, just enjoy.” You could see them visibly relax after the tension of figuring out USB cable connections and learning the significance of new terms like “modem”, “space bar”, “application”, and “return key”. I kept the initial tour to just the basics – my avatar waved to them, walked around a bit, sat down, stood up, and then flew (“Oooohh!!!”) over the area that will be their new virtual “home”.
* We break the training into two successive sessions – it’s too much to swallow all at once. Ideally, it would be three sessions. Almost everything our subjects are learning is completely new to them – it’s overwhelming.
* Modeling the fabulous instructional comics made by my colleague, Liz Dorland, I created very simple tutorial comics (sample: Talk.WIC) that review the basics of talking, moving, and finding objects in inventory. I kept to one page per skill, used very simple language, and printed them out. These went over very well.
* Our subjects identify with their avatars very quickly. Evidence of this….one who walked into the water, “Yikes! I don’t want her to get wet!” Two who bumped into each other, “Watch where you’re stepping, girl!” One who didn’t like her look, “How do I change these clothes, I look awful.” And as we prepared to shut down, “Where should I put her? She needs a house to sleep in.”
Throughout the training sessions I could feel the constant push and pull of how much to instruct versus giving time and space to explore on their own. We think of didactic instruction as the most efficient method (pack it all in) for teaching complex information to a large group. In a budget-strapped grant such as ours, with a large population (100 subjects), time is of the essence. But we also know that exploration and discovery are keys to learning that sticks and a better overall experience for the learner. This idea is reinforced by a study, recently published in the journal Cognition, by Elizabeth Bonawitz (UC Berkeley) and Patrick Shafto (University of Louisville), where four and five-year olds visiting a museum were divided into four groups with four pedagogically different introductions to a novel toy. The toy had many “hidden” interactive elements in the toy (squeakers, musical panels, lights). Group 1 received a full-on demonstration. The experimenter with Group 2 pretended she was interrupted in the midst of her demonstration, suggesting that she didn’t finish demonstrating all that the toy could do. Group three’s demonstrator pretended that she’d discovered something the toy did by accident. And in Group four, the toy was simply given to the children with the comment, “Wow, see this toy? Look at this!” and then the demonstrator left the room. After the various introductions, the children were left with the toy and allowed to play. All of the interactions were videotaped and later analyzed. The results were crystal clear – the less direct instruction received, the longer the children played with and explored the toy. And it wasn’t just time spent on task, the fourth group (the group given less instruction) tried out more actions on the toy. In other words, direct instruction seemed to dampen the learners’ motivation for exploration and self discovery.
But how to balance that experimental truth against the fact that these adult learners have absolutely no context for this new world they are exploring? Their low literacy, limited experience with computers, and negligible understanding of the internet would cripple open exploration and only lead to frustration. For this program, we’ve settling on a hybrid model – a blend of direct instruction, followed by periods of exploration. What some refer to as guided inquiry. For example, we show them the gesture panel in Second Life, ask them to click on it, and then stop talking – indicating a gap in the “class” for exploration. We lead their avatars to an interactive dance floor, show them how to access the choser panel, and then stop talking – indicating another gap. The audio recordings of the training sessions reveal a rhythmic swell and lowering of sound, almost like a musical score, that trace the pattern of these “gaps”. The noise level gets loud and raucous during the exploration times then quietly hums during the direct instruction times. I’m coming around to understand that the chaotic noise level is, in part, a measure of success. But it is omportant to recognize that these learners (and most of us, for that matter) need both. They need context, shape, and guidance — and they need to explore and follow what interests them.
Designing sessions this way involves mental adjustments on the part of the instructors. For example, we must get used to a reduced emphasis on “covering the content”. One has to let go of anxiety over topics left out or diminishing time to “get to” key concepts. We are learning to measure success by the learners’ enthusiasm for the experience. If they end the session feeling like it was fun, that this virtual world was intriguing and they wanted to get in there again to play around, then the training session was a success. But that doesn’t mean that we don’t have a plan for the skills they must master in order to be succesful. Fortunately, we’ve built into the instructional design later opportunities for the learners to scoop up specific skills missed in front end training, if needed.
I have to say that my favorite moment of the training was when they first logged into SL and saw their avatar rez in front of them. “Ooooh! There she is!” These moments, when their delight and fascination over this completely foreign world, are contagious.