Which is to say, psychologically, giving a largely virtual and meaningless reward for largely virtual, meaningless activities makes sense. Doing so for real-world efforts, however, would tend to expose how arbitrary, empty (and quite possibly marketing-related) those rewards are.
My wife just recently completed the Susan G Komen 3-Day for the Cure. Fund raising, preparation, and training took up large parts of the past year. Her team of two raised just over $8000. She got her real-world feelgood out of actually raising $8000, and actually walking 60 miles in three days, and physically standing side-by side with others who did so. An electronic gold star (even if attached to some real monetary benefit, say) would not have enhanced the experience.
That's a good point and I wouldn't necessarily disagree with it. Which is why I think video games and that kind of thing are more likely to serve as excellent practice scenarios for skill development than direct models for a rewards system.
That is a virtual reality is more likely to allow practice for real life, and inspiration for real life, than a real world model rewards system.
However, just to address Ryan's point let me say this. I have a Second Life. (I rarely use it except as a communications and exploration platform. Occasionally I try out new VR vadding techniques and equipment there.) In certain virtual reality frameworks you can conduct real world business that can also actually allow you to develop real world benefits and rewards that overlap in both directions (reality, and virtual reality). Movement and achievements can move in both directions.
In an open VR architecture rewards can also change and develop over time, as opposed to a static VR architecture (like a non-networked video game) where rewards are always pre-programmed.
In other words I think any rewards based system should be open to modification over time, be adaptable enough to suit the individual personality of the user (and his/her particular motivational set), and also allow the end-user to build their own rewards structure as they see fit.
Then it will be more likely to be truly motivational, and ultimately useful in the real world.
So I would suggest any such rewards system be based upon an open VR architecture, and the reward system structures could evolve in both directions simultaneously to fit the needs of the end-user(so).
Or in other words, don't try and program the system functions themselves, just the general architectural framework. Leave the actual systems programming and design up to the end user. Let them develop their own individualized applications.
Other than those comments I agree with the work-ratio hypothesis. Anything of any real value requires real work development. However I also believe, due to personal experience and observation, that with the proper applications of certain technologies, the nature of work is changing. Sometimes radically. This seems self-evident, but not necessarily so, because I am speaking psycho-socially.
And work is becoming far less laborious (physically) over time (for vast numbers of people in developed nations at any rate), and work is producing both products and results that are far more flexible and innovative. Again, to a large extent, because of various technologies.
By this I mean, as just one example, video games are subtly changing ideas about work and reward structures, even if only sub-consciously at this point. But every generation is also sub-consciously (and sub-culturally) adopting new technology mindsets with each new idea-framework they adapt. Work nowadays is nothing like it was for my grandparent's generation, and I don't just mean that in regards to physical productivity and labor saving technological enhancements. I mean technology has changed whole cultural mindsets in regards to what work actually is, the types of rewards one can be expected to derive from work, and what work is likely to look like in the future.