« Expanding My Footprint | Main | Interview for Changesurfer Radio »

Augmented Fluid Intelligence

Can we survive the multitasking era?

Okay, multitasking is hardly up there with global warming, pandemic disease and asteroid strikes as a civilization threat, but it's becoming increasingly clear that multitasking reduces overall effectiveness and accuracy. Yet we're forced to juggle more and more simultaneous activities in our work, in our social networks, even in our play. As a result, simple tasks take longer, and we're far more likely to make errors. In short, as our world gets more complex and we face greater challenges, we're becoming less able to respond successfully.

Theorist Linda Stone calls this overtaxed ability to focus "Continuous Partial Attention" -- a name that's much cooler than multitasking, you have to admit -- and she describes it as an "artificial sense of constant crisis." But in many ways, the world we're moving into is even worse than this, because we're becoming so accustomed to the constant interruption that we're starting to find it hard to focus even when we've achieved a bit of quiet. It's an induced form of ADD -- a "Continuous Partial Attention Deficit Disorder," if you will, ADD via CPA.

Our ability to handle simultaneous complexity is governed by what cognitive scientists call "fluid intelligence," commonly defined as the ability to find meaning in confusion and to solve new problems. Fluid intelligence can be exercised, and in fact appears to be increasing. If Steven Johnson's argument in Everything Bad is Good For You is right, we're seeing this gradual increase in intelligence precisely because our cultural and social expressions are increasingly taking forms that are stimulating to our fluid intelligence.

But this process will inevitably have limits. Eventually, we'll hit a ceiling in the ways in which we can improve our fluid intelligence naturally. At that point, we'll face a hard choice: make major changes to our work and social cultures, so as to reduce the degree of simultaneous attention-grabbing activity; or develop augmentation systems that enhance our natural fluid intelligence by recognizing, from moment to moment, what needs our actual focus, and what can be handled by proxies. The wise choice would be the first one. It should come as no surprise, then, that I suspect that we'll do the second.

As it happens, we're already working on devices that will do just this. The problem is, these systems aren't quite done -- and at present, actually tend to make matters worse.

If you haven't heard of Twitter, count yourself lucky. It's an application that lives somewhere in the interzone between blogging and text-messaging, and went from being nearly invisible to nearly ubiquitous in less than a week in early March. [This explosion was likely due to the combination of overlapping tech-fests (TED, SXSW, GDC) with concentrated early-adopter attendance and Twitter's complete dependence upon network effects for utility (i.e., the more people have it, the more useful it becomes). Expect other network-dependent apps to try to artifically reproduce this perfect storm next March.]

Twitter allows you to send quick and easy messages about your various activities to people who have selected to receive them; the current joke is that where regular blogging let you give daily reports on your cat, Twitter lets you give minute-by-minute updates. During busy periods, it's quite easy to be overwhelmed by the volume of incoming messages, the vast majority of which will be of only passing, mild interest at best.

But that leaves the tiny minority of truly useful and interesting posts, ones which have particular value due to their timely arrival. At present, finding those requires wading through the mass of "my kitty sneezed!" or "I hate this taco" messages; of course, this is exactly the kind of low-complexity activity that we'd habitually perform via Continuous Partial Attention.

Imagine, however, if Twitter had a bot that could learn what kinds of messages you pay attention to, and which ones you discard. Perhaps some kind of Bayesian system, more complex than current spam filters, but not outrageously so. Over time, the messages that you don't really care about would start to fade out in the display, while the ones that you do want to see would get brighter -- an adaptation of the "ambient technology" concept. These bright headlines would stand out against the field of gray, drawing your attention only when you would desire it. If this worked reasonably well, you'd have reduced the overall demands on your fluid intelligence by outsourcing some of the rote filtering to a device.

These kinds of bots -- attention filters, perhaps, or focus assistants -- are likely to become important parts of how we handle our daily lives. We don't want to have the information streams we've embraced taken away from us, and every decision to scale back how frequently we check email or stock tickers or combat results or the like raises the spectre of our competitors choosing to tough it out. As these information streams become more and more important to our professional and personal lives, the harder it will be to pull away. So rather than disconnect, we'll get help.

We'll be moving from a world of Continuous Partial Attention to one of Continuous Augmented Attention.


For an even more futuristic solution, read my short story titled "Multi-Tasking"

The point where the focus of AugCog changes from gathering and increasing the information available to someone, to filtering and reducing it, is a really fascinating concept.

There's been a fair amount of fiction about it, and of course applied examples exist... An immediate example for any of us that have been online for more than a few years is to remember how badly web search was starting to suck when Google came along with PageRank, changing the way sites were filtered.

my kitty sneezed and I hate this taco

for more of a social media twist on managing twitter...what if tweets could be voted on or tagged and the tweets of users with the most votes or tags would show up the brightest?


Creative Commons License
This weblog is licensed under a Creative Commons License.
Powered By MovableType 4.37