Quantified Self: Attention Tracking

This post is part of my liveblogged account of a conference. Two disclaimers: Liveblogging is hard, and I often get things wrong. If I did, please feel free to correct me via email or in the comments and I’ll make changes when appropriate. Second, the opinions expressed in these sorts of posts are those of the speakers, rather than mine.

Matthew Trentacoste, a PhD student at the University of British Columbia, leads a breakout session at Quantified Self on “attention tracking”. He introduces himself as a person who’s been diagnosed with ADHD. As someone who’s easily distracted, he’s interested in figuring out strategies to reduce these distractions.

One piece of advice doctors give is to manage your environment: work in a quiet room with few things around. Matt has tried this, and concludes that he can distract himself even within a quiet room. And the internet, of course, is not a quiet room. He’s been building tools that help him focus in online environments. Tools like Rescue Time track what window is in focus on your computer, and tells you how much time you spent on email versus web browsing. But there’s a big difference between spending five minutes focused on email at the end of an hour, rather than 5 minutes, in 15 minute quick checks of email, which can be very distracting.

His tools offer simple data on how long he spends looking at a specific window and how often he changes focus, and a score. The score attempts to communicate and monitor the frequency and number of distractions within a given time.

This leads to a discussion of how we might define “attention” – Matt’s definition centers on resistance to distraction. But other definitions around a room full of brainstorming, frequently-distracted people, focus more on productivity, and specifically on achieving “flow”, as described by Mihaly Csikszentmihalyi. Some of the techniques suggested in the room for increased productivity (the Pomodoro time management technique, for instance), seem to trade flow for productive interruption. Matt suggests that we consider Paul Graham’s “maker schedule” versus “manager schedule” – making requires large pieces of uninterrupted time. In extreme programming, for instance, time is allocated in half-day blocks, and work with a partner helps counter the tendency to overfocus on interruptions and achieve flow. But, as another speaker suggests, there’s a close link between AHDH and flow – one definition of ADHD is focusing intensely on unhelpful or inappropriate tasks.

One of the participants asks whether we can really study attention in a quantifiable fashion. Like sleep tracking, there’s a danger of interrupting attention by tracking attention. One of the participants works at NeuroSky, a company that manufactures inexpensive brainwave monitors, suggests that it’s possible to measure alpha and beta waves and understand focus and relaxation. Olympic archers, she tells us, are both very focused and very relaxed. In more traditional cognitive tasks, like doing math problems, we show high focus, but much lower relaxation. Alpha waves correlate to states of relaxation and beta waves to focus. We can understand some of the differences in focus based on the amplitude of waves – we see higher beta waves when doing math problems than when we read, for instance. With athletes, we see a mix of alpha and beta waves, suggesting both relaxation and cognitive engagement.

Naveen Selvadurai (of Foursquare) suggests that we might be missing the point in simply trying to optimize our focus or productivity. There are physical aspects to attention – you’re often more distracted when you’re hungry. He suggests that we might think about our day in terms of blocks where we’re focused on different activities. Matt suggests that he’s got different capabilities at different times of day – deep thought seems to happen best between 11pm and 3am.

The discussion moves to possible metrics to study attention. One participant has built a tool that periodically flashes random numbers or characters in the center of his computer screen and somewhere in the periphery – he can measure his attention by checking his peripheral response abilities. Someone suggests the “sustained attention response task”, clicking after each number that’s not a three, for instance.

Other participants are building Google Chrome extensions that track when we open and close tabs – one participant is interested in “tab churn” as a signal of productivity, opening and closing tabs when we need them, or leaving them orphaned when we are distracted. Matt has built a tool that watches all windows on MacOS and outputs times of focus in terms of millisecond.

There’s a suggestion that we start using information about physical states to inform conversations about attention. One option might be to incorporate information from pedometers to track when we are leaving our desks and how long we take to get refocused. I find myself thinking that, given Naveen’s observation about focus and hunger that tracking blood sugar might be helpful in terms of enhancing subjective data about focus with physical data.