Yesterday I wrote about how anthropologists might describe my day, today I'll give a psychologist's perspective.

One aspect of cognition that interests me is how our attention changes. Psychological theory suggests two competing orientations-deliberate and reactive. When acting deliberately, we have an idea of something we want to do. So we perform a "means-ends" analysis: To accomplish this goal, I first need to do this other thing, and to do that, I need to do something else, and so on. When acting reactively, instead of being focused on a high-level goal, we jump about according to whatever ideas enter our mind, often affected by something we see, hear, or feel.

I'm especially interested in our reactive behavior. How can we jump about so quickly from one objective to another, and maintain track of what we really need to get done? We can be interrupted by internal ideas ("Oh, I'm thirsty") or by something outside ourselves ("Bill, do you know where we put the small carrots?"). With each interruption, we shift our attention, at least momentarily. We have an ability to disregard an interruption ("I will not read my email until I finish writing this report"), which psychologists call "meta-cognition"-an ability to control what we are thinking about and doing. Some scientists believe that this ability to reflect and deliberately control ourselves, especially by using language, is the essence of human-like consciousness. This aspect of human intelligence makes us quite different from other conscious animals, such as cats and dogs.

Yesterday I talked about "chunks" or "activities," broad ways of thinking about what we are doing, which persist for several minutes or hours. A reactive behavior, such as reaching for a drink to quench your thirst, might just last for a few seconds or minutes.

So today I logged everything I did or thought about between 1000h and 1030h. Here is the list:

• Stop working on email: Do first things first.
• Idea: Must set up the audio recording system on the upper deck.
• Need to go to the bathroom.
• On the way downstairs, see the medical waiver form on the table, grab it, and return to stateroom to fill it out.
• Realize I'm thirsty and grab my water
• Idea: Calculate the diesel fuel consumption rate.
• Go downstairs to the toilet & return to my stateroom.
• Grab my drink again.
• Calculate the fuel consumption, note it.
• Abby says: "Everyone please remember to fill out the medical waiver this morning."
• Must refill the internal water tank; flip switch; set timer
• Where's Liam? Who's going to cook tonight?
• Return to my room to fill out the medical waiver
• Reach over to activate email, but decide not to read it now
• Start to fill the medical waiver, give it to Abby and tell her a story
• Brent says, "The webcam is set up..."
• I see Liam: "Are you going to cook dinner?"
• Reset the timer as the tank is not full yet (it's now 1020h)
• Drink
• Notice that the temperature is 45.7F
• Stop and think: Now what? Look at my list, it says, "Set up audio."
• Get the cables stashed in a duffle in my stateroom
• Notice email has arrived from my brother, don't read it now.
• Tell myself: Focus on the cables
• Remember to check the water level, decide to carry the timer with me
• Start plugging cables into the audio mixer
• Brent is nearby, fiddling with the webcam in the south window
• Check the water level again; turn it off
• Log the water level (tell Abby it took 15 minutes to fill 45 gallons)
• Brent says that the webcam is too bright, even with the sunglasses
• Return to the audio mixer; the speakers need a plug adapter
• Check the webcam image on the computer screen; joke with Brent that he's not done if he hasn't used any duct tape yet
• Notice Maarten in his stateroom to my left, hear his printer; wonder if I could connect my computer to it
• Notice Liam taking lamb out for dinner, wonder if freezer space has opened up.

I then focused on the audio work over the next 20 minutes with no further interruptions. After looking at my email, I spent a half hour getting the audio debugged. I then spent, with lunch in between, three hours mostly working on photographs.

So let's see what was happening. I refocussed my attention about 34 times during those 30 minutes, which is much more than every minute when you consider that it took at least three minutes to make a roundtrip visit to the toilet on the lower deck.

These 34 shifts of attention involved 11 different topics. With two exceptions (the temperature reading and the printer idea), I was juggling all of these in my head during the half hour. The main topic was setting up the audio, which I returned to six times. But interestingly, there were six topics that I returned to three times each: Email, drinking, webcam, water filling, medical form, dinner. I also turned to two topics twice: Fuel calculation and bathroom. So I'd be doing something, and in less than a minute of focused activity, remember one of the other nine things I needed to be doing. Less than one minute between interruptions!

Of those 34 shifts of attention, seven resulted in deciding not to change what I was doing, including email (don't do it!) and bathroom (not yet!). Six times my attention was captured by something outside myself, something I saw or heard: Abby or Brent saying or doing something, my hearing the printer, my seeing the temperature, seeing Liam. Nineteen times something came to mind that I felt to be of greater priority than what I was doing, so I changed my focus: bathroom (once), medical form (3), fuel calculation, drinking (3 times), setting up the audio (5), handling the water (3), dinner (3). So more than half the time, when I thought of something else, I changed what I was doing, about every 90 seconds. That's reactive behavior!

Curiously, with all of this flitting around, once I actually stopped and asked myself, "What should I be doing now?" I looked at my list of what I needed to do. This was a very different kind of shift, mediated by an "external representation" (writing on paper), and not just an idea in my head.

Okay, so what can we conclude? Is this good? Had I drunk too much coffee? I think the main point is that I needed to get multiple things done-the medical waiver, the water refilling, getting someone assigned to make dinner, and most importantly, setting up the audio system. I also needed to attend to biological realities, and to attend to my crew. I used several strategies to manage my time: 1) Allowing my attention to shift, so I could "time share," managing multiple tasks at once, 2) Deliberately deciding not to do certain things, and 3) Looking at a list of objectives.

You're probably worn out just reading this, and I guess I was, too. I spent most of the next three hours sitting in front of my computer. But the same shifting continued between emails and photos and looking at web sites. Only now, the action was mostly in my fingers.

Maybe what interests me most about this kind of analysis is what it reveals about how quickly and effortlessly we shift attention. Yet when you ask people to report on what they were doing, they are unaware of these patterns. For example, I noticed at Haughton (Devon Island, 1999) that people usually spent less than a minute in the work tent-for it was a storage tent as much as a place to work. And I found when we were driving the Humvee at Haughton (2003) that we stopped on average every 3.7 minutes to re-calibrate our direction, for two entire hours! In MDRS5, the crew said interruptions were the greatest problem in getting things done. To quantify this, and to understand better how cognition works and what the mind accomplishes, we need to make records as I did today. Knowing this, we can better understand how schedules affect a crew, why people can't predict how long tasks will take, and indeed, how to develop robots that can keep up with our strange way of working.