Explore before Exploit
The psychologist Alison Gropnik has summarized one behavioral difference between human babies and adults. Babies and young children explore their environments while adults exploit their environments:
“…you’ve really got these two different creatures. So you’ve got one creature that’s really designed to explore, to learn, to change. That’s the child form. And then you’ve got this other creature that’s really designed to exploit, as computer scientists say, to go out, find resources, make plans, make things happen...”
Gopnik notes the tension between exploring and exploiting.
“…if you’re thinking about intelligence, there’s a real genuine tradeoff between your ability to explore as many options as you can versus your ability to quickly, efficiently commit to a particular option and implement it. And it turns out that even if you just do the math, it’s really impossible to get a system that optimizes both of those things at the same time, that is exploring and exploiting simultaneously because they’re really deeply in tension with one another.”
“…the way that computer scientists have figured out to try to solve this problem very characteristically is give the system a chance to explore first, give it a chance to figure out all the information, and then once it’s got the information, it can go out and it can exploit later on. So, explore first and then exploit.”
(quotes from https://www.nytimes.com/2021/04/16/podcasts/ezra-klein-podcast-alison-gopnik-transcript.html accessed 21 May 2023).
Gopnik offers the image of an exploring mind as a lantern, the exploiting mind as a spotlight. The lantern provides a bit of light over a wide area, the spotlight lives up to its name and puts light on one spot and leaves the rest of the environment in shadows.
Over a lifetime, infants and children start with lantern minds; adults develop spotlight minds. It’s tough for adults to get back in touch with exploration.
Teaching and coaching Plan-Do-Study-Act
I sense the tension between exploration and exploitation as I coach individuals and teams to test ideas that might improve outcomes for patients with less stress for the care team.
Over the past 20 years, I’ve tried to teach people to test ideas purposefully, using the Plan-Do-Study-Act method embedded in the Model for Improvement. I advise on ways to plan the test: define the question(s) the test might answer, state one or more predictions, outline the steps to run the test, and so on. I ask people to try the change idea on a small scale.
Learning to use PDSA is not easy for most of the people I’ve tried to teach.
I wonder if I am missing something in my approach. While testing has the flavor of exploring, the testing approach I’ve promoted is already heavily shaped by exploitation. Could I reduce the learning burden by starting with more exploration?
Testing a change to my teaching approach
Next month, we’ll start planning a learning community to launch in the fall. We’re recruiting a new set of teams from federally qualified health centers. In the community, we’ll work together to improve the integration of oral health and medical care for patients with diabetes.
In the past three years, three cohorts of FQHC teams have identified a core set of changes that appear to be useful and effective in delivering integrated care.
One of my jobs is to teach and coach teams in the learning community teams. The teams should learn to use PDSA to test and incorporate the changes into regular practice. After initial practice with formal PDSA testing, in the past we then introduce an activity called ‘Go See’, to encourage team leaders and supervisors to observe how people are handling changes to their work.
My idea for the new community: have teams Go See before they try an explicit change using PDSA.
In other words, let’s first ask for a bit of exploration, without an agenda that is shaped strongly by an exploitation mind-set.
Follow-up to initial Exploration
After the initial exploration experience, we’ll encourage formal testing. Like in past years, I will ask teams to design a test cycle using the Plan-Do-Study-Act method. My guess is the initial exploration will make it easier for the teams to ask questions and to make predictions as they plan a sequence of tests. If the initial formal test increases the degree of belief that the change is useful, more test cycles can move the change into implementation.
A Sketch: Transition from Exploration to Exploitation
Here’s a sketch to show my idea. The grey wedge presents the proportion of exploration mind at any given point on the horizontal axis. The blue wedge is the proportion of exploitation mind at any given point on the horizontal axis. The black lines represent my sense of effective operating zones that balance exploration and exploitation.
Zone A represents the exploratory Go-See activity. Zone A is not 100% exploration because the Go-See activity is part of a project to learn specific changes that aim to improve the health of patients with diabetes.
Zone B represents my usual approach, inviting teams to start with a small formal test of a change and then increase the size and scope of their tests—see this post on PDSA “ramps” that shows a similar wedge growing to the right.
The schematic indicates that exploration should never disappear; even as a team implements a change, it pays to go exploring from time to time, in the spirit of ‘sand castle maintenance.’