What Science Actually Says About Movement in Workshops (Including Three Things It Doesn't Say)
Two weeks ago I published the 10 movement patterns hiding in every workshop. Last week, the 3 remote patterns where people actually move. In both articles I dropped research numbers without slowing down to explain them. A few of you asked. Fair enough. This one is for you.
I read the studies so you don't have to. Here are the five findings solid enough to build on, and, because I like you, the three claims facilitators keep repeating that the research does NOT back up.
The five solid findings
1. Walking makes people generate more ideas. A lot more.
The study everybody half-quotes: Oppezzo and Schwartz, Stanford, 2014. Four experiments, 176 participants, published in the Journal of Experimental Psychology.
The findings that matter for us:
- 81% of participants generated more creative ideas while walking than while sitting.
- The effect held on a treadmill facing a blank wall. So it's the walking itself, the scenery is a bonus.
- The boost persisted after people sat back down. Walk first, harvest after: the design writes itself.
- The effect applies to divergent thinking. Generating options, brainstorming, exploring. On convergent tasks, single-correct-answer problems, walking helped little or nothing.
That last point is the one facilitators skip, and it changes everything about when you schedule a walk. Ideation phase: walk. Decision phase: sit.
2. Screens hurt divergence, not convergence.
Brucks and Levav's 2022 Nature paper gives the walking result an almost perfect mirror. In a laboratory study and a field experiment across five countries, pairs collaborating by videoconference generated fewer creative ideas than pairs meeting in person.
Eye-gaze and recall measures, plus language analysis, pointed to the mechanism: the screen captured people's visual attention, narrowing their cognitive focus and the associative thinking behind idea generation.
But when pairs had to select which idea to pursue, videoconferencing groups were not less effective. The researchers even found preliminary evidence that they might be better, although that smaller effect was uncertain.
Put this beside Oppezzo and Schwartz: walking helps divergent thinking but not convergent thinking; videoconferencing hurts divergence but not convergence. The studies did not test the combined sequence, but the practical design rule is irresistible: ideate away from the screen, decide on it.
3. Doing beats listening. It's been measured 225 times.
Freeman and colleagues, 2014, published in PNAS. A meta-analysis of 225 studies comparing traditional lectures with active participation in university courses.
Students in lecture-only formats were 1.5 times more likely to fail. Exam performance rose by about half a standard deviation with active formats. In plain words: sitting and listening is measurably the worst learning format we keep using.
Workshops are usually better than lectures already. But every time you catch yourself presenting slides at people for 20 minutes, this is the meta-analysis whispering that a gallery walk of those same slides would do more.
4. If they act it out, they'll remember it.
Psychologists call it the enactment effect: performing an action while learning beats hearing or reading the same content. A 2022 meta-analysis measured the effect size at g = 1.23. If you don't speak statistics, know this: in learning research, anything above 0.8 is considered large. 1.23 is enormous.
This is my favorite argument for the human spectrogram. When someone walks to a spot in the room and stands on their opinion, they enact it. Months later people forget what was said in a meeting, but they remember standing alone in the "strongly disagree" corner while everyone watched. That memory does work no slide ever will.
5. Proximity decides who talks to whom. Since 1950.
The oldest and maybe the most underused finding of the list. Festinger and colleagues studied a student housing complex at MIT: physical closeness crushed every other factor in predicting who became friends. Neighbors connected. People a few doors apart, much less. Different building, almost never.
Modern replications keep confirming it. Randomized classroom studies show sitting next to someone multiplies the odds of becoming friends by 3 to 5. A randomized experiment in Hungarian schools showed assigned adjacent seating raised mutual friendships from 15% to 22%, including across ethnic groups.
Now look at your workshop. People sat down next to their colleagues at 9am and your seating chart has been silently deciding who exchanges ideas ever since. Every movement pattern that reshuffles proximity, the musical jungle, the speed rotation, the table dispersal, is you taking that decision back.
The three things science does NOT say
This is the part that might sting a little.
1. "Standing makes you think better." No.
A randomized controlled trial had office workers alternate seated and standing work and measured attention, processing speed, and working memory. Result: no cognitive difference. None. And one study even found that standing desks degraded attention on demanding tasks, because keeping your balance quietly eats cognitive resources.
Standing meetings are still great for other reasons: they're shorter, and they cut sitting time, which your cardiovascular system appreciates. But claim they boost thinking and you've left the evidence behind.
2. "A quick energizer restores attention." Not the jumping part.
A clever Swiss study split kids into four groups: movement with high cognitive demands, movement with low cognitive demands, no movement with high cognitive demands, and neither. Then they measured attention.
The result surprised me: physical exertion alone did nothing for attention. Cognitive engagement was the active ingredient. The group that moved AND thought did great. The group that just moved? Same as the control.
So the "everybody stand up and shake it out" break might feel nice, but if you want attention back, the movement needs a brain attached. Position yourself on a scale. Find a partner matching a statement. Sort yourself into a corner. Movement plus a decision.
Every pattern in my first article passes this test. That was on purpose.
3. "Walking meetings are the future." The trials disagree.
Here's an honest one. A large workplace trial in Denmark introduced both standing meetings and walking meetings across multiple offices. Standing meetings were adopted, loved, and cut sitting time by about an hour per workday for regular users. Walking meetings? Almost nobody actually did them, and the few who tried found them impractical.
The lesson for facilitators: spontaneous walking fails, designed walking works. That's exactly why the walk & talk in a workshop needs a route, a prompt, a fixed duration, and a harvest step. Structure is what separates the Stanford result from the Danish one.
What I do with all this
Four design rules fall straight out of the evidence:
- Do divergent work walking and away from the screen; return to the screen to evaluate and choose.
- If they'll need to remember it, make them enact it.
- Reshuffle proximity early and often; the seating chart is not neutral.
- Every movement needs a thinking task attached. No brainless jumping jacks.
And one meta-rule: when a claim about movement sounds too good, check whether it's the walking study being stretched past what it measured. It usually is.
Next time someone tells you movement in workshops is just fluff, you have a stack of studies, meta-analyses, and randomized trials in your back pocket. Use them while standing on a chair if you like. It won't make you smarter, but it will make the point.