How Many Things at Once

Writing down when, where and how you’ll do something is one of the most reliable tricks in behavioural science. Over five days it takes people from 40% success to 61%.

Give the same person six goals instead of one, and have them make the same plan for each. Success: 43% — below the people who made no plans at all.

The planning didn’t get diluted. It went negative.

This is about the question underneath that result: how many things you’re allowed to be working on. There are two piles of evidence on it, they point in opposite directions, and both of them are right.

The hour clock

The paper at the bottom of “don’t multitask” is Sophie Leroy’s, from 2009, usually about four retellings down. Two lab experiments: give someone a task, interrupt them or let them finish, hand them a second unrelated task, measure how they do.

Part of your attention stays behind on the first task and degrades your performance on the second. That’s attention residue.

The part that doesn’t survive the retelling is the interesting part. Finishing the first task was not sufficient. People completed the thing and still carried it. What predicted a clean switch was time pressure while working on it — pressure forced closure, and closure is what let them put it down.

The thing you carry into the next hour isn’t unfinished work. It’s unresolved work. Those aren’t the same, and you can finish something without resolving it.

Two lab experiments, student-shaped tasks, US. Nobody has run the version that matters: what the residue looks like when the first task is something you’ve been building for eight months.

The decade clock

The opposing pile gets quoted at anyone who says they’re focusing on one thing. Darwin and his barnacles. Einstein’s 1905. And unlike most of that genre, there is a study under it.

Root-Bernstein, Bernstein and Garnier, 1993, in Creativity Research Journal, working from a cohort followed since 1958: forty American research scientists, all men, self-selected, spread across physics, chemistry, biochemistry and biology. The researchers graded them on citation impact, classified every paper by topic, and counted how often each scientist moved between topics across their first hundred publications.

Group	Major topics	Topic switches
Long-term high impact (n=6)	5	43 (range 35–51)
Short-term high impact (n=6)	3	16 (range 10–22)
Low impact (n=17)	2	17 (range 7–27)

Hold it loosely. The long-run group is six scientists. The comparison group is six. The significance test is reported with a single degree of freedom. That is the evidentiary foundation of an idea you will hear in a hundred keynote talks.

Two things inside it are better than the headline.

First: frequency separates nothing. The low-impact scientists switched seventeen times. The short-term high-impact group switched sixteen. Same number. Switching a lot does not predict anything on its own.

Second: it isn’t juggling, it’s returning. Forty-three moves across five topics is not five things sampled and abandoned — you cannot return forty-three times to something you dropped. It describes a small number of long interests circled repeatedly over decades. The number that reads as restlessness is the opposite of it.

And one caveat that belongs on the record: the topic categories were assigned afterwards, by the researchers, sorting papers into groups they describe as resembling separate textbook chapters. That’s the only way to do the analysis, and they say plainly how they did it. It also means the ruler is held by someone with a view. Nudge the boundaries and the forty-three moves.

They were never arguing

The two piles don’t contradict each other. They’re measured on different clocks.

Leroy’s clock is the working hour — what happens in the minutes after your attention moves. On that clock, switching is expensive and reliably so.

The 1993 clock is the decade — how many intellectual territories a person occupied across twenty years of publishing. It says nothing whatsoever about what those scientists did between nine and eleven on a Tuesday. For all that dataset knows, every one of those forty-three moves took a summer.

So the advice gets mangled in both directions. Never switch is an hour-scale finding sold as a life plan. Keep many irons in the fire is a decade-scale observation sold as a daily schedule. People feel torn between them because they’ve been handed two true statements with the units filed off.

Which leaves the real question — one project or three? — completely unanswered. Nobody has run that trial.

The number was never the variable

The best thing available is hiding in the study at the top of this piece: Dalton and Spiller, 2012, Journal of Consumer Research. Three studies.

Study 1: 67 business-school staff and MBA students, five days, one goal or six. One goal with a plan, 61% against 40% without. Six goals with a plan, 43% against 48%. Study 2 repeats the shape with 216 undergraduates at one goal versus three.

Their explanation is not that you ran out of capacity. Planning six things forces you to confront how hard doing six things would be, and that perception undermines commitment to all of them. You aren’t tired. You’ve been shown the bill.

Then Study 3. Same six goals, 107 students, a photography assignment — except one group has the six goals framed for them as manageable. Nothing about the workload changes. The planning benefit comes back.

The number of goals was never the variable. The felt weight of them was. And they moved the felt weight with a sentence.

Which means the question that’s actually consistent with all three studies isn’t how many things do I have. None of them counts projects. Leroy counts switches inside an hour; the 1993 paper counts territories across twenty years; Dalton and Spiller count how heavy your own list looks when you stare at it.

The narrower question that survives is: how many of them are asking you for something today. That’s the number that sets the weight — and it’s largely independent of how many things you’ve got.

What this doesn’t license

It is not a technique. Deciding fewer things are urgent, feeling lighter, doing more — that is 107 undergraduates and an afternoon of photography. Whether you can do it to yourself, deliberately, about something you actually care about, sustained over a year, is untested.

All three studies are American. One cohort was recruited in 1958 and was entirely male, which is not a neutral fact about a study of how scientists work.

And none of it is permission to be scattered. Five topics, forty-three returns. If your five things have one visit each, you don’t have the thing that was measured — you have its opposite, and six dead men borrowed to justify it.

---

Sources: Leroy, “Why is it so hard to do my work?”, Organizational Behavior and Human Decision Processes 109(2), 2009 · Root-Bernstein, Bernstein & Garnier, “Identification of scientists making long-term, high-impact contributions,” Creativity Research Journal 6(4), 1993 · Dalton & Spiller, “Too Much of a Good Thing,” Journal of Consumer Research 39(3), 2012.