The multitasking myth: why the men-and-women-at-work gap isn't what the lab tests suggested

The setup looked, at first glance, like a particularly aggressive open-plan office. Five computer stations, a chat window blinking in the corner, a video to be tracked, a map to be memorised, a financial decision queued for sign-off, and a continuous stream of calendar prompts. On 10 July 2026, researchers writing in Scientific Reports described what happened when 48 men and 48 women were asked to keep all five plates spinning for 22 minutes at a stretch. The answer, in one sentence, was the opposite of what the popular literature has been selling for two decades: when the conditions resembled real life, the cognitive gap between the sexes shrank to almost nothing.

The finding matters because the received wisdom — that women are better multitaskers, that the difference is hard-wired, that it explains the gender composition of certain front-line jobs — rests almost entirely on self-report. Ask people whether they are good at juggling several things at once, and a consistent pattern emerges: women rate themselves higher, men rate themselves lower, and the media treats that as a fact about cognition. The new study suggests it is, at best, a fact about confidence.

What the lab actually measured

The protocol, designed by a team at the University of Essex and reported by Medical Xpress on 10 July 2026, mimicked what the researchers called "everyday life multitask". Participants had to keep five parallel tasks alive simultaneously: a general-knowledge quiz, a card-sorting exercise, a target-tracking visual task, a simulated financial decision, and a continuous conversational exchange delivered through a chat interface. Performance on each task was scored separately, and an aggregate score captured overall throughput.

On the aggregate measure, men and women performed almost identically. The headline divergence surfaced only when the researchers looked at how the five tasks were weighted: men, on average, deprioritised the conversational task more than twice as often as women did. They dropped the social thread to keep the more "instrumental" tasks running. The result is not that women are better at juggling. It is that, when forced to choose, men and women cut from different parts of the cloth.

That is a more interesting finding than the one it displaces, and a more politically inconvenient one. It implies that the widely-cited "female multitasking advantage" is, in the conditions that actually resemble a working day, a near-statistical dead heat, with the residual difference concentrated in prioritisation rather than raw capacity.

The counter-narrative: the self-report problem

For two decades, the dominant public claim — that women outperform men on multitasking — has leaned on a slender evidence base: the 2012 Stoet, O'Connor, Conner, Laws "girls outperform boys" paper and its descendants, almost all of which used self-report questionnaires or, in the lab, asked participants to do two tasks at once rather than five.

The Essex design is, by design, hostile to that tradition. The conversation task is, in many ways, the most demanding of the five because it requires continuous comprehension and response; the visual tracking task, by contrast, can be performed in bursts. A subject who lets the conversation slide but keeps the tracker at 100% will, on a crude two-task benchmark, look like a high-performer. A subject who does the opposite will look like a low-performer. Until the conversation is scored as a first-class task, the sex difference is invisible — or, worse, runs the wrong way.

In other words, the apparent female edge in the older literature may be an artefact of which tasks the experimenters thought to score. The new finding does not overturn two decades of work so much as it suggests the older work measured the wrong thing.

What the result does, and does not, imply

The temptation, on reading a study like this, is to extract a clean policy message: stop hiring on the assumption that women will be the better jugglers, or stop telling men they are constitutionally worse at the job. The evidence here does not support either cleanup. It supports something narrower and more useful.

First, on average, men's and women's ability to coordinate multiple streams of work under realistic load is comparable. Variance within each sex is large; variance between the sexes is small. This is consistent with a broader body of cognitive-sex-difference research in which the largest gaps appear in what people choose to focus on, not in how well they execute.

Second, the study identifies a behavioural rather than a cognitive gap: men, in this sample, were more willing to drop the social thread. The researchers are careful not to over-interpret this. It may reflect task design (the conversation was a chat interface, not a face-to-face encounter). It may reflect differential motivation. It may reflect prior experience with similar work environments. Without further replication — and the authors explicitly call for it — the prioritisation finding should be treated as suggestive, not dispositive.

Third, the practical implication is methodological. Studies that rely on self-report of multitasking ability are measuring something that is, at best, a thin proxy for performance. Employers making selection decisions on the basis of those instruments are, on this evidence, deciding on the basis of a number that has weak external validity.

What the evidence still can't tell us

The 96-participant sample is small for a study making a claim that cuts against a popular narrative, and the authors flag that caveat themselves. The age range of the sample, the cultural setting (a UK university department), and the specific task mix all bound the generalisability of the result. A replication in a working-age cohort, with tasks that more closely resemble a real knowledge-work job — Slack threads, email triage, a video call — would test whether the prioritisation pattern survives the move from the lab bench to the desk.

There is also the awkward question of why men in this sample dropped the conversation task more often. The paper offers hypotheses but not a mechanism. The most parsimonious reading is that the conversational task was, for the male participants, the lowest-marginal-utility item in the bundle, and they were acting rationally in deprioritising it. A reading more consistent with the popular framing is that women have a stronger baseline preference for maintaining social exchange. The data, as reported, cannot separate the two.

For now, the cleanest summary is the one the researchers themselves offer: men and women perform comparably on a real-world-style multitasking battery, and the residual sex difference in the data is a difference in priorities, not in capacity. The idea that women are the better multitaskers is, at best, half-true; at worst, a story that self-report data have been telling us for twenty years because nobody bothered to build a test that looked like a working day.

The next move belongs to the replicators, and to the employers who have, until now, been willing to make selection decisions on the back of questionnaires that the field now has a defensible reason to distrust.

Desk note: Monexus treats the new multitasking study as a methodological correction rather than a metaphysical claim. The wire coverage — Medical Xpress, 10 July 2026 — is the primary source; popular summaries of the older "women multitask better" finding were not relied on for any factual claim in the article.