Can the Mental Load Be Measured? What the Research Says

In short. The mental load resists classic measurement: it takes no visible time, it happens while you are doing something else, and everyone judges it differently. Yet over forty years, research has learned to describe it, break it into stages, and then measure it with questionnaires. Mental Loadless's Mental Load Index (MLI) builds on those findings and turns them into a measure of how the load is shared: published, quantified, checkable. This article covers what the research says, each MLI design choice that follows from it, a full calculation for an example household, and the limits we openly accept.

Sunday evening, Camille is thinking about the little one's vaccine booster. Sam, meanwhile, did the grocery shopping, cooked the dinners, dropped off the swim bag and took the little one to the doctor. On a to-do list, Sam did everything. Yet in Camille's head, it has been a long week.

Both stories are true. The problem is that our measuring tools only see one of them.

The problem: work that takes no visible time

For decades, housework was measured in minutes. Time-use surveys ask people to fill in a diary: at 6:10 p.m., I am cooking; at 6:40 p.m., I am bathing the kids. The method is precise for physical tasks. It is how we know that in France, in 2010, women did 71% of the housework and 65% of the parenting tasks [2].

It is, however, almost blind to the rest. Thinking about the vaccine while cooking shows up nowhere: the diary says "cooking." Mentally checking that the swim bag is ready during a meeting shows up nowhere either. This work happens in parallel with something else, in fragments of a few seconds, at any hour. It leaves no trace in a time diary.

The intuition is not new. The French expression charge mentale is generally credited to the sociologist Monique Haicault, who published a 1984 article in Sociologie du travail on handling working life and home life at the same time [1]. In 1993, an article in the Journal of Marriage and the Family separated, in its very title, two things surveys had lumped together: doing the tasks and managing the household [3]. The word was there, and so was the distinction. What was missing were the tools to measure it.

What research has managed to measure

Describe before you count: interviews

The breakthrough came from a qualitative method. In 2019, the sociologist Allison Daminger published a study in the American Sociological Review based on 70 in-depth interviews with members of 35 couples [4]. She proposed an operational definition of household cognitive labor, in four steps:

1. Anticipate a need before it shows up.
2. Identify the options for meeting it.
3. Decide between those options.
4. Monitor that the need has actually been met.

Execution, the physical and visible part, comes on top of that. Daminger describes this work as taxing, often invisible to both partners, and a source of conflict. In her sample, women do more of it, especially anticipating and monitoring. Deciding is shared more evenly [4]. That finding matters a lot for measurement: it says that not all stages are equal when it comes to inequality. (We wrote a whole article on Daminger's cognitive labor model.)

Capturing thought in the moment

Other researchers tried to catch this work as it happens. In 2014, Shira Offer used experience sampling: 402 mothers and 291 fathers received random signals during the day and noted, on the spot, what they were thinking about [5]. More than 27,000 signals in all. The finding is subtle: mothers and fathers think about family matters just as often, but those thoughts weigh on well-being only for mothers [5]. In other words, frequency is not the whole story: the same mental work does not cost everyone the same.

Defining exactly what is being measured

In 2022, Liz Dean, Brendan Churchill and Leah Ruppanner proposed a theoretical synthesis of the mental load [6]. They define it as cognitive labor plus emotional labor, with three properties: it is invisible, boundaryless (it spills over into work, leisure and sleep) and never-ending (it is never done). They also call for large public surveys to include it: no standard measure exists yet. That definition explains why measuring in minutes fails: you cannot time work that has no boundaries.

Measuring at scale: questionnaires

The next step was larger samples. That is what questionnaires are for, and they have multiplied.

• Lucia Ciciolla and Suniya Luthar surveyed 393 partnered American mothers on 13 household management tasks [7]. 88% say they are mostly the ones who organize the family's schedule; 70% describe themselves as "the captain of the ship." Feeling solely responsible for this management is associated with lower satisfaction, with life and with the partner, though the effects are modest [7].
• Jenny Wayne and colleagues built, across five studies, a validated 9-item scale of "invisible family load," with three dimensions: managerial, cognitive and emotional [8].
• Ashley Weeks and Leah Ruppanner put 21 cognitive tasks to 3,000 American parents [9]. The analysis reveals two distinct families: daily cognitive labor, centered on the family's well-being and mostly carried by mothers, and episodic labor (home maintenance, finances), more often carried by fathers.

In 2023, a systematic review of 31 studies took stock [10]. The phenomenon is described consistently, but methods remain heterogeneous: interviews, questionnaires, observation, time estimates. There is no single reference instrument.

The self-report trap

Almost all of these methods rely on what people report. And self-reports have well-documented biases that hit the mental load head-on.

The first is psychological. In 1979, Michael Ross and Fiore Sicoly described the egocentric bias, including among married couples: each person remembers their own contributions better than the other's, and so credits themselves with a larger share [11]. It is not bad faith; it is how memory works.

The second is measurable. When questionnaires are compared with time diaries, husbands and wives both overestimate their contribution [12]. In US data, questionnaire estimates run about 50% higher than diary estimates [13]. And around a birth, the gap changes the conclusion itself: according to diaries, women add more than two hours of work a day, versus 40 minutes for men; with questionnaires alone, that inequality would not have been visible [14].

The third is narrative. In 2020, Daminger showed that couples committed to equality describe how they divide things up as neutral, and put an unequal outcome down to circumstantial trade-offs [15]. You can believe you are at 50/50 without being there, in good faith.

That has a direct consequence for any measure: a single estimate, from a single person, is not enough. The work has to be broken down into observable pieces, and perspectives have to be compared. The MLI does both.

From finding to instrument: the MLI's design choices

The Mental Load Index does not invent a theory of the mental load. It turns these findings into a calculation. Each design choice answers a specific finding. Here they are, one by one.

Choice #1: count stages, not tasks

A to-do list credits a task to whoever executes it. The MLI splits each item into five stages, Daminger's four steps plus execution: anticipate, identify, decide, execute, monitor. Each stage can have a different owner.

What the research says: execution is only the visible part of the work [4][6]. A 2025 study of 322 mothers rated, separately for 30 tasks, who plans and who executes: the split is more unequal for planning than for execution [16]. Counting only who does a task means measuring the most balanced part.

Choice #2: different weights by stage

In version 1.0, the published method assigns these weights:

Anticipating weighs the most, executing the least.

What the research says: anticipating and monitoring are the two steps where Daminger finds the clearest gap between partners; deciding is the most evenly shared [4]. Execution is already measured everywhere else, by time-use surveys and to-do lists alike.

What we openly accept: the order of the weights follows from the research. The exact values (1.5, 1.25…) are a modeling choice, not statistically estimated coefficients. We say so because that is the difference between an honest method and one that pretends to know more than it does.

Choice #3: a joint decision weighs nothing

When a decision is marked as made together, it is credited to no one. A decision made by two is no one's load alone.

What the research says: deciding is the most evenly shared step [4]. Crediting a joint decision to one partner would skew the measure; crediting it to both would inflate it.

Choice #4: the occasional weighs more than the daily

An episodic item (renewing an insurance policy, planning a birthday, a vaccine booster) is multiplied by 1.3. A daily item keeps a factor of 1.0.

What the research says: Weeks and Ruppanner identify daily and episodic cognitive labor as two distinct dimensions [9]. What we infer from it: a daily item eventually becomes a routine. A rare item has to be pieced together again each time, and it sits in the back of your mind long before the deadline. The 1.3 factor is, again, a modeling choice. And it does not lean either way: in Weeks and Ruppanner's data, episodic labor is more often carried by fathers.

Choice #5: emotional weight counts

Each item has an emotional weight: light (×0.5), neutral (×1, the default) or heavy (×1.5). Thinking about a child's vaccine does not weigh the same as thinking about the cat's food.

What the research says: the mental load has an emotional dimension, not just a cognitive one [6][8]. The work of thinking about someone close to you, of looking after their well-being, cannot be reduced to logistics.

Choice #6: a share, not a grade

For each stage of each item, the contribution is calculated as:

stage weight × cadence × domain × emotional weight

The domain factor is 1.0 for all eight life domains in version 1.0. Each adult's MLI is then their share of the household total, out of 100. The MLIs in a household always add up to 100.

Why a share: research provides no absolute scale for the mental load [6][10]. A share does not compare your household to the neighbors'. It does not say whether your load is "big" or "small." It says how it is shared in your home, and that is the question a measure can honestly answer.

Choice #7: a separate index for anticipating and monitoring

Alongside the MLI, the equity index looks at only two stages: anticipate and monitor. Never execution. It weights each stage by its contribution, not by a simple count of items.

What the research says: these are the two steps where the gap between partners is clearest [4]. Isolating them makes it possible to measure the sentence "I do it all in my head, even when the other person does it."

One household, six items: the full calculation

Back to Camille and Sam. Six items took up their week. For each one, here is who held each stage.

A sample calculation: anticipating the vaccine booster weighs 1.5 (anticipate) × 1.3 (episodic) × 1.0 (domain) × 1.5 (heavy emotional weight) = 2.925. Executing it weighs 0.5 × 1.3 × 1.0 × 1.5 = 0.975.

Doing the same calculation for every stage gives each person's contribution, item by item:

The household total is 30.31. The result:

• Camille's MLI: 59.9. Sam's MLI: 40.1.
• Equity index (anticipate and monitor only): Camille 72.5%, Sam 27.5%.
• A to-do list, meanwhile, would have credited all six tasks to Sam: 100% to 0%.

Three readings of the same household, three different answers. None of them is wrong. The to-do list measures execution, and Sam does execute everything: that is real time and real energy. The MLI measures the whole cycle. The equity index isolates the two stages where gaps hide best.

Note also what the calculation does not say. Sam carries two items entirely, the dinners and the insurance, from anticipating to monitoring. The MLI does not single anyone out. It shows how the load is shared, and it shows where it is already balanced.

Comparing perspectives: three numbers instead of one

Knowing about the egocentric bias does not make it go away. So the MLI does not rely on a single estimate.

Once a month, each adult uses two sliders to estimate the share they think they carry and the share they think the other person carries. The app then sets these estimates against the measurement. Say Camille puts Camille's own share at 75% and Sam puts it at 50%: the screen then shows "own view 75% · partner's view 50% · measured 59.9%."

Why three numbers? Because the gap between perspectives is a finding in itself. A 2026 study surveyed both partners of 235 couples separately: the more they disagreed on whether their split was equal, the lower the quality of their relationship [17]. Putting these perceptions side by side turns a disagreement ("I do as much as you do") into something both partners can look at together, from the outside.

Measuring without judging

Measuring how a couple shares the load is a delicate thing. Three principles govern how the MLI is presented.

No grade, no winner. The app ranks no one and uses no competitive vocabulary. A load gets shared; it does not get won.

No imposed ideal. Equity is not measured against a mandatory 50/50. The couple sets its own target. A 70/30 that was chosen and is kept is a contract honored. A period of imbalance (a new baby, a big work project) can be declared, for two weeks to six months, and the app then stops flagging the gap.

No medical label. The MLI measures how the load is shared, not a state of health. It measures and organizes; if you wish, a professional interprets. For that, a monthly PDF export can be brought to a couples therapist or a family mediator (our page for practitioners explains what it contains).

What the MLI does not claim to be

A serious method also says what it does not do.

1. It is not a validated psychometric instrument. The MLI has not been through a validation study on a sample, unlike a scale such as the one by Wayne and colleagues [8]. It applies published findings; it does not replace them.
2. Its weights are a modeling choice. Their order follows the research, their exact values do not. That is why they are published, and why every measurement records the version of the method that computed it. If a weight changes, the version changes.
3. It measures a share, not a volume. Two households can both be at 60/40, one with three items in progress, the other with thirty. The MLI does not say which one is busier. Yet volume matters too: a study of 265 couples found that hours spent on fixed-schedule tasks (meals, laundry, everyday shopping) are associated with lower well-being, and that it is their absolute amount that matters, not each partner's share [18].
4. It is still a self-reported measure. It depends on what the household enters, and on how tasks are worded: we know that question wording shifts reported contributions [9]. So as not to ask for five settings per item, the app suggests default owners, which anyone can correct. An item never entered does not exist for the MLI.
5. It does not measure time. A task that takes long to do and a quick one weigh the same, at equal stage, cadence and emotion. Time-use surveys remain the right tool for that.
6. The research it builds on has its own limits. Samples are mostly North American, middle-class and heterosexual [7][10], and the measured effects are often modest [7]. These studies describe associations, not causes.

What is left to measure

Research on the mental load is moving fast, and many questions remain open. How does the split change after a birth, a separation, shared custody? Does handing over an entire domain hold up over time? Do households that set an explicit target get closer to it?

Households that wish to can contribute to an anonymous collective barometer: explicit consent, revocable at any time, and no statistic published below ten households per category. It is our way of giving a little back to the research that gave us so much.

Going further

• The full MLI method, with its weights and vocabulary.
• The 12-question mental load test, which applies the MLI weights to your answers, in two minutes, without sending anything.
• The definition of the mental load and the key statistics.

In the app, the MLI is calculated item by item and over time, from what your household actually does. The method can be read in the app before you have entered a single item.

Sources

1. Haicault, M. (1984). La gestion ordinaire de la vie en deux. Sociologie du travail, 26(3), 268-277. doi.org/10.3406/sotra.1984.2072
2. Champagne, C., Pailhé, A., & Solaz, A. (2015). Le temps domestique et parental des hommes et des femmes : quels facteurs d'évolutions en 25 ans ? Économie et Statistique, 478-479-480. insee.fr
3. Mederer, H. J. (1993). Division of Labor in Two-Earner Homes: Task Accomplishment versus Household Management as Critical Variables in Perceptions about Family Work. Journal of Marriage and the Family, 55(1), 133-145. doi.org/10.2307/352964
4. Daminger, A. (2019). The Cognitive Dimension of Household Labor. American Sociological Review, 84(4), 609-633. doi.org/10.1177/0003122419859007
5. Offer, S. (2014). The Costs of Thinking About Work and Family: Mental Labor, Work-Family Spillover, and Gender Inequality Among Parents in Dual-Earner Families. Sociological Forum, 29(4), 916-936. doi.org/10.1111/socf.12126
6. Dean, L., Churchill, B., & Ruppanner, L. (2022). The mental load: building a deeper theoretical understanding of how cognitive and emotional labor overload women and mothers. Community, Work & Family, 25(1), 13-29. doi.org/10.1080/13668803.2021.2002813
7. Ciciolla, L., & Luthar, S. S. (2019). Invisible Household Labor and Ramifications for Adjustment: Mothers as Captains of Households. Sex Roles, 81(7-8), 467-486. doi.org/10.1007/s11199-018-1001-x
8. Wayne, J. H., Mills, M. J., Wang, Y.-R., Matthews, R. A., & Whitman, M. V. (2023). Who's Remembering to Buy the Eggs? The Meaning, Measurement, and Implications of Invisible Family Load. Journal of Business and Psychology, 38(6), 1159-1184. doi.org/10.1007/s10869-023-09887-7
9. Weeks, A. C., & Ruppanner, L. (2025). A typology of US parents' mental loads: Core and episodic cognitive labor. Journal of Marriage and Family, 87(3), 966-989. doi.org/10.1111/jomf.13057
10. Reich-Stiebert, N., Froehlich, L., & Voltmer, J.-B. (2023). Gendered Mental Labor: A Systematic Literature Review on the Cognitive Dimension of Unpaid Work Within the Household and Childcare. Sex Roles, 88(11-12), 475-494. doi.org/10.1007/s11199-023-01362-0
11. Ross, M., & Sicoly, F. (1979). Egocentric biases in availability and attribution. Journal of Personality and Social Psychology, 37(3), 322-336. doi.org/10.1037/0022-3514.37.3.322
12. Press, J. E., & Townsley, E. (1998). Wives' and Husbands' Housework Reporting: Gender, Class, and Social Desirability. Gender & Society, 12(2), 188-218. doi.org/10.1177/089124398012002005
13. Bianchi, S. M., Milkie, M. A., Sayer, L. C., & Robinson, J. P. (2000). Is Anyone Doing the Housework? Trends in the Gender Division of Household Labor. Social Forces, 79(1), 191-228. doi.org/10.2307/2675569
14. Yavorsky, J. E., Kamp Dush, C. M., & Schoppe-Sullivan, S. J. (2015). The Production of Inequality: The Gender Division of Labor Across the Transition to Parenthood. Journal of Marriage and Family, 77(3), 662-679. doi.org/10.1111/jomf.12189
15. Daminger, A. (2020). De-gendered Processes, Gendered Outcomes: How Egalitarian Couples Make Sense of Non-egalitarian Household Practices. American Sociological Review, 85(5), 806-829. doi.org/10.1177/0003122420950208
16. Aviv, E. C., Waizman, Y. H., Kim, E., Liu, J., Rodsky, E., & Saxbe, D. (2025). Cognitive household labor: gender disparities and consequences for maternal mental health and wellbeing. Archives of Women's Mental Health, 28(1), 5-14. doi.org/10.1007/s00737-024-01490-w
17. Coundouris, S. P., & Henry, J. D. (2026). A dyadic examination of the division of household labor and relationship quality from a tri-load perspective. Scientific Reports. doi.org/10.1038/s41598-026-60727-z
18. Barnett, R. C., & Shen, Y.-C. (1997). Gender, High- and Low-Schedule-Control Housework Tasks, and Psychological Distress: A Study of Dual-Earner Couples. Journal of Family Issues, 18(4), 403-428. doi.org/10.1177/019251397018004003