Date Night Research: Why Novelty Beats a Pleasant Evening

Key takeaways

Across three experiments, couples randomly assigned to seven minutes of a novel and arousing shared task showed greater increases in experienced relationship quality than those given a mundane one.
A no-activity control established the effect came from the task, not time together, and it appeared in ratings of videotaped discussions, not only self-report.
The mechanism named is boredom, not affection: reported shared “exciting” activities predicted satisfaction, and the association was mediated by relationship boredom.
Self-expanding activities were associated with higher sexual desire and greater satisfaction, effects not attributable solely to positive affect, time spent interacting, or closeness during the activity.
The comparison was novel-and-arousing against mundane, not good against bad, so the restaurant you both love is the control condition.

The standard prescription for a long relationship that has gone quiet is a nice evening. Book somewhere. Light a candle. Put the phones away. Make time for each other.

Researchers have run that comparison directly, against the alternative, in a laboratory. Pleasant lost.

Seven minutes, and a task nobody would call romantic

A 2000 paper reports a newspaper questionnaire, a door-to-door survey and three laboratory experiments. It is the experiments that matter, because they can do what a survey cannot.

Couples were brought in and randomly assigned to spend seven minutes on a shared task. Some got something novel and arousing. Some got something more mundane. A third group did no activity at all.

Relationship quality was measured before and after.

7 minutes

In all three experiments, the novel and arousing task produced greater increases in experienced relationship quality than the mundane one. The no-activity control established that the effect came from the task itself, not from having spent time together — and the same effect appeared in independent ratings of videotaped discussions before and after.

Two details do most of the work there. The videotape rating means this was not only couples reporting they felt better; observers watching them talk saw a difference. And the no-activity control rules out the explanation everyone reaches for first, which is that any time together would have done.

The correlational half of the paper points the same way, and names the mechanism it runs through. Reported shared “exciting” activities predicted relationship satisfaction, and the association was mediated by relationship boredom.

Why pleasant is a different category

The distinction the research keeps drawing is not between good evenings and bad ones. It is between two kinds of good.

The theory is self-expansion: people are motivated to broaden their sense of self, and the rush of a new relationship is partly the experience of rapidly taking on another person’s perspectives, interests and capabilities. That process is fast at the start and inevitably slows. What is left is not a worse relationship. It is a relationship that has stopped expanding.

A pleasant evening is enjoyable and it does not expand anything. A novel, slightly demanding, mildly arousing activity does, and that is the variable being manipulated when the laboratory hands a couple something awkward to do together for seven minutes.

Which reframes the entire complaint. “The spark has gone” is usually treated as a loss of feeling to be recovered by better atmosphere. On this account it is a loss of novelty, and atmosphere does not supply novelty.

What these studies can and cannot say

Seven minutes in a laboratory is a short-term manipulation, and an immediate lift in reported relationship quality is not the same as a durable change in a long relationship. The 1993 field study extended the idea over ten weeks, but the strongest evidence here remains experimental and brief.

The samples are also the usual ones: volunteer couples, mostly Western, recruited for research. And “novel and arousing” was operationalised by the experimenters, not chosen by the couples — which is a real gap between the finding and the advice, because in life you have to pick your own.

The 2019 work rules out the obvious alternatives

The most useful recent addition tested the objections rather than restating the theory.

Across three studies using dyadic, daily-experience, longitudinal and experimental methods, engaging in self-expanding activities with a partner was associated with higher sexual desire — and that desire, in turn, with greater relationship satisfaction, an increased likelihood of having sex, and more satisfying sex when they did.

Then the important part. The researchers showed the benefits were sustained over time, and that the effects could not be attributed solely to three things any sceptic would raise first:

Not just positive mood. Doing something enjoyable puts people in a good temper, and a good temper improves how you rate your relationship. Controlling for positive affect did not remove the effect.
Not just time together. The activities take time and couples who spend more time together do better. That is not what this is.
Not just closeness during the activity. Feeling close while doing something is the most obvious mechanism, and it does not account for the result on its own.

That is an unusual amount of ruling-out for this field, and it is why the finding survives the reasonable objection that all of this is just “have a nice time and you will feel better about each other.”

What this changes about what you plan

The practical difference is not effort or expense. Novel and arousing is often cheaper than pleasant.

Choose unfamiliar over enjoyable. The comparison in the experiments was novel-and-arousing against mundane, not good against bad. The restaurant you both love is the mundane condition.
Accept being slightly bad at it. Mild arousal and a bit of awkwardness are the active ingredients. An activity you are already competent at is a pleasant evening with extra steps.
Seven minutes is the demonstrated dose. The laboratory effect came from seven minutes. Whatever the right amount is for a real relationship, it is not “a weekend away we cannot arrange until spring.”
Treat boredom as information, not as a verdict. The mediator in the survey data was relationship boredom. It is naming a fixable state, not diagnosing the relationship.
Do not wait to feel like it. The design was random assignment. Nobody chose the novel task because they were in the mood for it, and it worked anyway.

None of that requires the relationship to be in difficulty, and none of it requires an occasion. The variable being manipulated in every one of these studies is unfamiliarity, which is available on a Tuesday.

Try this

This week, replace one thing you would both enjoy with one thing neither of you has done. Not better — unfamiliar. A route you have not walked, a dish neither of you can cook, something with instructions.

The test is not whether you had a lovely time. It is whether at some point you were both slightly out of your depth at the same moment.

That is the condition the experiments manipulated, and it is the one a nice dinner is specifically designed to avoid.

The parts of this that are genuinely hard

Novelty is not equally available. A couple with small children, shift work, illness or no money have far less access to the unfamiliar than the volunteer couples in these studies, and advice that assumes otherwise is advice for people whose lives already have room in them.

There is also a real difference between a relationship that has stopped expanding and one that is in trouble. Self-expansion research is about the ordinary decline in intensity after the early period. It is not a treatment for contempt, for a breach of trust, or for a partner who does not want to be there, and applying it to those is a way of avoiding the actual problem with an activity.

And the honest limit on the evidence is duration. The strongest results are short-term and experimental; the longer-term work is correlational or observational. What has been demonstrated cleanly is that seven minutes of novelty moves something a pleasant seven minutes does not. What happens over five years is a weaker inference.

The bottom line

The spark is not a feeling to be recreated with atmosphere. In the experiments that separated the two, a novel and arousing shared task raised relationship quality where a pleasant one did not — visible to outside observers, not explained by having spent the time together, and mediated by boredom rather than by affection.

The later work rules out the obvious alternatives: not mood, not hours logged, not closeness in the moment. Which leaves an unglamorous instruction to end a corpus on. Do something together that neither of you is any good at, and do it before you feel like it.

Sources (3)

Every factual claim in this article comes from one of these sources.