Most people are feeling lonely these days. Our behavioral science experiment found an unexpected way of boosting connection

Irrational Labs
5 min readMar 24, 2021


COVID-19 has prompted record levels of loneliness. This behavioral science study found an unexpected way of boosting connection

Most people are feeling lonely these days. This behavioral science hack could help

As behavioral scientists, we wanted to see if we could help people feel less lonely. Our experiment had a surprising result

By Richard Mathera, Kristen Berman and Bhavya Mohan

Feeling lonely lately? You’re not the only one.

COVID-19 has provoked unprecedented feelings of isolation and loneliness. It has cut us off from loved ones as well as places that once brought us comfort and connection, like our favorite coffee shop or gym.

The truth is, the majority of people in the U.S. already felt lonely before the pandemic. But surveys indicate that since March, at least a third of Americans feel even lonelier than before.

This is not trivial. While we’re trained to think that feeling lonely isn’t really a “big deal,” the fact is that it’s quite literally a matter of life and death — loneliness and social isolation increase the risk of early mortality.

So what do we do about it?

In behavioral science, we look for ways to close the intention-action gap — the difference between our desire to do something and us actually doing it.

An important question therefore becomes: At a time when we’re all craving connection, why don’t we take more steps to change our situation?

In a purely rational world, everyone who wants or needs more connection would simply ask for it. If you’ve ever been in need of help or lonely — you know it’s not that easy. Asking for help and connection makes us vulnerable. It may be that we don’t like to expose ourselves to others as weak and needy, or that we don’t want to admit to ourselves that we need it.

So we conducted an experiment. Our basic question was, can we make it more likely for people to opt into “getting help” but do it in a way that doesn’t require people to actually “ask for help”?

The study

Our team tackled this through a service that both accepts requests for help and allows for volunteering: a helpline. Many organizations offering these types of services have seen jumps in demand since the start of the pandemic.

For our study, we recruited 500 American adults ranging in age from 18 to 76. We showed all participants an identical description of a helpline that would provide “the opportunity for non-judgmental support and connection.”

Then we asked people whether they’d like to participate, with one important difference: half of all participants were asked about calling the helpline, while the other half was asked about volunteering with it.

Why volunteering? Volunteering is sneaky (in a good way). The core benefit appears to be helping others. However, an entire body of research suggests that volunteering has positive effects on mental health, and that this might be driven by increased levels of social connectedness. It’s possible, then, that people can get the same (if not more) social connection by offering to help others than by accepting help from others.

In each condition we measured how likely a person would be to participate (to call or volunteer at a helpline), for how long they’d participate, and how much connection they predicted would occur. Most importantly, we gave people the chance to actually sign up by providing their email address.

The results

Immediately after the study, one participant wrote to us, saying, “Many years ago I was part of a team that voluntarily helped mothers suffering from postnatal depression, it was very rewarding … I would be happy to do something like this again.”

Her enthusiasm parallelled our quantitative findings.

We found that participants were significantly more likely to participate in the volunteer condition: Only 8% signed up to receive help, while 16% signed up to volunteer. People were twice as likely to engage as volunteers.

Participants also said they’d spend significantly more time volunteering (73 minutes on average) versus receiving help (22 minutes on average).¹ Those ready to volunteer were prepared to spend over 3x as long doing so.

Why is this?

One hypothesis is that asking for help may signal something socially undesirable — that we need help. It’s possible that this negative social component may impact the likelihood of asking for help at all.

That’s, in part, what we found. We asked participants to rate the level of community respect they thought they’d feel volunteering versus asking for help. People predicted they’d gain significantly more community respect as a volunteer, compared to asking for help.

We also asked participants in both conditions to predict how socially connected they thought they’d feel to the first person they spoke to on the helpline. People predicted that they’d feel far more socially connected as a volunteer, as opposed to a helpline caller.

We found that both these factors — predicted social connectedness, and perceived community respect — helped to explain our main finding, which was that people are more likely to opt into a program if they’re providing help versus asking for it.

If you want to help people feel less lonely, don’t lead with loneliness

Services that help people with loneliness exist (i.e. mental health apps), and they’re fairly linear: You need help? We help you.

But is that the best way to drive adoption? Likely not. In designing services to reduce loneliness, it’s possible we’re missing the biggest variable: how to get people to actually use them.

Our study suggests that it may be easier to get people to feel connected when they see themselves as providers of help rather than recipients. Plus, volunteering has other

benefits and may actually be more meaningful, leading to an experience that’s both richer and more deeply fulfilling.

We think we’ve found a clever way of driving adoption of a service everyone is so desperately in need of. It’s almost like a backdoor into connection — framing the whole thing as what you can give rather than what you can get.

So whether you’re designing an app to reduce loneliness; trying to get an older relative to feel more human connection; or want to strengthen your own sense of connectedness, you’re likely to be most successful if you find a way to have people see themselves as the source of help and compassion for others instead of the one in need of it.

[1]: Our results even held when we controlled for a number of factors — the age of our participant, their income level, their education level, their gender, how much they volunteer, how much they talk on the phone, and their current feelings of loneliness.



Irrational Labs

Irrational Labs is a product design company that creates behavioral change for good.