Confession: I don’t know how to pick a ripe mango. And I can’t tell if Korean pears are sweeter than Chinese ones, or whether firmer plums are tastier than softer ones. 

But the fruit uncle at my neighbourhood wet market does. So every Sunday, he wraps up an already skinned and trimmed pineapple before I even ask, tells me what else is good, and I trust him.

Part of it is fatigue—it’s always my last stop and by then, I’m done thinking. I’ve already negotiated vegetables, chosen cuts of meat, and tried to make sense of fish I don’t quite recognise (snapper and seabass fillets look exactly the same to me). I don’t think I have the bandwidth left to also divine which mango is the sweetest.

Mostly though, I just don’t know enough. 

But Uncle does. 

And he’s been right often enough that I’ve stopped second-guessing him. After years of going to the wet market, you’d think I would have learned how to pick a ripe mango on my own. But I haven’t; I’ve learned something more useful: who to ask.

Not everyone has this relationship with their fruit seller, I know. This only works because I’ve been going back for years. If this were my first visit, I wouldn’t just accept whatever he suggested. But it isn’t. He’s been right about the pineapples, right about plums, right about those strawberries he warned me about that one time (I didn’t listen; he was right). The trust developed slowly until one day, it just became easier to ask, “Uncle, what’s sweet?” than to decide for myself.

In behavioural science, there’s a term for this: delegated cognition—handing a decision to someone else when they know more than you do. It’s less about laziness and more about recognising when your own signal is weak and someone else’s is stronger.

It sounds technical, but you know that feeling. You ask a friend to order at a restaurant you’ve never been to but they have. You let the hairdresser decide how much to cut (this one requires real faith). You trust the fruit uncle who has, week after week, been right.

And you’re happy to, because it works.

If you enjoy Jom’s work, do get a paid subscription today to support independent journalism in Singapore.

Which brings me to the supermarket where I do have more information. And more choice. 

My neighbourhood supermarket stocks eight varieties of apples. They are red, green, pink, and come from China, New Zealand, and the US. Sometimes, if I’m lucky, their packaging also tells me that they’re “crisp”, “tart”, or “Great for baking!” 

And yet, even if the labels say they’re sweet, I wouldn’t really know. 

I could guess. People tell me Envy apples are the sweet ones, and that Granny Smiths are not. I could read the packaging, but they don’t tell me if the New Zealand apples are crunchier than the ones from Washington. Does shinier mean better? Or just waxier?

Eventually, I just pick whichever looks prettiest, or is on sale, or vaguely resembles something I might have bought before. And so I do leave with apples, but without the confidence that I chose correctly.

Classical economics says this shouldn’t happen. More choice and more information are supposed to improve decisions, and confidence in them. But this assumes the information actually helps you decide. Labels like “Great for baking!” don’t tell me if an apple is sweet. They’re merely marketing copy designed to make the apples appear appealing, not actual guidance designed to help me choose.

And even if the information were genuinely useful, psychologists have found that abundance can backfire. This is the paradox of choice: the more options you have, the harder it becomes to decide. When you don’t know how to evaluate what you’re looking at, eight varieties of apples feels overwhelming, not empowering.

To be clear, I’m at a supermarket more often than I’m at the wet market. During Covid, when I wanted to minimise my time in the supermarket, I’d plan my route based on its layout. I’d start off: fruit, vegetables, meat, fish, seasonings, and then the long stretch along the wall—dairy, frozen foods, and maybe an ice-cream treat—before the final lap: rice, oil, cashier. I choreographed my grocery runs like a minor Olympic event—never won anything, but every trip, I’d shave minutes off my personal best…

That predictability is incredibly comforting. When I know what I want, the supermarket is unbeatable.

The problem shows up when I don’t.

The supermarket gives me more choice and more information. But I’m doing all the evaluation work myself—reading, comparing, second-guessing. And I’m less certain than I would be if my fruit uncle had simply said, “This week the Fuji apples are sweet.”

Which is where online grocery stores come in, trying to give me what the supermarket doesn’t: not just information, but guidance. Labels and reviews, yes, and also algorithms that learn what I like and guide me toward what I might want. It’s almost like my fruit uncle, but on my phone.

I use an online grocery store all the time. I get paper towels, rice, bottled sauces, canned goods—standardised things that don’t change. The app works brilliantly for that. 

But I don’t buy fresh food from them. Not anymore, anyways. I tried once—ordered chicken breasts that looked fine in the photo and had decent reviews. But they arrived…not fine. I complained, got my S$6.95 back, and felt vindicated for about 30 seconds.

Then I realised the refund didn’t actually fix anything. It solved the transaction, sure. But there was no one to tell me which brand of chicken to try next, or whether I should have stuck to chicken wings instead. So I stopped trying.

The online shopping algorithm can learn that I buy jasmine rice every month and soy sauce every few weeks. And I appreciate that. But it can’t look at a cut of meat and tell me if it’s fresh. 

Another thing—the online grocery store doesn’t need me to come back next week. Neither does the supermarket, for that matter. I’m one customer among thousands and if I stop showing up, the system carries on.

My fruit uncle’s world is smaller than that though. Interactions don’t reset. They build on each other.

Robert Axelrod, a political scientist at the University of Michigan who I once heard speak, called this “the shadow of the future”. In studying how cooperation emerges—even between people who might otherwise compete—he found something strikingly simple: cooperation doesn’t require kindness. It only requires people to expect they’ll meet again. When you know you’ll see someone next week, treating them well today isn’t just altruism, it’s also strategy.

Uncle knows I’ll be back next Sunday, which means it matters that I leave happy this Sunday. If the plums disappoint, I’ll hesitate the next time he recommends them. If I hesitate often enough, I might stop coming altogether. His reliability isn’t just about being nice—it’s also self-interest shaped by the fact that we see each other week after week.

The online grocery store can’t replicate that. Neither can the supermarket. They can offer convenience, scale, much more choice. But they can’t give me someone who knows that next week I’ll be standing in front of them again, deciding whether to trust them one more time.

I’m still at the supermarket most weeks. I still order rice online. But when I make it to the wet market on a Sunday morning, I’m reminded that I don’t always have to figure things out myself.

Like this morning.

I was at the fruit stall, looking forward to my weekly pineapple.

“Today, pineapple not sweet,” Uncle told me, almost apologetically.

I really wanted pineapple. And briefly contemplated walking to the supermarket for it.

But I didn’t. So till next Sunday then, Uncle.

Serene Koh is a behavioural scientist and director of the Behavioural Insights Team in Singapore. She also teaches behavioural science at the National University of Singapore.

Letters in response to this piece can be sent to sudhir@jom.media. All will be considered for publication on our “Letters to the editor” page.

If you enjoy Jom’s work, do get a paid subscription today to support independent journalism in Singapore.