Week 7: Who Goes Next?

Why the way a queue is arranged can lower how fair it feels to wait in it — even when nobody was actually served out of turn.

Imagine two checkouts are open. One arrangement: a single line snakes toward both, and whoever reaches the front takes whichever counter frees up next. Another arrangement: two separate lines, one per checkout, and you pick one when you arrive. You’re in the second arrangement. The other line moves faster — not because anyone cut in, just because that checkout happens to be quicker this time. Someone who arrived after you gets served first. Nobody broke a rule. It still doesn’t feel that way from where you’re standing.

The idea#

In the study we’ll look at, both structures used first-come-first-served ordering within their queues. What differed was the structure — one shared line versus several separate ones — and structure changes something the rule alone doesn’t: how easy it is to actually check the rule is being followed. In a single shared line, everyone’s position relative to everyone else is visible the whole time. In separate lines, all you can see is your own line; whether the whole system is still running first-come-first-served is something you have to take on trust, or infer from a small and possibly misleading sample.

This week calls that gap legibility — this course’s own term, not one borrowed from a paper, for whether a person waiting can verify the rule from where they’re standing. It’s informed by an older observation from Larson (1987), who argued that first-come-first-served has long functioned as a kind of social-justice principle for how queues are designed and defended, and that people’s attitudes shift when the order they observe seems to violate that principle — whether or not it actually did. Larson’s paper is a framework piece, built from observation and argument rather than an experiment; treat it as the source of that idea, not as evidence for it.

What a controlled comparison found#

Rafaeli, Barron and Haber tested this in a controlled, computer-simulated laboratory queue: participants experienced an actual timed wait, played out on a screen, structured either as one shared queue or as several separate queues feeding the same set of service points, with arrival rate, service speed, and everything else held equal (Rafaeli, Barron & Haber, 2002). The single- versus multiple-queue structure was the only thing that varied.

Three measures matter here. Participants in the single-queue structure reported greater predictability than those in the multiple-queue structure; predictability rose over the course of the single-queue wait but stayed flat in the multiple-queue condition. Participants in the single-queue structure also reported higher arousal, while arousal declined over the wait in the multiple-queue condition. And participants in the multiple-queue structure reported lower perceived justice than participants in the single-queue structure.

That last finding is more specific than it first sounds, and the specific version is the one worth keeping. The researchers separated multiple-queue participants into those who had actually experienced a first-come-first-served violation (someone arriving later reached the front first) and those who had not. Even the multiple-queue participants who experienced no actual violation — who were treated exactly as first-come-first-served as anyone in the single queue — still reported lower perceived justice than single-queue participants. Queue structure alone lowered the reported sense of justice, with no rule ever broken.

What this doesn’t establish#

This was one lab simulation, not a real physical queue: participants watched their progress on a screen and could not leave their assigned queue once they’d entered it — which means the study says nothing about queues where switching lines is possible, a common feature of real supermarket or bank queues. “Perceived justice” here means exactly what participants reported feeling on a survey, not a verdict on whether the queue was actually fair. And “legibility” is this course’s own reading of why structure might produce that reported difference — Rafaeli et al. tested structure and measured its effect on reported perceived justice; they did not test, or claim to explain, legibility as the mechanism behind it.

Why this matters#

Week 6 put you in a room with other people, without yet asking about the order they were moving in. This week asks exactly that: not whether a space felt fair, but whether the system directing who goes next was legible to the person inside it. It extends the course thesis into the social: waiting is an experience, not merely a duration, and Rafaeli et al. found lower reported perceived justice in the multiple-queue structure, including among participants who had not experienced an actual first-come-first-served violation — evidence that this experience can be shaped by structure even when a rule was never broken.

Once we understand the rule that structures who goes next, what happens when different people are governed by different rules? That’s next week.

What you’re doing this week#

Find one real queue, work out what rule seems to govern it, and check that rule against what you actually see happen. The session walks through how.