Focus Test

The everyday picture of focus is a beam you point at things. The measurement tells a different story: focus is mostly suppression. The distractors are not outside your attention waiting to be let in — they are already in, already generating a response, and the work is holding that response back.

That is exactly what this test isolates. When the flankers point the same way as the target, they help. When they disagree, they have already begun preparing the wrong keypress and you have to override it. The difference between those two conditions — your interference cost — is a direct measurement of how much that override costs you right now.

Which reframes the practical question. “I need to focus more” usually means “I need to suppress better,” and suppression is a resource that depletes with fatigue and stress in ways that trying harder does not fix.

The headline number is not your raw speed — it is the gap between congruent and incongruent trials.

Two people with identical average speeds can have very different interference costs, and that gap is the interesting part. Someone slow overall with a tiny interference cost has excellent filtering and a slow motor response. Someone fast overall with a large cost is quick when nothing competes and falls apart when something does — which describes a lot of players in a messy teamfight.

Because it is a difference score, it also cancels out most hardware latency: display and input lag apply equally to both trial types and subtract away. That makes interference cost unusually comparable across devices, unlike almost everything else on this site.

Selective attention shows one of the largest sleep-deprivation effects in the literature, and it degrades before things you would notice. Your raw speed can look nearly normal while your interference cost has doubled — meaning you feel fine, and you are specifically bad at exactly the situations with competing information in them.

The mechanism is that suppression is effortful and effort is what fatigue rations. Stress does the same thing through a different route: it narrows the pool of control available for overriding prepared responses. This is the measurable core of what players describe as everything happening too fast — nothing sped up, your filter got more expensive.

It is also why this sits in the daily check. A measurement that degrades before you notice is worth taking, precisely because self-assessment fails there. See tilt, explained.

Partly, and with a real caveat.

• Task performance improves with practice. Interference costs shrink as you become familiar with a specific task. Some of that is genuine control, some is the mapping becoming automatic.
• Transfer is the contested part. Whether training a flanker task makes you better at filtering distraction generally is one of the most disputed questions in cognitive training, and the honest answer is that far-transfer evidence is weak. We are not going to tell you otherwise.
• Removing distractors beats out-suppressing them. Unglamorous and highly effective. A second monitor playing video is a permanent interference cost you volunteered for.
• Sleep and stress management are the real levers. They do not raise your ceiling; they stop you playing far below it.

The useful framing is not training this number upward but watching it. A day when your interference cost has doubled is a day to expect worse decisions in chaos — which is information you can act on immediately, without training anything.

A fight is a flanker task with money on it. The target you need is surrounded by four things generating competing responses — particles, health bars, a minion wave, someone else’s ultimate — all of which are already pulling your click toward them. Landing on the right one is suppression, not speed.

This also explains a specific familiar failure: playing well in lane and badly in fights. Lane is a low-interference environment and fights are a high-interference one. A player with a large interference cost is genuinely two different players in those two situations, and no amount of mechanical practice addresses it. What does help is knowing which day you are having before you queue.