title: "Why your reaction time ceiling is neurological, not practice"
slug: why-reaction-time-ceiling-is-neurological
date: 2026-04-18
category: Science
readTime: "5 min read"
description: "Raw reaction time has a hard ceiling set by nerve conduction, not training volume. Here is what actually moves, and what does not, when you drill reactions."
keywords: reaction time ceiling, simple vs choice reaction time, nerve conduction gaming, reaction time training, cognitive performance gaming

Why your reaction time ceiling is neurological, not practice

Most players who grind reaction trainers are chasing a number they cannot reach. Simple reaction time, the time from a stimulus appearing to your finger responding, has a floor that is set before you ever pick up a mouse. The floor is biological. Training does not lower it by much. What training does do is different, and worth understanding, because the wrong model leads to the wrong practice.

Here is the blunt version. Your simple reaction time ceiling is neurological. Your choice reaction time ceiling is cognitive, and that one is trainable. Most of the gap between an average player and a top-tier player lives in the second number, not the first.

The anatomy of a reaction

A reaction is not one event. It is a chain. Light hits the retina, which encodes a signal and sends it up the optic nerve. The signal passes through the lateral geniculate nucleus, enters primary visual cortex, and gets routed to higher processing areas for identification. A motor plan is assembled in premotor and motor cortex, sent down the corticospinal tract, and crosses a neuromuscular junction to actually move a finger. Each of those steps takes time, and most of the time is not spent thinking. It is spent in transit.

In humans, nerve conduction velocity in myelinated fibers tops out around 70 to 120 meters per second. That sounds fast until you remember the signal has to travel from your eyes to your visual cortex at the back of the skull, then forward to motor areas, then down to your finger. The round trip, including synaptic delays, takes roughly 120 to 150 milliseconds on a good day. That is the physical minimum. You cannot train nerve fibers to conduct faster than they are built to conduct.

Donders and the reaction chain

The reason we know this is trained methodology, not guesswork. In the 1860s, the Dutch physiologist Franciscus Donders developed a subtraction method that separated reaction time into its cognitive components. He showed that adding a decision step, what would later be called choice reaction time, added a consistent chunk of time on top of the simple reaction baseline. Every experimental finding in the 160 years since has reinforced the same picture. Simple reaction time is a mostly fixed biological floor. Choice reaction time sits above it, and the gap is where thinking happens.

This matters for training because the simple floor is the part players obsess over and the choice component is the part that actually decides gunfights.

What twin studies and baseline data say

Reaction time is one of the most heritable cognitive traits we can measure. Across multiple decades of twin studies, heritability estimates for simple reaction time land in the 0.4 to 0.6 range, meaning roughly half the variance between people is genetic. Practice tightens your distribution and reduces variance, but it does not typically move the fastest trial you can produce by more than 10 to 20 milliseconds.

If your honest, rested, caffeinated simple RT floor is 210 ms, a year of dedicated training might push it to 200 ms. It will almost never push it to 170 ms. The players at 170 ms were at 180 ms before they ever trained. They got the genetic version that came with thicker myelination, shorter axonal paths, and faster synaptic transmission. It is not fair. It is also not the relevant number.

What actually trains

The ceiling that matters in ranked play is choice reaction time under load. That is the time to resolve "which of several stimuli is this, and which response is correct." Choice reaction time scales with the number of alternatives in the logarithmic relationship described by Hick's Law. Going from one option to four can add 150 ms. That is an order of magnitude more time than anything you could save on the simple floor.

Choice RT is trainable because the cost lives in stimulus identification and response selection, not nerve conduction. Both are plastic. Professional players do not have faster neurons. They have fewer branches to evaluate. Through repetition, decision trees get compressed into pattern matches, and pattern matches run close to the simple RT floor. That is why a pro looks faster than a rookie with better reflexes. The pro has pruned the tree.

You can see the same effect in the Posner cueing paradigm. When a valid cue tells attention where a stimulus will appear, reaction times drop by 30 to 50 ms versus an uncued trial, even though the motor requirement is identical. Attention already pointed at the right spot is the closest thing to a free millisecond in cognitive science.

Practical implications

Three things follow from this.

First, stop measuring your ceiling with simple reaction time tests. That number is mostly a biological readout. It tells you if you are healthy, rested, and caffeinated. It does not tell you whether you will win the next duel.

Second, train choice reaction time in conditions that match the load of real play. Drills with multiple possible responses, cognitive noise, and clear outcome feedback move the needle. Drills that flash a single dot on a blank screen do not.

Third, stop comparing your best trial against someone else's best trial. The comparison that matters is your distribution, specifically the width of your distribution under pressure. If you have read our prior piece on 7 evidence-based ways to improve your reaction time, you have seen the same point from another angle. The floor is fixed. The variance is not.

What you can actually measure

The NeuroRank combine separates simple RT, choice RT, and go/no-go performance, and records the distribution under fatigue and distraction, not just your best trial. If you want to know whether your ceiling is a neurological floor you cannot move or a cognitive bottleneck you can train, that split is worth running.

Take the combine and see which half of your reaction profile is trainable.