What is the worst NHRA reaction time?

How is reaction time tested?  (both in the past and present)?

• So on online tests, I routinely get reaction times of around 240-250 ms when undistracted and not on stimulants [1]. But that lags the supposed average of college students of 210 ms, if you use http://biae.clemson.edu/bpc/bp/Lab/110/reaction.htm#Type%20of%20Stimulus as a reference. All that past literature was done on RTs before the computer age, though. So I wonder if RTs are slower when one tests them on a computer? (maybe it could also be that all the reaction time tests use small stimuli - if the stimuli covered the whole screen my RTs would be shorter. I got 227 ms when one of the websites had a larger circle) I'm also curious about http://www.dailymail.co.uk/sciencetech/article-2323944/Were-Victorians-cleverer-Research-indicates-decline-brainpower-reflex-speed.html#ixzz2UouozYTR - I really do wonder if this (if true) is simply an artifact of measurement methods, which can matter A LOT for RT. I find it hard to believe that Charles Galton would have the apparatus to measure RTs with such precision. [1] http://www.humanbenchmark.com/tests/reactiontime/index.php

• Answer:

  The measurement device makes all the difference here. From the time that your brain registers the presence of a stimuli, a lot of stages need to occur before the reaction time (RT) is actually recorded. On your end, you need to program and execute the button press and the efferent signal needs to travel down the spinal cord to the appropriate hand. From there, any delay is going to be mechanical, and you'll get a lot of variability from device to device. How stiff and how tall is the button that needs to be pressed? How often does the OS poll the keyboard to determine whether a button press has occurred? If the RT test is online, what's the latency between your computer and their server? Moreover, what's the delay between the stimuli being sent, and the target refreshing on your monitor? Each of these variables can add a delay of 10ms or more to the RT measure, and modern computing equipment is actually a lot worse for these delays than computers 30-40 years ago. With so much noise, cognitive psychologists typically don't put much stock in interpreting raw RTs. Instead we usually compare RTs across two conditions (hard vs easy) and we assume the equipment delay will remain constant across both. If you really want a raw measure of response latency, you'll need to do the following: Get an old-school CRT monitor and some presentation software that can time-lock targets to the monitor's refresh-rate. Next you'll need to record electrical activity (EMG) directly from the muscles of the response hand. The lag between the monitor refresh and the onset of the EMG pulse is going to be you best bet, and even this won't be perfect. Long story short, there's really no such thing as raw "response time" and definitely no way to compare them in different labs using different sets of equipment.