I'm thinking about writing an essay on the comparison between the human nervous system (reaction time) and a neural network that does the same reaction time test. I am very new in this area, so I was wondering if I can build a neural network that can perform a test like this: https://humanbenchmark.com/tests/reactiontime

I just wanted to know how I should approach this problem, and what would be the best way to compare it to the human nervous system.

I have thought about maybe using an image classification neural network, and have it looks for different colors and such, but not too sure about its technical aspects as of yet. Any help is appreciated.


1 Answer 1


Basically, what you want to do is an anomaly detection at the fastest speed possible. You need to sample the image at two timesteps and if there is a difference you click. The smaller the difference between your timesteps the better it is.

But, you would not need a neural network for it. Since, it is a single colour that will change. A color is represented by a 3-tuple (RGB) so you just need to compare them and if they are different then you click. A single if condition will suffice for it.

  • $\begingroup$ Thank you. But would this still fit into the general bubble of AI, and if not, in which scenario (regarding reaction time) would I actually be able to implement a neural network. $\endgroup$ Feb 17, 2021 at 21:01
  • $\begingroup$ To fit your problem into the AI bubble, you need to look at change point detection. Detecting any kind of change in data, occurence of a person in CCTV video, that is something you would require neural network for. Your data should be complex enough to warrant the use of a neural network. $\endgroup$ Feb 18, 2021 at 8:53
  • $\begingroup$ How would I be able to test that against a human reaction time, as that is my primary goal. Compare the reaction times of neural networks and humans. Also, could I still implement a neural network but not necessarily needing it?... Thank you $\endgroup$ Feb 18, 2021 at 17:50
  • $\begingroup$ The reaction time for a human comprises of looking at the picture, processing the picture and clicking the button. Most of a human's time will go in clicking. While the neural network will be able to do the click very fast than a human. This is the going to be the biggest challenge. $\endgroup$ Feb 18, 2021 at 19:26
  • $\begingroup$ You can use ESM (Experience Sampling Method). You can create an app which will at different times in the day ask the user to do the reaction time task. You will then also the neural network to do the same at backend and log that data. You can then compare. This is a possible way to do it. But, the challenge I mentioned in above comment remains. $\endgroup$ Feb 18, 2021 at 19:28

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .