Visual Reaction Time

Measuring VRT 
Summary 
       This experiment attempts to investigate the effects of omnidirectional, directional and distractor 
sounds have on human visual reaction time. This experiment is broken down into 2 parts: measuring a 
person's visual reaction time and then analyzing the data.
Part1: Measuring a person’s visual reaction time 
       The experiment here was to a test a person’s visual reaction time by having a white 20mm 
square appear on a black computer background at random times ranging roughly from 3-5 seconds. The 
square can appears randomly on the left side or right side of the screen. The left side is approximately 
25% of the screen width and the right side is 75% of the screen width.  When the square appears 4 
different sound conditions can happen.  The conditions are: no sound is played, the directional sound is 
in the same direction as the target, omnidirectional sound is played and lastly a distractor sound is 
played that randomly picks a sound side and is played at a random time that is irrelevant to the target 
display but must be played before the target is displayed.  The sound clip that will be played is a 1 kHz 
sinusoidal tone with duration of 200ms. The left sound side location is 1m away from the cube and 30 
degrees of the horizontal to the left side and the right sound side is located 1m away from the cube and 
30 degrees of the horizontal to the right side.  If the tester takes longer than one second to react than 
his reaction time for that square is assumed to be 1000 ms.  
       The software Unity was used to create an application able to test the experiment above.   
Unity’s audio spatializer was used in order to simulate the directional sound.  Headphones must be used 
in order to get the directional effect, Bose Around the Ear headphones were used during this 
experiment. The sound conditions test inputs were ordered in a randomly generated Latin square 
meaning each test set had 16 tests, 4 of each condition and that will all play at least once before they 
can play again. Start times for the targets were exponentially distributed time, given from this formula: 
T=LN(1-u)/(-alpha).  The values of u ranged from 0.5 to 0.7 and alpha values ranged from 0.2 to 0.33. 
The distractor start times are given by the target time minus the randomly generated exponentially 
distributed time. The Unity application logs the experiment for a trial to a csv file. They can be viewed in 
the Trials folder. The symbols used within the data: “R” is right side, “L” is left side, “N” is no side , “O” is 
omnidirectional. All time used within the data  is in milliseconds. The same monitor and mouse were 
used in A1. 
Part 2: analyzing the data
System latency calculated in the last experiment was about 69 milliseconds and it will be subtracted 
from the reaction time. The data is sorted by their sound condition, than averaged and placed within a 
bar graph.  The standard deviation is calculated and displayed on the graph as error bars.