You have two different separate groups to plot. That could mean you have two bars for a bar chart or two lines for a graph of a function related to the two groups, or two different color of dots for a scatter plot of some kind.
Basically you will have two sets of data, one for your L group (low video games) and one for H group (high video games). Plot the two sets of data in a way that lets us see the information separately, and lets us compare the two groups.
So, one way is to have the X axis show which group the data point is from, and the Y axis show the reaction time data. This can be done by putting a dot at every Y value (or some other shape, a square or triangle). Since you only have one dimension of Y data you don't need to worry about graphing a line or a 2-dimensional scatter plot (well I'm kind of describing a 2 dimensional scatter plot already, where the X axis tells you which group it was).
In other words, horizontal distance separates the two groups, and vertical distance separates the reaction times and shows the quantities (how fast the reaction times were).
Then another way to depict the data points instead of using individual dots can be to use the statistical properties of the two groups of data to draw a box and whisker diagram or box plot diagram like it says here:
http://en.wikipedia.org/wiki/Box_plot
Another suggestion I thought of for this experiment is to ignore a middle group. You could have a "less than 1 hour" group and a "more than 3 hours" group and then toss out the ones who play 1-3 hours a week. That can end up giving a more definite result. If you want to test everybody, you could use 2 groups instead of 3.
Another aspect of experiments (probably more than you want to get into right now) is to try to account for or eliminate confounding factors in your data like whether boys play video games a whole lot more than girls, and whether video game players are on a sports team too.
You might find a statistical result that reaction time and video game playing are related, but then the issue of cause and effect might not be clear. To determine cause and effect you might have to get the video game players to stop playing video games for a month, and see if their reaction times drop! It gets expensive, which is why medical trials are expensive I think.