This experiment will require audio software which is capable of generating two sine tones through separate tracks. There are plugins for various commercially available packages which can do this.
Create two mono tracks, each with a signal generator. Pan both tracks to the centre. Mute one of the tracks and set the sine tone signal generator frequency of the other track to 1200 Hz, at a just comfortable listening level. Then set the other generator to 2400Hz, and bring its level down to inaudible. Now unmute it and slowly bring the level of the 2400Hz tone up, noticing when you can hear it. Now mute both tracks and reset the 2400Hz tone to 600Hz, at an inaudible level and repeat the experiment. At what level does the 600Hz tone appear? Repeat this with various tones, and observe what happens. Repeat this with the 1200 signal generator set to a square wave. Set the other signal generator to a 2400Hz sine wave to a level where it can be heard through the square wave (probably just round 6dB lower than the square wave). Now switch the second tone back and forward between a sine and square wave. What happens?
This is a fairly simple way of investigating a complex acoustic phenomenon. Acousticians would take such experiments much further, but this is a useful exercise in its own right to demonstrate an aspect of masking.