What tel explained is correct. I wanted to add that we actually see pure light in a composite way too.
Note for example that the M receptor has a peak at around 550nm, and a symmetric response curve. So, if you shine a 540nm light, the response of the M cone is the same as the response at 560nm. So, how do we know that it's a cool, bluish green and not a warm, yellowish green? By the response of the other cones. The ganglion cells in the retina combine the response of several cones to represent the wavelength of the light perceived by the photoreceptors (as well as some other attributes of the image within their receptive field).
As for "tricking" the eye, we're doing it all the time. Higher levels of visual processing will interpret colors depending on everything else perceived in the picture, such as light intensity, texture, and even cultural knowledge about colors. A mostly unbiased representation of wavelengths is only present at very early stages of visual processing.