Present one image to your left eye and a different image to your right eye simultaneously. The brain doesn't blend them — it suppresses one while the other dominates, then switches. You cannot hold both in view at once. You cannot decide when to switch. You cannot stop it.
The Wilson model captures this: two neural populations inhibit each other, and each slowly fatigues under sustained activation. When the dominant population has fatigued enough, the suppressed one breaks through. The switch is driven by dynamics, not decision.
Levelt's revised second proposition: increasing one eye's input strength primarily shortens the OTHER eye's suppression — it speeds up the rate at which you return, not the rate at which you leave. Try moving one slider up and watching what happens to the competing color's return time.
This simulation cannot create actual binocular rivalry — that requires separate images presented to each eye (stereoscope, mirror stereoscope, or red-green anaglyph). What the Wilson model shows is the switching mechanism: mutual inhibition plus slow fatigue produces oscillatory alternation without any external pacemaker.
The suppressed image in real rivalry is not absent. It is still being processed: semantically (primed words appear faster), affectively (emotional and meaningful stimuli break through suppression sooner — the continuous flash suppression paradigm). The signal is present and active; only the access route is blocked. The image you are not seeing is doing something.