[...]
On further thought, and the dress imbroglio was a prime case, also manipulated by the media, it *was and is* blue (not gold, false pics were circulated to feed the hysteria) another factor at play is *subtractive* product. This happens with colour, and was/is a huge factor in rendering of colour in mediums: Additive or Subtractive colour. The same could occur with sound recognition, save that it would take some form of reference to 'add' or 'subtract' against.
Which brings us to the manipulation of bait and switch. Let's just dwell in the analog sphere for a moment. Stereo sound multiplexing. Long before digital broadcasting was thought of, let alone used commercially, a technique was used called "subsidiary broadcast channelling". It used a sub-carrier for Musak and other channels on top of regular radio channels. Rather than my explain this, read here: (not perfect, but good explanation)
https://en.wikipedia.org/wiki/Subsidiary_communications_authority
This was a form of 'multiplexing'...
Now fast forward to 'stereo' broadcasting. When listened to in mono....it could be used, by the use of a 'pilot tone' (that you wouldn't hear, it was either super or sub sonic) to switch between one channel or the other with a 'hidden switch', such that a broadcaster could substitute a word, phrase or dildo in your ear to fool you into thinking you're having great sex, an unlikely scenario for some, I admit....but none-the-less, the gullible masses would gobble it up, and claim that aliens had experimented on them since they felt so guilty having an unaccredited orgasm.
Now digital: (What's doubly interesting is that it is now claimed the voice was real, not synthesized, which has virtually zero affect on digital tricks available to manipulate the listener. The *medium* was/is digital.)
Now think about digital recording. How many tracks can you record on *one* digital channel? Approaching infinite depending on sample rate and the bandwidth of the medium, assuming the ability to code and decode of the equipment either end.
Accepting that, and the proclivity to fool ears and eyes with additive and subtractive sound (the latter, btw, actually being used for years to decode FM stereo multiplex broadcasts until the 'switching' method gained traction using the pilot sub-carrier times two, it was a crude form of digital discrimination but I digress, won't explain the math here), it is therefor incredibly easy for a technically adept person, probably a digital recording engineer, to encode an algorithm in digital stream to flip "Yammy" (it's not "Yanni") or "Laurel" either at a timed interval, or randomly to further the agenda of a wildly hysterical public acting like controlled puppets.
Which brings us to AggregateIQ, elections, and Rob Ford....