In the universe of Pandemonium, humans have been genetically modifying themselves (more specifically, their offspring) for many years, in order to give them a variety of more advantageous traits.
One of these traits is horns that grow on the head that contain radio organs, that allow the person possessing them to transmit and receive (and crudely direction-find) radio signals across a fairly broad section of the radio spectrum.
People with horns are able to use their horns to communicate with others via radio rather than via vocalisation... and unlike vocalisation, a person could hold as many as two or three simultaneous conversations, rather than just one.
Obviously, children aren't born with horns, they have to grow them later, and they continue to grow throughout a person's life, gradually becoming more capable, so people would have to vocalise to their children. Not everyone would have horns either, so people would have to vocalise if they didn't have them or the person with whom they are communicating didn't have them.
My question is that I'm looking for a reason why two or more adults, all with horns, would usually vocalise to each-other if they were close enough to do so, and only radio each-other at a distance, when out of line-of-sight, or when unable to vocalise for some reason.
EDIT
To clarify matters:
Not all humans have radio horns. Around 75% of transhumans have them, and around 75% of the population are transhuman.
The horns have multiple transmitter nodes (the number depending upon age and length) that each transmit variable-power FM signals across a broad spectrum of frequencies. This system is capable of three-dimensional beamforming, and effective main beam ranges would vary from as little as 10-20m to as much as 5-10km, though there would still be significant transmission sidelobes.
The horns also have multiple receiver nodes (again depending upon age and horn length) that allow for range and direction finding.
Persons possessing horns effectively 'vocalise silently' into them. The neural structures added by the horns' genes translates the imagined sounds into an appropriate RF signal, and the person can choose direction, channel and power.
Persons possessing horns 'hear' radio signals translated into equivalent audio signals, and can percieve direction to as little as 6° azimuth and elevation, increasing in precision with age from about 22° in early childhood.
Horn radio communication is language-dependent, though with a broader range of possible sounds than a human voice, since transhumans can transmit any sound they can imagine.
The nature of many transhuman traits (including radio horns) is such that they are transmitted to the next generation upon artificial chromosomes that will effectively self-duplicate if necessary after fertilisation so that they will always be diploid, even if a transhuman has a child with a non-transhuman. Incompatible artificial chromosomes can also edit themselves out, and some can be selectively transmitted.
So, because there is no reason why a transhuman cannot successfully breed with a non-transhuman and pass on their horn genes to their offspring, making mating dependent upon mutual radio capability would be counterproductive.
Because people with horns 'hear' all radio signals that their horns can receive, they will hear nearby conversations, much as if they were a normal human in a room full of talking people. Horn-mediated radio communications are directed to a recipient by choosing appropriate power levels and beamforming. Recipients can tune out extraneous noise to a degree, much as normal humans can hear one audio conversation in a room full of talking people, with the added advantage that RF comms allow different carrier frequencies to be selected.
This is in contrast to modern digital radio communications, where the recievers do not usually direction-find, and different streams of data are identified by the recipient address in the data packets.