Signal-transfer theories propose that some sort of physical carrier wave, analogous to how electromagnetic (EM) waves carry radio signals, transports psi information.  For many decades this was an appealing explanation for telepathy because we know that the brain generates EM fields, and we know that EM fields can carry information around the world at light speed.  In 1899, the physicist Sir J. J. Thomson proposed in an address to the British Association for the Advancement of Science that EM fields might be the physical carrier of telepathy.  The title of Upton Sinclair's book, Mental Radio, reflected the enthusiasm once held for signal-transfer models.

The most problematic issue for signal-transfer theories, assuming that the carrier is at least analogous to EM fields, is that for all known physical fields the strength of the field drops off quickly with increasing distance.  If psi were mediated through any form of ordinary physical field, we'd expect psi accuracy to drop off rapidly with increasing distance.  But psi experiments conducted under conditions of heavy EM shielding and at long distances do not show sharp declines in accuracy.  This is not to say that the question of distance has been conclusively settled; as we've seen, there are inklings that distance (or possibly knowledge of distance) may play a role in some circumstances.

There is an exception to the rule about distance drop-offs.  At extremely low frequencies (ELF, 0.3-1 kHz), EM field strength is sustained over long distances because it passes through barriers that would absorb or block higher frequencies.  In the 1960s the Russian physicist I. M. Kogan proposed that telepathy is carried by ELF waves.  This was a novel solution to the problem encountered with higher frequency EM, but it too contains a problem.  Like any method relying on EM, it cannot easily account for the apparent time independence of psi.  So-called advanced wave solutions for EM could, in principle, allow for a form of retroactive signaling.  But the time scales for such signaling would be limited to the speed of light, which means one could obtain a precognitive glimpse of only one nanosecond per foot as light travels.  That is far too restrictive to account for precognition effects observed in life and lab, where future information can apparently be sensed from milliseconds to months or more in advance.  It also doesn't easily account for retrocognition, the flip side of precognition, in which (hidden) information from the past can be perceived.

Signal-transfer theories have included proposals based on tachyons (hypothetical particles that travel faster than the speed of light), antimatter (particles that can be interpreted as traveling backward in time), neutrinos, and gravitons.  Unfortunately, all of these models suffer because the effects would be limited in space or time, and none of them account for clairvoyance.  In addition, signaling theories don't provide a plausible explanation for telepathy, specifically in how signals sent from one brain might be "decoded" by another brain.