Developed in 2002 by Frank Sumption, the electronic device known as the “Ghost Box,” or “Spirit Box” or simply “Frank’s Box” has been used by many paranormal investigators. What Frank did was modify an AM radio so that it was constantly scanning through different stations. The idea was that a spirit who wishes to communicate could use the white noise and partial sounds of the scan to communicate in real time with investigators. That’s much better than having to wait to review your recordings to try to find quiet, whispery EVPs, right? So why don’t we use it?

The Practical and Scientific Reasons: To put it plainly, the Frank’s Box is too subjective to be considered a scientific tool for communicating with spirits. In Feb 2015, NPI (Nyack Paranormal Investigations) did an experiment on the validity and subjectivity of the Ghost Box when it comes to paranormal investigations. This experiment was a double blind experiment conducted with 98 participants. All participants were sent an MP3 of the same 1 minute recording from the Ghost Box. However, Group B (half of the volunteers) received a plain recording whereas the second half received a recording that someone had recorded and added to the beginning someone asking, “What is your occupation?” with no other manipulations to the original recording made. Group A received the recording without the question at the beginning.

The results showed that the use of a Ghost Box is inherently flawed and subjective, and that there is also a tendency for confirmation bias to skew the listeners toward inferring certain responses based on the question(s) asked. Another interesting conclusion they discovered is that many different people heard many different things when listening to the same section of audio. At a point that several participants agreed that a word was spoken, they heard many different and different-sounding words, including actor, blackburn, backwards, and whatever. If we can’t all hear the same thing, how can we be sure that it is the evidence that we seek?

Another psychological phenomenon that is at play here is called Pareidolia. Pareidolia is a type of apophenia, which is a more generalized term for seeing patterns in random data. Some common examples are seeing a likeness of Jesus in the clouds or an image of a man on the surface of the moon. Similar to confirmation bias, Pareidolia negates the validity of the Ghost Box because there is just too much subjectivity and margin for error created by the human brain when listening to one. Even though the radio is scanning channels quickly, there are still full words and even partial sentences that can be heard, which adds to the mind’s desire to create a pattern. When you are asking a certain question of the Ghost Box, your mind is far more likely to force a pattern which isn’t necessarily there into the response.