There were a total of 21 sittings within the study period. Recordings were reviewed for EVP and a tally was made of how many direct responses or irrelevant comments were isolated. From the results in Fig. 2, it appears that relatively few direct responses were captured in any one sitting but if considered overall, 11 percent of the total amounts of EVP clips were of interest. It should be noted that a number of these clips were grade C and therefore an unpractised ear may well arrive at a less impressive figure.
Fig. 2 Séance recordings during the study period, from August 2009 to March 2010; Columns A, B and C contain the amount of irrelevant comments for that sitting graded by their clarity (A – clear, B – ambiguous, C – very ambiguous). Column Tn is the total number of irrelevant comments in grades A - C.
As well as looking at the amount of direct responses, I was keen to establish if EVP could be captured on more than one device at a time. In the sittings where two recorders were used, there are examples of the same EVP present on both recordings. Whilst these clips are significant there are relatively few; further duplicated recordings are taking place to try to replicate the results. If duplicated voices on more than one device are consistent, it might suggest that EVP is a physical resonance within the local environment but this poses an obvious question; if that is correct, why can we not audibly hear voices at the time of recording?
I am unable to answer that question at the moment, and the illustration in Fig. 3 only makes it all the more confusing. The amplitude and waveform of EVP clip v10, “That bastard saw me” is similar to that of the group member. Whilst a difficult to hear clip might be ambiguous, this voice is obvious and hard to dismiss as anything other than a voice present in the room. Hopefully further recordings will provide more information on the mechanics of transmission and reception.
At the outset of the study, the amount of available hours for performing analysis, dictated how many recordings I could cover before the next sitting. At most this worked out to be two recorders, on two laptops per meeting. The difference in the ability of the recorders to capture EVP was obvious from very early on. Both Olympus VN2100 devices out performed the Sony and Olympus LS10. The Olympus LS10 captured the least amount of EVP; I would expect this because of the low residual noise on recordings. The Sony, whilst having a fair amount of residual noise is not easily controlled for playback and has a manually set volume control. This often resulted in recordings being uncomfortable to listen to on playback. The Olympus VN2100 is a relatively
modest digital recorder with a reasonable amount of residual noise even on the highest quality setting but it is easy to use, reliable and well controlled for playback and uploading data. By the 8th recording, it was obvious to me that the Olympus VN2100 was best suited to EVP work and whilst I was aware that duplicating recordings was important, the amount of time I had to complete recording analysis was limited and a decision was made to continue with only this recorder.
ll computers performed equally well. The four laptops produced very similar waveforms when playing back the same recording; I was confident enough not to duplicate recordings on the same laptop with two recorders. I now upload data and recordings to just the Toshiba laptop as I find the sound quality superior.
Fig.3 Audacity waveform of clip v10