Sie sind auf Seite 1von 4

EEG Event-Related Spectral Signatures associated with Psi-conducive States

Thomas F. Collura, Ph.D., P.E


BrainMaster Technologies, Inc.
Oakwood Village, OH
Norman S. Don, Ph.D.
Kairos Foundation and the The University of Illinois Chicago
Wilmette, IL
Presented at the 35th Annual Meeting,
Association for Applied Psychophysiology and Biofeedback
Colorado Springs, CO
April 1-4, 2004
SHORT ABSTRACT
This report describes the results of an analysis of EEG recorded from a high-scoring
subject on a series of tasks associated with psi states. EEG was recorded during a forcedchoice, five-response task. Using offline analysis of the EEG, it was possible to identify
a spectral signature that was found (p < 0.001) when the task was performed at an abovechance level. The details of this signature will be shown, and implications for
neurofeedback training of associated states will be discussed.
Key words: EEG, Event-related potentials, Psi states
LONG ABSTRACT
This report describes the results of an analysis of EEG recorded from a high-scoring (S)
subject on a series of tasks associated with psi states. EEG was recorded during a forcedchoice, five-response task. Event-related EEG was recorded over an 8000 millisecond
epoch prior to each response, from 5 EEG channels, one EOG channel, and one EMG
channel. Using offline analysis of the EEG, it was possible to identify a spectral
signature that was found (p < 0.001) when the task was performed at an above-chance
level. Bonfonerri-corrected t tests were used to determine the frequency bands that were
significantly larger for correct versus incorrect responses. Interchannel correlations were
also computed, and found to higher (mean of 0.89) for correct responses than for
incorrect responses (mean of 0.02). The event-related spectral signature was found to
show significantly higher energy in motor strip, at the following frequencies: 13, 18, 19,
24, 32, 36, 37, 39, 43, 46, 48. All t tests were significant at the level p < 0.001 at these
frequencies. An engineering analysis of these frequencies and their associated
amplitudes suggests that when correct responses were forthcoming, the subject was
producing bursts of EEG with a center frequency near 40 Hz, and at a rate of 6-7 bursts
per second. This is consistent with the concept that a 40 Hz binding rhythm is being
produced, and that this rhythm occurs in bursts at a rate near the low alpha or high theta
range.

METHOD & RESULT


EEG was recorded during a forced-choice, five-response task. Event-related EEG was
recorded over an 8000 millisecond epoch prior to each response, from 5 EEG channels,
one EOG channel, and one EMG channel. Using offline analysis of the EEG, it was
possible to identify a spectral signature that was found (p < 0.001) when the task was
performed at an above-chance level. Bonfonerri-corrected t tests were used to determine
the frequency bands that were significantly larger for correct versus incorrect responses.
Interchannel correlations were also computed, and found to higher (mean of 0.89) for
correct responses than for incorrect responses (mean of 0.02).
SPECTRAL SIGNATURE
The event-related spectral signature was found to show significantly higher energy in
motor strip, at the following frequencies: 13, 18, 19, 24, 32, 36, 37, 39, 43, 46, 48. All t
tests were significant at the level p < 0.001 at these frequencies. An engineering analysis
of these frequencies and their associated amplitudes suggests that when correct responses
were forthcoming, the subject was producing bursts of EEG with a center frequency near
40 Hz, and at a rate of 6-7 bursts per second. This is consistent with the concept that a 40
Hz binding rhythm is being produced, and that this rhythm occurs in bursts at a rate
near the low alpha or high theta range.
SUMMARY
These findings suggest that there is a particular EEG event or process associated with
successful performance of a Psi task. This state is characterized by a particular type of 40
Hz rhythm, and has a unique spectral signaturesuggesting a rate of between 6 and 7 bursts
per second. This suggests the possibility of using neurofeedback to develop the ability to
recognize, and produce, these states.
IMPLICATIONS FOR NEUROFEEDBACK
Under the proposed model, it might be possible to detect and feed back EEG signals
associated with psi-conducive states by delivering positive (reward) feedback when the
EEG exhibits a pattern of harmonic activity, for example, the presence of 7, 14, 21, 28,
35, 42, Hz peaks in spectral power. This model provides a specific experimentally
derived rationale for using this type of EEG reinforcement, for the development of psirelated abilities.

ILLUSTRATIONS
Expected frequencies from 6-7 Hz Rate Model
1.2

Expectancy (normalized amplitude)

0.8

0.6

Series1

0.4

0.2

64

61

58

55

52

49

46

43

40

37

34

31

28

25

22

19

16

13

10

0
frequency (hz)

x
T values (x 1000) for correct trials
5000
4500
4000

3000
2500

Series1

2000
1500
1000
500

frequency (hz)

64

61

58

55

52

49

46

43

40

37

34

31

28

25

22

19

16

13

10

0
1

T value x 1000

3500

x
Significance (1/p) values for correct trials
1200

1000

significance (1/p)

800

600

Series1

400

200

63

61

59

57

55

53

51

49

47

45

43

41

39

37

35

33

31

29

27

25

23

21

19

17

15

13

11

0
frequency (hz)

REFERENCES
Collura, T.F. (1987) "A Transient-Event Model for EEG Power Spectra," Proc. 40th Ann.
Conf. on Engrg. in Med. & Biol., Niagara Falls, NY., Sept., 1987, 247.
Don, N.S., Warren, C.A., McDonough, B.E., and Collura, T.F. (1987) Event-Related
Brain Potentials and a Phenomenological Model of Psi-Conducive States, Research in
Parapsychology, Matuchen, NJ: Scarecrow Press.
Don, N.S., Warren, C.A., McDonough, B.E., and Collura, T.F. (1988) State Specificity
and Psi Testing, Proceedings of the Parapsychology Association, June, 1988

Das könnte Ihnen auch gefallen