University of Nevada
Reno
Experiential States Associated With the Production of a
Summed-Amplitude, Alpha Biofeedback Signal
A dissertation submitted in partial fulfillment of the
requirements for the degree of Doctor of Philosophy
by
Wayne Edward Carr
Duane varble, Dissertation Advisor
April 1992
The dissertation of Wayne Carr is approved:
_______________________________________________________
Dissertation Adviser
_______________________________________________________
Department Chair
_______________________________________________________
Dean, Graduate School
University of Nevada
Reno
April 1992
acknowledgements
I would like to thank the newest members of my committee, Dr. Robert Solso and Dr. bill Wallace, for their help and support so late in the process. I especially feel a debt of gratitude to Dr. Duane Varble for his understanding and efforts. I also wish to thank dr. bill danton and dr. john peacock for "hanging in there" over the long haul.
outside of my committee, I greatly appreciated the advisement of dr. Paul tyson, dr. ronald pekala, dr. willard day, dr. tom harrington and dr. les fehmi. I especially wish to thank Dr. joe kamiya and dr. young koh for their wisdom, time and patience.
i want also to acknowledge the all-around support of sandy goldstein and linda himmel who showed a great deal of faith in me when I was finishing. lastly, I thank all others who read my prospectus, as well as, the research participants, who gave a strong commitment and a hefty amount of time.
This dissertation is dedicated to my mother, barbara carr, and brother, robert carr, whose emotional and financial support made this long-awaited accomplishment possible.
abstract
twenty-six males, 13 "right-movers" and 13 "left-movers," were given biofeedback training using an auditory signal from a five-channel EEG developed by Dr. lester fehmi. the signal reflects the algebraic sum of the alpha amplitudes from five scalp sites and is a measure both of alpha amplitude and of phase-synchrony.
Each of ten weekly biofeedback sessions, was composed of three baselines and two training conditions. In both enhancement and control conditions, the subject attempted to increase the volume and occurrence of a feedback tone. In the enhancement condition, the tone increased in volume as alpha amplitude and phase-synchrony increased. In the stabilization (control) condition, the tone increased in volume as alpha amplitude approached the average amplitude of the subject's daily baseline.
following the training, subjects took questionnaires assessing their experiences of each condition. Several "covariate" dimensions (e.g. perceived success and task difficulty) were also assessed. It was predicted that subjects would have significantly different experiential scores between conditions. For the enhancement condition an increased sense of "merging" and of attentional "widening", rated on several subscales, was also predicted.
Only "left-movers" had significantly different experiential scores between conditions and had significantly different alpha amplitude scores between conditions. no significant between-condition differences, however, were found for any specific subscale measure of attentional "widening" or "merging". Most measures of "attentional
widening" correlated negatively with amplitude scores in the enhancement condition, which was opposite to the direction predicted. the reported experience of feeling "separate" from ones experience correlated negatively with enhancement scores, lending support to the "merging" hypothesis. Significant correlations between alpha amplitudes and experiential scores occurred almost exclusively for the enhancement condition. This indicates a general connection between alpha enhancement and changes in the measures of experience.
table of contents
page
signature page ............................................... i
acknowledgements ............................................. ii
abstract......................................................iii
table of contents ............................................ v
list of tables ...............................................vii
introduction ................................................. 1
advantages of present study............................ 2
hypotheses ............................................. 4
review of the literature ..................................... 7
early research ......................................... 7
factors that may effect the level of alpha activity..... 12
experiential measures................................... 23
synchronization ....................................... 29
fehmi's research........................................ 32
lateral eye movements .................................. 34
multichannel eeg ....................................... 36
fehmi's theory of attention ............................ 40
method ....................................................... 43
subject selection ...................................... 43
apparatus .............................................. 45
design ................................................. 47
step by step procedure ................................. 53
results ...................................................... 64
scoring ................................................. 65
statistical tests ...................................... 68
discussion ................................................... 80
conclusion.............................................. 97
bibliography .................................................105
appendices ...................................................126
appendix a .............................................126
Tables ...........................................126
appendix b .............................................157
hypotheses summary ...............................157
definitions ......................................158
general categories of experience .................158
Categories of experience .........................159
general take-home instructional material
for subjects .....................................163
eye movement preference questionnaire ............184
daily pre-baseline questionnaire ..................190
Instructions for questionnaire E-1 ................196
questionnaire "E-1" ..............................199
PCI questionnaire ................................215
FOD questionnaire (covariates) ...................223
questionnaire "T" (covariates) ...................226
list of tables
TABLE PAGE
1. names of baseline scores, baseline maximum scores 126
and training scores
2. enhancement scores used in comparisons127
3. abbreviated names of experiential dimension scores 128
for "A" and "B" conditions from questionnaire E-1
4. "standard" covariates for "A" and "B" conditions129
5. four types of correlations130
6. differences in EEG scores among training scores and 131
baselines
7. differences in experiential dimension scores between 132
conditions
8. interactions of eye-movement preference by conditions133
on experiential scores
9. differences in experiential scores between conditions134
for "left-movers"
10. differences between "right-movers" and "left-movers" on135
experiential scores
11. covariate differences between conditions136
12. differences in enhancement scores between conditions137
13. differences in enhancement scores between conditions138
for "right-movers" and "left-movers"
14. summary of correlations of training scores and 139
experiential scores controlling for baselines
15. summary of positive and negative contrasting 140
correlations
16. possible pattern of significant correlations for141
different groups
17. Significant covariates for between-condition142
differences in experiential scores
18. correlation of enhancement scores with covariates143
(not controlled for baseline)
19. summary of correlations of enhancement scores 144
and experiential scores for left- movers, controlling
for baselines and standard covariates
20. summary of correlations of enhancement scores 145
and experiential scores for right- movers, controlling
for baselines and standard covariates
21. correlations between EEG baselines and experiential146
dimensions for E-1 and PCI
22. correlation of enhancement scores with covariates147
(controlling for baselines)
23. correlation of enhancement scores with experiential148
dimensions from questionnaire E-1 (not controlled
for baselines)
24. correlation of baseline and maximum scores with 149
experiential scores
25. correlation between enhancement scores and 150
experiential dimensions with baselines as covariate
26. correlation between enhancement scores and 151
experiential dimensions for "left-movers", with
baselines as covariates
27. correlation between enhancement scores and 152
experiential dimensions for "right-movers", with
baselines as covariates
28. correlation between enhancement scores and 153
experiential dimensions with baselines and standard
covariates used as covariate
29. correlation of E-1 and PCI score154
30. correlation PCI with enhancement scores155
31. correlation of constant factors with enhancement156
scores (controlling for baselines)
Introduction
The purpose of this study was to determine if specific experiential states are associated with the learned production of a multi-channel, phase-sensitive, electroencephalographic (EEG) alpha biofeedback signal (Fehmi 1974).
Most previous research attempting to connect learned increases of alpha activity with subjective experience has been relatively limited or inadequate in the following ways:
(1) Only certain areas of the brain have been studied. Biofeedback signals have almost exclusively been from occipital sites. No published studies have been done attempting to connect experiential states with biofeedback signals derived from large areas of the brain.
(2) Usually only a few general dimensions of subjective experience have been focused on. Questionnaires used to assess experiences have been limited to a few questions on general aspects such as relaxation, pleasantness, visual attention, anxiety, alertness, body sensation, and flow of thought (Plotkin, 1978, 1977; Travis, Kondo & Knott, 1975). Many studies have narrowly focused on whether or not there is a roughly predefined "alpha experience" associated with learned increased occipital alpha activity (De Good, Elkin, Lessin & Valle; Glaros, 1977; Plotkin 1979; Travis, Kondo & Knott, 1975).
This predefined "alpha experience" has been summarized by Walsh (1974) as: "awake but not alert, reduced body awareness, relaxation, floating, lightness, neutral to pleasant emotions, content, peaceful, slightly high, thought undirected or free flowing, thoughtless, eyes defocused, visual inattention, vision or eyes blurred, crossed or unfocused." Later research however, has suggested that there is no one predefined alpha experience but rather a wide range of possible
experiences (Tyson, 1979). (3) The research has not simultaneously controlled for, or taken into account, three very important factors that influence subjective experience: expectations, sensory deprivation with sustained internal focus of attention, and perceived success. Studies by Plotkin (Plotkin, 1978, 1979, 1980; Plotkin, Mazer & Loewy; Plotkin & Rice 1981) have shown that any study validly claiming evidence for true alpha-dependent experiential states must take the above three factors into account.
(4) Limited attention has been given to important factors that may influence alpha activity or its measurement. These factors are: (a) experimenter personability (e.g. "friendly" eye-contact and tone of voice), also called the "person factor" (Taub & School 1978), (b) the subject's experience of the qualities of the biofeedback signal (Tyson, 1982), (c) the subjective "control strategy" the subject uses to enhance or suppress alpha activity, (d) the training times used (Hardt, 1975a, 1975b), (e) the baseline procedures used (Alpha "enhancement scores" are usually determined by comparing increased or decreased alpha activity to a baseline).
Manipulation of the above factors in ways that increase the likelihood of above-baseline alpha activity is important for the following reasons. First, it is not yet clearly established if subjects can even enhance their alpha activity above eyes-closed baseline (Plotkin, 1979). Second, alpha-dependent experiential states may be more likely when alpha activity is maintained as high as possible (Hardt, 1975a, 1975b). (5) Only a limited number of EEG characteristics have been studied. Integrated amplitude (the area under the alpha wave amplitude curve over a given time) or percent-time (the amount of alpha activity above an amplitude threshold) have usually been
the only EEG characteristics studied relating to subjective experience. There are at least fifteen other EEG characteristics, including intrahemispheric and interhemispheric phase-synchronization, which have been studied very little in regard to experiential states.
Advantages of present study
The present research expands on the previous research in the following ways:
(1) Training involved right and left occipital, parietal, temporal and frontal lobes instead of just occipital lobes.
(2) Questionnaires used included scales covering dimensions of subjective experience different from, and more extensive than, most previously used.
(3) The previously mentioned important factors of expectation, sensory deprivation, and perceived success were controlled for or assessed.
(4) Factors that may influence the amount of alpha activity and the scoring of alpha activity were given careful attention in the following ways:
(a) The experimenter attempted to optimally use the "person factor".
(b) An assessment of the subject's emotional experience of the biofeedback signal was taken on the first sessiom.
(c) total training time for alpha enhancement was five hours, (in most studies, total training time has been less than an hour).
(d) Daily eyes-closed optimal and average baselines was taken according to a procedure similar to that used by Plotkin (1978), the most out-spoken critic of baseline procedures in alpha studies.
(5) The present study used a unique EEG amplitude measure developed by Fehmi (1975) that simultaneously reflects intrahemispheric and interhemispheric alpha phase synchrony. Alpha amplitudes from five channels are algebraically summed to produce one feedback signal. All channels add amplitudes to the degree they are in phase and cancel to the degree they are out of phase, giving a "global" sensitivity to alpha phase synchrony. Similar phase-sensitive biofeedback devices, designed by Fehmi are currently used in a variety of clinical settings (Fehmi, 1987) (Fritz, 1985) (valdez, 1985) to treat stress-related symptoms and to break patient habits of rigidly and obsessively attending to certain feelings, thoughts or behaviors. Rigid attending is theoretically replaced with more "flexible" attending. Fehmi labels this flexible attending "Open Focus". Most research up till now, using similar phase-sensitive equipment, has not been well-controlled or specifically directed at subjective experience. It was thus the experimenter's intention to do a well-controlled study of biofeedback training using Fehmi's five-channel phase-sensitive signal.
Hypotheses
(A) Primary Question:
The primary objective was to test whether a subject trained to increase Fehmi's five-channel summed amplitude signal (higher synchronous alpha) will have significantly different scores on measures of subjective experience than when trained to stabilize this signal within a certain amplitude range close to his average eye-closed baseline (lower synchronous alpha). According to Fehmi subjects should report different experiences. Subjects were trained to stabilize their alpha amplitude level roughly around their daily baseline in order to equate control and experimental (enhancement) tasks as much as possible.
(B) Secondary Questions:
Based on his clinical observations and a few limited studies, Fehmi (1974, 1978, 1979, 1980) suggested two hypotheses regarding biofeedback training with his summed amplitude alpha signal. These two hypotheses were tested in the study.
Subjects trained to increase the feedback (enhancement) will be more likely to have the following basic changes in subjective experience than in the control contingency:
I. Merging With One's Experience:
A. An increased subjective sense of "merging" with
being "absorbed" into, "dissolved" into the
content of their experience (thoughts, feelings,
visual images, sights, sounds, etc.).
Example: Feeling like one "becomes" the
feedback tone.
B. A decreased subjective sense of being distant,
being separate from, thinking "about" as an
"object", the content of their experience (thoughts,
feelings, visual images, sights, sounds, etc.).
Example: Feeling less like the feedback tone
is an objective emotion-free sound that is
totally "out there".
II. Widening of Attentional Focus:
A. Widening of "within-sense" attentional focus:
Increase in experience from the same general
category of experience. [See appendix, "General
Categories of Experience" list, page 163 (not the "categories
of experience" list)] Some examples
of general categories of experience are:
sounds, tactile sensations, feelings or
emotions, and visual images.
Example: Being aware of greater numbers of
sensations on the skin simultaneously.
B. Widening of "between-sense" attentional focus:
Increase in experiences from different
categories of experience.
Example: Being aware of: sight, sound, touch,
and feeling simultaneously
C. Increased "integrated" awareness"
1. Increased experiencing of separate
experiences as increasingly larger, more
inclusive or "global". In other words,
people will report experiences that are
more like multisensory "images" or
perceptions, "wholes", "gestalts",
"fields", "totalities" or "unified
constellations" of different sensations.
Example: Experiencing a moment, person,
idea, event, one's surroundings or "every
thing" as one clear reality or image
involving sight, sound, feeling, body
sensation, etc.
2. Increase tendency to have a relaxed focus
on a larger "whole" of experience as
opposed to an isolated point in experience.
3. Decreased experiencing of the content of
one's experience as "fragmented".
4. Increased awareness of, and allowing of,
"opposite states" e.g., anger and love
or relaxation and tension.
D. An overall widening of attentional focus
1. Increased tendency of attention to be
"inclusive" as opposed to "exclusive".
2. Decreased narrow concentration or "narrow
focus" of attention.
Review of the Literature.
Early Research
Early studies by Kamiya (1968, 1969, 1974; Nowlis & Kamiya 1970), Brown (1970, 1971), and others (Peper 1970; Travis, Kondo & Knott 1975) demonstrated that subjects who learned to increase percent-time of alpha were more likely to describe their experiences during training as: calm, peaceful, relaxed, floating, letting go, free flowing thought, low in anxiety and pleasant. These studies, however, did not control for the effects of subject expectation.
Later Plotkin (1976a, 1976b, 1978, 1979, 1980; Plotkin & Rice 1981) and others (Glaros, 1977; Stroebel & Glueck, 1973; Valle & Levine, 1975), who included control groups, found that the subjective experiences connected with increased alpha activity could be partially, if not totally, attributed to subject expectations. Subjects who were led to expect certain experiences by instructions, implicit suggestion, or prior knowledge were more likely to have "alpha" related experiences during training than control subjects with different or with no
expectations.
Despite the strong expectation effects found, an actual connection between "alpha experiences" and alpha activity has received some support from other studies which have also manipulated subject expectations. De Good, Elkin, Lessin and Valle (1977) found that subjects with prior positive expectations for alpha enhancement, who were led to believe they were suppressing alpha when they were actually enhancing alpha, "nevertheless provided self-reports consistent with their relatively high alpha density." In another study, Walsh (1975), found that alpha experiences were reported only when both high alpha activity and appropriate expectations were present. At least 26 studies provide some support that learned alpha activity may be connected with a range of subjective experiences. At least 23 studies have found no such support or have been able to attribute associated subjective experiences to expectation related phenomena.
Plotkin's Criticisms
In a comprehensive critique of the connection between "alpha experience" and alpha activity, Plotkin (1979) concludes that no such direct connection exists, and that increased alpha activity means only that one is awake (or not drowsy), "not looking" (often attending to "nonsensory" mental activities to avoid "looking"), and little else. He outlines alternative explanations of the "alpha experience" which can generally be classified as: (1) expectation or placebo effects, (2) sensory deprivation and "introspective sensitization" effects, and (3) perceived success.
Subjects' expectations shaped by past knowledge and present instructions are further strengthened in the biofeedback situation itself because the subject can attribute his experience to both the use
of an "objective" feedback device and to his own active efforts (Plotkin, 1980). For similar reasons, Stroebel & Glueck (1973) have called biofeedback the "ultimate placebo".
Sensory deprivation settings and EEG biofeedback research settings are similar according to Plotkin. EEG biofeedback settings, often require a subject to sit, eyes-closed, relatively motionless in a comfortable chair, in dark, sound-attenuated surroundings, and attend to a monotonous feedback stimuli for a period of time. Many subjective effects may thus be attributable to sensory deprivation (Plotkin, 1979).
In "introspective sensitization" (Hunt & Chefurka, 1976) the mere act of paying attention to one's internal experience and "raw" sensations for a period of time, has been shown to alter one's subjective experiences. Sensory deprivation and introspective sensitization effects, however can be controlled for by placing the control subjects in nearly identical settings with similar tasks.
Perceived success has also been shown to be another powerful factor. Plotkin (1979), manipulated subjects' perceived success during training by adding to their actual feedback score (read to them every two minutes during training) an additional two percent every two minutes. These subjects were more likely to report the subjective experiences described in their pretraining instructions. At the same time, no significant relationship between alpha enhancement scores and experiential reports was found.
In a later study on alpha activity and anxiety reduction, Plotkin found similar results: "Anxiety reductions were highly correlated with the trainee's ratings of perceived success at the feedback task but were unrelated to either direction or magnitude of changes in that alpha activity."
In addition, Plotkin's subjects, with one exception (1980), have consistently failed to raise alpha activity above baseline, as Hardt and Kamiya claim their subjects have done (1978, 1970, 1976). Plotkin doubts their claims and asserts "there is absolutely no published evidence that the increases in alpha activity that are frequently seen during alpha training have ever constituted an unequivocal case of actual alpha enhancement above optimal prefeedback baseline levels." At least thirteen published studies however, have reported above eyes-closed baseline alpha enhancement.
Kamiya and Hardts' Defense
Representing the other side of the controversy, Kamiya and Hardt have discounted Plotkin's claim that there is no real alpha-dependent experiential state, because plotkin's subjects have not achieved above eyes-closed baseline alpha activity. Kamiya and Hardt claim their subjects have achieved above baseline alpha activity. Kamiya states simply, "Plotkin cannot talk about what he cannot achieve." (personal communication, April, 1983).
Hardt (1975a, 1975b) claims that for the alpha experience to be strong and be more independent of expectations, a subject must achieve above eyes-closed baseline levels and that this usually does not occur until at least two hours of training have occurred. Hardt (1975a) states, "After 6 or 10 hours, or whenever we get doubling and quadrupling of resting alpha levels, an alpha 'high' may be as inevitable as an LSD high after 1000 micrograms. Setting and expectations may influence the quality of the 'high' but not the fact of it."
Kamiya and Hardt however, have presented only a limited amount of well documented evidence to support their claims that their subjects do produce above-baseline alpha. In one study showing a reduced anxiety correlation with increased alpha production, Kamiya and Hardt (1978) illustrated how two of their subjects averaged from 16% to 21% above average baseline. These increases were maintained for more than two hours. The study showed that high anxiety subjects reduced their anxiety in proportion to alpha activity increase. Unfortunately, they did not control for the effects of perceived success, which Plotkin's replicative study demonstrated can have similar equally strong effects on anxiety reduction. As previously mentioned, Plotkin found no such connection between alpha activity and anxiety reduction.
At this point it can't said whether the contradictory results of Plotkin and Kamiya and Hardt are due to the lack of a control for perceived success on Kamiya and Hardts' part, or due to Plotkin's failure to get his subjects to achieve above baseline alpha activity as Hardt claims is necessary.
Hume (1978), when referring to the Kamiya - Plotkin debate, concludes, "perhaps someone should do an experiment where the expectancy of the experimenter is manipulated, as well as the expectancy of the subject, since it is becoming apparent that you either believe in feedback-mediated change or you do not, irrespective of the evidence. The only valid position at present is to sit on the fence, but many people find this uncomfortable." This debate, abandoned by most researchers, who have been polarized to one position or the other, has yet to be resolved. The present study did not attempt to resolve this specific debate since the traditional occipital EEG measure was not used, however, the use of a different EEG measure along with the avoidance of many of the past methodological shortcomings appeared provide a logical avenue of exploration.
Possible Explanations
There are several possible explanations for the lack of strong consistent evidence for alpha-dependent experiential states: (1) The level of alpha activity (integrated amplitude or percent-time) achieved by experimental groups in different experiments has varied and may be too low in most cases. (2) There are differences and limitations of the experiential measures used. (3) Control groups have not been maximally equated, in terms of tasks, to the experimental group. (4) Only limited EEG characteristics have been studied. The last three points will be covered in more detail later.
Factors That May Affect The Level of Alpha Activity
The level of alpha activity is crucial if it is true that alpha-dependent experiential states only occur when alpha activity is above an ambient level. These differences in measured levels of alpha activity may be affected by: (1) the use of different baseline procedures, (2) the use of different EEG sites, (3) the use of different equipment, (4) differing experimenter personability ("person factor"), (5) type of subjects selected, (6) the possible use of less effective subjective control strategies used to increase alpha activity, (7) the use of different types of biofeedback signals, (8) the use of different training times.
(1) Baseline Procedures Hardt, Kamiya, Tyson, and others claim finding alpha activity above eyes-closed baseline. The majority of other investigators have not found sustained, above eyes-closed baseline, alpha activity (Lynch & Paskewitz, 1971; Plotkin, 1976).
The same baseline procedure, however, was not used in these studies. The procedures used to determine baseline are important, since enhancement scores are determined relative to baseline. Some researchers however, have not clearly stated their baseline procedures, making comparison between studies difficult.
Baseline periods have varied between 1.5 minutes and 48 minutes, however, most are three minutes (Ancoli & Kamiya, 1978). For example, Hardt and Kamiya (1978) took an average daily eyes-closed baseline averaged over an eight minute period, and a total average baseline averaged across four days.
Tyson (1982) used two daily averaged 1.5 minute baselines and two daily averaged three minute baselines with feedback tone on. Ancoli and Green (1977) took one 48 minute baseline. Valle and Levine (1975) used the last two minutes of an eight minute eyes-closed rest period as baseline.
Plotkin (1978) uses daily optimal eyes-closed baselines instead of average baselines. In this case, twice the highest one-minute of integrated alpha activity over a ten-minute baseline period is used as the optimal baseline. Enhancement scores are then determined relative to baseline.
Plotkin determines enhancement scores (E) by dividing the integrated amplitude for a two-minute epoch (X) by the optimal baseline (Bo) so that E = X/Bo. From this information, the average enhancement scores for each day can be determined.
Although Plotkin's optimal baseline is certainly the most demanding, it also appears arbitrary. Declaring that subsequent enhancement scores below the "optimal" baseline (the highest one minute score taken over the ten minute baseline period) are not true enhancement risks being too demanding. In attempting to avoid getting an overly low reference baseline, Plotkin risks getting one too high.
(2) Electrode Sites Plotkin has unfortunately used occipital sites. The highest enhancement scores however, have been found by Hardt and Kamiya (1978) to be from a central site (C3). In their study of anxiety, they state," ...the relation between alpha enhancement and reduced state anxiety was seen centrally at C3, while alpha suppression at occipital sites (Oz, O1) was related to anxiety increases." The original studies done by Fehmi, which led to his development of his five-channel phase-sensitive EEG, were often done from a midline frontal site.
Tyson (1982) states: "electrode placement appears to be the simplest explanation for alpha enhancement since the cited studies used occipital locations and the evidence (in his study) for alpha enhancement is over the parietal region." The occipital lobes may thus be among the least desirable areas for study.
(3) Equipment Factors The use of different EEGs with different filters and sensitivities makes comparison difficult. Many studies have used equipment manufactured by Grass or Beckman. Plotkin used Biofeedback System equipment. Both Hardt (personal communication, April, 1984) and Fehmi (personal communication, March, 1984) have criticized the sensitivity of other researchers' equipment (particularly Plotkin).
(4) Person Factors Taub and School (1978), while training subjects to raise hand temperature, found that "impersonally" treated subjects were only able to raise mean hand temperature 1.3 degrees Fahrenheit. The experimenter was impersonal in that he used last names only, discouraged extraneous conversation, and avoided eye contact. Personally treated subjects, on the other hand, were able to raise hand temperature 4.2 degrees.
In a previous study, an impersonal experimenter could only get 2 of 22 subjects to raise skin temperature! Taub and School conclude, "It seems highly probable that the person factor is equally critical for the success of other types of biofeedback training." Thus, if experimenter negative expectations are conveyed to subjects, through subtle impersonal behaviors, then such subtle impersonal treatment of both experimental and control subjects can affect outcomes, even in double-blind research.
On the other hand, Kamiya has been noted for his warm "bedside manner" in the treatment of his subjects (Hardt, 1976). Plotkin claims also to have treated his subjects with positive regard in a pleasant tone, and states he desired for them to succeed.
In terms of subject treatment, Plotkin however, differs from Hardt, Kamiya and Tyson, in one overt way: During training, every two minutes, he verbally reported to his subjects over an intercom their average amplitude score. If the achievement of above baseline alpha is an "intimate" operation sensitive to the "person factor," then listening to the experimenter's voice every two minutes reading his score may be more likely to be perceived as a "watching," evaluation, or intrusion. A subject who feels constantly "watched" or evaluated or is affected by subtle cues in the experimenter's tone of voice may have more difficulty attending to the task and hence not be able to raise his alpha activity.
(5) Control Strategies Control strategies are the emotional, cognitive, and behavioral operations used by the subject to attempt to increase or decrease the feedback tone. An example of a behavioral control strategy to increase alpha activity might be oculomotor relaxation (relaxation of eye muscles used in "looking" behavior). A more cognitive control strategy might be to "let go" of all expectations.
Control strategies may be relatively efficient or inefficient. It is unreasonable to assume that all control strategies are equally effective in the learning and disinhibition of alpha activity. Some control strategies may thus be more effective than others in increasing alpha activity. It is possible that some studies have recommended to subjects inefficient control strategies, hence their subjects have not produced above-baseline alpha activity. In addition, very different control strategies may be equally effective in increasing alpha activity and yet produce very different experiential effects (Tyson, 1979a, 1979b).
Tyson (1979a) found a large variety of subjective experiences associated with increased alpha activity. He states, "The variety of reported experiences could also reflect different strategies for controlling alpha activity and may argue against the description of the 'alpha experience' as a unitary phenomenon."
Tyson proposes that it is possibly the interaction or combining of control strategy and above average baseline alpha activity that may determine the subjective experience.
As previously mentioned, some control strategies may be more efficient at raising alpha activity than others. Plotkin (1976) himself has found that control strategies involving oculomotor "relaxation" were the most effective in increasing or in disinhibiting alpha activity. As previously mentioned, it is possible that some studies have given to subjects less efficient control strategies, hence have not resulted in above-baseline alpha activity.
Control strategies for alpha enhancement have traditionally involved "passive volition," "passive awareness," "letting go" without active deliberate trying, pushing, or effort (Kamiya, 1968; Hume, 1976; Johnston, 1982). Alpha suppression, on the other hand, has traditionally involved the "reverse" strategy of active, deliberately directed thinking, trying, and attending.
Plotkin (1978), in one study, paradoxically recommended a control strategy more associated with alpha suppression to subjects who were learning alpha "enhancement". In these instructions to his subjects he included statements that "...they could expect to succeed if they kept trying and if they continuously search for more effective control strategies; that they should never give up, since some persons can not 'figure it out' until the tenth session." Words and phrases like "keep trying," "continuously search," "never give up," "figure it out" are neither neutral instructions, nor instructions that allow for passive awareness; in fact, they discourage passive awareness. Plotkin has essentially given his subjects a control strategy to continuously keep trying and search for more effective control strategies, possibly up to the last session. In this way he may have confounded some of his results.
Oculomotor Control Strategies The blocking of occipital and parietal alpha activity has been shown to be strongly related to oculomotor "looking" behavior (Goodman, 1976; Mulholland, 1973, 1981; Peper, 1971; Wertheim, 1974, 1981). These studies have shown that mere "seeing" alone or mere oculomotor movements alone do not necessarily block alpha; rather the interlocking of visual and oculomotor systems in "looking" behavior appears to block alpha. Wertheim (1981) concludes, "the amount of alpha activity in occipital EEG is a measure of the degree to which retinal and extra-retinal activity information are used in oculomotor feedback, irrespective of visual information processing."
Peper (1971) suggests that the same relationship between occipital alpha and oculomotor activity exits between central area alpha and movement responses. Tyson (1979) generalizes to suggests that the general disconnection between sensory and motor systems increases alpha. Successful control strategies may thus involve any activity which successfully "disconnects" the two systems.
Plotkin (1976a) found a major control strategy, which he claims was to avoid "looking," was to concentrate on "nonsensory" (thoughts, feelings and images) activity. Plotkin however, was only making an inference, without direct evidence, that the strategy was to avoid "looking." In any case, this general strategy did not allow subjects who used it to exceed above eyes-closed baseline alpha activity. In addition, it is debatable that thoughts, feelings and images are "nonsensory." Fehmi (personal communication, August, 1984) claims all such experiencing is sensory and is experienced in terms of one or several sense modalities. There is, however, no objective research specifically along these lines.
(6) Subject Expectations Although no supportive research has yet been done, the following two points concerning subject expectations are reasonable possibilities that should be taken into account:
First, control strategies may be inseparably linked to subject expectations (and of course subject expectations have been shown to be strongly linked with the kind of subjective experience one has during training (Plotkin, 1979)). For example, if a subject's chosen or suggested control strategy is to "open one's self to experience," then it is reasonable that he be more likely to expect more feelings of "openness" during training. It is also reasonable to assume that he would be more likely to choose or alter his control strategy to "seek
out" such experiences. Instructions should thus be worded very carefully so as not to unintentionally suggest control strategies and, or alter subject expectations in undesired directions. As mentioned previously, Plotkin (1978) may have made such unintentional suggestions.
A second factor that may influence control strategy and thus possibly alter expectation is the experiential questionnaires used. If a subject repeatedly answers the same questions from session to session, it appears reasonable that the subject would be more likely to attend to those aspects of his experience during training and incorporate such attending into his control strategy. For example, if a subject was asked repeatedly about his experience of time, he may be more likely to attend to his experience of time in the following sessions. In this case, the subject may be more likely to develop expectations and control strategies that involve the experience of time.
Similar to Tyson's suggestion that alpha enhancement training and control strategy may interact to produce a wide variety of subjective experiences, Walsh (1974) found that alpha enhancement training and expectation may interact in the same way. Walsh found that "alpha experiences" were reported only by subjects who were given alpha enhancement training and were told to expect alpha experience. Neither condition alone was sufficient.
Control strategy, subject expectation, instructions given to subjects, questionnaires used, and the degree of alpha activity thus may all be related and interact to influence subjective experience. In the current study, these factors were attempted to be used optimally or an attempt was made to hold them relatively constant between control and experimental conditions . The same instructions, control strategies and questionnaires were given to both conditions. Any difference in
experiential scores between control and experimental conditions should thus be due to actual differences in alpha activity.
(7) Subjects Used The majority of studies on experiential states and alpha activity have used beginning psychology students. Some investigators have suggested or approved the use of subjects from different categories (J. V. Hardt, W. B. Plotkin, personal communications, April, 1984).
In reference to subject selection, Fehmi (1975) states, "much of the first three years of (his) research resulted in negative results simply because we, in our desire to hold the strictest lack of bias in selecting subjects, chose to use beginning psychology students." On the other hand, Fehmi further states, "persons having practiced some discipline over a number of years such as meditation, an art form, or some athletic or martial art practice do exceptionally well in producing large amplitude waves, controlling frequency and amplitude, and controlling the phase relatedness of their brain wave activity." Kamiya (1968) has made similar observations about subjects in similar categories. There is as yet, however, no study specifically comparing the learning of alpha of subjects in these categories to those in other categories.
(8) Feedback Signal The wave characteristics, frequency, delay, and modality of the feedback signal itself has been shown to dramatically affect alpha enhancement.
Tyson (1982) found that subjects who received a "sine wave" shaped auditory feedback signal had dramatically better enhancement scores than subjects listening to a biofeedback tone with a rough" sawtooth" shaped wave. The use of ineffective or alpha-inhibiting feedback tones, according to Tyson, may account for many of the failures to achieve
above baseline alpha.
Visual vs. auditory feedback has also been shown to alter alpha activity. Fehmi (1974) found that "right-movers", given auditory feedback, have better control of mid-frontal alpha activity, while "left-movers" have better control of mid-frontal alpha activity when given visual feedback. The frequency of the biofeedback tone is yet another factor. In an unpublished study of feedback tone frequency, Fehmi (1975) found that more frontal alpha activity was produced by subjects who listened to a tone that was amplitude modulated at an alpha frequency than was produced by listening to a tone at a theta frequency (5Hz). Conversely, more theta activity was produced when subjects listen to a tone at a theta frequency. Fehmi concluded, "This suggests the superiority of rhythmical amplitude modulated feedback training for the achievement of increased amplitude."
Lastly, the phase delay of amplitude modulated feedback can effect enhancement scores. Using auditory feedback, amplitude modulated by the subject's own alpha waves, Fehmi (1976) delayed this feedback from 12.5 to 372.5 degrees of phase angle for each subject. This delay time varied from 0 to 100 milliseconds. It was found that subjects showed a subjective preference for a delay of 192.5 and 282.5 degrees. Significantly more alpha activity was produced for the 282.5 degree delay than for the 192.5 degree delay. A phase delay of 282.5 thus appears most desirable for training.
(9) Training Times Apparently most investigators have not seriously explored the question: Can stronger subjective results be achieved, and can above baseline alpha enhancement take place in more than one or two hours total training time? Given that some meditative states require months or years of practice, it seems unreasonable to
expect instant above baseline alpha with accompanying strong experiential changes.
Fehmi (personal communication, 1974) himself, states he did not learn to enhance his own integrated amplitude above average baseline until after about 25 hours of training. T. Scully (personal communication, December, 1984), a researcher of bilateral occipital synchronization, recommends at least l6 hours training. Hardt (1975) claims: "the first 100 minute increase may only be adaptation and habituation....Beyond 140 minutes there may be direct alpha learning." Only about ten studies (six are listed) (Paskewitz & Orne, 1971; Regestein, Pegram, Brigitte & Cook, 1973; Hardt, 1978; Hardt & Kamiya, 1978; Plotkin, 1978,1980) have attempted long term training. These studies vary greatly in training trial length and vary in total training time from 4 to 24 hours, and have had varying results.
Limitations of Experiential Measures
The limitations of experiential measures is yet another factor that may have contributed to the lack of strong and consistent evidence for alpha-dependent experiential states. Questionnaires have varied from study to study, making comparisons difficult. Three difficulties with experiential measures are: (1) vague descriptions of subjective experiences, (2) a lack of measures of "simultaneous" experiences and of "attentional biases," (3) inadequate definition terms.
Some studies have allowed subjects to remember, in their own words, their experience when the feedback was at its highest levels. This has the advantage of not prejudicing a subject as to what subjective experiences to pay attention to. Such descriptions, however, can be very ambiguous. For example, one of Millay's (1981) subjects stated "sustained synch [occipital alpha synchronization] elicits feelings of
being grasped in a beam of cosmic rays, an immobilization, an electric shock of energy like being center stage in a spotlight."
On the other hand a typical experiential questionnaire requires more specific responses. Questions are usually responded to by selecting a point on a multipoint semantic differential scale that briefly covers such general dimensions as: body awareness, alertness, experienced body weight, experienced body size, amount of thought, physical relaxation, unusualness, pleasantness, profundity, awareness of surroundings, awareness of time, happiness, value, energy level, degree of emotionality (Plotkin, 1978). If the change in subjective experience is not covered by the dimensions measured then no change will be detected.
Another difficulty with measuring subjective experience is that descriptions of possible alpha experiences, given prior to training, can sensitize subjects to attend to only one aspect of their total "available" experience, and thus encourage an attentional bias. Likewise the questions given after training can encourage one-sided descriptions reflecting only one aspect of many simultaneous experiences. After answering such post-training questions, a subject may retain such attentional biases in future sessions. Attentional biases that discourage awareness of simultaneous experiences is another factor that has been ignored in studies of subjective experiences and alpha.
An example of such bias is reflected in multipoint bipolar differential scales. The subject is not usually permitted to state if he simultaneously experiences both ends of a bipolar scale. For example, on a pleasure-pain scale, it is possible that the subject may experience a strong sense of pleasure and a strong sense of pain
simultaneously without having to be in the "middle." Likewise the subject may experience a sense of tension and relaxation simultaneously. These experiential scales may thus bias subjects against experiencing and articulating simultaneous experiential states by forcing them to chose one point on a bipolar scale. If avoiding such attentional bias is crucial for accurate reporting of experience, and such "simultaneity of experiences" is a main effect of training using Fehmi's equipment, then current experiential measures are far from adequate.
Different experiential states are mistakenly viewed as mutually exclusive. For example, in one study, Plotkin (1979) described to his subjects one of two supposedly "opposite" or mutually exclusive states: In one state "the mind slows down considerably... thought is free flowing... a serene state...beyond emotion...an egoless state." In the supposedly "opposite" state "The mind becomes more efficient... we are able to direct our thoughts...thoughts often entirely relevant...you become precisely aware of who you are as a person." Instead of being mutually exclusive, these two descriptions could very well be two aspects of the same state.
Another underlying difficulty with measures of subjective experience is that language has never been developed to describe or "map" subtle, unusual, or "unexplored" experiential states. Terms are often difficult to define.
Experiential Measures
There are few established questionnaires of subjective experience that do attempt to cover a large range of the content of an individual's awareness. Unfortunately, most questionnaires focus primarily on factors such as mood, anxiety, content of daydream, or factors related to hypnotic susceptibility. Many questionnaires have been made up by
the experimenter to be used only for one or two studies.
Fortunately, a questionnaire has been developed that attempt to cover a much wider breadth of conscious experience and for which some research has been done confirming both good reliability and validity. Pekala (Pekala, 1983; Pekala & Levine, 1982, 1983; Pekala & Wenger, 1983) has developed the "Phenomenology of Consciousness Inventory" (PCI) (see Appendix).
The PCI covers such predefined dimensions labeled as "volitional control," "rationality," "altered experiences," "attention" among others. These questionnaires have not been used as yet with EEG biofeedback studies.
Pekala (Pekala et al., 1982) has shown that the PCI, given to subjects after just sitting and "introspecting" for a time, has high internal consistency, test-retest reliability, and validity.
In regards to the PCI, Pekala concludes, "Phenomenological experience can be reliably and validly assessed via self-report questionnaire completed retrospectively in reference to an immediately preceding period of subjective experience."
Equating Control and Experimental Conditions
Evidence showing a difference between subjective experiences between experimental and control conditions is considerably weaker when control and experimental conditions are not maximally equated in terms of settings and tasks. Many studies have used "yoked" control groups in which subjects were given noncontingent or false feedback recorded earlier from an experimental counterpart. Strayer, Scott and Bakan (1973) however, have shown that subjects receiving such noncontingent alpha feedback are more likely to become drowsy, thus potentially confounding results. There also exists the danger of the noncontingency
being recognized by the control subjects. Tyson (1982) also suggests that yoked feedback may not actually be non-contingent in that such feedback, at times, may actually correlate with the alpha production of some control subjects. Hatch (1982) suggests control subjects be given feedback in an "altered contingency," one which is as similar to the experimental contingency as possible. For example, the present study gave subjects the task of stabilizing the same EEG signal they are attempting to increase. This procedure should greatly increase the similarity of tasks between contingencies. Tyson (1979a) used the altered contingency of giving the experimental group alpha training and the control group theta training.
Few EEG Characteristics Studies
Another possible reason for the lack of strong consistent evidence is that only a limited number of EEG characteristics have been studied. The characteristics studied may not be the ones most related to subjective experience. Plotkin himself (personal communication, April, 1984) refers to the occipital alpha biofeedback that the vast majority of researchers have used as: crude, simplistic and global. Plotkin states he is not surprised that no unequivocal results were obtained supporting alpha-dependent experiential states.
Most research, as previously mentioned, has been limited to occipital lobes and has only used integrated amplitude or percent-time as a measure of alpha. Integrated amplitude has replaced percent-time in the majority of later research (Travis, Kondo, & Knott 1974; Hardt & Kamiya, 1976). Integrated amplitude is the total area under the alpha wave curve. Integrated amplitude and its square integrated energy allow for constant analogic feedback. Percent-time, on the other hand, only allows the subject to know if his alpha activity is above or below a
certain amplitude threshold.
Unfortunately, most research has not been concerned with expanding to other EEG measures and to more comprehensive experiential measures but with whether or not there is a predefined "alpha experience" associated with subjective experience.
Some EEG characteristics usually not included are: (1) phase synchrony: the simultaneous rising and falling of alpha waves between sites within or across hemispheres, (2) coherence: the stability of phase angle, (3) alpha burst characteristics, (4) frequency or power spectrum, (5) subharmonics in the frequency spectrum (Don, 1977), (6) topographical location, (7) the spread of changes in EEG characteristics to other cortical and subcortical sites, (8) the shift of high amplitude waves, and of coherency and phase synchrony to other frequencies, (9) direction of wave propagation, (10) shifts in phase angle over time, (11) the shift in EEG characteristics between lobes and hemispheres, (12) specific patterns of EEG characteristics between lobes and hemispheres.
The EEG measure to be used in the current study reflects both amplitude and phase synchronization and is covered in more detail later.
Meditation Research
One study on Zen meditation (Kasamatsu & Harai, 1966) showed that, during eyes open meditation, for experienced meditators, there is a marked increase in alpha amplitude in midline frontal and central regions as well as in the parietal region relative to occipital regions. Later on in meditation there is a decrease in alpha frequency, and finally, the appearance of a rhythmical theta train.
Hardt and Kamiya (1976) found, in advanced Zen meditators, an increase in activity and coherence that spreads forward as meditation deepens. Beginning meditators, on the other hand, showed low coherence most of the time. All the above changes took place when the meditator had his eyes open.
Research on T.M. (Transcendental Meditation) (Banquet, 1973, 1972; Levine, Herbert, Haynes & Strobel, 1975; Wallace, 1970) has found that for experienced meditators, during meditation, alpha amplitude increases and interhemispheric and intrahemispheric alpha phase synchrony spread to other frequencies and spread forward from posterior to frontal areas. This phenomenon is referred to by Banquet as "hypersynchrony" (Banquet & Sailhan, 1974). All lobes of the brain and possibly subcortical areas eventually become involved during this type of meditation.
Both T.M. and Zen research thus show involvement of large areas of the brain and not just the occipital lobes. In fact, for Zen meditation, the occipital lobe may be involved the least in alpha activity changes!
Glueck and Stroebel (1976) have shown, when comparing T.M., EMG biofeedback, and Alexander relaxation technique, that specific patterns of coherence and synchrony between lobes may exist for each technique used. The possibility thus exists that each meditation strategy may have its own EEG "thumbprint" in terms of phase relationships.
Coherence, Synchronization and Attention
Some studies have related coherence and phase synchronization to attentional processes. In a study of children, Martinius and Hoovey (1972) found that increases in bilateral synchrony were correlated with superior performance on a task involving attending to tones. They conclude that alpha increase is "an expression of a more general activation process whereas the improvement of inter-occipital alpha
synchrony was clearly related to the partial function of attention tested." In a study of alpha biofeedback, Hord, Tracy, and Naitoh (1974) draw a similar general conclusion when they speculate that "the phase angle of alpha activity in frontal and occipital regions is an electrophysiological correlate of attention mechanisms."
Coherence measures have also been used to distinguished between groups of individuals with different attention related traits or pathologies such as field dependence (Colter and Shaw, 1982), dyslexia (Sklar, Haley and Simmons, 1972), and hyperkinesis (Montagu, 1975).
Synchronization and Experiential States
Research correlating biofeedback learned synchronization with experiential states is limited to a few poorly controlled studies (Johnston & Millay, 1982; Mikuriya, 1979, 1980; Millay, 1975, 1976, 1980, 1981; Scully, 1978). Mikuriya (1979) trained subjects to increase bilateral occipital and temporal synchronization. Bipolar electrodes were used. Feedback was based on the production of in-phase waves of 5 Hz from matching right and left lateral sites. Subjects trained with this feedback tended to report a state "being free from visual images and for some was perceived as a mental stillness and space between thoughts that appeared to be 'a portal' to the perception of internal physiological states....Several subjects who were talented physical scientists recognized ...a problem-solving mental state devoid of visual and auditory images that was useful in the contemplation of abstract problems...One subject described interhemispheric alpha synchronization as a Zen meditative state and did not know that synchronization of alpha rhythm had been described in Kasamatsu and Hirai (1966) as characteristic of Zen meditative states." Mikuriya concludes "a preliminary impression is that interhemispheric alpha synchronization
represents a cognitive breath holding, a suspension of usual thought process."
Millay (1982) trained subjects to increase bipolarly recorded bilateral occipital coherence and synchronization. Subjects reported "with few exceptions, a state comparable to other disciplines such as meditation, various types of centering exercise as Aiki-do."
An example of one of Millay's subject-reports is "I need to put the center of my attention on the top of my head... I usually think in images...a constant flashing of different images which I need to allow to change and to flash in order to remain in synch.... Analysis usually takes me out of synch. In later runs I have discovered I do not only 'see' to think but I also can hear and feel...a constant flow...much like a Zen meditation."
Another subject reported "as long as I allow the thought to flow I seem to remain in synch. When I became aware of a thought I didn't like or a thought that would disturb me, I went out of synch." Another subject reported his experience as "similar to an Aiki-do exercise projecting energy out of the forehead."
Unfortunately, both Mikuriya and Millay used no control groups or control conditions, hence it is not possible to directly attribute experiential reports to synchronization. Also such single reports have very limited value and may only suggest possible relevant dimensions of experience that may be measured in future questionnaires. In the above research, the flow of "thought" and "images" appear to be important factors.
Fehmi's Research
Fehmi's (1971) original research on synchronization involved monopolarly recorded bilateral occipital phase synchronization. A control group with yoked feedback was included. Experimental subjects, who received phase synchrony training, tended to report that "the maintenance of phase synchrony is tiring and difficult, and that it is associated with concentrating on a relatively stable mental image." The reports also indicated that "phase disparity is associated with permitting one's mind to wander....Control subjects gave comments which referred to the difficulty of the task and the fact that no behavior consistently controlled the tone." It was not inquired of control subjects, however, if there was any detection of the noncontingency of the feedback tone. Also, the important effect of perceived success was not measured.
Concentration Without Tension Interestingly, the difficult concentration reported by experimental subjects in Fehmi's study was not accompanied by higher EMG activity, suggesting "concentration without tension." In a study of Zen meditation, Kasamatsu and Hirai (1966) state: "(in) well achieved meditation it will be said thus 'concentration without tension'...is going into the utmost world of psychic life."
Fehmi also noted that pilot subjects reported that their ability to increase phase agreement required that they first concentrate with effort and then completely let go of any attempt to focus and direct attention. They described a kind of "rebound" into an effortless but more total concentration.
The reports of Fehmi's subjects of fatigue and "single minded concentration upon a mental, visual image," is remarkably similar to the report of Plotkin's (1980) one subject who was able to achieve above optimal baseline alpha activity (Plotkin does acknowledge
well-documented, above optimal baseline alpha activity at least in one case).
Plotkin's subject could raise his monopolar midline occipital alpha activity up to 867% above optimal eyes-closed resting baseline! As will be discussed later, alpha activity recorded from a midline site, according to Fehmi, is also a measure of phase synchrony between adjacent lateral sites.
Plotkin's subject reported, similar to Fehmi's subjects, that during enhancement, "there was intense concentration at the time ..not worried, but not relaxed either. If I just relaxed, the alpha waves wouldn't be strong." Plotkin reports that the subjects emphasized that the enhancement was "not relaxing while he was doing it (in fact, 'it was almost an energy drain.')...He definitely did feel more relaxed immediately after the session....In general, concentration (either visual or kinesthetic) on a particular spot or on a mental image, was the deciding factor achieving alpha enhancement, although he had to lean forward to succeed." One of the several strategies the subjects used was to concentrate on a space between his eyebrows and nose.
Plotkin's subject reported "internally tensing" his entire body while concentrating, however, similar to Fehmi's findings, no EMG artifact was detected. The report of intense effort and concentration "is especially interesting in the light of the fact the he had been led to believe by us that alpha enhancement would be profoundly relaxing." The research by Plotkin and Fehmi suggests that future research should include scales measuring such experiential dimensions as stability or flow of mental images, effort required, concentration, and involvement of posture and areas of the body.
Lateral Eye Movements (LEMS) Selzer & Fehmi (1974) found that "right-movers" were better able to suppress frontal alpha than "left-movers". On the other hand, "left-movers" were better able to enhance frontal alpha than "right-movers", who actually suppressed alpha when asked to increase alpha.
The "right-mover" - "left-mover" phenomenon was first studied by Day (1964). He found that most people divert their eyes to the right or left when initially reflecting on an answer to a question. In addition, he found some people predominantly moved to the right while others predominantly moved to the left. These movements he labeled conjugate lateral eye movements (CLEMS). The direction of eye movement supposedly indicated that the contralateral hemisphere is being predominantly used to process the question (Bakan, 1971; Kinsbourne, 1973). This theory is further supported in that "right-movers" perform better on tasks more associated with left hemispheric functioning while "left-movers" perform better on tasks more associated with right hemispheric functioning (Weiten & Etaugh, 1973).
Fehmi (1974), found that strong "left-movers" have twice the baseline frontal alpha activity as "right-movers". "Left-movers" however were not able to reduce alpha levels below baseline, while "right-movers" were not able to enhance alpha activity. Fehmi states, "Integrating these results with other findings, I am led to describe the differences in strategy in this way: right-lookers narrow focus and objectify their experience as a general strategy in approaching new tasks. Left-lookers diffuse their focus and enter into their experience as a general strategy".
The nature of the question or task can also influence which hemisphere the subject uses and hence the direction of eye-movement. Questions and tasks more associated with left hemispheric cognitive processing (e.g.. questions about words) are also more associated with right lateral movements. Questions and tasks more associated with right hemispheric cognitive processing (e.g.. "spatial" questions) are also more associated with left lateral eye movements (Gur & Gur, 1974; Gur, Gur & Harris, 1975).
Ehrlichman (1978) has questioned the strength of the association of eye movement direction with hemispheric activity. Later research, however, using positron emission tomography (PET) provides more conclusive evidence (Gur, Gur, Rosen, Warach, Alavi, Greenberg & Reivich, 1983). This study found "Lateralized task effects were also obtained in the frontal eye fields, supporting a hypothesized neural network linking cognitive activity with motor orientation."
The research thus shows that lateral eye movements can be effected by both the subjects "habitual" hemispheric preference and the type of question or task given. Which of the two factors is stronger at a given time may be related to whether the experimenter giving the questions is located behind or in front of the subject (Gur, Gur & Harris, 1975). Gur et. al. suggest "Because the face to face situation, being more personal, may be more threatening and anxiety-provoking, S falls back on characteristic and preferred modes of response." Hiscock (1977), however, did not find anxiety to be the strong factor in the experimenter location phenomenon. In any case it appears that subjects are able to process questions according to the nature of the question and not only according to their habitual hemispheric preference.
It may also be that, in the "experimenter behind the subject" condition, some subjects are more "habitual" in preferred hemispheric use while others may be more "flexible" in that they are more likely to favor the hemisphere most appropriate to the question.
Fehmi (1974) found that strong "left-movers" were good alpha enhancers and poor alpha suppressors, while strong "right-movers" were good alpha suppressors and poor alpha enhancers. On the other hand, more "flexible" subjects may be better able to both enhance and suppress alpha. More "flexible" subjects might better be able to use different attentional strategies and be able to get a better "sense" of control for both enhancing and suppressing. No research, however, has been done to test this idea.
Subjects who show no marked hemispheric preference, and who also do not show any particular pattern in eye-movement direction in relation to question type, were not used. This is because it has been suggested by some researchers (Kuzma & Jamieson, 1978) that there may be hemispheric "competition" in these subjects. In other words, these subjects may not "know" "which way to go" in terms of which hemisphere to use.
Multichannel EEG
Fehmi currently trains his subjects and patients to enhance alpha activity using his five-channel phase-sensitive EEG. Experiential reports associated with the use of this equipment appear to be different than reports associated with only occipital synchronization. Comparisons between studies, however, are difficult since there are no common experiential measures used.
Midline Sites Fehmi's five-channel EEG instrumentation records monopolarly from three midline sites (Oz, Pz, Fz) as well as from two temporal sites (T3 and T4) which comprise separate channels. Fehmi (1978) has demonstrated that midline sites (e.g., Oz) record an EEG alpha signal that is roughly the algebraic sum of the alpha signal taken from two lateral sites on the same lobe (e.g.: O1 and O2). In other words, Fehmi has demonstrated that signals from the two lateral sites
that are in phase, add to each other's amplitudes when measured at the midline. Signals from the two lateral sites that are out of phase, cancel each other's amplitudes when measured at the midline.
Midline sites, according to Fehmi, thus are measures of amplitude and phase synchrony of alpha activity between adjacent lateral sites on each lobe. The more phase synchrony there is, the stronger the midline signal. In addition, using Fehmi's equipment, signals from all five sites are algebraically summed so that these signals cancel or add to each other to the degree of their phase synchrony or phase disparity. Fehmi's feedback signal is thus a "fuzzy", global measure of phase synchrony and amplitude.
Fehmi's more recient eight channel EEG, which was not used in the current study, does not use midline sites, rather EEG signals are recorded monopolarly and laterally to the right and left of midline for each lobe (O1, O2, P3, P4, T3, T4, F3, F4), to make a total of eight channels. These eight channels are algebraically summed in the same way to provide another global measure of alpha phase synchrony and amplitude that simultaneously reflects both intrahemispheric and interhemispheric synchrony.
Disadvantages One possible disadvantage of the use of Fehmi's five or eight channel EEG is that a subject may learn not to produce high amplitude, in-phase alpha from all five or eight sites. Rather, the subject may learn to produce very high amplitude alpha activity from one site, or very high amplitude in-phase alpha activity from two or more sites. At the same time, the subject may learn to suppress alpha activity at other sites that are not be in phase with the stronger signal. In such cases, the total summed signal would still appear to be increasing, but only due to increasing alpha at a few sites, and not
from all sites.
According to Fehmi (personal communication, July, 1984), the above mentioned possibility has not happened to any significant degree. In the present study it is unreasonable however, to expect that all lobes contributed equally to the summed alpha signal since some lobes tend to initially produce alpha activity of greater amplitude than others. For the majority of people the occipital lobes usually produce the largest, most stable alpha activity (Rubin, 1938). The purpose of the present study was not to test the exact distribution of alpha apmplitude or synchrony but to test the attentional hypotheses posed by Fehmi using the equipment already in wide use and which was similar equipment used to originally formulate his hypotheses.
Monopolar vs. Bipolar Recordings Fehmi (1975) (Fehmi & sundor 1989) states that monopolar recordings using linked earlobes as reference are superior for measuring global phase synchrony. Bipolarly recorded alpha activity will actually show a decrease in amplitude as the alpha phase synchrony increases between the recording site and the reference site. For example, a bipolar recording of an EEG signal from an occipital site equal in amplitude and in phase with the EEG signal from the reference sites will cancel and appear as a flat signal.
Experiential Effects Using Multichannel EEG Biofeedback
In an unpublished study by Fehmi (1974) involving middle level management executives, twelve subjects rated their experience before each session on 45 semantic differential scales. Both enhancement and suppression trials were included in each session. After 20 sessions, experimental subjects rated themselves as "more calm, less depressed, more able to concentrate, more self initiating, more detached from their experience, more observant, more personal (as opposed to formal), more
in oneness (as opposed to separateness), more insightful, and more satisfied with life. the control subjects rated themselves as being less accepting of praise, less satisfied with life, more distant (as opposed to absorbed), less self-conscious and more even-tempered that they were at the beginning of training."
Such reports are difficult to compare with experiential reports of other studies because they are more in terms of self-perceived trait changes, as opposed to actual experiences during training.
In the above study, control subjects received yoked feedback which may lead to boredom as previously mentioned. No measure was taken of frustration or boredom with task, drowsiness, suspicion of non-contingency of feedback, and perceived success. Baselines were taken that determined that actual above-baseline enhancement had taken place.
Experimental subjects during initial sessions received feedback from only one lobe. In subsequent sessions, additional lobes were "added" until only the last three sessions were done with feedback from four lobes. Because of this procedure it is difficult to attribute experiential changes to purely five-channel feedback.
In another study, Fehmi (1978) found that subjects asked to visually scrutinize their surroundings in a critical manner produced less alpha activity in temporal, frontal and occipital areas than when asked to observe their environment in an "appreciative "manner. Thus, "appreciativeness" and "criticalness" may be two other experiential dimensions that should be included in experiential measures.
Serious well-designed research has yet to be done addressing the actual possible effects of multichannel phase-sensitive biofeedback on experiential experience.
Fehmi's Theory of Attention
Fehmi's six previously mentioned hypothesis concerning five-channel alpha enhancement include: (1) Subjects are more likely to have the subjective sense of "merging", "immersing", or being "absorbed" into the content of their awareness (as opposed to feeling separate and distant from such content). (2) These subjects will also be more likely to become simultaneously aware of greater numbers of thoughts, sensations, perceptions, or feelings. (3) This type of awareness involves subjects being aware of experiencing "reality" in terms of a relatively greater number of senses.
These and his other previously mentioned hypotheses that were tested are predominantly based on Fehmi's clinical observations working with clients in biofeedback clinics, on personal experiences while enhancing his own alpha activity over a period of years, and on the interpretation of some of his previous studies.
The previous study of middle management executives (1974) supports Fehmi's previous hypothesis involving the "merging with experience" in that subjects rated themselves "more in oneness (as opposed to separateness)". Control subjects, on the other hand, rated themselves as "more distracted (as opposed to absorbed)".
In his clinical experience, Fehmi (1979) observed that "extreme goal oriented individuals who habitually function with a narrow scope of attention find it very difficult to adopt the gentle approach necessary to control the feedback." According to Fehmi, once an individual adopted more of a "relaxed" type of attention and allowed their attention to "diffuse" or "expand " to encompass more sense modalities and subjective experiences simultaneously, and to immerse themselves in experience, they were better able to increase the signal.
Open Focus From such observations and studies, Fehmi has proposed that attention can be viewed in terms of four poles: (1) "narrow focus" of attention or "one-pointedness" of attention and (2) the opposite pole, a widened "diffused focus" of attention or "many-pointedness" of attention, (3) "absorbed" attention or merging with experience and (4) the opposite pole, remote or "objective" attention "about" experience and separate from one's "sense of self."
The ability to attend to one's experiences "narrowly," "diffusively," "objectively," and "absorbently" without a preset bias toward any one of these four modes of attention, Fehmi calls "Open Focus". In Open Focus, attention is given equally to how one attends and to what one attends to. Most previous research however, has focused primarily on what the subject is attending to and not to how he is attending. For example, Plotkin (1976a) has focused on whether subjects were attending more to "sensory" or "nonsensory" stimuli.
Attending Without Bias Fehmi postulates that through attending without preset bias (Open Focus) toward one any of the four "attentional poles" one can better encompass in his attention seemingly opposite or contradictory experiential states and modes of attending. These states of awareness and modes of attending can thus be included into one "conscious field" or "field of awareness".
This attending without bias, appears similar to what Tyson (1979) calls attending without expectation. Attending without expectation Tyson claims, is one successful control strategy to increase alpha activity. Tyson states "A meditative attitude that quiets the predictive process (is)...having no expectations. Most participants, however, find this strategy considerably more difficult than oculomotor relaxation." Thus, according to Tyson, as one has fewer expectations he stops predicting what to attend to, and increases the possibility of
raising alpha activity.
The Simultaneous Incorporation of Many Mental States The Open Focus concept of simultaneously incorporating many mental states is also remarkably paralleled in the T.M. research literature. Banquet (1974), in referring to the state of deep meditation, concludes: "thus the phenomenon of awareness seems to become independent of any mental state. It rather functions as a background, support, or witness to these different states. These results, if confirmed, would indicate that the brain can simultaneously integrate and harmonize different or even apparently opposite modes of attention in a unique state of awareness".
In addition to multi-channel EEG training, Fehmi has developed a series of "awareness exercises" consisting of questions asking one to experience the "space" in various areas or his body. Later on in the exercise, questions are asked that require one to include more and more sense modalities in experience: "at the same time that you are aware of feeling space, emotions, and other body experience can you also be simultaneously aware of any tastes, smells, thoughts, sounds and imagery that might be present."
These awareness exercises, used in conjunction with multi-channel EEG training Fehmi labels "Open Focus training" (OFT). Open Focus training has been used to clinically treat trait anxiety, epilepsy, optimalization of function in athletes and others, stress, tension, obsessive compulsive disorders, headaches and other forms of pain (fehmi 1987).
Because of the widespread clinical use of Fehmi's Open Focus training and attentional model, it is important to provide experimental evidence that either supports or casts doubts on Fehmi's hypotheses.
Method
Subject Selection
twenty-six right-handed male subjects were procured with an ad (see appendix) in various local publications. Responding men showing a strong interest in meditation or similar discipline when talked to on the phone, were asked to come in for a screening interview. Subjects were between the age of 24 and 55 and lived within a 70 mile radius of princeton, New Jersey. They were from a wide variety of professions from engineering to sales, and about one-third were married.
Health Questionnaire Screening:
Potential subjects were interviewed by the experimenter and given a health questionnaire (see appendix). Men with apparent or reported neurological problems, learning difficulties, serious emotional problems, and other health problems that might interfere with concentration and sitting, were not selected as subjects. If a potentially serious health problem had been detected that the potential subject was not aware of, (none were), he would have been referred to his physician.
The health questionnaires were evaluated by the experimenter. if any ambiguities arose (none did) as to how much subject's particular health problem might actually interfere with participation, an MD. would have been consulted.
The purpose of questionnaire evaluation however, was not to "diagnose" specific problems or interpret overall scores but only to roughly eliminate extreme health problems and other problems that would interfere with attention and sitting. For example, subjects who reported problems such as frequent headaches or anxiety, or frequent
poor bladder control were not used because such problems could interfere with concentration on the task or with the ability to sit quietly and comfortably.
Health questionnaire answers were simply looked at one at a time and certain potential subjects were eliminated on the basis of certain responses to certain single questions. No overall scores were taken.
Potential subjects were then tested for eye-movements. During the procedure they were asked 42 questions. Direction and duration of lateral eye gaze after the question was recorded on a video camera focused on the subject's face through a hole in the curtain in front of him. The subject faced a homogeneous visual field and the experimenter was seated about four feet behind the subject.
All eye movements were counted that moved roughly more than ten degrees to the right or left that occurred within five seconds after the question was asked. Only People who moved to one side over 60% of the time were counted as right movers or left movers and were selected as subjects.
Apparatus
EEG was recorded using five felt, saline solution soaked, electrodes designed by Fehmi. Electrode sites were always Oz, Pz, Fz, T3, T4, with the right ear lobe as reference and left ear lobe as ground.
The EEG was designed by Fehmi and was manufactured by Biofeedback Computers Inc. The EEG had an isolated output channel with a continuously tuneable filter, threshold control, feedback phase delay control, feedback volume control. EEG biofeedback could be presented from any single channel or a combination of up to five channels. Each single channel had its own variable gain control
Amplifiers had an input noise less than 1.5 microvolts P-P. Common mode rejection was greater than 110 db+. 60 Hz filtering had 40-50 db attenuation. Low frequency response had a 24 db/ octave roll off below 2 Hz.
The experimenter received a digital display of integrated amplitude per unit of time, from the "averager" output. The averager algebraically summed and averaged filtered EEG from across the five channels. Epoch length for recording average amplitude was variable from .5 to 16 minutes.
Raw and filtered EEG from the combined channels were monitored on a two channel oscilloscope for wave form and artifacts. Epochs during which muscular, eye movement, or electrical artifacts were detected on the oscilloscope were not used in determining a subject's enhancement score. By pressing the "abort" button the computer automatically discards all the information collected in the epoch the artifact occurred in and the information for that epoch was re-collected during an additional subsequent epoch.
Feedback was controlled by a separate unit especially built for the research. The unit contained a volume control, an "A-B" switch to switch feedback between enhancement (A) and stabilization (B) contingencies, and threshold controls to determine the upper and lower part of the amplitude "window" which would permit stabilization feedback to occur within the desired amplitude range. Additional controls adjusted how quickly the tone increased in volume as the subject's alpha amplitude approached the center of the stabilization window.
Feedback was a "chirping" sound produced on a Sonalert device. The "chirps" were amplitude and frequency modulated by the averager output for the combined five channels. Feedback was also delayed by 50-75 milliseconds as recommended by Fehmi (1976). Feedback was contingent upon integrated amplitude in the alpha frequency (8-13 hz), summed and averaged across five channels, above or below certain microvolt thresholds.
During baseline and feedback sessions subjects sat in an upright comfortable chair in a 5'x 7', dark, sound attenuated room.
Data Collection EEG output was connected by cable to an American Biotechnology (ABC) Unicomp JV/F isolator and AC to DC converter. The converter was in tern connected to one channel of an ABC eight channel multiplexer (MPX-1). The MPX-1 connected to "John Bell" / Extended technology (ETI) interface card in an Apple 2e computer. Using ABC Unicomp "Bio-trac" software the EEG amplitude signal was sampled, averaged, and displayed on a monitor during data collection.
A printout of integrated amplitude levels from the averager output of the sum of all five channels was obtained for all sessions for either two minute epochs (for training) of one minute epochs (for baseline). This printout also included the average integrated amplitude level for the entire 30 minute training session as well as its standard deviation, and it's minimum and maximum one (or two minute) epoch.
Design
The group was given both enhancement and stabilization trials in each daily session. Training was extended over ten sessions. Neither experimenter nor subject knew the order of the enhancement (A) stabilization (B) training received during the training session.
Variables Independent variables were: (1) biofeedback contingency (enhancement vs. stabilization), (2) training time (session number).
Dependent variables were: (1) baseline scores, (2) "enhancement" and "stabilization" scores, (3) questionnaire scores.
Enhancement Condition Subjects were trained through feedback to produce auditory feedback levels as high as possible. For each subject, enhancement training was done 30 minutes per weekly trial for ten consecutive weeks.
The intensity of the feedback signal increased as the subject approached the stabilization point. The amplitude thresholds of the stabilization point were set during the second baseline relative to that baseline. The stabilization point or "window" was set as close to the subject's second daily baseline as possible. This was done by roughly adjusting upper and lower amplitude thresholds. This procedure was repeated between the first and second trial (contingency) during a short period when the subject was just sitting and not attending to the feedback. This procedure is explained in more detail in the "threshold and switch setting procedure".
Enhancement and stabilization conditions were equated in terms of instructions, time, surroundings, which controlled for expectation and placebo effects, and the effects of sensory deprivation. Attempting to roughly equate the two contingencies in terms of task difficulty controlled to some degree for the effects of perceived success. Subjects were asked to rate task difficulty and perceived success in the questionnaires given at the end of each session in order to assess the contingency differences.
Instructions for increasing the feedback All subjects were given "neutral" instructions that did not implicitly or explicitly suggest a control strategy other than just increasing or decreasing the feedback (see appendix).
Words such as "enhancement", "suppression", "stabilization" or "Open Focus" were not be used in instructions to subjects. In addition, words such as "see", "hear", "view", were avoided to prevent biasing a subject towards any particular sense modality.
First Training Session A day prior to the first session all subjects were requested to abstain from drugs (including caffeine and alcohol) and to attempt to get a good night's sleep on the days prior to and during training (see "Take Home Material" in appendix B). If a subject missed a day, the session was continued the following day or as soon as could be scheduled. If more than three successive weeks were missed the subject was dropped from the study.
A prebaseline questionnaire was given each day prior to baseline. The prebaseline questionnaire inquired as to the subject's previous night's sleep, recent use of drugs, current drowsiness and "energy level", and any new unusual circumstances or stresses.
At the first session the subject was comfortably seated in the experimental room. Electrodes were then applied, and instructions for participating in the baseline were given and explained to the subject. The instructions asked the subject to sit comfortably upright with eyes closed, to "gently" limit eye and head movements but to not rigidly "freeze", to avoid going to sleep, and to not deliberately use any particular meditative discipline.
All subjects were treated with a positive "person factor" as recommended by Taub and School (1978). This means the experimenter attempted to adopt a "supportive" attitude and tone of voice, using first names and including occasional "friendly" but not "intrusive" eye contact.
Optimal And Average Baseline Lights were turned out and a 10 minute eyes-closed baseline was taken. A procedure for obtaining optimal eyes-closed baseline employed by Plotkin (1978) was used because of its strictness.
The subject sat with his eyes closed for three minutes just prior to collecting data for the ten minute baseline period. This was done because some subjects tend to have initially high alpha immediately upon closing their eyes, thus over-inflating baseline scores (Plotkin, personal communication, April, 1984).
Using Plotkin's procedure, the one-minute epoch of the baseline period with the highest integrated amplitude level was used as the optimal baseline. In contrast, during training integrated amplitude levels were taken over two-minute epochs.
Optimal baseline and average baseline were the references used in the determination of enhancement and stabilization scores. In determining average baseline, the average two-minute integrated amplitude score over the ten minute baseline period was used.
After a two minute break and a three minute sitting, another baseline was taken for a ten minute period with the feedback tone on and adjusted to a comfortable intensity. This gave one measure of the effect of the feedback signal itself on subject alpha production. During the baseline the feedback was alternately switched between A and B each two minute period to prevent any learning of, or getting used to, one particular contingency. Also, during this period, thresholds were adjusted to set the stabilization point close to the baseline (see "Threshold and Switch setting Procedure")
Enhancement Scores: Enhancement and stabilization scores ("A" and "B"), were determined two ways. The first way used an optimal baseline as reference; the second used average baseline as reference.
Enhancement ("A") and stabilization ("B") scores using baseline maximum scores (max) as reference were determined by dividing the integrated amplitude levels of the two-minute epochs during training (C) by the maximum baseline score (max) for that session. (Thus Amax = AC/max and Bmax = BC/max).
Enhancement scores for the enhancement conditions("A") and for the stabilization condition("B") using average (v) baseline as reference (Lv), were determined by dividing the integrated amplitude training score (C) by the average baseline (b) (thus Av = AC/b and Bv = BC/b).
Average amplitude (I) scores were recorded during the ten minute periods after a five minute "free-time" period. The "free-time" (see section on "free-time" that follows) was not counted as part of the recording period. Ten minute recording periods were spaced with five minute "free-time" periods between them in order to give the experimenter a break and to maximize experimenter vigilance while monitoring of the EEG for artifacts via oscilloscope during the recording period.
Training After the daily baselines, all subjects received two 30 minute trials, one "enhancement" trial and one "stabilization" trial. The order of presentation was reversed each day. One trial consisted of one five minute free-time followed by one ten minute training trial. This sequence was then repeated one to complete the 30 minute trial.
A five minute break was given between training trials A & B. During the break the subject was asked how he was feeling and then if he was drowsy or experiencing any discomfort.
"Free-time" During training, if the subject wanted to informally "play" with the feedback signal to explore what makes it go down (suppression) as well as up (enhancement), he could do so. The five minute "Free-time" was given prior to each ten minute recording period.
"Suppression" was suggested to subjects as an option during "free time" as suggested by both Plotkin and Fehmi (personal communications, April, 1984), "to give the R.P. (research participants) a better feeling for brain wave control" (Plotkin 1973).
Experiential Questionnaires Immediately following training, electrodes were removed and the subject was given the "in office" questionnaires. Most of these questions pertained to the recalled times during the session when the subject experienced the feedback at its highest levels. The "in office" questionnaires included: (1) the Free-Response ("FR") Questionnaire, which is a written one-quarter to one-half page description of the subject's experience in his own words, and (2) the first experiential ("E-1") questionnaire.
When the subject had finished the "In Office" Questionnaire he was given the "take-home" questionnaires (see appendix) and was asked to complete it at home by the next morning. The "Take-Home" questionnaires included: (1) the "E-2" questionnaire on subjective experience, (2) the Phenomenology of Consciousness Inventory (PCI), (3) and the "Final Overall Daily Questionnaire" ("FOD"). The "FOD" questionnaire included questions on covariates such as "perceived success", the experience of drowsiness, boredom, and any frustration during training. At the end of the last session, a "post-study" questionnaire was included asking for the believed purpose of the study and the credibility of all instructions given.
Step by Step Procedure:
III. FIRST MEETING
A. Meeting with new potential subject about one hour
1. Pass out "take home material" which explains
study, terminology and gives guidelines for
participation. This material
includes:
a. Introduction and summary
b. General take-home instructions
c. General instructions concerning the
training session:
(1) Session format
(2) Instructions concerning the
feedback
(3) General guidelines for the
trial
d. General instructions for all experiential
questionnaires
e. Categories of experience
f. Instructions for questionnaires
2. Briefly go over "take-home material" and remind
them to study it at home.
3. Give Initial Questionnaires
to be filled out. This includes:
a. "Initial Questionnaire"
b. "Initial Health Questionnaire"
c. "Standard Health Questionnaires"
d. "Handedness Questionnaires"
e. Consent form
4. Eliminate inappropriate subjects with:
a. Interfering medical problems
b. People who appear:
(1) slightly disoriented
(2) under excessive stress
(3) primarily looking for
treatment of a disorder
5. apply electrodes to potential subject and turn on
biofeedback tone for ten
minutes total time.
a. Electrode sites are always: OZ,
Pz, Fz, T3, T4, with right ear
lobe as reference and left as
ground.
b. Tell subject the purpose is to
become acquainted with the equipment
and to decide if it interests him.
c. Tell subject the sound he will
experience may or may not be the same
type of feedback he will learn later.
d. Tell subject not to try to increase
or decrease or modify the sound in any
way but just to get used to it.
e. for the first five minutes leave
the feedback on, with the volume and occurrence
of feedback tone reflecting the actual amplitude
(the enhancement [A] contingency).
f. for the second five minutes leave
the feedback on with, the volume and occurrence
of the feedback reflecting amplitude
stabilization (the stabilization [B]
contingency).
IV. EYE MOVEMENT "SCREENING" PROCEDURE
A. Test for eye movements: (approximately 30 min.)
1. Check equipment: mic, tape, focus, chair
positions.
2. Make friendly contact with subject.
3. Seat subject in front of curtain.
4. Sit three feet behind subject.
5. Read and explain Eye movement instructions
to subject (see appendix).
6. Start video recording.
7. Ask questions from eye movement question list
(see appendix) and wait for answer after each
question.
B. At end of screening:
1. Remind subject to review take-home material.
2. Answer questions on the exercise.
3. Schedule initiation session and training
sessions.
C. Scoring:
1. Review eye movement video tapes.
2. Count as eye movements all eye movements
roughly more than ten degrees to the right
or left that occur within five seconds after
the question is asked.
3. Only people who move to one side over 60%
will be counted as right movers or left movers
and will be used as subjects.
V. SECOND MEETING: ("INITIATION SESSION"):
A. Greeting and friendly contact.
B. Establish trust, contact, and rapport.
C. Answer any questions up till now on take-home
material.
D. Go over and answer questions about take-home
material in detail:
1. General take-home instructions
2. General instructions concerning the
training session:
a. Session format
b. Instructions concerning the
feedback
c. General guidelines for the trial
3. General instructions for all experiential
questionnaires.
4. Categories of experience
E. Go over Experiential Questionnaires
(FR, E-1, E-2, PCI, FOD.)
VI. SEQUENCE ASSIGNMENT
A. Assign each subject a code number from a list of "right
mover" or "left mover" codes.
B. Code numbers are pre-assigned equally to one of two
sequences (see sequence sheets 1 & 2). The sequence
of presentation of Trial A (enhancement) and Trial B
(stabilization) will be varied for each session for
each subject according to the sequence sheet assigned
to that subject. The experimenter will not know the
sequence and hence will be blind to whether the subject
is attempting enhancement or stabilization.
VII. TRAINING SESSION: (inCLUDES: bASELINE, iNCREASE tRIAL, and
qUESTIONNAIRE pERIOD) (ten sessions [days] per subject)
A. Pre-baseline
1. Call subject and confirm appointment.
2. Greet and make friendly contact
3. Seat subject in office and have him go
over:
a. General guidelines for the trial
b. Categories of experience
4. Have them also go over the material from
their "assigned packet": (See appendix)
a. Strategy instructions
b. Switch instructions
c. Sequence sheet
5. Subject goes over sequence sheet without
experimenter present. Subject keeps
sequence sheet covered when finished.
6. Give daily prebaseline questionnaire.
(See appendix)
B. Baseline
1. Seat subject in 6 x 6 booth.
2. Apply sensors, adjust equipment, setup
software.
3. Give baseline instructions to subject
(see appendix) and explain aloud.
4. Have subject sit with eyes closed for
three minutes.
5. Take baseline with no feedback: (ten minutes
a. Monitor oscilloscope for artifacts.
"Abort" or do not count 1 min periods
that contain artifacts.
b. At the end ask subjects how they feel.
6. <