52. Доказательства существования неосознаваемой психической деятельности, даваемые анализом вызванных потенциалов: обзор литературы Г. Шеврин (Evoked Potential Evidence for Unconscious Mental Processes: A Review of the Literature. Howard Shevrin)

52. Evoked Potential Evidence for Unconscious Mental Processes: A Review of the Literature. Howard Shevrin

(Paper prepared for discussion at International Symposium on the Unconscious, Tbilisi. Georgia. USSR, 1978)

University of Michigan Medical Center. Ann Arbor, Michigan, USA

This review will draw together a small, scattered but theoretically significant group of evoked potential studies bearing on the evidential base for the psychoanalytic concept of the unconscious and its relationship to consciousness. These studies, most of which were not undertaken with the psychoanalytic concepts of the unconscious in mind, nevertheless speak to its possible neurological correlates. It will first be necessary to sketch the main outlines of the psychoanalytic idea of the unconscious and its relationship to consciousness, following which the experimental studies will be described and evaluated. The paper will conclude with a discussion of the implications of this research for the psychoanalytic concept of the unconscious.

The Psychoanalytic Concept of the Unconscious

Every science is defined by its interest in a particular phenomenon whose understanding comprises its fundamental task. In physics, the phenomenon is matter in motion; in chemistry, the transformation of physical substances; in biology, the activity of living organisms. Psychoanalysis began with a series of observations on behavioral, emotional, and cognitive changes following emotionally significant reminiscence. The subsequent history of psychoanalysis can be characterized as an elaboration of these observations on the basis of a clinical method developed to explore the conditions under which this phenomenon occurs and the conditions which impede its emergence or modify its character. For example, Freud discovered that the patient's way of experiencing his relationship to the analyst (transference) often signalled the nature of the experiences affecting his current life of which he was unaware and which were related to childhood events. By calling the patient's attention to this relationship and "interpreting" its significance, which meant demonstrating its current impact and its genetic roots, the patient was then led to recall heretofore forgotten experiences which marked the beginning of the particular maladaptive pattern under scrutiny.

For our purposes it makes sense to return to the earliest observations because they are least influenced by subsequent conceptualizations and theories and we may thus gain a clearer idea about the nature of the phenomenon itself. In the case of Anna O. Breuer reported that odd behavior, appearing sporadically in the course of her daily preoccupations, often disappeared following the recall of certain experiences. This recall was made possible by an altered state of mind in which she had remarkable access to memories apparently unavailable to her in her normal state (Breuer and Freud, 1895). To Breuer this altered condition appeared similar to hypnosis and he referred to it as a state of autohypnosis. The account of his largely improvised treatment contains many illustrations of this process. At one time, Anna O. complained to him that a dress she was wearing looked blue to her although she knew it was in fact brown. Later that day in her altered, or autohypnotic state, she recalled that a year before she had made a dressing gown for her ill father out of the same material as her dress except that the material for the dressing gown was blue. Her relationship to her father was of considerable emotional significance to her and had already been implicated in other symptoms. Once this recall occurred the false perception disappeared. How can we account for this phenomenon? A cognitive disturbance in one state is altered by a reminiscence in another. Breuer hypothesized that the memory of the blue dressing gown was somehow active during the normal state and produced the cognitive disturbance: it was in a condition of unconsciousness, but nevertheless actively creating a disturbance.

Once this hypothesis is entertained it raises new questions: 1) Why is the memory not available to normal consciousness especially if it is emotionally significant?, 2) What happens to a memory when it is unconscious?, 3) How does it affect conscious experience0 Efforts to answer the first question led to the concept of repression and its more general form, the defense mechanisms; efforts to answer the second question led to the concept of primary process ideation: the mental life that is kept from becoming conscious follows different principles of organization than mental life readily available to consciousness; efforts to answer the third question led to models for symptom formation, dreams, parapraxis, humor, and psychotic thinking.

As Rapaport (1967) has pointed out the concept of the unconscious in psychoanalysis is a corollary of a basic assumption: the existence of psychic continuity. Anna O's misrecognition was not simply a random, non-psychological event; rather, it could be understood once we become aware of all that was involved for her psychologically, which was supplied by her in her altered state. Psychic continuity is by definition individual and historical: it is always one particular individual's continuity unique to him and made up of a unique sequence of events. What is discontinuous in conscious experience (e. g., a brown dress "looking" blue) is actually continuous once we become aware more completely of the individual's experience. Therefore, the concept of the unconscious follows from this assumption of psychic continuity: it refers to those mental events not currently available but which must be assumed to exist in order to account for the underlying continuity.

There can be different conceptions of the nature of psychic continuity. For example, Breuer believed that what was unconscious were dissociated altered states, much as he observed in his patient, and that hysteria could be accounted for by the examination of these etiologically significant hypnoid states. By contrast, Freud emphasized motives, insisting that a person had to have a "reason" for keeping something out of consciousness and that this "reason" was a motive. He would have supposed that tor Anna O. it would have been disturbing to acknowledge the nature ot a close relationship to her father; therefore, she had to keep the memory of the dressing gown out of consciousness but that it persisted in some way and influenced at one point her conscious experience. From Freud's point of view, a motive was a necessary but not sufficient condition for repression. It made repression necessary, but repression itself required a mechanism of some kind for bringing about unconsciousness.

The assumption of a psychological unconscious is a distinctive and unique characteristic of psychoanalysis. However, psychology has done without this assumption before psychoanalysis and in many quarters still does. William James (1890) was scathingly critical of the notion of an unconscious and suggested a point of view involving rapidly alternating states of consciousness as in multiple personalities. Further, he denied the postulate of psychic continuity by asserting that what appears nonsensical and unaccounted for by the usual rules of thought (e. g., seeing a brown dress as blue) was likely due to some temporary aberration in the nervous system. In fact he accounted for the bizarre connections in dreams as due to deficiencies in the nutrition of the underlying neurophysiological processes. There was thus no psychic continuity, but an intrusion of physical events directly on the mental; at best there is a complex psychophysiological continuity in which different principles, working independently (e. g., nutrition of the nervous system), randomly interfere with each other. In this system there is no place for an individual, historical consideration of mental events. Modern behaviorism is essentially the heir of this viewpoint: the way to modify behavior is to extinguish a response. Extinction is a neurophysiological concept having to do with a property of the nervous system not unlike James' "nutrition". In this system there is no need for a psychological continuity. It is thus of central importance to see if the psychoanalytic assumption of an unconscious rejected by other psychological systems can receive support from other quarters not concerned necessarily with issues of psychopatholgy or psychic continuity itself. For the psychoanalyst the unconscious is an intrinsic part of his method and as such untestable by the method istelf. It would be significant if there were independent support for the existence of unconscious mental phenomena. In addition, the use of other methods may throw a new light on the nature of unconscious processes of which the psychoanalytic method itself is incapable.

Unconscious Processes and the Evoked Potential

In 1967, Libet et ah, reported that a somatosensory evoked response can be obtained at the cortex when the subject is stimulated by a subthreshold electrical current to the skin. Although the subject consistently reported no sensation the evoked potential revealed a decided cortical response (Libet, Alberts, Wright and Feinstein, 1967). The brain could show cortical responsiveness in the absence of awareness. Libet and his colleagues believed their findings showed an early, primary evoked potential which they hypothesized was associated with the subthreshold stimulus. They further speculated that consciousness as such was associated with a later, secondary component. The exact nature of the findings and the investigators' interpretation is best captured in their own words:

With subdural recordings at a point on postcentral gyrus and with the lccus for the skin stimulus matched to the referral ficm this cortical po;it, it was found that an a\erage evoked potential could be detected with stimulus strengths that weie distinctly below the lower limit for threshold-c (e. g.. consciousress). That is, evoked potentials could be detected with skin stimuli which were well belcwthe iarge of uncertainty and which never elicited any conscious sensory responses. . . . the evoked potentials picdnced by such single subthreshold-c pulses of the skin consist mainly of an initial or primary surface positive component... (1973, pp. 769-770).

It was also discovered by these investigators that a single pulse of relatively high current delivered to the thalamic postero-ventral nuclei will not produce a conscious sensory experience although a large evoked potential appears in the appropriate area of the somatosensory cortex (Libet et al., 1967). Libet concludes:

It follows from these considerations, that the preserce oremplituce of an evoked potential, particularly of the primary component, cannot be assumed to be an indicator of the occurrence or intensity of subjective sersoiy experience witl out other validation under the conditions of study... (1973, pp. 770-771).

Below we will consider other theoretical implications of Libet's work with respect to the conditions for conscious sensory experience itself.

In the same year as the Libet findings appeared, Riggs and Whittle (1967) published some provocative observations which have been largely ignored in the interim. The investigators were interested in finding out what accounted for the subjective disappearance of a stabilized retinal image. When eye movements are prevented experimentally from shifting the retinal image from one set of receptors to another there is a gradual fading and often a complete disappearance of a previously clear pattern. Their intent was to discover if disappearance of the visual pattern would be accompanied by a loss of electrical responses at the retina and at the occiput. The cortical evoked potential was obtained by the use of bipolar occipital electrodes, each placed 2.5 cm. to the right of the midline, one at the level of the inion and the other 6 cm. higher on the head. The stimulus was made up of black and white stripes alternating 47-4 times per second. Their surprising finding was that both the retinal and occipital EP's revealed no differences in latency, amplitude or rate under stabilized and unstabilized viewing. Peripheral and central electrical activity remained constant whether or not the subject was aware of the stimulus. As with Libet's finding, awareness and cortical evoked activity are not necessarily correlated. Riggs and Whittle summarized their findings as follows:

The results with image stabilization may be summarized with a statement that no diminution was found to occur in the electrical responses evoked in either the retina or the occipital cortex. This generalization holds despite the fact that the subject reported a marked difference in the subjective appearance of the stripes under the two conditions. The stabilized condition produced a fading of the image that sometimes amounted to virtually complete disappearance. The persistence of the electrical activity was also found under a variety of conditions of stimulation with respect to frequency, intensity and wavelength (p. 445)... it is... natural to suppose that... signals are blocked at levels higher than that from which we record (p. 449).

Some years later Riggs (1976) speculated that these results pointed to a higher centre for consciousness. His speculations took him back to Descartes and the pineal gland. In an amazing leap over the centuries, Riggs vaulted over the Freudian concept of the unconscious for which the notion of cortical activity without consciousness would constitute no surprise.

In an earlier investigation of the stabilized retinal image relying, however, on theEEG and not on evoked potentials Lehmann et al., (1965) found convincing evidence that the subjective disappearance of the image was very likely brought about by a centrally paced signal taking the form of bursts of alpha. Just prior to the time the subject reports the image as faded or disappeared, there are bursts of alpha occurring in the EEG. The investigators conclude that "periodic fading of the stabilized retinal image is the result of central changes" (1965, p. 342). When we combine the findings from these two studies dealing with the stabilized retinal image, we can hypothesize that movement of the stimulus across the retina is a necessary condition for awareness of the stimulus but not of its information processing.

In a series of studies bearing directly on the relationship between the evoked potential and the psychoanalytic idea of the unconscious, Shevrin and his co-workers have reported a number of findings in which the VEP was used (Shevrin and Rennick, 1967; Shevrin and Fritzler, 1968a, 1968b; Shevrin, Smith, and Fritzler, 1969, 1970, 1971; Shevrin, Smith, andHoobler, 1970; Shevrin, 1973). The paradigm for this research was based on the phenomenon of subliminal perception (Dixon, 1971). A pair of visual stimuli matched for area, color, shape and contour but differing in attributable meaning was presented to subjects at subliminal speeds (1 ms; 3mlam luminance). It was hypothesized that the experimental stimulus, a picture of a pen pointing at a knee, would evoke a subliminally greater amplitude of response than the matched control, which was made up of abstract shapes lacking any meaningful characteristics. This hypothesis was based on the assumption that an intrinsically interesting stimulus would excite greater attention; a number of evoked potential studies had already shown that attention was associated with increased evoked potential amplitude (Hillyard, 1977). It was found that a bipolar electrode display, occiput (2.5 cm. to the left and above the inion) to frontal (at the midline close to hairline), revealed a consistent difference at subliminal levels in the hypothesized direction. The differentiating amplitude was a positive going component peaking at approximately 160-180 ms. poststimulus. Note should be taken of the fact that this component is a ]late, secondary amplitude and does not fit with the Libet finding of an early primary component as the one associated with subthreshold stimulation. Nevertheless, in the absence of any awareness of the stimulus the subject showed a clear, differentiating evoked potential.

In addition to the subliminal findings the same stimuli were presented at 30 ms. under the same illumination. Subjects were able to differentiate the stimuli; although amplitudes and latencies were greatly increased, the same difference between the two stimuli was found. Thus at both sub- and supraliminal exposure times consistent amplitude differences were found that accorded ]with previous findings on attention.

The experimental paradigm employed by Shevrin and his colleagues was distinguished by a unique assessment of associative thought processes instigated by the subliminal stimuli. This was important for two reasons: 1) the psychoanalytic concept of the unconscious was not simply a physiological one, but distinctly psychological - unconscious mental processes were involved, 2) as Uttal (1965) has pointed out the mere presence of a neurophysi-ological response may or may not be of cognitive relevance: the evoked potential may simply signal the existence of a process but not actually be related to its encoding. For both of these reasons it was necessary to show that the experimental stimulus elicited subliminal cognitive effects.

Fortunately, a considerable body of research on subliminal perception (Dixon, 1971) had already established that associations can be used to demonstrate subliminal cognitive effects. In particular, it has been found that the association networks activated by a subliminal stimulus are broader and more loosely connected than the networks activated by the same stimulus presented supraliminally. It was thus possible to hypothesize that the picture of the pen and knee would elicit more associations subliminally than supraliminally, while no such difference should be found for the control stimulus. A reliable standard procedure for identifying such associations was developed. The hypothesis was supported in several experiments (Shevrin and Luborsky, 1958; 1961; Spence and Holland, 1962; Spence, 1964). More associations related to the experimental stimulus appeared sub- rather than supraliminally. Thus, it was demonstrated that the subliminal stimulus was not only signalled by a differentiating evoked potential but that the evoked potential was associated with cognitive effects related to the meaning of the stimulus.

As described in the opening section of this paper, the psychoanalytic concept of the unconscious requires that a) there be evidence of some difference in the nature of the unconscious thought processes as compared to conscious thought, b) that there be evidence of repression or defensive activity. With respect to the first point theShevrin investigations already showed, consistent with other subliminal studies, that the association networks would be broader and looser in the subliminal condition. The experimental stimulus was constructed to permit another level of thought organization to be identified: this new level was based not on the conceptual use of language but on the concrete, phonemic use of words. Luria and Vinogradova (1959) have demonstrated that the semantic field changes from the conceptual to the phonemic under the influence of a mild sedative in normal subjects and in the usual functioning of the mentally retarded. Freud long ago posited that in a variety of conditions, psychological and normal, (dreams, the formation of neurotic symptoms, psychotic thinking, in early childhood, slips of the tongue, humor) the conceptual character of language is often lost and is replaced by concrete, phonemic association links. These phonemic links were instances of what he referred to as the primary process in ideation. In the picture of the pen and knee it is possible to trace in the association process words conceptually related to the stimulus (e. g., pen: ink, paper; knee: leg, bone), but also phonemically related (e. g., pen: pennant, happen; knee: any, neither). Furthermore, the two words (pen and knee) form a new word (penny) unrelated conceptually to its constituents. It was found that conceptual and phonemic associations correlated with different parameters of the subliminal evoked potential. Conceptual associations correlated with a differentiating amplitude, while the clang level associations correlated with bursts of alpha-like activity approximately 1500 ms. post-stimulus. Primary process and conceptual or secondary process cognitive activity appear to be related to different parameters of the subliminal evoked potential. It was tempting to interpret this difference as paralleling the psychoanalytic distinctions between preconscious and dynamically unconscious mental processes insofar as the former are considered to be more characterized by conceptual thinking and the latter by primary process ideation.

One last finding from these investigations of some relevance to the distinction between a descriptive unconscious (preconscious) and dynamic unconscious remains to be described. It was found, as hypothesized, that subjects rated as highly repressive on the basis of the Rorschach had smaller subliminal evoked potentials to the experimental stimulus and associated fewer stimulus related words. There was thus some reason to believe that repressive forces may have been at work although the evidence was still indirect and inferential (Shevrin and Fritzler, 1968b; Shevrin, Smith, and Fritzler, 1969; 1970).

In an effort to replicate the Shevrin findings Schwartz and Rem (1975) introduced a number of differences from the original experimental paradigm: 1) the exposure time was increased to 3 ms. above the absolute threshold, 2) stimuli were presented in pairs rather than in blocks, 3) the stimuli within the pairs were presented in a fixed time interval, one second apart, 4) no association data were obtained, 5) the task was changed from one requiring attention to a single stimulus presentation and description of what was seen to a discrimination task in which the subject was asked to judge whether the stimuli within a pair were the same or different. Subjects were presented with four different types of pairs: experimental-control; control-experimental; experimental-experimental; control-control. The same stimuli were used as in the Shevrin experiment. There were two findings, one negative and one positive. Schwartz and Rem did not find the amplitude difference reported by Shevrin and his co-workers. However, they did find that cross-correlations between the stimuli within, pairs showed that the evoked potentials were significantly smaller for the experimental-control, control-experimental pairs than for the control-control, experimental-experimental pairings. These findings indicated that even though subjects were not making a better than chance discrimination (at 3 ms.), the evoked potentials were still differentiating between the experimental and control stimuli. Schwartz and Rem interpreted these differences as likely due to the minor structural differences between the stimuli rather than to the differences in content or meaning. In the absence of any means for determining meaning (i. e., no association method) this interpretation was open to doubt.

Shevrin (1975) pointed to a number of difficulties in the Schwartz and Rem experiment: in addition to significant differences in procedure, there were also some statistical difficulties that might account for the failure to find an amplitude difference. Schwartz and Rem, in a subsequent re-analysis of the data, found that indeed there was an amplitude difference between the experimental and control stimulus but that it was in the opposite direction from that reported by Shevrin: the amplitude for the control stimulus was larger than for the experimental stimulus. It is of some interest to note that Rietveld (1966) has reported that with visual stimuli attention may have an anomalous effect on evoked potential amplitude depending on the difficulty of the task: when the task is difficult the evoked potential associated with the attended to stimulus diminishes, while in the simpler task it increases over the evoked potential to the unattended to stimulus. By presenting the stimuli in pairs, one second apart, Schwartz and Rem may have significantly increased the difficulty in the attentional task and thus reversed the amplitude relationships found by Shevrin for the experimental and control stimuli. In any case, the null hypothesis cannot as yet be rejected that no difference exists between the visual evoked potential amplitudes for the experimental and control stimuli presented either below the detection threshold (1 ms.) or below the discrimination threshold (3 ms.).

There is another group of visual evoked potential studies that cast some light on the relationship between the VEP and consciousness, although none of these studies was undertaken with this end in mind. Begleiter, Porjesz, Yerre, and Kissin (1973) employed an ingenious experimental paradigm to investigate the effect of expected stimulus intensity on the VEP. Following a training session in which subjects were asked to anticipate a bright and dim flash signalled by a different tone, medium intensity flashes were distributed randomly among bright and dim flashes depending on the preparatory signal. Half the medium flashes were preceded by the "bright" flash signal and half by the "dim" flash signal. Electrodes were placed at the midline 2.5 cm. above the inion and at the vertex. The left earlobe was used as reference. It was found that the evoked potentials recorded at the vertex were correlated in amplitude with the expected stimulus intensity (bright or dim) and not with the actual stimulus intensity (medium); while n о difference was found at the occiput. The investigators interpreted these results to mean that a memory of a previous stimulus may be released by an expectation. Why this was not found at the occiput was left unexplained. The difference between vertex and occiput is of more than passing interest:" appears that at the occiput, that is, at the sensory receiving area an accurate evoked potential is obtained, while at the vertex an expected (and inaccurate) EP is obtained. What was the subject "seeing" - the accurately recorded stimulus at the occiput or the expected stimulus at the vertex? Unfortunately no data is provided as to what the subjects actually reported seeing. The entire context of the experiment leads one to believe that they "saw" the expected stimulus. If so, then the occipital EP remained inaccessible to awareness and thus unconscious.

In a subsequent experiment, Begleiter and Porjesz (1975) employed a procedure that left little doubt that the subjects saw a bright or dim light and that their vertex evoked potentials were correlated with this perception even though they were responding to a flash of medium intensity. Subjects were rewarded for correct discrimination of actual bright and dim lights among which the medium flashes were interspersed although subjects were led to expect only bright or dim flashes. Thus, they reported seeing a bright or dim flash when only a medium flash was presented. In this experiment no occipital electrode was employed so that replication of the occipital finding was not possible.

Three other experiments support the interpretation that evoked potential activity in the front of the brain is associated with what the stimulus "means" to the subject and what he may in fact be aware of. Johnston and Ches-ney (1974) presented subjects with a visual stimulus which could be seen as the number thirteen or the letter В depending on the context (e. g., flanked by the numbers 17 and 12 or by the letters P and E). Following a familiarizing series in which either letters or numbers were presented in random order, in an experimental session subjects were told that they would be seeing either numbers or letters and that their task was to name the numbers or letters as rapidly as possible. With the exception of one subject the remaining subjects had electrodes either 2.5 cm. above the inion at the midline or 2.5 cm. above the nasion. For one subject electrodes were placed at both locations. Differences between the same stimulus perceived as either letters or numbers were only found at the frontal location. Again, correct information available ooccipitally had no effect on what the subjects reported. Some process must keep this information from affecting higher levels of cognition and apparently awareness as well.

Brown, Marsh and Smith (1973) found that the same word, "fire" evoked a different auditory potential over the speech area when it was used as a noun (e. g., "Sit by the fire") as compared to when it was used as a verb (e. g., "Ready, aim, fire!"). No such difference was found over the right hemisphere.

In an earlier study Weinberg, Walter, and Crow (1970) investigated the evoked potential to actual and imagined stimuli with the use of electrodes implanted in orbito-frontal and cingulate cortex as well as on the surface of the superior frontal cortex. They found that the evoked potential to actual flashes and clicks were similar to the evoked potentials when the subjects expected the stimuli but were in fact not stimulated. There was some evidence that what the investigators called the emitted (as contrasted to evoked) potential had a shorter latency by some 25 ms. They conclude:

From what we know about the nervous system there is no reason to assume that the experience of an image depends on the establishment of those processes originally involved in the registration and coding of the stimulus. However, an interesting implication of our observations is that an assumption of this nature might be a reasonable working hypothesis. The fact that emitted potentials correlate well with potentials which are evoked by stimuli suggests that the underlying electrophysiological processes which are initiated by stimulus presentation may be reactivated under appropriate conditions of motivation and set even in the absence of those stimuli (p. 8).

Again in this study no electrodes were placed in the sensory receiving areas, so we do not know if there would have been no emitted potentials found there.

There are two reports in the literature dealing with the evoked potential and hysteria. Hernandez-Peon, Chavez-Ibarra, and Aguilar-Figueroa (1963) obtained evoked potentials from a fifteen-year-old girl suffering from a glove and sleeve analgesia and thermoanesthesia of the left arm. Electrodes were located in the midline of the vertex and one each in the parietal region at a midpoint between the ear and the midline. Whereas the unaffected right arm showed a clear evoked potential response to a pin prick, the hysterically affected left arm showed no clearly evident evoked potential, (although inspection of the curves Isee figure 2, p. 890] supports a more ambiguous interpretation insofar as some low amplitude activity appears to be present in the evoked potential for the left arm). Nevertheless, activity in the right arm evoked potential is considerably greater in amplitude. Following the administration of a mild barbiturate (Kemithal) the left arm evoked potential now showed a large amplitude. Unfortunately, it is not noted in the article wheth. er the evoked potential was from the parietal, vertex or both electrode sites. In view of the finding reviewed above concerning the difference between vertex and occipital leads it would have been good to know from which electrodes the findings were reported. If the potentials were recorded at the parietal electrodes (paralleling the occipital electrodes for visual stimuli) then it would mean that the inhibition was at an earlier point in the afferent pathway. If the inhibition were at the vertex it would mean that the inhibition occurred between the sensory receiving areas and other cortical areas. Further it might clarify with what aspect of the evoked potential consciousness itself is associated. Shevrin, for example, found that a subliminal stimulus in normal subjects with mild repressive tendencies elicited a reduction in amplitude of a later secondary component. The electrode display was bipolar - occipital to frontal - and thus makes it difficult to separate out contributions from the different cortical areas. All that can be said is that in both studies a reduction in evoked potential activity was found in the presence of repressive characteristics.

Of special importance is a question concerning the subject's awareness, of the stimulus when the mild barbiturate lifted the electrical inhibition. The answer to this question is of some importance because of Libet's findings that a primary amplitude is associated with an absence of awareness for a somatosensory stimulus, leading Libet to hypothesize that a secondary component may be associated with awareness; on the other hand, Shevrin found that a secondary amplitude was present in the absenсе of awareness for a visual stimulus. Further, a number of investigations (Johnston and Chesney, 1974; Begleiter et al., 1973; Begleiter and Porjesz, 1975; Weinberg et al., 1970; Brown et al., 1973) indicate that a late secondary component in the anterior of the brain is associated both with meaning and awareness. If Hernandez-Peon's hysterical subject was unaware of the pin prick in a mildly sedated state while her evoked potential showed a clear secondary wave this would then be comparable to the Shevrin finding that a secondary component can be present in the absence of awareness. Note should also be taken of the fact that in the Shevrin repression studies for repressive subjects the secondary component did not disappear but was diminished relative to less repressive subjects. We may be dealing with a threshold factor: a secondary amplitude below a certain strength would not elicit consciousness, or the secondary component must last for some critical time duration. This latter alternative would fit well with the Libet findings (to be discussed below) on the existence of a cortical factor in consciousness that is a function of stimulus duration. The tentative hypothesis would be that the presence of a secondary component in and of itself may not coincide with consciousness, but that its persistence needs to be of a certain critical duration. In the studies on EP correlates of meaning the stimuli were all supraliminal and thus the time duration of the secondary amplitude would be beyond the critical time postulated.

In a second study Potts and Nagaya(1969) investigated the visual evoked potential in three cases of hysterical amblyopia. They found that the visual evoked potential to a red stimulus subtending a visual angle of 0.6 degrees elicited a normal evoked potential in both the normal and the hysterically affected eye. From the context it would appear that the visual evoked potentials for the hysterical subjects came from the occipital to parietal electrodes although this is not clearly indicated either in the accompanying figure or in the text. Here again it would have been interesting to have a reported data from the Fz and Cz electrodes which were also employed in this study. In the actual amblyopic patients their frontal leads showed less difference between the normal and amblyopic eye than at the more posterior leads, suggesting that what little sensory input was received at the occiput was nevertheless relayed forward to the frontal regions and underwent some further processing there. Would the hysterical patients have shown little such frontal activity for the pseudo-amblyopic eye?

This review will conclude with a consideration of Libet's intriguing findings concerning the adequate cortical stimulus for a conscious experience. Libet (1973) has been able to demonstrate that in order for a subject to experience a somatosensory stimulus a critical stimulus pulse duration is necessary. For most subjects it is in the 0-5 to 0-6 second range although the pulse duration may be as little as 0-4 seconds and as long as 1.0 seconds. Further, the pulse duration value is relatively constant for a given subject. Since Libet had already demonstrated that a somatosensory evoked potential can be detected at the cortex with single pulses well below the threshold of conscious experience, it would appear reasonable to suppose that a peak duration is necessary for consciousness itself to appear. The interesting paradox emerges that a single pulse lasting 0.1 second delivered to the skin can evoke a conscious experience while it takes a pulse lasting at least 0.4 seconds to elicit a conscious experience at the same sensory site with direct cortical stimulation. Is it possible that a natural stimulus has a different requirement for consciousness as compared to the artificial stimulus delivered directly to the brain? Libet has demonstrated that this is not the case. First, he cited the well-known fact that even though the peripheral stimulus may be brief its central effects may be measured in hundreds of milliseconds. Secondly, he employed a novel application of the masking technique to show that indeed the later activity evoked by the peripheral stimulus is necessary for conscious experience to occur. The masking procedure is based on a well-known phenomenon that when a second stronger stimulus follows a previous weaker stimulus within a certain critical time interval the first stimulus will not be perceived although it is above threshold strength for conscious experience when presented alone. Libet used as his first stimulus a single pulse delivered to the skin which was at threshold strength for conscious experience. A second or masking stimulus was delivered directly to the somatosensory cortex in the region including the area stimulated peripherally. The direct cortical stimulus has the advantage of not confounding peripheral and central factors as is often the case in the usual masking experiment where both stimuli are delivered peripherally. The direct cortical stimulus consisted of a brief train lasting about 0,5 seconds of pulses presented at 60 pulses per second. The peak current was 1-1,5 times the strength needed to elicit a conscious experience. Libet found that for most subjects if the interstimulus interval was between 125-200 ms., and for one subject up to 500 ms., consciousness of the first stimulus was absent. The subject was only aware of the second stimulus. Further, the cortical stimulus had to last a certain minimum duration; single cortical pulse did not work. Libet concluded:

Retroactive masking of the peripheral (Si) sensation by a later stimulus (S₂) directly to somatosensory cortex, with S₁-S₂ intervals of up to 200 ms. or more, could only be due to interference with some late components of the brain responses to S₁ that are necessary for the mediation of a conscious sensory response.... the extent of retroactive masking that could be demonstrated provides powerful support for the hypothesis that a relatively long period of suitable cerebral activations is a necessary feature of the experiences (1973, p. 775).

If we combine Libet's two findings, the first showing that a cortical primary response is present for a subliminal stimulus and the second that a cortical pulse train is necessary for conscious experience, we can conclude that somatosensory stimuli can register cortically without consciousness and that conscious experience itself is a function of some additional system having its own properties. Of critical importance for consciousness is the train duration. As suggested above, this train duration may be more critical for consciousness than the presence of a secondary amplitude which persists for less than the critical time.

Summary and Conclusions

Although the studies reviewed are few and replication is largely absent, the evidence is generally favorable for the psychoanalytic concept oi the unconscious and, in particular, supports, the following propositions:

1) As predicted by psychoanalytic theory, cognitive activity can go on without benefit of consciousness. Between conscious experience and neuro-physiological processes there is no leap as assumed by James, for there are unconscious cognitive processes that can be detected by a neurophysiological indicator, the evoked potential. The assumption of unconscious thought processes helps us to explain phenomena presented to the psychoanalyst from the early observations of Breuer and Freud in the Studies on Hysteria to the modern application of the free association method in psychoanalytic treatment.

2) Unconscious cognitive activity can be divided into two classes: cognitive activity that is no different in organization from conscious cognitive activity, referred to in psychoanalysis as the preconscious; and cognitive activity that is characterized^by a different organization from conscious cognitive activity, referred to in psychoanalysis as the dynamic unconscious. There is some evoked potential evidence to support this proposition; namely, the work of Shevrin and his colleagues on the differences between conceptual and clang associations which apear to be related to quite different parameters of the VEP.

3) The dynamic unconscious implies the existence of defenses that actively inhibit the appearance in consciousness of certain contents. Here too there is support from evoked potential studies. Shevrin and his colleagues have shown that repressive subjects had a smaller VEP to subliminal stimuli and; associated fewer stimulus related words. Hernandez-Peon et al., were able to demonstrate a comparable reduction in the somatosensory evoked potential in the left arm of an hysterically anesthetized girl. Potts and Nagaya have demonstrated that in hysterical amblyopia the occipital evoked potentials were indistinguishable for the affected and non-affected eyes, strongly suggesting that the inhibition occurred anteriorly, very likely in frontal areas.

4) Finally, the psychoanalytic view leads to the conception of consciousness as a separate system having its own conditions of activation. The full range of cognitive activity can go on without benefit of consciousness. Consciousness itself is activated under certain special conditions. Libet's work speaks directly to this point. Only when a stimulus can activate a central cortical process lasting a certain minimal time will consciousness occur. Further, that the presence of a later secondary component need not be accompanied by consciousness unless it lasts a certain critical time. It need only be added that a variety of factors may keep this activation from taking place: a) a sec-cond stronger stimulus interfering with the later activity of a previous stimulus, very likely a common and normal occurrence, b) a subliminal stimulus too weak to elicit the necessary central activity but capable of activating the full scale of cognitive activity, c) some centrally controlled inhibition likely based on motivation as in hysterical repression. Investigations of the latter condition should make it possible to discover the nature of such defensive mechanisms as repression.

Bibliography

Begieiter, H. and Porjesz, В., "Evoked Brain Potentials as Indicators of Decision-Making", Science, vol. 187, pp. 754-755 (1975).

Begleiter H., Porjesz, В., Yerre, C., Kissin, В., "Evoked Potential Correlates of Expected Stimulus Intensity", Science, vol. 179, pp. 814-816 (1973).

Breuer, J. and Freud, S., Studies on Hysteria, Standard Edition, London: The Hogarth Press, vol. II (1895).

Brown, W. S.. Marsh. J. Т., and Smith, J. C, "Contextual Meaning Effects on Speechevoked Potentials", Behavioral Biology, vol. 9, pp. 755-762 (1973).

Dixon, F., Subliminal Perception: The Nature of a Controversy, London: McGraw-Hill (1971).

Hernandez-Peon R., Chavez-Ibarra, G., and Aguilar-Figueroa, E., "Somatic Evoked Potentials in one Case of Hysterical Anaesthesia", Electroencephalography and Clinical Neurophysiology, Amsterdam: Elsevier Publishing Co.. vol. 15, pp. 889-892 (1933).

Hillyard S. A., "Electrophysiological Assessment of Attentional Processes in Man", (Paper presented at Conference on Event Related Brain Potentials in M a n, April 28, 1977, Washington D. C).

James, W., The Principles of Psychology, New York: Henry Holt and Co., vol. I (1890).

Johnston, V S., and Chesney, G. L., "Electrophysiological Correlates of Meaning", S с i e n с e, vol. 186, pp. 944-946 (1974).

Lehmann, D., Beeler, G. W., Jr., and Fender, D. H., "Changes in Patterns of the Human Electroencephalogram During Fluctuations of Perception of Stabilized Retinal Images", Electroencephalography and Clinical Neurophysiology, vol. 19, pp. 336-343 (1965).

Libet, В., "Electrical Stimulation of Cortex in Human Subjects and Conscious Sensory Aspects", Handbook of Sensory Physiology, vol. 2, pp. 744-790 (1973).

Libet, В., Alberts, W. W., Wright, E. W., Jr., Feinstein, В., "Responses of human somatosensory cortex to stimuli below threshold for conscious sensation". Science, vol. 158, pp. 1597-1600 (1967).

Luria, A. R., and Vinogradova, O. S., "An Objective Investigation of the Dynamics of Semantic Systems", Gen. Psychol., vol. 50(2), pp. 89-105 (May, 1959).

Potts, A. M., Nagaya, Т., "Studies on the Visual Evoked Response: III. Strabismus Amblyopia and Hysterical Amblyopia", Dосum. Ophthal., vol. 26. pp. 394- 402 (1969).

Rapaport, D., "The Scientific Methodology of Psychoanalysis", Collected Papers of David Rapaport, New York: Basic Books, Inc., Ed. by M.M. Gill, vol. 14, pp. 165-220 (1967).

Rietveld. W. J., TORDOIR, W. E. M., and HAGENOUW, J. R. В., "Influence of Atten-tiveness of Vigilance Task Difficulty, and of Habituation on Cortical Evoked Responses and on Artifacts", Acta Physiol. Pharmacol. Neerl., vol. 14, pp. 18-37 (1966).

Riggs , L., "Human Vision-Some Objective Explanations", Am. Psychol., vol. 31(2), pp. 125-134 (1976).

Riggs, L. A., Whittle, P., "Human Occipital and Retinal Potentials Evoked by Subjectively Faded Visual Stimuli", Vision Res., Great Britain: Pergamon Press, vol. 7, pp. 441-451 (1967).

Schwartz, M., and Rem, M. A., "Does the Averaged Evoked Response Encode Subliminal Perception?", Psyhophysiology, vol. 12, pp. 390-394 (1975).

Shevrin, H., and Lueorsky, L., "The Measurement of Preconscious Perception in Dreams and Images: An Investigation of the Poetzl Phenomenon", Journal of Abnormal and Social Psychology, vol. 56, pp. 285-294 (1958).

Shevrin, H., and Luborsky, L., "The Rebus Technique: A Method for Studying Primary-process Transformations of Briefly Exposed Pictur?s", Journal of Nervous and Mental Disease, vol. 133, pp. 479-488 (1961).

Shevrin, H., and Rennick, P., "Cortical Response to a Tactile Stimulus During Attention, Mental Arithmetic and Free Associations", Psychophysiology, vol 3, pp. 381-388 (1967).

Shevrin, H., and Fritzler, D., "Visual Evoked Response Correlates of Unconscious Mental Processes", Science, vol. 161, pp. 295-298 (1968a).

Shevrin, H., and Fritzler, D., "Brain Response Correlates of Repressiveness", Psychоlоgiсal Reports, vol. 23, pp. 887-892 (1968b).

Shevrin, H., Smith, W. H., and Fritzler, D., "Repressiveness as a Factor in the Subliminal Activation of Brain and Verbal Responses", Journal of Nervous and Mental Disease, vol. 149, pp. 261-269 (1969).

Shevrin, H., Smith, W. H., and Fritzler, D., "Subliminally Stimulated Brain and Verbal Responses of Twins Differing in Repressiveness", Journal of Abnormal Psychology, vol. 76, pp. 39-46 (1970).

Shevrin, H., Smith, W. H., and Hoobler, R., "The Direct Measurement of Unconscious Mental Processes: Average Evoked Response and Free Association Correlates of Subliminal Stimulation", Reprinted from the Proceedings, 78th Annual Convention, American Psychoanalytic Assn., Miami, Florida (Sept, 6, 1970).

Shevrin, H., Smith, W. H., and Fritzler, D., "Average Evoked Response and Verbal Correlates of Unconscious Mental Processes", Psychophysiologv, vol 8 (2), pp. 149-162 (March, 1971).

Shevrin, H., "Brain and Free Association Indicators of Developing and Decompensating Repressions", Presented on panel, Research Perspectives on Treatment and Diagnosis (Chairman: Howard Shevrin), Bulletin of the Menninger Clinic, vol. 35 (6) (Nov., 1971).

Spence, D. P., and Holland, B.f "The Restricting Effects of Awareness: A Paradox anp Explanation", J. Abnorm. So с Psychol., vol. 64, pp. 163-174 (1962). UTTAL, Win. R., "Do Compound Evoked Potentials Reflect Psychological Codes?", Psychоlоgiсal Bulletin, vol. 64 (6), pp. 377-392 (1965).

Weinberg, H., Walter, W. Grey. Crow, H. J., "Intracerebral Events in Humans Related to Real and Imaginary Stimuli", Electro., vol. 29, pp. 1-9 (1970).