Research on the Relation of Psychoanalysis and Neuroscience

This report, which inaugurates a new format for JAPA, is presented to honor the work of Howard Shevrin, who since 1962 has contributed to the psychiatric, neuroscientific, and psychoanalytic literature on the interface between clinical meaning and empirical science. Highlights of his work include his 1968 publication in Science, “Visual Evoked Response Correlates of Unconscious Mental Processes” (Shevrin and Fritzler 1968), his 1993 plenary address to the American Psychoanalytic Association, “Is Psychoanalysis One Science, Two Sciences, or No Science at All? A Discourse among Friendly Antagonists” (Shevrin 1995), and distinguished work on primary process, memory, conscious and unconscious processes, and the value and power of a comprehensive psychoanalytic theory. Shevrin’s Dream Interpreters: A Psychoanalytic Novel in Verse (2003) moves us dramatically through the inner realms of his characters, giving his readers an introduction to the psychoanalytic experience.

Some psychoanalysts say that interdisciplinary work such as Shevrin’s cannot address clinical meaning. Some cognitive scientists say that this kind of work cannot rise to the level of neuroscientific proof. The research reported here counters both statements.

Howard Shevrin and Michael Snodgrass, both of the University of Michigan (where Shevrin, a member of the Michigan Psychoanalytic Institute, is director of the Ormond and Hazel Hunt Laboratory for the Study of Conscious and Unconscious Processes), led an animated discussion of their research on unconscious processes. For many years Shevrin and his colleagues have engaged in empirical studies of psychoanalytic theories. They have quantitatively demonstrated that unconscious processes exist in the brain and that these processes have cognitive, affective, and motivational properties. Further, they have demonstrated that some of these unconscious processes are subject to inhibition—repression—that is both individual and motivated unconsciously. They have also provided empirical assessment of the two different modes of mental functioning postulated by Freud (primary and secondary process thinking). Their research convincingly instructs us on how “meaning” can be studied scientifically. This kind of integration is not easy to accomplish and, when it can be achieved, is especially valuable; the work presented here is complex to describe, but rewarding in its powerful implications.

While Freud and other psychoanalysts have argued that the results of the psychoanalytic method provide an adequate test of its theories, others, notably the philosopher of science and critic of psychoanalysis Adolf Grünbaum, argue that empirical support must come from methods independent of the clinical setting. Grünbaum’s critique of Freud’s tally argument (1984, 1986a,b), calling for extraclinical proofs, still shapes the debate. As the afternoon meeting commenced, Shevrin announced with pride that Grünbaum had conceded that he and his colleagues successfully satisfied the standards for research capable of scientific verification.

Following this exciting announcement, the afternoon’s discussion focused primarily on four significant findings:

Preparing the audience to understand their research, Shevrin and Snodgrass described some of the reasoning that informs it. To adequately research unconscious processes, it is useful to think in terms of the “equivalence” of mind and brain processes, rather than to view them as “either/or.” (For one reporter, the notion of equivalence brought to mind the philosophical theory of dual-aspect monism, which suggests that the mental and the physical are two aspects of, or perspectives on, the same substance; see Solms and Turnbull 2002, pp. 56–58.) Thus, while neuroscientific investigation of the neural correlates of unconscious processes is indeed independent of the clinical setting (satisfying one of the conditions for valid research), neuroscience has all too often ignored the roles of unconscious conflict and personal meaning. In so doing, neuroscientific research ignores central components of psychoanalytic theory, particularly the role of dynamic repression. Shevrin and his colleagues have been able to address this central “aspect” of mind using their notion of equivalence.

In other words, the ways the mind works (as investigated through psychoanalysis) and the brain works (as investigated through neuroscience) show correlations, and neither can be fully understood in the absence of the other. This bold statement distinguishes Shevrin and Snodgrass’s work from that of reductionists who argue that mind and brain are separate or that one perspective can be reduced to the other. Shevrin and his colleagues have developed a tripartite research paradigm that integrates psychodynamic hypotheses, cognitive subliminal processes, and psychophysiological measures ¹ to address both psychoanalytic and neurodynamic matters.

Shevrin reminded us that minds demonstrate enormous fluidity; they are always active and always changing. The relationship between conscious and unconscious processes in the mind is fluid, personal, and highly variable. Further, the brain demonstrates a high baseline activity that supports total mental activity, and every mental activity involves the entire brain. (One could say that every conscious act involves an unconscious one and vice versa.) The intrinsic activity of the brain—neuron-to-neuron connections—represents by far the greatest share of brain activity. Any psychological event always takes place within the context of the entire brain’s activity. The context of meaning in which brain activity occurs is as important as the context of the brain activity in which subjective experiences are processed.

The researchers sought evidence for a cause-and-effect relationship between unconscious conflict and conscious symptom experience. Whereas Shevrin’s studies date back to the 1960s, the presentation to this discussion group focused on the most recent findings—those in support of the causal effect of repression on conscious processes.

The first experiment described involved subjects who met DSM-III-R criteria ² for social phobias. Shevrin and his colleagues set out to explore specific psychoanalytic propositions regarding social phobias: that symptoms are associated with unconscious conflict unique to each individual, and that repression operates outside of conscious awareness to suppress awareness of these conflicts such that there is no obvious connection between the unconscious conflict and the conscious symptoms.

The research was conducted using the following paradigm: Each subject was given a thorough psychodynamic evaluation including psychological testing and clinical interviews. (Psychoanalysts hold that a psychodynamic interview can explicate causal connections between conflict and symptom that prove to be highly specific for each individual.) The transcripts of the interviews were evaluated separately by clinical judges who wrote psychodynamic formulations for each subject. Next, a specific unconscious conflict was identified for each subject. On the basis of the psychodynamic formulations, judges selected words and short phrases that represented two categories: (1) words that best reflected the patient’s conscious experience of his symptom; (2) words that best reflected the putative unconscious conflict. Word lists were then created to contain words that reflected conscious and unconscious knowledge, as well as control words with universal connotations of goodness or badness.

By way of illustration, Shevrin presented the case of Mr. C., which is elaborated in detail in Conscious and Unconscious Processes, written with James A. Bond, Linda A. W. Brakel, Richard K. Hertel, and William J. Williams (Shevrin et al. 1996). The patient, a male college student, suffered from an acute onset of social phobia. Presenting as a frightened and nervous young man, he described his inability to eat in a restaurant where others were present. As he described his difficulty swallowing foods in public, the subject elaborated on when the symptom had first developed. He recounted sitting with a couple of male friends in his college dormitory cafeteria. One of the friends sliced open a piece of fish, from which emerged a foul odor. Mr. C., revolted, feared he would vomit; his problems with eating and swallowing in public immediately ensued. He also described headaches, shortness of breath, racing heartbeats, and nausea, all of which disappeared as soon as he left the cafeteria (though the fear of swallowing in public persisted).

Psychodynamic interviews revealed that the young man was struggling with an unconscious conflict around yielding to a homosexual seduction. (Later information confirmed that this young man had been sexually “seduced” by an uncle in childhood.) A dream that occurred during the course of the interviews further linked the unconscious homosexual wishes and his fear of swallowing in public.

From transcripts of the interviews, words and phrases associated with this subject’s particular conflict were selected. Conscious symptom words/phrases included rotten fish, shorter breaths, swallowing, and nauseous. Words/phrases associated with unconscious conflict included stab, John, ripped apart, and men hugging.

According to psychoanalytic theory, here confirmed by neuroscientific research, one would anticipate that evoking the unconscious conflict (by presenting words associated with the subject’s presumed unconscious conflicts) would be met with an inhibitory (repressive) response, and indeed this is what the research demonstrated. The findings of Shevrin’s team demonstrated the dynamic action of unconscious processes.

Using a tachistoscope, words were presented to each subject for a subliminal duration (one millisecond) or for a supraliminal duration (30 milliseconds). At the one millisecond exposure, subjects have no awareness of having seen a stimulus. EEG recordings measured the electrical responses known as event-related potentials (ERPs) that occurred in response to the different word presentations. The findings were subjected to an analysis that quantified the relative contribution of various frequencies to the EEG signal. There is a rich neuroscientific literature supporting the link between the power within a particular frequency range and a specific emotional or cognitive state of function. The alpha frequency (8–13 cycles per second) serves an inhibitory function of brain activity in both normal and pathological activities.

Alpha power—a measure of the frequency associated with inhibition—was increased when words associated with unconscious conflict were presented subliminally. Here the researchers identified a physiological correlate of repression. Their research indeed demonstrated increased alpha power in response to the subliminally presented conflict words; this effect was not present when the conscious symptom words were presented subliminally. Using a sophisticated time-frequency analysis, it was possible to correctly categorize unconscious conflict words versus conscious symptom words based on differential patterns of brain response. Increased repression/inhibition was demonstrated by increased alpha power associated with unconscious conflict words. This was not the case in response to the conscious symptom words. Repression appears to have been effective in keeping the unconscious conflict out of awareness. Interestingly, if the conflict words are presented supraliminally (in other words, if the words are experienced consciously), the relationship disappears. The unconscious conflict is not recognized when the conflict words are presented within the subject’s awareness. In addition, neutral control words also failed to elicit an inhibitory response.

In another part of the experiment, subjects were shown the conscious symptom words either subliminally or supraliminally. Unlike the unconscious conflict words, the conscious symptom words were recognized by subjects more readily in the supraliminal condition. These findings strongly support the causal role of repression in keeping unconscious conflict out of awareness.

It was after reviewing these experiments that Grünbaum acknowledged that Shevrin and his colleagues had shown the presence of unconscious conflict, thus demonstrating the scientific rigor Grünbaum demanded. However, Grünbaum asserted that the work done up to that point did not demonstrate the causal efficacy of unconscious conflict in producing psychological symptoms. After considering Grünbaum’s critique, Shevrin agreed that he and his colleagues had not yet proved such a causal link. To address this difficulty, he devised another experiment, using unconscious conflict words, conscious symptom words, and generic unpleasant words as priming stimuli preceding the subject’s exposure to a conscious symptom word (or target word). The unconscious conflict words generated inhibition of response to the target word, but only when the unconscious conflict words were presented subliminally. The conscious symptom words and the generic negative words did not induce inhibition, whether presented subliminally or supraliminally. Shevrin et al. (2013) concluded that “from this pattern of converging experimental and control results, we are in a position to infer that only the unconscious conflict stimuli selected a priori by psychoanalysts from clinical data causally link clinical inferences based on psychological meaning (unconscious conflict over emotionally incompatible desires) with brain processes (patterns of electrophysiological inhibition). If so, then to our knowledge this is the first psychophysiological evidence for Freud’s unconscious conflict theory of psychopathology.” This experiment found a cause-and-effect relationship between the existence of an unconscious conflict and its inhibitory repressive effect. Again, this inhibitory effect was seen only with subliminal (unconscious) presentation of the conflict words. Shown these results, Grünbaum wrote back, “I am convinced.”

Considering the question “Why is it that the conflict words presented supraliminally do not elicit an inhibitory effect?” Michael Snodgrass took the microphone and asked, “How do we know what is truly unconscious? How can we meaningfully relate what is psychological and what is physiological?” There is evidence in brain functioning, he said, that the effects produced by unconscious conflict are unique and different from the effects of conscious understanding. Thus, the presenters and their colleagues have brought scientific rigor to the study of the unconscious.

The obstacle of proving that a stimulus is completely unconscious has been resolved, as the researchers have developed psychophysiological techniques that demonstrate an objective threshold for each subject such that there cannot have been even fleeting conscious awareness of the stimulus. It would not be possible for subjects to consciously extract any word meaning under these conditions.

The presenters then returned to the striking finding from their recent research. The effects from unconscious stimuli occur only when the stimuli are presented so that people are not aware of them. When the stimuli are presented to consciousness, the effects do not occur. What does this suggest? The presenters infer that if the stimulus is presented consciously, defensive processes damp down or suppress the impact of the unconscious conflict. We see this clinically when patients are talking about something meaningfully related to their unconscious conflicts and then suddenly stray from the topic; psychoanalysts recognize a defense against recognizing unconscious conflict operating to redirect the patient’s associations. In the laboratory, similarly, consciousness offers a defensive bulwark that mutes the impact of the unconscious conflict on behavior. In the clinical situation, defensive processes may sometimes be overridden, resulting in dramatic clinical phenomena such as lapses of memory, autonomic responses, and emotional outbursts.

What lies ahead? What should research further address? Shevrin is especially interested in two areas: investigating the rules of primary process thinking as it applies to unconscious conflict, and studying the role of motivation in defensive processes. Cognitive psychologists tend to ignore these areas. They have tended to think that unconscious processes occur very quickly, don’t last long, and don’t leave a memory trace, ideas contradicted by the research results on unconscious conflict obtained by Shevrin and his colleagues. Careful neurophysiological work demonstrates the presence of more protracted electrical activity that is correlated with complex processing of unconscious material such as clinical meaning. (In neurophysiological terms, this conclusion is demonstrated by the presence of a P300 response for unconscious stimuli as well as for conscious stimuli. The P300 response is an event-related potential that occurs approximately 300 milliseconds after a stimulus in connection with a subject’s process of decision making or other meaningful reaction to the stimulus.) Later components of unconscious brain activity are also present. There is evidence, therefore, that unconscious processes are not simple froth that the brain throws off but are complex and engaged in fundamental activities of the brain and the person. And although there is enormous individuality in brains, when subjects were challenged by individually tailored sets of words, the brain effects demonstrated that the physiological responses are universal.

Snodgrass made an impassioned plea for us to care about basic research. It is now possible to see the equivalences between what happens in the mind and what happens in the brain. Inasmuch as psychoanalysis is under attack and is fractionated, these research methods can help us demonstrate unconscious processes and determine which psychoanalytic theories are supported and which might not hold up. Scientific support for psychoanalytic ideas is crucial. Moreover, this research is clinically relevant, as demonstrated by experimental data presented by Snodgrass et al. (2014). All the subjects in the experiment met the criteria for simple phobia (e.g., fear of snakes or spiders). In the laboratory, an index was developed that measured the degree to which subjects were unconsciously repressing phobia-related information. Working on the assumption that unconscious repression is relevant to clinical practice, the researchers hypothesized that some subjects would repress a great deal, and others very little. In the actual experiment, some people were able to tolerate the phobia-related information, and others not.

The subjects all received cognitive-behavioral treatment for their phobias. For example, this treatment might have been exposure therapy, which has proven successful in treating some phobias. The subjects were asked to rate their fear of the phobic object at the beginning of treatment and again at the end. At the beginning some rated their fear at 80–90 on a 100-point scale; after exposure therapy, they rated it at a dramatically reduced level (40–50). There were, however, large individual differences. Snodgrass et al. hypothesized that the degree to which subjects unconsciously repressed conflict (as per the repression index mentioned above) would predict how well they would respond to cognitive treatment. The results were spectacular. Unconscious repression clearly impeded recovery, and those who exhibited the most repression benefited hardly at all from exposure treatment. It was hypothesized also that unconsciously these subjects were unable to tolerate the treatment and so could not benefit from it. Those who showed little repression not only could tolerate the treatment but greatly benefited from it. With the repression index, the researchers could explain two-thirds of the variations in treatment outcome. If these results can be replicated, the index can be used as a diagnostic test to indicate the best treatment option in a given case—who would benefit from CBT/exposure therapy and who would require psychodynamic psychotherapy.

The discussion group ended with an animated discussion that addressed the question of repression versus dissociation, the behavioral effects of unconscious conflict, and the impact of consciousness on conflict. It is hoped that Shevrin and Snodgrass will return in the near future with more of their exciting findings, helping psychoanalysts and cognitive scientists formulate and explore new research questions.

Shevrin’s novel in verse, Dream Interpreters, is a counterpart to Conscious and Unconscious Processes. The novel is concerned with meaning in the clinical context. Unlike the other book, which describes his research in quantitative terms, Dream Interpreters is an effort to convey the actual psychoanalytic experience. Certainly cognitive and affective neuroscientific research should rise to the challenges posed by the complexity and sophistication of that experience.