The Dreaming Brain 
by Allan Hobson.
Basic Books, 319 pp., $22.95, March 1988, 0 465 01703 7
Show More
Show More

Allan Hobson is a leading Harvard neuroscientist who has figured prominently in the breakthroughs which have occurred over the past three decades in the neurophysiology and neuropsychology of sleep and dreams. Long known within the field for his provocative views on the philosophical implications of sleep research, Hobson in this much-awaited volume addresses himself for the first time to a general audience. The heart of this work is an exposition of the widely-accepted Hobson-McCarley model of dream activity. Dreams, Hobson proposes, are the product of the synthetic, ordering activity of higher cortical brain areas responding to somewhat random internal stimuli generated by lower brain centres in REM (rapid eye movement) sleep. In Hobson’s hands, this activation-synthesis model offers a grand vision of the creative power of the human brain, and yet one set squarely in the mainstream of current cognitive science and neuroscience. It is thus the first encompassing alternative modern neurobiology has offered to Freud’s interpretation of dreams. Hobson is knowledgable about and sensitive to psychoanalytic theory, but does not hesitate to declare its shortcomings. The Dreaming Brain may be seen as an attempt, generally successful, to supplant the analysis of dreams that lies at the core of psychoanalytic theory with a radically different neurophysiological vision, a central challenge to the scientific and psychological foundations of the Freudian world-view.

The ubiquity of dreams has been common knowledge ever since Aserinsky and Kleitman discovered REM sleep and its association with subjective reporting of dream experience in 1953, and yet it still occasions surprise. We each of us dream in a regular and predictable fashion four or five times a night, and just as routinely fail to recall our dreams on awakening. In each evening’s sleep sustained dream scenarios, elaborated over periods lasting up to an hour, alternate with thought-like mentation in 90-minute cycles. Thus, as we traverse the seventy-year span that late 20th-century civilisation on average allots us, we will spend at least fifty thousand hours dreaming, a full six years devoted to oneiric life.

A central challenge for any theory of dreaming is to explain the bizarreness of dreams, to provide a rationale for the surreal shifts of person, place and thought that make dreams dreamlike. In characterising the fantastical qualities of dream narratives, Hobson enumerates five cardinal features of dreaming: vivid sensory impressions, often motor and visual (motor and visual hallucination); discontinuities of time, place and person, including flagrant violations of physical law (spatial and temporal cognitive distortion); uncritical acceptance of events, however unusual, as experientially real (delusional acceptance of hallucinoid experience); strong emotion (intensification of affect); and the tendency to fail to remember the experience once it is completed (amnesia). The parenthetic terms are those which result when Hobson, a physician as well as a scientist, subjects the dreaming mind to the neurologist’s or psychiatrist’s ‘mental status exam’. His disconcerting but illuminating conclusion is that by strict medical criteria each of us, when dreaming, is formally psychotic, delirious and/or demented. Dreams are indeed bizarre, and through them we each partake of a madness structurally similar to that which afflicts individuals with schizophrenia, organic dementia and other neuropsychiatric conditions. Insights gained from the study of the dreaming brain, Hobson gives us reason to hope, may shed light on the genesis of the numerous waking states with which it shares formal features.

Freud suggested that the bizarreness of dreams arises from a process of intra-psychic disguise and censorship. In sleep, he hypothesised, the sway the ego-superego holds over the unconscious, repressed impulses that comprise the id is lessened. Forbidden wishes threaten to intrude upon the conscious mind, but their disturbing quality would disrupt the sleep state and awaken the sleeper. Therefore, Freud suggested that another postulated psychic entity, the censor, disguises the unwelcome impulses through a variety of defensive transformations. The forbidden urgings are joined with the day residue of recent events through psychodynamic processes of displacement, condensation and symbolisation. Freud concluded that the bizarre manifest content of dreams must be dissected and interpreted to uncover the holy grail of all psychoanalytic endeavours: the latent content that is the motive force of the dream process.

Hobson lodges a number of objections to Freudian dream theory. Some are familiar: notably, the charge that Freud failed to specify findings that might invalidate his proposals, refusing to expose his insights to the peril of falsification, which is an essential, if unforgiving feature of all scientific theory. Hobson notes, for example, that Freud, having declared all dreams to be the fulfilment of wishes, had then to confront the quite common and seemingly contradictory occurrence of anxiety and other unfulfilling feelings in dreams. Freud explained away this incongruity by suggesting that dreams featuring such unpleasant, negative emotions were simply failed attempts at wish-fulfilment. Though disagreeable or even harrowing, when correctly analysed such dreams, too, could be seen to originate in the wishes of the dreamer seeking gratification. The apparently contradictory could always be reinterpreted as confirmatory, and failure to accomplish this feat indicated a defect, not in psychoanalytic theory itself, but in the interpreter-critic’s understanding and application of the theory.

Some of Hobson’s criticisms land only glancing blows. But most of them are telling, and none more so than his suggestion that Freud’s psychological theory was critically – fatally – dependent upon his own and his contemporaries’ now outdated neurobiological conceptions. Freud provided a complete outline of his neurobiological views in the Project for a Scientific Psychology of 1895, a work he later suppressed. There, in common with most of the leading brain scientists of his day, he envisioned the nervous system as a net into which energies flow via sensory channels and circulate until dissipated in motoric discharge, as against the modern understanding of neurons as elements of signalling networks, with a capacity not only to store but also to abstract, cancel or discard information. Freud’s antique neurobiological view of nerve cells as energy transmitters, Hobson argues, later re-surfaced in various guises in his psychodynamic theories. For example, it was, Hobson suggests, an unacknowledged source for Freud’s postulation of the existence of a dynamically-repressed unconsciousness constantly threatening to erupt into motor action and requiring an elaborate psychic system of checks and balances for its constraint. Freud’s dream theory, Hobson concludes, was deeply rooted in an explicit model of the brain, a model that has proved to be fundamentally unsound.

We have, however, heard all this, or something rather like it, before. To paraphrase Mark Twain, everyone has complained about Freudian theory but no one has done anything about it. Bloodied but unbowed by a century of critique, the Freudian outlook continues to be taught in our classrooms and applied in our clinics. Though they have disagreed with many of its particulars, detractors of Freudian theory have offered no fully-imagined rival in its stead, advanced no equally sweeping and embracing vision of the operations of the human mind. Hobson’s real achievement in The Dreaming Brain is to provide a glimpse of what such a ‘viable alternative’ might look like: a brainbased, neurophysiologically-sophisticated, modern ‘depth psychology’.

Consider then the activation-synthesis model of dreaming. Its genesis lay in basic research in cellular neurophysiology carried out in the Sixties. It was then that Hobson and his long-time collaborator Robert McCarley decided to pursue their interest in the regulation of sleep and dreaming by intensively studying neurons located in the brainstem, the phylogenetically-older brain structure which sits at the base of our hypertrophied cerebral hemispheres. Ideally positioned to modulate diffusely distributed cerebral activity, the brain-stem was known from lesions naturally occurring in man and experimentally inflicted on animals to play a central role in sleep-wake alternation. Hobson and his colleagues used micro-electrodes to record firing patterns intra-cellularly from individual neurons in the brainstems of sleeping and awake cats. They searched for and succeeded in identifying a ‘REM-on’ population of cells which were uniquely active in REM sleep periods. Then in a classic Pasteurian chance-favouring-the-prepared-mind discovery, they also stumbled on a corollary group of nerve cells which reliably fell silent during REM sleep. Spurred on by these findings, Hobson and McCarley proposed that REM sleep and dreaming are the outcome of a change in the operating state of the brain which is mediated by two reciprocally-interacting populations of brainstem neurons. Inhibiting one another in turn, the activity of the two neuronal groups generates the cyclic appearance and disappearance of REM activity in sleep. In The Dreaming Brain Hobson recounts these early discoveries and also discusses his most recent experimental substantiation of this model: the demonstration that REM sleep can be experimentally triggered by the application of the stimulatory neurotransmitter acetylcholine to the ‘REM-on’ population of cells. Like Morpheus of Graeco-Roman legend, and a thousand other gods of a thousand other human faiths, sleep researchers can now actively bring dreams to men (or at least to feline household pets).

It is on this solid foundation in neurophysiology that the activation-synthesis model of dreaming is erected. The brainstem generators of dreaming, Hobson and McCarley suggest, orchestrate a complex set of functional alterations in many other neural systems which cumulatively result in an entirely different mode of mental activity. The ‘REM-on’ cells drive the eyes into motion (resulting in the ‘REM’ of REM) and simultaneously send excitatory pulses to the higher visual and association cortices that process sensory information. They thereby produce, in the language of systems theory, an automatic, internally-generated information signal. At the same time, access of input from the external world to the auto-activated brain is prevented by the inhibition of primary sensory neurons and by the competition of the internally-generated messages. Just as crucially, the motor commands generated, and experienced, by the higher cortex in dreams are blocked by inhibition of motor-system relays in the spinal cord.* Auto-activated, disconnected, auto-stimulated, the dreaming brain, Hobson proposes, struggles to make sense of its internally-generated stimuli, integrating them into a dream plot which it synthesises in the light of an individual’s past experiences, ongoing concerns, settled character and cognitive style.

The bizarreness of dreams is given a straightforward – and compelling – explanation in this activation-synthesis model: the form of dreams is related to, isomorphic with, the form of brain activity in sleep. The sensorimotor hallucinosis of dream experience is the natural concomitant of a specific activation of sensorimotor neural centres: it is because visual and motor circuits are preferentially stimulated that we see vividly, move, even fly in our dreams far more often than we taste, smell or touch. Improbable dream scenarios are delusionally accepted as experiential reality because the brain is deprived of external cues with which to structure experience and construct an orientational framework. The distortions of time, place and person that abound in our dreams reflect the unconstrained nature of the internal stimulus source, only incompletely endowed with representational coherence and thematic unity by sorely-taxed analytic-synthetic systems. The intense feelings which mark our dream experience in part reflect a direct activation of the brain’s emotional systems; and the amnesia we have for the bulk of our dream experience results from a corresponding failure to activate long-term memory circuits.

It would be appropriate to note a few of the salient points of divergence between the Freudian disguise-censorship theory of dreaming and Hobson’s activation-synthesis hypothesis. For Freud, the driving energy behind the dream process is psychic conflict; for Hobson, it is a neurally-determined change of psychophysical state. Where Freud construes the direction of information-processing in dreams as backward-looking, in Hobson’s view it is elaborative and progressive. In the Freudian outlook, bizarreness is a secondary outcome of defensive transformations; in the Hobsonian vision, it is a primary feature of the operational state of the dreaming brain. And crucially, whereas Freud sees the meaning of dreams as opaque, having been obscured and bowdlerised by the censor’s activity, Hobson views the significance of dreams as transparent. He suggests that the content of individual dreams reflects the persistent concerns, conflicts and challenges of our daily lives which cortical centres draw upon in trying to make sense of unpatterned incoming dream stimuli. For Hobson, the significant content of dreams is the manifest content, redolent with meanings which can be read directly, without decoding. In dreams, Hobson suggests, not only is a cigar generally a cigar, but anxiety about an upcoming promotion is generally anxiety about an upcoming promotion. In a spirit which is psychodynamic without being psychoanalytic, the activation-synthesis theory encourages us to continue to look to our dreams as revealing issues of great personal import and urgency, and to interpret for ourselves their meaningful, often highly conflictual themes without recourse to an arcane calculus of symbol transformation known only to a few adepts.

The activation-synthesis theory offers a view of mental activity which is at once rigorously scientific and non-reductionist. In this, it is an accurate reflection of its progenitor. Hobson’s integrative, synthetic, interdisciplinary spirit is palpable on every page. His interpretations of neurophysiological findings yield an enriched view of mental function:

the activation-synthesis theory supposes an open system of information-processing, which is capable not only of reproduction and distortion of stored information but of the elaboration of novel information. Activation synthesis thus includes creativity among its assumptions. This theory sees the brain as so inexorably bent upon the quest for meaning that it attributes and even creates meaning when there is little or none to be found in the data it is asked to process. In this sense, the study of the dreaming brain is the study of the brain-mind as an autocreative mechanism.

I have already referred to Hobson’s likening the dreaming mind to various forms of mental disorder. Again and again in this work, however, he recurs to another, more positive human mental state as an analogue of dreaming. The creativity of the dreaming brain, he suggests, shares a deep kinship with that of the artist: ‘Since dreaming is universal, it stands as testimony to the universality of the artistic experience. In our dreams, we all become writers, painters and film-makers, combining extraordinary sets of characters, actions and locations into strangely coherent experiences.’ Hobson’s is a scientific materialism with a human face.

The reader of this volume receives a thoroughgoing education in the history and substance of the modern brain sciences. Cajal and the neuron doctrine, Sherrington and the reflex concept, Hodgkin and Huxley and the action potential: the discoverers and their fundamental findings are reviewed, placed in contemporary perspective, and fitted together in Hobson’s central story to culminate in the generation of a neurobiological model of the dreaming brain. However, the quest to understand dreaming has led Hobson far beyond the bounds of traditional neurobiology. Certainly he draws on elements from all levels in the hierarchy of the conventional neurosciences, including the molecular (neurotransmitters), the cellular (neurophysiological recordings) and the gross anatomic (studies of naturally occurring lesions). But in the multidisciplinary array of materials he adduces in support of the activation-synthesis hypothesis, computational theory (information processors), neuropsychology (right-left hemispheric differences in cognitive style) and philosophy (brain-mind isomorphism) are also represented. Hobson takes an approach to investigating brain activity which, in Daniel Dennett’s terms, is ‘bottom up’ and ‘top-down’ at one and the same time. In Hobson’s own phrase, it is ‘both/and’ rather than ‘either/or’.

Perhaps the most compelling evidence Hobson produces for the activation-synthesis model, however, is drawn from a set of data unique to the field of dream studies: the subjective dream report. True to his conviction that introspection is as valid a technique of neuropsychological inquiry as, say, single-cell recording, Hobson has for decades kept a journal at his bedside, and, on awakening, recorded his own night-time visions. Excerpts from his dream journal are interspersed throughout this volume, recounted and dissected to illustrate various aspects of activation-synthesis theory. They not incidentally reveal Hobson himself, replete with ambitions, anxieties and passions, carrying on, in much altered but no less fruitful form, the psychoanalytic tradition of self-analysis.

Hobson, though, is too well-trained a scientist to rest his case for the activation-synthesis model on anecdotal interpretations of individual dreams. He and his colleagues have sought to make dream analysis a genuine science. Over the past decade, they have pioneered new methodologies for the formal and quantitative investigation of systematically-collected dream reports. In one paradigm, a cross-sectional image of human dream activity is gathered by instrumentally awakening subjects in sleep laboratories precisely at periods when physiological recordings show them to be in the REM state. By applying descriptive statistics to reports collected from many such subjects, Hobson and his colleagues have identified and catalogued characteristic modes of perception, thought and action in dreaming. He has painstakingly built a grammar of dreams, a catalogue of the typical incongruities, distortions and discontinuities that constitute the language of dream narratives. From such a robust body of psychological data strong and novel correlations may be drawn with the known physiological features of the dream state.

Of course, dream reports obtained in the artificial setting of a sleep laboratory may be unrepresentative. In a corollary research programme, Hobson and his colleagues have sought to capture dreams ‘on the hoof’ in their natural environment. They have examined collections of home-based dream narratives recorded in bedside journals by a few subjects in the course of their everyday or, more appropriately, everynight lives. It is out of this work that the aesthetic and emotional climax of The Dreaming Brain emerges – Hobson’s analysis of the dream life of the Engine Man. Recognising that his own and his colleagues’ dreams might be contaminated by their knowledge of the hypotheses under investigation, Hobson sought a ‘dream informant’ who could provide detailed but theory-neutral dream material. Unexpectedly, in a catalogue of medical books, he came across a manuscript of 233 dreams recorded in a three-month period in 1939 by a Washington scientist-scholar. The Engine Man – it is the fascination with trains evident in his journal that causes Hobson, in provocative imitation of psychoanalytic tradition, to christen him thus – recorded meticulous descriptions, and occasional illustrations, of his dream experiences. Hobson subjects these accounts to a systematic micro-analysis. The narratives are carefully scrutinised, for example, for instances of dream bizarreness, and each is categorised by domain (plot, thought or state of feeling) and processing type (discontinuity, incongruity or uncertainty). Out of a plethora of such techniques, Hobson constructs a complete cognitive portrait of one individual’s dream-mind, the curvilinear trajectories of his fictive dream movements, the vivid colours of his visual dream images, the frequency and whimsy of his fanciful dream vehicles. It is an interpretative approach which illumines, not the individual psychopathology of the dreamer, but the universal rules of dream cognition as realised and applied in a singular individual’s dream life. This feat of scientific, literary and aesthetic interpretation culminates in Hobson’s remarkable, sustained analysis of the Engine Man’s ‘Customs Building Dream’. Hobson here proves himself a practitioner of the art which A.R. Luria and Oliver Sacks have called ‘romantic science’, forging a rigorously scientific vision of man that retains the full richness of living reality.

I do have a few quibbles with The Dreaming Brain. A firmer editorial hand would have elided the redundant passages which rehash in each of the different sections of the work material already fully covered elsewhere. Moreover the initial chapters, which review the early history of dream research, are not a total success. Hobson has made a heroic effort to rescue the 19th and early 20th-century pioneers of dream science from the oblivion to which they have been consigned by the overwhelming success of the antagonistic Freudian tradition. One of the figures Hobson reclaims is the English psychologist Mary Arnold-Forster, whose Studies in Dreams of 1921 anticipated current investigations of ‘lucid dreaming’, in which the dreamer consciously controls dream events. To place these theorists in context, Hobson must introduce disconnected bits and pieces of the modern outlook on sleep and dreams, which receives its full exposition only later in the work. The result is an uneasy blend of history and exposition.

Much remains to be learned about dreaming. Not least the solution to the dream researcher’s mystery of mysteries: why does dreaming take place at all? What biological purpose do sleep and dreams serve? In his closing chapter Hobson rehearses many of the speculations that have been advanced in response to this disconcertingly elementary question. Sleep as rest for neurons, a time, perhaps, for neurotransmitter repletion. Sleep as a means of ‘staying off the streets’, of reducing the organism’s activity during the hours of greatest danger. Dreaming as a means of stimulating and maintaining in functional readiness brain circuits not used in everyday activity. Dreaming as a means of reviewing daily experience, ‘offline’ information-processing. Alternatively, in Francis Crick’s turning-the-tables proposal, dreaming as a means of eliminating unwanted information, detoxifying the cognitive apparatus, ‘dreaming in order to forget’. Hobson puts his own spin on each of these suggestions, and offers a few that are characteristically his own: dreaming as creative elaboration, generating new feelings toward, and new solutions to, daytime problems; dreaming as entertainment, providing a ‘sybaritic and delight-enhancing function’. A proponent of evolutionary, if not psychodynamic over-determinism, Hobson suggests that sleep and dreams, like many other biological traits, may serve several purposes simultaneously. He recognises, however, that even collectively the theories which have been propounded to explain the teleological function of dreaming still fall short of the mark.

The purpose of dreaming may so far have eluded scientific investigation, but Hobson, in the theory outlined in The Dreaming Brain, has provided a fully satisfactory initial model of the mechanism for the generation of the dream state. This is not all he has done. The revolution in neuroscience is proceeding apace in laboratories throughout the world. But the full impact of the discoveries and conceptualisations of the new neurobiology has yet to be felt beyond the confines of the laboratory bench. As Hobson observes, the remaining years of the 20th century will undoubtedly be marked by the promulgation of innumerable new neurophysiologically-informed visions of human nature. By offering in the activation-synthesis hypothesis a coherent alternative account of dream activity within a neurocognitive framework, Hobson has helped loosen the Freudian shackles which have hitherto hobbled mind-brain theorists.

A hundred neurophilosophies may now bloom. Even more important, by forging a brain-based theory of dreaming which invigorates and uplifts the human spirit, he has created an exemplary model for a unified theory of all human mental activity of which the activation-synthesis account of dreaming will form one small but endlessly fascinating part.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 10 No. 15 · 1 September 1988

Jeffrey Saver’s review of Allan Hobson’s The Dreaming Brain (LRB, 4 August) was instructive as to the habit of old behaviourism masquerading – or deluding itself – as new. So the ‘activation-synthesis’ theory is ‘non-reductionist’? This begs the question that, relative to human experience, all mere mechanisms, neurological or otherwise, are reductionist as total ‘explanations’. Stilt, while on the subject, we may note how Saver pooh-poohs as antiquated Freud’s (1895, later suppressed, as Saver admits) ideas on neurology, while extolling rather breathlessly currently hallowed neurological notions. The difference in these circuitry mechanisms being, if not between Tweedledum and Tweedledee, then surely not more than between a diesel and a petrol engine – interesting to engineering but disappointing as psychology. Except to a behaviourist, that is.

The whole article is a tissue of neurological naiveties. So implanting electrodes in a cat’s brain allows us to give it dreams? Big deal, supposing we’d never observed that heavy suppers or the presence or absence of a loved one also achieved this objective – the first being, if you like, a neurological reason, the second perhaps rather more than this. Saver further tells us, as though we didn’t know it already, that dreams are bizarre; then he discovers for us through Hobson that ‘dreamers are artists’ – as though Freud, and how much more interestingly, had had nothing to say about this. The most offensive aspect of his review is, indeed, the reductionism, amounting to travesty, to which he subjects Freud in comparing his scheme of things with the scientific certainties of Hobson. That of which he does make passing mention is overschematised and oversimplified, utterly failing to take into account the richness of the human dimension in Freud’s psychoanalysis – that is, the very aspect of Freud notably absent, or specifically excluded, from behaviourism. But behaviourists never learn, except dubiously about rats and cats.

Saver seems actually to know this, as he dodges eagerly, if unprofitably, from the hard facts of neuron stimulus and response to the softer-than-software territory of dreams themselves – not any of Freud’s, heaven forbid, but Hobson’s undetailed ones or, still more anonymously, those of a ‘Washington scientist-scholar’ from which he elicits a figure – ‘The Engine Man’ – that has a strange, attenuated resemblance to some of the dream-characters of Freud; here, with regard to behaviourism and its mechanical explanations, there is even something risible.

Dreamland, though, is marsh ground and Saver soon leaps back to the drier regions of neurology. Respectable science as it may be, neurology has yet, pace the entertaining anecdotes of Oliver Sacks, to tell us anything profound about ourselves as human and social individuals – how could, indeed, a mere uncovering of circuits and their responses ever do so? The argument of ‘activation-synthesis’ describes itself as still that of the reflex machine – without even an explanation of its primum mobile: i.e. the mechanism that causes the brainstem to fire its random signals, later to be cortically elaborated into dreams. Not that the last question is particularly interesting in itself, since an answer would immediately beg the next more fundamental question: it is at least possible that ‘activation-synthesis’ is the mechanism ‘causing’ dreaming – but does this tell us anything illuminating about the significance of our dreams? Saver acknowledges that Hobson has not, after all, managed to explain ‘why’ we dream – which rather disposes of his own easy dismissal of Freud, who certainly did seek to give, in ‘The Psychoanalysis of Dreams’, a (non-neurological) explanation, and in this and later writings integrated it into an overall picture of humanity in all its biological, social, personal-developmental and historical richness. More or less mistaken, as final explanations, these ideas are certain to be, but they do take on board what behaviourism, to its discredit and ultimate triviality, always leaves out.

Ron Taylor
London SW19

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences