The Typology of Hypnotic Suggestion (Eysenck)

The psychologist Hans Eysenck (1916-1997) published an important analysis of experimental data on hypnotic suggestion, often overlooked by hypnotherapists, in his book Dimensions of Personality (1947).  The fifth chapter of Eysenck’s book is entitled ‘Suggestibility and Hypnosis’ and reviews evidence relating to the classification of different types of suggestion and the relationship between suggestibility and personality types.  In the preface to the 1998 edition of the book, presumably written just before his death, he neatly sums up his findings on suggestion,

I was trying to extend my experimental approach to psychiatric concepts, and chose certain specific statements from psychiatric textbooks for testing.  It had been almost universally claimed that hysterics are suggestible, but there was no experimental evidence.  I applied a number of standard tests of suggestibility to groups of hysteria and anxiety states, as well as non-neurotic controls, and found that there were at least two kinds of suggestibility which I called “primary” and “secondary”.  Hysterics did not differ from anxiety states, but neurotics as a group differed profoundly from normals, being much more suggestible.  When I showed the results to Sir Aubrey Lewis, my boss, he immediately summoned all the patients I had tested to make sure the hysteria had been correctly diagnosed – he agreed that they had.

In plain English, by means of careful statistical analysis Eysenck found, contrary to previous authors, that measures from many hundreds of subjects indicated those who suffered from phobia, anxiety, depression, and related problems, tended to be considerably more responsive to hypnotic suggestion than average, even more so than so-called “hysterical” patients who suffered from dissociation, amnesia and psychosomatic symptoms. 

Typology of Suggestion

Eysenck’s research, as we have seen, led him to conclude that it was necessary to make a distinction between two main factors or types of suggestion.

1. Primary suggestion. (Direct “ideo-dynamic” suggestion.)

This is the form of suggestion primarily associated with traditional hypnosis. A direct command or instruction is given to the subject in which the response desired is explicitly stated, e.g., “Your eyes feel sleepy and are beginning to close.” It employs the classic ideo-dynamic suggestion effect which James Braid, the founder of hypnotherapy, identified as the central mechanism of hypnotism.  Braid later described hypnotism as “mono-ideo-dynamics”, i.e., focused attention upon a single (“mono”) dominant idea. 

The term “ideo-motor reflex” refers to the seemingly causal effect of suggested ideas or images upon the muscular movements of the subject.  The term “ideo-dynamic reflex” was soon adopted by Braid to widen this concept by encompassing all bodily responses to ideas.  ”Ideo-dynamic” basically means the “power of the mind” over the body, as Braid put it.  This wider concept includes, e.g., not only motor (muscular) responses but also secretory responses such as the salivation caused by imagining sucking a lemon.  As Eysenck notes, the muscular suggestion tests used in most hypnotic scales, and traditional tests like Chevreul’s famous pendulum experiment, all function by means of primary suggestion.  Carpenter’s ideo-motor reflex can therefore be seen as the original psychological theory of hypnotic suggestion.

2. Secondary suggestion. (Indirection.)

This is also known as suggestion by indirection or indirect suggestion. Indirection entails a kind of manipulation of the subject’s beliefs and expectations. Eysenck explicitly states that by “indirection” he means misdirection or deception.

In other words, the experimenter uses indirection (i.e., deceit or trickery, as the dictionary defines the term) in order to give the subject an impression which the latter then claims as his own. (Eysenck, 1947: 167)

The classic example of this method would have to be the placebo effect, in which subjects are misled into believing they are receiving a potent medication when, in reality, they are being given a sugar pill or some other inert substitute.  For example, the Soviet researcher Platonov, following Bekhterev the founder of Russian psychotherapy, uses the term “indirect suggestion” to refer to the placebo effect, writing, “In indirect verbal suggestion the effectuation of the suggestion is, as a rule, related to a particular object or influence by means of which the suggestion must actually be effectuated.” (Platonov, 1959: 36).  In other words, it is “suggested” to a patient that a sugar pill contains a pain killer and their response is indirect insofar as it depends upon them subsequently swallowing the pill.

Eysenck gives the example of Binet’s progressive weight tests in which five progressively heavier weights are held by the subject followed by ten completely equal weights. The first five tend to create a “set” or “expectation” that subsequent weights will increase in the same direction and, by means of trickery or “indirection”, this leads to the illusion that all of the weights are becoming progressively heavier. Eysenck also refers to the use of deliberately leading questions in recall tests, to insinuate a false memory, as a form of suggestion by indirection. There are many other similar ways of testing and quantifying the effect of indirection which use manipulation, or illusion, to create certain expectations in the mind of the subject.

To this Eysenck adds a possible third category,

3. Prestige suggestion.

This form of suggestion employs the authority, charisma, or “prestige” of the speaker (hypnotist) to influence others. Many experimental studies demonstrate the tendency of individuals to imitate the attitudes and opinions of others whom they admire or identify with. The prestige traditionally attached by many people to medical doctors, scientific experts, or famous stage hypnotists facilitates a kind of social compliance and suggestibility.

Hence, Eysenck identified three main forms of suggestion and suggestibility, on the basis of direct suggestion, indirection, and prestige.  These notions can all be seen in Braid’s original theory of hypnotism which clearly recognised the ideo-dynamic reflex as the central mechanism of hypnotic suggestion.  Braid also documents many primitive experiments in which he employed (harmless) deception, in the style of the placebo effect, to elicit suggested responses.  For example, Braid frequently told “sensitive” Mesmeric subjects that he was holding a magnet over them, when in fact it was merely a piece of wood or some other inert object.  However, Braid also very frequently refers to the “law of sympathy and imitation” in hypnotism, a concept linked to the notion of prestige suggestion, although it encompasses imitation of one’s peers as well as persons held in high esteem.

Direct Suggestion versus Indirection (Deception)

Over a decade earlier, the behavioural psychologist Clark L. Hull had published the first major textbook on experimental hypnosis, Hypnosis & Suggestibility (1933).  Unlike Eysenck, Hull equated direct suggestion (the ideo-motor response) with prestige suggestion.  He dedicated a whole chapter of his book to reviewing the experimental data bearing on the relationship between direct and indirect suggestion, which he termed “prestige” and “non-prestige”.  Like Eysenck, however, he also concluded that experimental results, although scarce, failed to show that indirection worked in a similar manner to hypnosis or prestige suggestion. 

Meanwhile, the writer is proceeding on the tentative assumption that the final verdict will be negative, i.e., that non-prestige [indirect] suggestion is essentially different from – indeed, quite without significant relation to – hypnosis and direct prestige suggestion in general.  If this hypothesis could be firmly established it would aid greatly in clarifying the present lamentable confusion as to the fundamental nature of both hypnosis and suggestibility.  (Hull, 1933: 385)

Eysenck came to the tentative conclusion that prestige suggestion probably needed to be distinguished as a form of suggestion in its own right.  However, by reviewing his own research and that of others, he concluded, like Hull before him, that primary (direct) and secondary (indirect) suggestion were definitely not correlated.  He found, in other words, that they were different processes and that people respond to both types of suggestion to different degrees.  Moreover, he found that indirect suggestion did not seem to be directly related to hypnotism.

Throughout several studies, Eysenck consistently found no correlation between responses to direct and indirect suggestion. For example, in a study of 60 neurotic patients, he found the internal correlation of different direct suggestion tests to be quite high (+0.5) but there remained no correlation between direct and indirect suggestion. Measures of hypnosis itself, moreover, correlated very highly with responses to direct suggestion (+0.89). Eysenck’s statistical analysis suggested that although primary suggestibility could be regarded as a unitary trait, the same probably did not hold true of suggestibility to indirection which was less consistent in its effects.  He concludes,

In two factorial studies of altogether 16 different tests of suggestibility, it was shown that these tests define two entirely different and separate types of suggestibility: (1) Primary suggestibility, characterised by dependence on ideo-motor action, and (2) Secondary suggestibility, characterised by dependence on indirection. Primary suggestibility was shown to be closely related to hypnosis; secondary suggestibility showed no such relation. (Eysenck, 1947: 201)

Eysenck clearly explains that he uses the word “indirection” (like “misdirection”) to mean a form of “deceit or trickery“.  However, as used in hypnotherapy, the term “indirect suggestion” has come to mean something vague and is confused with the notion of being “permissive”.  Modern hypnotherapists, especially followers of Milton Erickson, sometimes describe permissiveinstructions such as “Just allow your eyelids to feel heavy” as being ”indirect suggestion”.  In their recent review of 29 studies on the subject, Lynn, Neufeld and Mare (1992) complained that terms like “permissive” and “indirect” had degenerated into “buzz words” which were so poorly defined that it was difficult to pin down their meaning (‘Direct versus indirect suggestions’, IJCEH, XLI, 2, 124-152).  Eysenck’s research, however, stands in a tradition of scientific research, like Hull’s, which attempted to properly distinguish the factors involved in different types of suggestion.

Indirect versus Indirection

Eysenck,  Hull, Binet, and earlier hypnotists mean “indirection”, in the sense of deception, when they say speak of experiments on “indirect suggestion.”  The Oxford English Dictionary (OED) defines the word “indirection” as follows,

1. Indirect movement or action; a devious or circuitous course to some end; round-about means or method. by indirection, by indirect means, indirectly.
b. In literal sense: An indirect or devious way.

2. Want of straightforwardness in action; an act, or practice, which is not straightforward and honest; deceit; malpractice.

However, I can discern two meanings of “indirection” relevant to the function of suggestion which are closely-related but should probably be distinguished.

1. Suggestion by roundabout or indirect means.  Where some external cause, including direct suggestion, evokes second order responses by association in the mind of the subject.
2. Suggestion by deception.  Where the subject is fooled into believing something by trickery, manipulation, or illusion.

To the best of my knowledge, the earliest definition of “indirect suggestion” was given by the German psychiatrist Albert Moll in his book Hypnotism (1889).

Auto-suggestion may be called up by some external cause; this [cause] may affect the person from outside, and thus induce an auto-suggestion.  Charcot refers some isolated traumatic paralyses to some such originating mechanism – though this point is still in debate.  According to this view a violent blow on the arm, following on certain disturbances of sensibility, may produce in the persona concerned a conviction that he cannot move his arm.  As the conviction was called up by the blow, this case stands somewhere between external suggestion and auto-suggestion.  We will call all cases in which the auto-suggestion did not arise spontaneously, but was the secondary result of something else, such as a blow, indirect suggestions; as opposed to direct suggestion, which arouses a certain idea immediately [...] (Moll. 1889: 59)

Moll emphasised that through habitual association the ideas evoked indirectly by suggestion could become involuntary and unconscious responses.  The neurologist Charcot’s patients had deceived themselves into believing that they were paralysed because they expected the severe physical trauma they’d suffered to result in that kind of  injury, a negative placebo effect resulting from the expectation of harm, which modern authors call the “nocebo” effect.  These “indirect” responses, or associations, may be triggered by an initial direct suggestion or by any other event.  If the subject does not realise that false or misleading associations are being evoked in their mind then indirection, in the sense of deception, has taken place.

Compare this with the example above (“Just allow your eyelids to feel heavy”) in which there doesn’t appear to be any significant deception taking place. Indeed, this is simply a direct, albeit permissive, instruction to imagine certain things and though associations are naturally evoked, such as the idea of sleepiness, the subject is conscious of them.  This isn’t just a semantic quibble. If this kind of suggestion works it probably does so because the client consciously interprets it as an instruction to imagine the response and not because they have been manipulated into believing something. This kind of permissive suggestion, or instruction to imagine, is probably better described as direct or primary suggestion rather than ”indirect” suggestion or “indirection”. It works mainly because the client imagines the response (primary suggestion) rather than because he has been deceived into expecting a response (secondary suggestion or indirection).

All direct suggestion necessarily evokes secondary suggestions, associations, spontaneous additions, as they are variously known, in the mind of the subject.  However, the placebo effect, the progressive lines/weights experiments of Binet, and other examples of “indirection” in Eysenck’s sense depend upon a lack of awareness on the part of the subject, a fact which makes them more akin to trickery, deception, or illusion than to direct suggestion, which openly addresses the conscious imagination.  Perhaps we should say that “indirect” suggestion becomes what Eysenck calls “indirection”, or deception, when the subject is not actually aware that suggestion is the cause of their response.  It seems to me that the presence of deception, manipulation, or lack of awareness, probably determines the difference in the way these two processes function.

This is an important point because we hypnotists should be clear about the kind of suggestion we’re using and the factors which are likely to make it succeed or fail. Those factors are fundamentally different for primary (classical) suggestion and secondary (indirect) suggestion.  Hull and Eysenck found, as we have seen, that direct and indirect suggestion responses did not seem to correlate.  One important observation in this respect, made decades earlier by Binet and confirmed by Hull, is that indirect suggestion seems to generally become less effective with repetition, whereas direct suggestion moves in the opposite direction, becoming more effective over time.   Eysenck later summarised his findings by writing,

Not only are these two types of suggestibility not related to each other; they behave quite differently in a number of ways.  Thus, primary [direct] suggestibility is not at all correlated with intelligence; intelligent people are no less suggestible than dull ones.  Secondary suggestibility [indirection], however, does correlate [negatively] with intelligence.  Here the more intelligent ones are less suggestible and the dull ones are more suggestible.  Again, primary suggestibility is closely related to hypnosis; a person who is highly suggestible also tends to be easily hypnotisable.  This is not true of secondary suggestibility. [...] These are only some of the differences observed between these two types of suggestibility, but there are many more, and they leave no doubt at all that we are dealing here with quite different traits. (Eysenck, Sense & Nonsense in Psychology , 1957: 215-216)

On the basis of his own psychological experiments, Binet concluded,

Finally it was to be noted that a repeated suggestion is less effective the second time than the first.  This weakening is undoubtedly particular to those indirect suggestions in the waking state which do not, properly speaking, involve any constraint upon the intelligence of the individual [i.e., no conscious attention].  In hypnotic experiments, on the other hand, the suggestibility of the hypnotised individual increases with the number of hypnotisations. (Binet, 1900, quoted in Hull, 1933: 376)

Following Binet, Eysenck found that direct suggestion was fairly unrelated to intelligence (IQ) but, perhaps unsurprisingly, that indirect suggestion by means of deception was more effective with subjects of low intelligence.  In plain English, the influence of deception or trickery (indirection) tends to weaken with repetition, whereas the power of conscious imagination (in response to direct suggestion) naturally tends to strengthen with practice.  Indeed, Hull consistently found that ideo-motor responses tended to follow a standard practice curve, i.e., to become rapidly easier with practice.  Hence, indirection may work better with new clients, and wane afterwards, whereas direct suggestion tends to grow in effectiveness with practice.  In the work of many hypnotherapists, however, there is a conflation of these two processes and they often merge somewhat chaotically with one another.  It should also be noted, incidentally, that insofar as indirection involves deception it is difficult to imagine how it could possibly be used in self-hypnosis, i.e., all self-hypnosis must probably depend primarily upon direct autosuggestion.

Of course, Ericksonian hypnotists make more extensive use of indirect suggestion and it has always been questioned whether they are right to call this “hypnosis” or whether, for the sake of clarity, they should call it something else.  For example, Andre Weitzenhoffer, one of the most influential researchers in the field of hypnosis, often questioned the extent to which Erickson and his followers could legitimately be described as employing hypnotism.  In The Practice of Hypnotism (2000), he writes,

That which has come to be known in recent times as “Ericksonian hypnotism,” or the “Ericksonian approach” to therapy, questionably belongs in this work, for such hypnotism as may be involved in it, often appears to be minimal, if at all there. (Weitzenhoffer, 2000: 588)

In conclusion, it seems that this issue of the different types of suggestion will continue to vex hypnosis researchers and clinicians.  There is no simple answer.  Many factors combine to produce the effects found in hypnotism and they probably vary from one client to the next.  However, there do seem to be grounds for making a clearer distinction between direct and indirect forms of suggestion on the basis of Eysenck’s research on indirection.  It may even be that hypnotism should be conceived as essentially a mode of primary (ideo-dynamic) suggestion and distinguished from the use of indirection because of the differences in functioning between these two types of intervention, although they are invariably combined in practice.  In other words, hypnotherapists may frequently use indirection although, from this point of view, it is perhaps not really hypnotism as traditionally understood but rather an important adjunctive technique in hypnotherapy.