2
Habituation: the
difference between
familiar and unfamiliar
stimuli

 

 

Reading this book and understanding it should be an example of quite an advanced form of learning. But if reading it merely helps you to fall asleep, that is an example of a much more elementary form of learning, habituation. As a simple form of learning habituation is being studied now, in primitive animals rather than readers of books, by scientists interested in the essential physiological changes that enable learning to occur. The usual form of habituation is the repeated exposure to the same stimulus that makes for very sluggish responses to that stimulus, but in some cases the lack of responsiveness does indeed blend into a state of sleep. In this chapter habituation will serve as an example in several issues which apply more widely: the use of the terms stimulus and response; the difference between behavioural processes and physiological processes; and the co-existence of elementary with more elaborate forms of learning.

 

Stimulus (S) and response (R)


Stimulus and response are descriptive units for behaviour, often abbreviated to S and R. If someone sticks a pin in you and this makes you jump, the pin-prick may be termed the `stimulus' and your jump the `response'. The fact that people normally jump when stuck with pins might be referred to as a`stimulus-response relationship'. Even in this simple example there are

20

many questions that can be asked about what is really going on: What exactly is the stimulus? Is it the pin, the insertion of the pin, the activation of nerve-endings in the skin, or the subjective feeling of the pin-prick? Is the response the twitching of particular muscles or any sudden body movement? Different answers are possible, and the terms `stimulus' and `response' are used in a variety of specialized ways. But the most common use is merely as a convenient distinction between things an organism may be confronted with and activities which it might display. Most psychologists would accept this distinction in the case of pins and jumping: `people normally jump when stuck with a pin' would be accepted as a stimulus-response relationship. The bone of contention comes in when more complicated cases are seen in stimulus-response terms. `If you talk to someone, they will usually talk back' is true enough, but `the stimulus of talking to someone will usually lead to the response of their talking back' still raises some hackles. However, with most cases of habituation, and other simple forms of learning, it is practically impossible to give an account of what happens without distinguishing between things which are there in the environment (stimuli) and things which the subject does (responses).

Concentrating on stimulus-response relationships is typical of behaviourists, and is useful in the study of learning when learning is judged to be a change in behaviour. The kinds of learning which are easiest to measure are those identified by a change in a single stimulus-response (S-R) relation. In habituation the change is that an animal or person ceases to react to a stimulus, and in classical conditioning (Ch. 3) the subject begins to react to a stimulus. Because very systematic measurements can be made of these elementary changes in stimulus-response relations, such S-R relationships are often taken as basic units of learning, as if they were atoms or molecules which can be assembled into more elaborate skills. In some cases theorists such as Hull (see p. 15) have assumed that all kinds of behaviour, however complex, are composed of S-R units. The difficulty in this is that in order to encompass any reasonably intelligent behaviour, it has to be assumed that there are hypothetical S-R units at work inside the head, which cannot themselves be measured. That spoils the original advantage of the S-R description, which was that it was based on accurate observations. In any case, it has turned out that S-R theories do

21

not apply very well to complicated activities like problem-solving. But, although S-R theories have limited application, this does not mean that it is necessary to abandon S-R descriptions. The topic of habituation demonstrates this point rather well: habituation cries out to be described as the decline in responsiveness to a stimulus, but one of the main theories of habituation says that this ultra-simple behavioural change comes about because of exceedingly complicated goings-on inside the head, which have little to do with S-R units (see p. 25). We can continue to talk of stimulus-response relations without committing ourselves to any particular explanations as to what physiological mechanisms are doing the job of governing the vigour of responses to a particular stimulus input.



Changes in response when a stimulus is presented repeatedly:
habituation and `warm-up' effects

The most rudimentary case of changed response to a repeated stimulus occurs with single nerve fibres. A nerve fibre can be made to respond, producing an impulse which travels down the fibre, by a standard electrical stimulus. But immediately after this, it remains completely unresponsive (the absolute refractory period) and then goes through a phase of being less responsive than usual (the relative refactory period). Obviously this is far removed from learning and is more a matter of taking time to recover the physical ability to make a response. Decline in response due to muscular fatigue has some similar characteristics; if you jumped when someone said jump, the vigour of your jumping would wane if the stimulus were frequently repeated, if for no other reason than muscular tiredness.

A second kind of drop in response with repeated stimulation is sensory adaption. The best example of sensory adaption is the reaction of the eye to strong lights : the pupil contracts to let less light into the eye and the retina of the eye becomes less sensitive. The effect takes some time to wear off so that, if you move a light into a much darker environment, things get gradually clearer and clearer: if for instance you go into a cinema on a sunny afternoon it is much harder to see the way to the seat at first than it would be ten minutes later.

The decline in response called habituation occurs over and above any response decrements due to sensory adaptation or

22

muscular fatigue and is much more interesting psychologically than these two effects. In most cases habituation can be conceived as a lessening of response produced by the increasing familiarity of a stimulus. If a friend stops putting up counterarguments when you state your pet political theory, this does not mean necessarily that he has been temporarily deafened or that he is too tired to talk, it may be because he has heard it all before, and it has become so familiar that it fails to stir him. You might say that he had become habituated to your theory (even if he still disagrees with it).

There is in fact a wealth of examples from a wide range of human and animal behaviour where isolated responses wane as the stimulus which provokes them recurs, without the intervention of muscular fatigue or sensory adaption (Hinde, 1970). However, the scope of habituation is so comprehensive, embracing almost every type of behaviour, that as Hinde says the `underlying processes' must vary enormously.

At one extreme, investigations of the sea-slug Aplysia have revealed that the habituation of its gill-withdrawal reflex depends to a large extent on the changes at synapses between sensory and motor nerve cells. A jet of sea-water squirted at a sensitive area near the gills would normally make the gills withdraw, but this reflex habituates after five to ten squirts given within a few minutes of each other. The independence of this habituation from adaptation or fatigue can be demonstrated because a strong squirt of sea-water to another part of the body brings back the full gill-withdrawal response to local stimulation (this is called dishabituation). Essentially the same habituation and dishabituation could be observed in decapitated slugs or, by taking electrical recordings from sensory and motor neurons, in a surgically isolated Aplysia abdominal ganglion (Castellucci et al., 1970). This work tells us something about the minimal nervous system equipment which can accomplish the behavioural changes involved in habituation. The same kind of behavioural changes, in terms of increases and decreases in responsiveness to stimuli, can also be brought about in animals with much more complicated nervous systems and indeed with ordinary human subjects, as the next section shows. This is a very striking example of how similar stimulus-response relationships can arise from different underlying processes. It may be perfectly true to say that `the response declines when the stimulus is repeated' or `the response habituates' in a wide

23

variety of situations, even if the reasons behind the habituation are not always the same.

Surprise, alertness and the orienting reflex Despite the variety of responses which are covered by the behavioural definition of habituation, it is possible to isolate some general features of habituation in what Pavlov called the `what is it?' reflex, which has been studied in man and other mammals. Any very unfamiliar or unexpected stimulus will produce the `orienting reflex'. A fairly intense stimulus such as an explosion or thunderclap may make one literally jump or lead to flinching of the whole body. Less extreme stimuli may nevertheless produce easily visible responses such as turning or looking towards the source of the stimulus. Animals may `prick up their ears' especially to strange sounds, and people `look surprised' when something unexpected happens.

Physiological expressions of surprise. The immediately visible reactions to strange stimuli are not the only ones available for the psychologist to measure. Indices of autonomic nervous system activity such as pulse rate and skin resistance, as well as general muscle tension and electrical brainwaves all show changes when a new stimulus occurs. A good deal of research has been done, especially by scientists influenced by Pavlovian traditions, such as Sokolov (1963), on the way in which these physiological manifestations of the orienting reflex change as stimuli are presented over and over again to a human subject. Sokolov uses three categories of response: first, adaptation reflexes, mainly specialized reflexes allowing for adjustments of sensory systems (such as the contraction of the pupil of the eye for high levels of illumination); second, the defensive reflexes to high intensity or painful stimulation; and finally, for lower stimulus intensities, the orienting reflex, distinguished by the fact that it occurs in response to new stimuli, and declines as a given stimulus is repeated to the same person. Components of the orienting reflex include an increase in sensory sensitivity, a drop in skin resistance, increased indications of arousal in the EEG (the electroencephalogram of brainwaves), constriction of blood vessels in the limbs, dilation of blood vessels in the head and a lowering of respiration and pulse rates.

The `neuronal model' of stimuli. The main point about the

24

orienting reflex is that it shows a progressive reduction as a stimulus gains in familiarity through repetition. But slight changes in the stimulus when the orienting reflex has completely disappeared will be sufficient to bring it back. That is, a familiar stimulus can become surprising again if there is a quite small change in it.

Sokolov has put forward a theory to explain this which assumes that a`neuronal model' or physical memory of each stimulus is built up by repeated experience. The orienting reflex is produced when a stimulus is given for which there is no model, but declines as the model for that stimulus is built up. Habituation results when there is a complete match between the stimulus received and the internal model. This theory explains why quite small changes in a familiar stimulus could lead to `surprise' in Sokolov's experiments. It also helps to ex.plain the `missing stimulus effect' Sokolov found. To show this effect, a subject is presented with the same stimulus at regular intervals until there is no flicker of response to be measured. Then the stimulus is missed out, and the subject shows the surprise reaction at the time when it was due. It seems as though there is some form of temporal conditioning (see Ch. 3) which can take place even during habituation, and which becomes part of the `neuronal model'. Thus when a stimulus presentation is missed out, there is a`mismatch' and the subject notices the absence. This is rather like being kept awake by the ticking of a clock, eventually falling asleep, only to be woken up when the clock stops.

 

`Warm-up' or sensitization effects

Although the neuronal model theory deals with the fact that familiar stimuli are ignored, while unfamiliar happenings rouse orienting reactions, it does not explain very well those cases where there are sensitization effects, which are opposite to the effects of habituation. Some experiments which help to clarify the distinction between habituation and sensitization have been conducted on the `startle response' evoked in rats by loud noise. In experiments of this kind, the rat is tested in a small soundproofed cage equipped with electronic devices which can pick up any sudden movements it makes. Habituation to loud tones can then be expressed in terms of changes in the number and vigour of the startle responses. Anyone who has lived for a time near a large airport knows that hearing a noise

25

in the 100 to 120 dB range can be both startling and disturbing. They are two opposite ways in which exposure to repeated loud noise can change this startle response. There is some degree of habituation or `getting used to it' over time, so that long-term residents near airports may be less disturbed by passing jets than their guests from a quieter area. But there is also a cumulative effect, so that at the end of a particularly noisy weekend, even hardened residents may be more `jumpy' than at the beginning. By using sounds in the `jet taking off' range, these two effects of cumulative exposure have been clearly identified in rats who were exposed to loud tones in eight thirty-minute sessions spaced out over four weeks (Davis, 1972).

gif
Fig. 2.1 Habituation and sensitization of startle. Amount of startle during eight sessions. Fifty very loud blips were given each ses sion, thirty seconds apart. (After Davis, 1972)

As can be seen in Figure 2.1 the rats `got used' to the noise in that they were less startled in successive sessions. But there was also a temporary sensitizing effect within each half-hour session, so that after the first one the rats were more startled at the end than at the beginning of a session. Habituation by degrees. The sensitization process may be responsible for another curious effect found with startle responses. This is that gradual increases in noise level serve as an ex- - ceptionally effective way of producing habituation to loud noises. If rats are exposed first to a moderately loud noise, and then to a sequence of gradually increasing loudness, they show much

26

less response to a very loud noise than animals given an equivalent period of training with either the very loud noise itself, or just a moderate sound, or a random series of noises. This kind of effect means that other factors, apart from the `neuronal model' or stimulus memory aspect of habituation, help to determine the level of response. The best guess at present seems to be that the `gradual increase' procedure allows the rats to keep in a relatively calm state of some kind or other, while habituation is taking place. In other words, a tolerance of noises in general can be built up without increased excitability or sensitivity entering in as much as usual. It is easy to think of the `calm state' as being lack of annoyance or anxiety, but this is another case where `state of excitability' may take several physiological forms.

Arousal, interest and sleep

That general states of arousal are closely connected with the orienting reflex and habituation was emphasized by Pavlov, who found that repetition of sounds of little interest to the dogs used in his experiments made the dogs go to sleep. Habituation to particular stimuli seems to reduce arousal, if no alternative interesting stimuli are available. Repetitious stimulation, obtained by counting imaginary sheep or reading dull books, is widely accepted as an aid to sleep, and similar effects have been found in laboratory experiments. But counting sheep will not be relaxing for a worried sheep farmer, because for him sheep are an important stimulus, having an arousing rather than sleepinducing effect. For particular people and species of animal there are different stimulus categories, some stimuli being danger signals, or aversive; others being positively interesting or rewarding; and others being relatively unimportant, provided that they are familiar. Habituation to the extent of decreasing arousal applies mainly to relatively neutral situations, that is stimuli which have little significance in the basic motivational systems. Thus for animals stimuli associated with food, sex or pain are less likely to habituate than random noises in the laboratory, and lectures or books that are boring for one person may be intensely interesting to another.

It has been suggested that curiosity, which represents the absence of habituation, is itself motivating. Although the first reaction of animals to strange stimuli is usually fear and withdrawal, the next reaction is some degree of curiosity, exploration,

27

and after sufficient investigation, `curiosity is satisfied', i.e. there is habituation. According to Berlyne (1969) level of arousal (in terms of alertness as opposed to drowsiness) acts as a motivating system because there is an optimal level of arousal. In a state of high arousal, familiar situations, or stimuli which have habituated, will be actively preferred, because they will bring arousal back to a more tolerable level, but in a state of low arousal novel and even dangerous stimuli will be sought out precisely because they raise the arousal level - or supply relief from boredom. As an extreme case, people will seek out dangerous activities such as parachuting or mountain-climbing for `excitement' as well as finding excitement more safely by watching films of disaster or horror. But for relatively neutral stimuli, the process of fashion in clothes and music provide examples of how important `novelty' can be in a stimulus. There are undoubtedly a great many complicating factors, but possibly habituation is one of the reasons why popular records decline in popularity when they have been heard over and over again. It is also often suggested that fads and fashions influence the popularity of psychological theories, so that even a fairly sound theory may be replaced by a new one because psychologists prefer novelty!

Habituation and satiation. Powerful motivating stimuli tend not to habituate, but this is probably more a case of the motivating effects masking any habituation which does take place. If you are starving, then your experience of food stimuli will not produce any noticeable reduction in your interest in food. However, most people at what we regard as normal levels of hunger would quickly habituate to particular foods. You may like pork-pies, but if you were to eat nothing else for a couple of weeks you would surely like them a good deal less. This is stimulus satiation for pork-pies. But it is worth distinguishing this stimulus factor from `after eating' satiation, which would occur if you at half a dozen pork-pies at once. Specialized physiological processes ensure that you would be `satiated', presumably for all foods, for a few hours after that. Food satiation is basically a short term physical influence, like muscular fatigue. Nevertheless, if you make yourself sick by eating chocolates you should be 'put-off' chocolates for some time afterwards. With important social and sexual stimuli, habituation and satiation may be entangled with other factors. `Absence makes

28

the heart grow fonder' may refer to a state of social deprivation, or to cases where affectional responses recover from habituation'as the object of the affections becomes less familiar. Eysenck (1973) has suggested that the search for social and sexual novelty is an aspect of the `extrovert' personality, which having a low level of arousal requires the maximum of excitement from external sources. An opposite kind of influence, which Eysenck wouldd say is more prevalent for introverted persons, is the desire to be reassured constantly by well-known social contacts and familiar surroundings.

One may conclude that habituation, sensitization and satiation, resulting simply from how often certain stimuli have been experienced, either in the recent past or over a long period, may be at the bottom of many emotional preferences, drives or reinforcers. Curiosity and a craving for novelty, or a need for the security of accustomed environments and habits are at opposite ends of a motivational system based on habituation, but long-past or recent experience of stimuli may modify a great many other motives.

Habituation in therapy

There are a number of techniques used in behaviour therapy (see p. 173) which involve an element of habituation. In the negative practice method (see below) repetition of response is the critical variable, in other methods the frequent or prolonged presentation of a stimulus may be a factor, and in the desensitization procedure (p. 30) the `gradual increase' sequence appears. These kinds of therapy are relevant here because they all make use of the repetition of stimuli and responses and an ingredient such as boredom, satiation or increased stimulus familiarity.

Negative practice. It has often been claimed that annoying habits such as tics could be treated by intensive repetition (Meyer and Chesser, 1970). A patient who wishes to be rid of a nervous tic may be told to produce it deliberately as accurately and frequently as possible for a period of several minutes. There are several reports of successful reduction of tic frequency after this simple expedient of deliberate repetition. Another problem which has been helped by this treatment is a rare neurological

29

condition involving pathological use of obscene words. In this case the patient was required to repeat an obscenity over and over again, in time to a metronome.

Stimulus satiation. Allyon (1963) treated a patient who had been in hospital for nine years and usually managed to keep twenty or thirty towels in her room despite the efforts of nursing staff to recover them. The treatment was for the staff to give the patient as many towels as she would accept, instead of removing them. She accepted 625 before beginning to refuse additional towels. It took four weeks to accumulate this number but during the next ten weeks the patient gradually discarded towels, until she finally kept only one or two. This more ordinary requirement was maintained during the subsequent year of observation. At the same time this patient was taught to refrain from her previous habits of stealing other patients' food and wearing several sets of clothes. The overall effect was to allow more normal social interaction between the patient and other people in the hospital.

Desensitization. This method of therapy has been in widespread use for about twenty years. The symptom most frequently treated by desensitization is a phobia or excessive fear of such things as spiders, traffic or heights. One of the main features of the therapy is that the therapist helps the patient to form a `hierarchy' or gradual progression of problem situations starting with stimuli that are just tolerable and progressing to situations which would cause panic or extreme anxiety and stress for the patient. When desensitization was first used, relaxation training was often a preliminary stage of therapy and the main stage consisted of the patient relaxing at the same time as imagining the problem situation, starting with the less severe forms at the bottom of the hierarchy, and working upwards until even the worst scene could be imagined without disturbance. In this case, using imagined problem situations, it was found that the relaxation training is an important part of the procedure (Paul, 1969) and it seems as though the conditioning of relaxation is the basis for success (see Ch. 3). Real-life (or "in vivo") events have been used instead of imagined scenes, and here relaxation training is not as important as `getting used to' feared stimuli (habituation), which plays a large part. For instance, a patient may be severely

30

distressed by fear of large hairy spiders, to the extent of not being able to go anywhere remotely likely to be harbouring a spider. Desensitization has been achieved in such cases by means of a collection of real spiders of various sizes. The patient first `gets used to' being at close quarters with a small hairless spider and then is confronted with a succession of increasingly hairy and monstrous ones. When techniques like this succeed it appears that graded exposure to real problem stimuli can be sufficient in itself to abolish extreme emotional reactions.

Flooding. There are some neurotic complaints where the 'little-by-little' desensitization method does not work, perhaps because the source of anxiety cannot be equated with a hierarchy of external stimuli like small to large spiders. Agoraphobia, fear of being in the open or fear of going out, is one problem where desensitization has been tried without much success, but it responds to an alternative procedure called flooding. In flooding a stimulus hierarchy is not critical, because the basis of the method is to maintain an extremely high level of anxiety until some process of exhaustion or stimulus satiation intervenes. It is obviously necessary to be very careful about using a procedure causing distress during treatment, but a number of investigations have indicated that flooding may be worthwhile both in agoraphobia and in obsessive-compulsive disorders.

Emmelkamp and Wessels (1975) suggested that agoraphobia may be relieved by flooding, especially if an in vivo method is used. This amounted to long walks taken by the patient, starting off from his home and taking a difficult route going directly away from home, as agreed previously with the therapist, without being allowed to take a dog or familiar item such as an umbrella to relieve the tension. For comparison, imaginal flooding consisted of sessions lasting as long as the walks (ninety minutes) in which the patient was asked to visualize as vividly as possible fear-provoking scenes described by the therapist. A behavioural test of the effect of these therapies was made by asking patients to stay outside for as long as they felt comfortable. Patients who panicked after a few minutes before treatment were able to stay outside for about an hour after the forced exposure of long walks, but those given the treatment by imagination showed relatively little improvement.

When flooding is used in therapy for chronic obsessive com-

31

pulsive neurosis, it is very important that the patient be prevented from engaging in his usual compulsive rituals in situations designed deliberately to provoke the rituals (Hodgson et al., 1972). A common obsession is the fear of contamination, which is connected with rituals to avoid contact with the feared source of contamination (animals, for instance) and compulsive hand-washing and cleaning. In the use of flooding for this obsession a patient might be encouraged to forbear avoidance or cleaning activities while staying in a room with several cats and dogs and having hamsters crawl over their clothes and hair. Exposure to the extreme case may enable the patient to cope with everyday situations, provided that the extreme case is tolerated without the usual compulsive actions. It seems that good results are also obtained, at less cost to the patient, if it is the therapist who must have hamsters in his hair, while the patient watches, at least for part of the treatment. This vicarious exposure to the problem situation (see Ch. 9) may help by making it easier for the patient to withhold his compulsive responses while becoming accustomed to previously disturbing events. But in one way or another, it looks as though flooding bears some similarity to a procedure for habituation; for it is a repeated exposure to a particular situation which leads to a decline in behaviours which were initially an inevitable product of the stimulus.

Summary and conclusions

The idle repetition of a stimulus in the environment which initially provokes some activity or alertness changes sensitivity to the stimulus. The main result is a shut-down of sensitivity to the stimulus, called habituation, but there may also be short-term increases in sensitivity, called sensitization. Habituation is a form of learning in so far as the current response, or lack of response, depends on previous experiences. When it is a matter of becoming familiar with stimuli, and being interested in the new as opposed to the old, habituation may reflect the learning of many complex features of familiar items. It is often useful to be able to ignore stimuli, especially as an alternative to being made over-anxious by them, and therefore some of the behavioural procedures for reducing anxiety - such as desensitization and flooding - mirror the habituation process.

32