How can best we account for the acquisition and maintenance of phobias?

The evolution of theories concerning acquisition and maintenance of phobias is marked by waxing and waning trends. Conditional theories arose from the rebuttal of psychoanalytical theory. Cognitive theories then emerged to challenge the acceptance of conditional theories. Each school of thought proposes therapies based their particular model it is important that a unified theory of the aetiology and maintenance of phobias be developed so that a universal approach to treatment may emerge. This essay will assess the contributions and relative importance of psychoanalytic, conditioning, cognitive and biological models of phobias. I will argue that no single proposition can completely explain all phobias. Finally, a unified approach will be attempted, based on the strongest evidence drawn from each school of thought.

The Diagnostic and Statistical Manual of Psychiatric Disorders (IV) classifies phobias one as seven anxiety disorders. Phobias have been defined as intense fears emerging in social situations (social phobia) or highly specific situations such as proximity to particular animals, heights, thunder, darkness (specific phobia). Although the triggering situation is discrete, contact with it can evoke a panic response (as in agoraphobia or social phobias). Specific phobias usually arise in childhood or early adult life and can persist for decades if they remain untreated. The seriousness of the resulting handicap depends on how easy it is for the sufferer to avoid the phobic situation.

After the work of Pavlov, which showed that behaviours could be conditioned in animals, theorists and clinicians alike sought to apply his ideas to human illness, including phobia. As early as 1917 Pavlovian models explaining the aetiology of phobia had been proposed. Yet they were only elucidated and championed after the Second World War as a consequence of the iconoclastic revision of Freud's psychoanalytic model by Wolpe and Rachman. In the first half of the 20th century psychoanalytic models were quite dominant. Unfortunately this popularity was more due to Sigmund Freud's notoriety than to the value of his ideas on phobia.

In keeping with psychoanalytic theory, anxiety is said to be the reaction of the ego to the dangers presented by the demands of the libido. Freud accepted as incontestable the idea that anxiety in the average human is closely connected with sexual restriction. The psychoanalytic model of phobia is best potrayed and best-criticised in Freud's archetypal case study of "Little Hans".

Hans first came to Freud's attention when, as a boy of four going five, he developed a fear of horses. Though Freud only met the boy once, he was able to offer him a diagnosis and a course of treatment through correspondence with Hans' father, a psychoanalysis enthusiast. Hans' fear of horses was said to be symbolic of his fear of his father. This fear arose because Hans had oedipal strivings for his mother, and wished to have his father out of the way so that he could sleep with her. Yet he feared his father's anger and retribution for his oedipal feelings, the worst expression of which may be castration.

Unconsciously, these sexual strivings were recognised as unacceptable to his ego and were suppressed. The castration anxiety associated with his father was transferred to a more 'acceptable' focus i.e. the horse.

To Freud, horses were clearly symbolic of Hans' father because the boy was afraid of 'what horses wear in front of their eyes and the black around their mouths'. This represented the father's eyeglasses and moustache. Fear of horses also gave the boy an excuse to stay at home where his mother was.
One of the treatments proposed was a ceremony in which the boy promotes his father into marriage with his grandmother. This was to remove the conflict between them over the boy's mother, and hence remove the boy's phobia.

One can only wonder at the little boy's bewilderment at being introduced to his sexuality in such an awkward, deviant way. Wolpe and Rachman's classic critique propose that Hans' own explanation for his own phobia, verified by his mother, was the better one. Their complete revision of this case study led to the demise of the psychoanalytic approach and the prominence of the behavioural approach. In my view, this demise was timely since the psychoanalytic approach is backed by modest empirical evidence. Furthermore, it is founded on uncertain factors, for example the boy's unconscious thoughts that he is unaware of but Freud is privy to by a mysterious process of deduction.

Though we may dismiss the psychoanalytic approach for its looseness of analytic inference, there is still some value in Freud's analysis. Freud recognised anxiety as an important component of neurosis. He also differentiated between objective (state) anxiety, which was rational and expedient, and neurotic (trait) anxiety, which was inexpedient, free-floating and excessive. Objective analysis could become neurotic anxiety in excess. This idea is valuable because it recognises that external stress factors or innate factors can determine levels or anxiety. Residual anxiety may be a vulnerability factor in developing phobias.

Furthermore, he anticipated Seligman's 'preparedness' concept of phobias by accepting that some fears that commonly arise in childhood (e.g. thunderstorms) may be accounted for by vestigial traces of congenital priming to meet objective dangers.

The concept of transference of anxiety from the real cause onto a more acceptable 'symbol' is also significant. Cognitive theory suggests that it is not the phobic object that the phobic tries to avoid, but the perceived consequences of such an encounter. Bandura's concept of self-efficacy also predicts that a lack of belief in the ability to cope may lead one to develop phobia. Fear of spiders may be representative of lack of confidence in novel situations. This idea of transference will be further discussed in the unified approach to phobia.

Rachman and Wolpe based their counter-explanation of Hans' behaviour on traumatic Pavlovian conditioning. Mowrer's two-stage theory was the basis from which they sprung. The two-stage theory elucidates the relationship between the phobic stimulus, anxiety and avoidance behaviour within the limits of classical conditioning.

The first step involves the contiguous presentation of the unconditioned stimulus (UCS) and the conditioned (phobic) stimulus (CS). Rachman proposes that potent UCS naturally cause fear, pain or loss of support. Varying degrees of each make up the unconditioned response (UCR). By a process of learning, the CS is associated with the USC and becomes capable of eliciting the UCR when presented on its own. Semantically, the UCR now becomes a conditioned response (CR) to a conditioned stimulus.

The second process involves the development of avoidance behaviour, which could be described as the phenotype of the phobia. Any action leading to avoidance of or escape from the phobic stimulus will lead to a decrease in fear or anxiety. This 'anxiety relief' reinforces the avoidance of escape behaviour. Hence fear has a motivating effect that may control behaviour.

Rachman and Wolpe's explanation of Little Hans' behaviour followed the classic conditioning model.

Support for conditioning as a mode of onset is based on trials in which physiological and motoric fear responses have been elicited from both animals and humans in laboratory conditions. However, it is unclear what frequency of laboratory phobias develops after one trial. There appears to be a need to use supertraumatic UCS to generate phobias successfully in one trial.

Some patients do trace the onset of their phobia to past traumatic experiences. Yet it is indisputable that traumatic conditioning experiences cannot account for all cases of clinical phobia. For example, Menzies and Clarke found that only 1 out of 50 cases of childhood water-phobia could be traced to a conditioning event at onset.

If classic conditioning accounts for phobia onset, we would expect an increased incidence of phobias in the population that experienced the air raids of the Second World War. This sharp rise was not found, though some air-raid veterans did develop related phobias. However, rather than this being a reason to dismiss conditioning, it helps elucidate the nature of potent UCS and CS. A successful CS needs to be highly predictive of the UCS. Air-raid sirens did not always lead to bombs dropping from the sky. In fact the exposure to sirens before, during and after the war in non-aversive situations may have inhibited the sirens' potential as a phobic stimulus. This phenomenon is called latent inhibition. Bomb shelters however were more predictive of the likelihood of an air raid. Hence contiguity is less important than contiguency in determining the phobic potential of a CS.

The demise of conditioning theory is mostly due to the failure of the equipotentiality premise. Wolpe and Rachman predict that any 'neutral' stimulus that happens to make an impact on an individual at about the time that a fear reaction is evoked acquires the ability to evoke fear subsequently. Yet, as Seligman points out, common phobic stimuli number approximately fifty. This set of common phobias in non-arbitrary. In fact, evolutionary psychologists tell us that the common phobias represent dangers to pretechnical man e.g. agoraphobia and fear of animals, insect, height and darkness.

Seligman proposes a modification of classic conditioning: the concept of preparedness accounts for the selective nature of common phobias. He argues that common phobias are vestiges of the reflexive defences of pretechnical man against pretechnical dangers. Phobias conferred a selective advantage to phobic individuals. Natural selection ensured that those acquiring phobias faster survived better. Studies have shown that evolutionarily relevant CS (e.g. spiders, angry faces) prove more resistant to extinction than neutral stimuli (e.g. flowers) when paired with the same UCS (e.g. electrodermal stimulation). However it has proved difficult to show that prepared phobias are acquired any faster than 'unprepared' ones, or that they are more resistant to extinction.

The common age of onset of childhood phobias suggests a genetic basis for certain fears. Infant Rhesus monkeys reared in isolation specifically showed fear to conspecific threat stimuli when exposed to slides of other Rhesus monkeys (a novel stimulus for the infants) in various poses. Separation anxiety and stranger fear has a similar age of onset in US, UK, Guatemala and Kung Bushmen children. Marks argues that this pattern of onset is under genetic control but needs interaction with the environment (maturation) for expression.

The 'visual cliff 'apparatus, formulated by Walk and Gibson, was used to test height-avoidance in infants of a variety of species with dramatic result. Evidence for maturation arises from the observation that human infants capable of self-produced locomotion were fearful and avoidant of the visual cliff while the prelocomotive infants showed no fear. Hence the 'background experience' of interaction with the environment while crawling was necessary for the maturation of the fear of heights. This draws attention to the value of prepared phobias. If only one trial is needed to condition a phobia, a trial event leading to falling off a cliff may be one trail too many. Seligman's insistence on at least one conditioning event is the Achilles heel of his version preparedness. In the aforementioned trial involving isolated infant Rhesus monkeys, they showed fear responses to conspecific threat though they had not had a prior aversive or even non-aversive exposure. Moreover, evidence for traumatic conditioning events among clinical phobics is thin. However the preparedness concept is still efficacious if considered outside the limiting framework of classical conditioning. It must be noted that this account is based on a post hoc analysis of survival pressures in our distant past that are difficult to test.

Has modern man lost the ability to develop novel phobias and remained with fears that hearken to premodern times? Agras and colleagues found that snake fear is two times more common than dental fear, despite the fact that modern man is exposed to more dentists than snakes. Levi argues that latent inhibition, caused by exposure to potentially noxious stimuli in non-aversive situations, increases resistance to traumatic conditioning for many modern stimuli. Often training is even given to prevent pain e.g. how to use a hammer safely.

Latent inhibition may explain why modern everyday objects e.g. spoons do not lead to commonly held phobias. However the explanation that noxious stimuli can be experienced in non-aversive situations is lacking. How can a dentists drill or the needle on a syringe be experienced without a measure of anxiety and pain?

Accepting that there is some genetic predilection towards certain phobias, how then do we account for the absence of phobias in every individual? Clarke and Jackson propose that habituation i.e. repeated, not-traumatic exposure to the feared object, results in extinction of phobias. Those who have common phobias are poor habituators or lack the opportunity for safe habituation. Willis found that 48% of parents of clinically water-phobic children claimed that the fear had emerged on the first exposure and had never really gone away. Clarke and Jackson propose another mechanism: dishabituation. This is a process by which innate fears are subdued only to re-emerge when the subject's levels of arousal are raised by a stressful situation. Menzies and Clarke found that 16% of height-fearful students traced onset to a non-conditioned traumatic event that coincided with severe stress or depression.

Recognition that many phobics do not report histories of direct Pavlovian conditioning led Rachman to propose the 'three pathways account' of phobia development. Phobias could develop directly, through classical conditioning, and indirectly through vicarious learning or transmission of information.
Children have been seen to share their parents' fear, which may be accounted for by both vicarious and transmitted learning. However Rachman's predictions on the differences arising from indirect and direct onset have drawn little supporting evidence. He proposed that direct conditioned phobias would be characterised by intense fears and more physiological and behavioural manifestations, while indirectly acquired phobias would be milder with more 'subjective' fears. Ost and Hugdahl have found direct acquisition to predominate in clinical cases and indirect onset to account for mild phobias. However their findings have been criticised on methodology and similar studies have not found supporting evidence.
Some studies do confirm that indirect onset is based on information that proves sensitive to cognitive solutions while direct onset is best treated with deconditioning. This has led to the increasing tendency for clinicians to match mode of onset and anxiety response patterns with treatment.

The cognitive approach differs from behavioural approach in the emphasis that is placed on learning. It is asserted that though conditioning may lead to phobia, this is not the only explanation. The cognitive model proposes that fearful responses arise from the overprediciton of threat posed by certain stimuli. Hence the basis of fear is not the phobic object itself but the beliefs of what shall be consequent to exposure. Thorpe and Salkovskis' experiment, detailed below, is illustrative.

Thorpe and Salkovskis devised a trial in which a variety of phobic beliefs were elicited from a wide range of phobic patients. Then they asked a mixed group of phobics and non-phobics to imagine that they were in the same room as a spider and re-rate the strength of various beliefs. Finally they exposed the mixed group to spiders and related stimuli and asked the group to re-rate their beliefs. Results showed that all phobics when thinking about their specific phobia held negative beliefs. These beliefs were categorised into harm, disgust and 'inability to cope' cognitions. The latter beliefs were the most commonly held. When asked to imagine a spider, the spider-phobics were much more fearful than other groups. Their measures of avoidance and fear were much higher than other-phobics and non-phobics. When exposed to spiders and related stimuli the spider-phobics were again the most fearful, although their re-ratings for various beliefs (except disgust) were slightly lowered.

This may indicate that the spider phobics had over-predicted how fearful they would be on exposure to spiders. Interestingly, the beliefs concerning 'inability to cope' correlated best with observations of fear and avoidance. This reinforces the cognitive argument that it is not the phobic stimulus itself that causes most anxiety but the imagined consequences. This study also confirms that the beliefs of spider-phobics are highly specific to this group and the situation. The lack of fear in other-phobics and non-phobics is due to a lack of negative beliefs about spiders. In the cognitive model, avoidance and escape are seen as 'safety seeking behaviour'. These behaviours arise from internally consistent logical cognitions that motivate escape from the perceived threat. The perception of threat may be over-predicted. However, the beliefs are neither non-cognitive nor unconscious, as Seligman proposed. It may be because maladaptive cognitions make avoidance seem highly advisable that phobias prove resistant to extinction.

Though over-prediction (distorted expectation) is a key tenet of this model, there have been few demonstrations of this phenomenon in phobias. Correlation between expectation and levels of anxiety has been found to be poor. Hence expectation is perhaps not the main determinant of anxiety. An illustration of this is that the anxiety felt by an insect phobic when looking at a picture of a cockroach does not correlate with any rational expectations. Perhaps this is indicative of irrational expectations i.e. the cockroach will come alive. In my opinion, this poor correlation is not suggestive or irrationality but shows clearly that the phobic stimulus itself is not the sole determinant of anxiety. The anxiety may be residual and may be caused by heightened arousal and trait anxiety.

Biological explanations of anxiety at present remain quite separate from psychological explanations. Evidence for biological theories comes from efficacy of medication. For example anti-obsessive drugs that increase serotonin seem to reduce obsessive compulsive disorders. However at present there is no effective drug for specific treatment of phobia. This approach may prove more relevant in the future if a biological explanation is found for state and trait anxiety. Drugs to alleviate anxiety in general may yet prove useful in improving specific phobias.

Having considered the aetiology of phobia, how then may we best account for its persistence? Classic conditioning proposes that avoidance and escape persist because the anxiety relief they provide reinforces them. While this proposition is predictive of lab-conditioned phobias, it does not apply well to natural phobias, which do not extinguish according to behaviourist predictions. In the laboratory phobias gradually decrease as CS is presented in absence of UCS. Clinical phobia persists despite numerous presentations of CS without the UCS. Superficially, Mowrer's approach might explain this: phobias persist due to failure of exposure (avoidance) rather than failure of extinction. However most phobics cannot organise their world to exclude all phobic stimuli.

Furthermore, repeated phobic episodes offer opportunities for corrective information to be taken on board. Lab-conditioned phobias show decline when the subject is merely informed that the UCS will not follow the CS. However natural phobias prove resistant to corrective information and hence persist. Gray moves gradually towards a cognitive approach by describing anxiety as a central state that mediates behavioural responses to stimuli that are either punishment or non-reward. Hence anxiety is a motivational response that guides animal behaviour. This 'behavioural inhibition system' does suggest that phobic behaviour will persist as long as the anxiety state lasts. However it offers no clues as to why phobias are resistant to de-coupling of the UCS and CS, or to corrective information. Cognitive therapy offers a solution to this problem.

Cognitive accounts argue that persistence is due to the remaining presence of maladaptive cognitions too. However, behaviours like avoidance can also drive maladaptive cognitions. The key to the phobia is the interpretation of the meaning and consequences of the phobic stimulus. Why don't phobic experiences where the predicted catastrophe does not materialise act to disconfirm maladaptive cognitions? The cognitive answer is that each phobic experience is judged as a 'near miss'. The subject considers their 'escape' a direct consequence of whatever avoidance behaviours took place. This is further motivation for active avoidance at the next occasion. It also protects their maladaptive beliefs from re-evaluation. This is an example of behaviour reinforcing maladaptive cognitions.

In fact many avoidance behaviours are designed to protect against the idiosyncratic beliefs and consequences expected. For example a spider-phobic may wear long sleeved shirts to protect from being crawled upon. This also protects them from the opportunity for disconfirmation because the have armoured against any such eventuality.

Clinically, cognitive therapy has proved a very successful adjunct to behavioural therapy. Behavioural therapy is based on graded exposure to the phobic stimulus, the end of which would be desensitisation by habituation. Traditionally any cognitive work would be of a general, biographical nature. However cognitive therapy that focuses on specific beliefs about the phobic stimuli and consequences of exposure has proved to have a multiplicative rather than additive effect on effectiveness of behavioural therapy. In short, patients exposed to a combination treatment achieve gains in fewer sessions.

Psychoanalytic and biological theories offer little insight on the persistence of phobias. Freud's assertion that phobias do not occur at all when the vita sexualis is normal is yet unproven. However the success of both cognitive and behavioural approaches to therapy show that an in-depth analysis of childhood experiences and sexual ideas is unnecessary for successful treatments.

Conclusion: Unified approach
Accounts of phobia are fragmented by default due to what seems to be the determination of each school of thought to produce complete and exclusive models. Onset of phobias cannot be completely explained by any extant model. The evidence suggests that a variety of mechanisms including conditioning, vicarious learning and simple transmission can account for onset. However there are certain factors that increase vulnerability to developing phobias. Genetic preparedness means that we are innately afraid of certain things until they are proven harmless. The state of anxiety or depression that one is in can make a person more vulnerable to dishabituation of innate fears or development of novel phobias. The state of anxiety can be affected by the environment (state anxiety) or can be characteristic to the individual (trait anxiety). The tendency towards anxiety can affect both onset and maintenance of phobias. Eysenck proposed a link between introversion and anxiety problems. In his two dimensional analysis of neurosis he proposes that extroverts show poor social conditioning and are more likely to develop behavioural problems. Introverts are easily conditioned so they tend to display anxiety problems. A reinterpretation of this from Clarke's cognitive perspective is that the focus of a person's attention on 'subjective' factors can lead to persistence of phobias. For example social phobics, who may try to be unobtrusive and never make eye contact, become inattentive to stimuli that may otherwise correct their negative self-evaluations.
Anxiety as a vulnerability factor for phobia may prove sensitive to modulation with drugs. The most important vulnerability factor is the presence of maladaptive beliefs. These can be acquired in ways similar to how phobias are acquired i.e. direct, vicarious or transmitted learning. However these maladaptive beliefs are the main reason for the maintenance of phobias. Cognitive behavioural therapy has proved effective because it provides opportunities for the patient and the therapist to re-evaluate maladaptive cognitions.

Maladaptive beliefs are less concerned with the object of fear and more to do with the person's belief in their ability to cope. This explains why phobias arise or re-appear in periods of stress and depression. The subject perhaps questions his or her own efficacy. This self-doubt may emerge as phobia because it is most acutely felt in situations where there is little control. This lack of control may be due to increased arousal coupled with the absence of a 'script', which is used to deal with the situation. This may account for the fact that modern noxious stimuli don't cause as much phobia as pretechnical phobic stimuli. We have scripts with which to navigate exposure to modern stimuli. Yet pretechnical stimuli cause increased arousal with no script to fall back on.

I propose that to test this 'lack of control' aspect one could contrast the incidence of snake phobia in cultures where there is little exposure and knowledge about snakes with a culture where snakes are commonly encountered and dealt with.

Rachman and others devised a procedure to test whether escape led to strengthened anxiety. They gave one group of phobics the option to escape when they had had enough exposure. To the other group the gave instructions to remain until their anxiety had declined to seventy percent. Of the group with the escape option, most did not escape. In fact they experienced a greater reduction of anxiety. This finding obliquely suggests that control is an important factor that deserves more attention.

Phobias may prove resistant because the real fear (of self-efficacy) is often transferred onto an object or situation where such fear would be more acceptable. Current therapy is aimed at desensitisation through exposure and providing corrective information through cognitive therapies. Perhaps strategies to increase confidence within a variety of novel situations (and not just the phobic situation) would be effective in reducing self-efficacy anxiety. Exposure may work even better if a script to deal with the phobic situation is formulated by both patient and therapist and is put into practise.

You write well and have a mature grasp of the language and topic. Bellissimo!

Some additional points:

Seligman's considered the type of stimuli that phobics fear and his theory of preparedness postulated that there must be some kind of evolutionary pressure which created a preference for certain stimuli that require less trials to have a fear association. These 'prepared' stimuli will have had some survival significance through human evolution and hence, a biological basis.

Ohman et al. (1975) conditioned subjects with a shock and either a neutral picture or one of a snake. All of the subjects acquired fear conditioning but after the shock was removed, the subjects who received the neutral pictures no longer exhibited fear to the stimuli whereas the patients who received had received the snake pictures still did. This is evidence for the preparedness theory, however, McNally (1981) found the opposite effect.

There is also little evidence to support the notion that stimuli which threatened human survival in the past (phylogenetic fear-relevant stimuli) would be more 'prepared' than "modern" stimuli (ontogenic fear-relevant stimuli) which though dangerous will not have acquired biological status.

The idea that phobias are harder to extinguish because of their biological basis receives the most support out of all Seligman's claims. It is also clear that there must be some biological basis to explain cross-cultural phobias with consistent onset times in development.
How can phobias be explained, and what is the best way to treat them?

How can phobias be explained, and what is the best way to treat them?

Phobias are defined as a disrupting, fear-mediated avoidance of an object or situation that do not present any real danger. They are often specific, such as arachnophobia (fear of spiders) or agoraphobia (fear of being outside), but may also be more general, such as social phobia. The traditional theory of the aetiology of phobias proposes that they are acquired via classical conditioning, with an object or situation associated with an averse event and therefore instilling fear, but this view has been much criticised. The evidence that phobias can be acquired without such a conditioning experience, and that some objects are more commonly feared than others seems to refute an entirely associative perspective on the acquisition of phobias, and some researchers have taken to explaining them in terms of Darwinian evolution instead, claiming that many phobias are biologically "prepared" for by ancient threats. These different theories naturally have an influence on the treatment of phobias, which has traditionally focused on the behavioural methods of exposure and prevention of avoidance or escape. Cognitive treatments have also been shown to be effective for some types of phobia, especially social phobias, and recent studies demonstrate that specific phobias can be effectively treated in just one session.

The Freudian psychoanalytic theory of phobias suggests that they are a defence against the anxiety produced by repressed id impulses, with the anxiety becoming displaced from the feared impulse (such as hatred towards one's father) and moved towards an object or situation which is symbolically connected to it. It is these objects, then, that become the phobic stimuli, and by avoiding these the individual can avoid dealing with his repressed conflicts. Arieti (1979), reinterpreted this theory as meaning that the repression is of a particular childhood interpersonal problem. According to this perspective, the child's initial trust of the people around him turns into a fear that adults are not reliable. In order to be able to trust people again, the child unconsciously transfers this fear onto a more acceptable focus, such as an object. The phobia then surfaces in adulthood when the individual undergoes a stressful experience. Fascinating as these speculations are, a psychoanalytic interpretation of the aetiology of phobias holds little water, and is even less useful in coming up with a treatment, since the phobia seen as symbolic and is therefore not dealt with directly.

A more acceptable theory of the acquisition of phobias is the classical conditioning approach. This assumes that fears are acquired by a process of associating a conditioned stimulus (CS), such as pain, with an unconditioned stimulus (UCS), such as a dog or a supermarket, via a traumatic experience. According to Mowrer's 2-stage theory, the fear develops motivating properties and any behaviour which reduces the fear, such as avoidance or escape, will become reinforced. There is quite a lot of evidence in support of the conditioning view of phobias, mainly from laboratory research in inducing fear in animals. Clinical observations, when patients recall a single traumatic experience triggering their phobia, also supports the conditioning theory, as do incidental observations from aversion therapy, in which participants rapidly come to associate nausea with alcohol. There are, however, some major problems with the classical theory, the main one being that many individuals experience extremely traumatic events, such as air raids in WW2, yet fail to acquire a phobia. Secondly, there has been little success in attempting to condition fears in humans, with most patients reporting indifference to the conditioning stimuli employed in ECT (Marks and Gelder, 1976). The conditioning theory also relies on the equipotentiality premise that all objects and situations have an equal capacity to be transformed into a fear signal, but the distribution of fears clearly demonstrates that this is not the case, with some things (such as spiders) being much more commonly feared than others (such as chocolate). Finally, there are often no plausible conditioning events recalled by patients, and some snake phobics acquired their fear without ever having seen a snake, which causes great problems for a conditioning theory of phobias.

In light of these findings, Rachman (1991), proposed that there are three pathways to fear - classical conditioning, vicarious acquisition and transmission of verbal information. The second of these assumes that we acquire much behaviour via the processes of observational learning and modelling, and that therefore fears can be acquired both directly and indirectly, by vicarious experience. Experiments into human electrodermal conditioning (Cook and Harris, 1937), for example, found that there are increases in the magnitude and frequency of skin-resistance responses to a cue after the subject has seen someone-else receive a shock. Finally, fears can also be acquired by the absorption of verbal information, as is thought to have occurred in South East Asia, when an epidemic of Koro (fear that genital shrinkage will cause death) started from rumours that the Viet Cong had poisoned food to produce impotence.

The discovery that non-contigious conditioning can take place, that is, that an association can occur even with a time lapse between the CS and the UCS, removed much of the difficulties with the classical conditioning theory of phobias, and this led to a revival in the form of the neo-conditioning theory (Rachman, 1991). Particular interest was given to the blocking effect, which states that a stimulus will not become a CS even if repeatedly presented before an UCS, if it does not have any informative value. This can explain why some stimuli are more likely to become the focus of phobias than others, since if a stimulus does not accurately predict the UCS, or it is already well predicted by another CS, then no conditioning will occur. The neo-conditioning theory can also explain why some individuals fail to acquire a phobia after a traumatic experience, since if they have had many positive experiences with that object or situation previously, one bad experience will not necessarily result in a fear acquisition. Davey (1989), has also proposed two ways in which a traumatic event might not lead to a phobia - firstly, if the CS is presented alone several times before it is paired with the UCS, then this latent inhibition makes it more difficult for an association to form. Secondly, even when a traumatic experience does establish a CS-UCS association, the strength of the conditioned response will depend on the individual's evaluation of the UCS. Fear can therefore be attenuated without weakening the association, if the individual re-evaluates the event favourably. Davey investigated these two factors with respect to dental phobics, and his findings strongly support a neo-conditioning theory of phobia acquisition. It was found that those people who had no dental anxiety had also experienced less traumatic events, and that those who had a trauma but did not acquire a phobia had had many un-traumatic dental experiences prior to the traumatic one (latent inhibition). It was also found that those individuals with good pain endurance were less likely to have a phobia, perhaps because this facilitates latent inhibition by providing more pain-free experiences. Finally, when latent inhibition should have weakened the acquisition of anxiety but did not, it was demonstrated that an extremely painful experience had attenuated the latent inhibition process. It seems, then, that the neo-conditioning theory can explain many aspects of phobia acquisition, but it still has one major drawback - how do individuals acquire a phobia to an animal, such as a snake, which they have never encountered?

To answer this question Seligman (1971) proposed an evolutionary based theory of phobias and pointed out four characteristics which distinguish phobias from conditioned fears - they are easier to acquire, irrational, highly resistant to extinction and have a quality of 'belongingness' between the fear and threats of a pre-historic origin. This led Seligman to propose a preparedness theory of phobias in which it is suggested that evolution predisposed organisms to easily learn those associations that facilitate survival. In this way, phobias are seen as instances of biologically prepared learning involving objects and situations which have threatened human kind throughout their history. This theory predicts that Pavlovian conditioned responses established to fear-relevant stimuli (such as snakes and spiders) should mimic the characteristics associated with phobias. The method most frequently used to verify this proposition is habituation (the stimulus is presented without the UCS), acquisition (CS+ trials are mixed with CS- trials) and extinction (non-reinforced presentation of the stimulus). The magnitude of the skin conductance response (SCR) is taken as the dependent variable. The evidence for this theory is inconsistent, however, with very little support for the notion that phobias are more easily acquired than conditioned fears. In fact, conditioning to fear-relevant stimuli is just as adversely affected by sub-optimal learning conditions as is conditioning to fear-irrelevant stimuli (Hugdahl and Ohman, 1980). It has also been shown that non-prepared stimuli can acquire aversive associations just as quickly as prepared stimuli (McNally, 1987), implying that phobias are not necessarily easier to acquire. The characteristic of irrationality was taken by Seligman to mean that prepared associations reflect a primitive, non-cognitive form of conditioning, and will therefore be held despite the knowledge that the phobic object will cause no harm. Support for this comes from Ohman et al's (1975) study in which neutral stimuli and pictures of snakes were shown to subjects, half of whom were shocked when the latter appeared, and half of whom when the former appeared. Both groups acquired the conditioned fear, but when they were instructed that they would receive no more shocks, this did not extinguish the fear of snakes. McNally (1981), however, reported the opposite results, suggesting that the concept of irrationality is far from clear cut. There is also no conclusive evidence that associations will develop to threatening stimuli relevant to mankind of the past, although it is true that there are many more spider phobics than people with phobias to motor cars, even though the latter are much more common in everyday life. Finally, the idea that phobias are more resistant to extinction has the most empirical support, and it has been found that responses to fear-relevant stimuli are slower to extinguish than those to fear-irrelevant stimuli. There are, however, alternate explanations for these findings, which do not require preparedness. Animatedness may mediate resistance to extinction, and it has also been found that ontogenetic stimuli, such as guns and electrical outlets, are also slow to extinguish, suggesting that they might be more relevant to the study of phobias than phylogenetic stimuli.

Recent advances in the preparedness theory of phobias have overcome some of these problems. Rather than viewing phobias as non-cognitive, researchers are looking at the manner in which the cognitive processing of fear information occurs (Ohman , 1985). Similarly, it is suggested that humans nay be biologically prepared to learn to fear perceptual features embedded in prevalent phobic stimuli, but these do not need to be associated with a UCS to become fear elicitors. Finally, it is clear that phobias persist because phobics avoid exposure to the feared object, not because they are intrinsically slow to extinguish. Another non-associative account has been proposed by Menzies and Clarke (1995), in an attempt to explain why fear of some stimuli can develop in the absence of any previous experience of the feared stimulus. Evidence for this Darwinian view that fear is independent of experience comes from a study by Sachett (1966), who found selective sensitivity of rhesus monkey infants who had been reared in isolation, to conspecific threat displays. This fear seems to develop in the absence of any traumatic conditioning. Similarly, separation anxiety in human infants is unrelated to the time they have spent with the carer, or to past aversive experiences during separation. Both these examples demonstrate adaptive behaviour which occurs without the need for associations to be learned, suggesting that they may be innate. A more striking example of this comes from Walk and Gibson's (1961) "visual cliff" experiments, which were used to show that fear of heights appears without any aversive association learning. Babies, chicks, kids and lambs all refuse to crawl or walk onto the "deep" side of the board, implying that they have an innate fear of heights. It is probable, however, that some locomotive behaviour is required for the development of this avoidance behaviour to occur, since pre-locomotive infants show little change in heart rate when placed over the cliff (Campos, 1984). If we are all afraid of heights soon after birth, however, why are relatively few people phobics in adulthood? It is possible that repeated habituation to the feared object results in failure to acquire a phobia, and it may also be the case that phobias arise through dishabituation.

Cognitive processes, although often left out of theories of the acquisition of phobias, are important in understanding the maintenance of the disorder. Clark (1999), has pointed out that there are several cognitive mechanisms which prevent phobic patients from disconfirming their beliefs that an object or situation is threatening. Firstly, phobics frequently engage in safety behaviours, such as escape or avoidance, which ease their anxiety and therefore prevent them from realising that the perceived threat is imaginary. In the case of social phobia, such behaviours can also inadvertently create some of the symptoms that the phobics fear, by drawing attention to themselves and influencing others in a way which partially confirms their fears. It has also been found that spider phobics demonstrate an attentional bias towards spider-related stimuli (Lavy and van den Haut, 1993), which may increase their fear. In contrast, it seems that social phobics tend to avert their attention away from other people in an attempt to reduce the threat, but this may actually increase their fears, since it prevents then from noticing that the other person actually likes them. Beck (1976) also suggested that spontaneously occurring mental images in which patients "see" their fears realised may play an important role in enhancing the perception of a threat. In support of this, Hackmann, Surawy and Clark (1998) found that 77% of social phobics reported spontaneously occurring, negative, observer-perspective images which they later admitted were at least partly distorted. This may explain why social phobics, who attend less to external social cues, still believe that they are coming across badly.

It has also been suggested that the beliefs that phobics hold are crucial to the maintenance of their anxiety (Thorpe and Salkovskis, 1995). It was found that phobics have a range of negative beliefs about their phobic object which form a logical framework for their fear responses. A patient with a bird phobia, for example, holds a different meaning of birds than a non-phobic individual has, and this is to do with the possible harm the bird could inflict, with the beliefs about their ability to cope and the efficacy of escape or avoidance. The importance of cognitive factors in phobias has been strikingly demonstrated by Watts et al's (1986) study into the colour-naming of phobia-related words. They found that spider phobics were extremely retarded on this task, but not on a conflicting colour-word Stroop or on a Stroop with more general threat words. Desensitisation of the phobics decreased the interference on the task. Far from being non-cognitive, therefore, phobics' beliefs about the phobic object and their reaction to it are extremely important in maintaining a phobia.

These different theories on the aetiology of phobias have led to different treatment methods, the most popular of which is the behavioural approach. Systematic desensitisation is used to reduce the anxiety of phobics, by getting the patient to imagine increasingly frightening scenes while in a state of deep relaxation. This is very effective, but even more important is exposure in vivo to the feared object or situation, although modeling techniques can also be used. In this method, fearful patients are shown a film or live demonstration of others interacting fearlessly with the phobic object. In order to test which therapy is the best for treating spider phobics, Ost et al (1997), randomly assigned 46 phobics to three groups - direct treatment, involving prolonged exposure to spiders, as well as participant modelling in which the patient copies the therapists actions, direct observation, in which the patients watch while another is exposed to the spider, and indirect observation, in which the participants watch a film of live modelling. The direct treatment was found to be significantly better than the other two methods, which did not differ. It has been suggested that direct treatment is the most effective because it gives the patient enactive information (Bandura, 1977) which affects his perceived self-efficacy in a positive manner, thereby resulting in a decreased fear of the phobic object.

According to Mowrer's 2-stage theory, agoraphobic patients should avoid leaving the feared situation until their fear has subsided, lest the escape/avoidance behaviour is reinforced. The habituation model further suggests that decrements in anxiety only begin after a prolonged period of exposure. To test these claims, de Silva and Rachman (1984), treated two groups of agoraphobics with one of two therapies - in the endurance condition, the patients stayed in the target situation until their self-rated anxiety had dropped by half. In the escape condition, they left the situation when their anxiety reached a high pre-set level. Surprisingly, both groups improved and there was no significant difference between the two therapies suggesting that, unlike Mowrer's theory implies, escaping while still fearful does not necessarily lead to increased fear and avoidance. It seems that an exposure strategy incorporating systematic escaping is as effective in treating agoraphobia as a purely exposure one. As is clear from this study, different specific phobias require different treatments, and research has also been carried out as to the best therapy for blood phobia (Ost, Fellenius and Sterner, 1991). 30 patients with this type of phobia were treated either with applied tension (AT), tension only (T) or exposure in vivo (E). Although all groups improved considerably, AT and T differed from E, suggesting that tensing the muscles to increase blood pressure and reduce the risk of fainting is the best way to treat blood phobia. Since tension only was as effective as applied tension (which involved exposure) and much more effective than exposure alone, this suggests that blood phobias are one of the few phobias which are not best treated with exposure.

In conclusion, it seems that the aetiology of phobias can be explained in several ways, with many phobias fitting into a conditioning theory of acquisition. An evolutionary theory of prepared or innate phobias may help explain those fears that develop without a conditioning experience or even without any experience of the feared object or situation at all. It appears plausible that there are numerous pathways to fear, as Rachman suggested, and that phobias can also be acquired vicariously or by verbal transmission of information. The treatment of phobias also depends on which type of phobia being dealt with, but in general, desensitisation to reduce anxiety and exposure to disconfirm fears of pain or trauma seem to be the most effective, although the notion that escape or avoidance need to be discouraged has not always been supported.

Another excellent essay.

Some additional points:

Seligman's considered the type of stimuli that phobics fear and his theory of preparedness postulated that there must be some kind of evolutionary pressure which created a preference for certain stimuli that require less trials to have a fear association. These 'prepared' stimuli will have had some survival significance through human evolution and hence, a biological basis.

Ohman et al. (1975) conditioned subjects with a shock and either a neutral picture or one of a snake. All of the subjects acquired fear conditioning but after the shock was removed, the subjects who received the neutral pictures no longer exhibited fear to the stimuli whereas the patients who received had received the snake pictures still did. This is evidence for the preparedness theory, however, McNally (1981) found the opposite effect.

There is also little evidence to support the notion that stimuli which threatened human survival in the past (phylogenetic fear-relevant stimuli) would be more 'prepared' than modern stimuli (ontogenic fear-relevant stimuli) which though dangerous will not have acquired biological status.

The idea that phobias are harder to extinguish because of their biological basis receives the most support out of all Seligman's claims. It is also clear that there must be some biological basis to explain cross cultural phobias.

You research the topics well and have a sound and mature grasp of the subject matter.