The incubation theory of fear/anxiety: experimental investigation in a human laboratory model of Pavlovian conditioning

The aim of this work was to test Eysenck's incubation theory of fear/anxiety in human Pavlovian B conditioning of heart rate (HR) responses. The conditioned stimuli (CSs) were phobia-relevant slides (snakes and spiders) and the unconditioned stimuli (UCSs) were aversive noises. The subjects were presented with two levels of noise intensity during acquisition and three levels of nonreinforced CS presentation (CS-only) in a delay differential (CS+/CS-) conditioning paradigm (2 x 3 x 2). Consistent with the incubation theory, conditioned HR acceleratory responses were sustained (resistance to extinction) for high-noise intensity and short-presentations of CS-only subjects. During the extinction phase, HR acceleratory responses quickly extinguished in low-noise intensity groups after the first presentations of CS-only. These findings were interpreted as support for the incubation theory of phobic fear.     

Electrodermal conditioning to potentially phobic stimuli: Effects of instructed extinction

It was hypothesized that electrodermal responses to potentially phobic stimuli “conditioned” through verbal threats about an aversive UCS should be equally resistant to instructed extinction as the responses obtained by actual CS-UCS pairings. Two groups of human subjects were exposed to pictures of either snakes and spiders (phobic CSs), or circles and triangles (neutral CSs) in a differential Pavlovian conditioning paradigm. Changes in skin conductance were recorded. Half of the subjects in each group were threatened with a shock-UCS while the other half were given shock-reinforced CS presentations. At the onset of extinction, all subjects in each of the four groups were informed that no more UCSs were to be delivered, and the shock electrodes were removed. All groups showed evidence of conditioning during acquisition. During extinction there was an immediate drop in responding in the two neutral groups, whereas the two phobic groups showed reliable evidence of resistance to extinction, with no differences between the threatened- and the CS-UCS-group. The observed resistance to extinction found in the phobic groups implies a similarity to the irrationality of real-life phobias. Furthermore, the data are in accordance with analysis of electrodermal fear-conditioning as a case of prepared learning.     

One-trial learning and superior resistance to extinction of autonomic responses conditioned to potentially phobic stimuli.

Human subjects were exposed to pictures of potentially phobic (snakes) and supposedly neutral (houses) objects as conditioned stimuli (CSs) in a Pavlovian conditioning experiment with shock as unconditioned stimulus (US), and skin conductance and finger pulse volume as dependent variables. The skin conductance responses conditioned to phobic stimuli were acquired after one CS-US pairing, and showed practically no extinction, whereas the responses to neutral stimuli showed very little resistance to extinction after both 1 and 5 reinforcements. The superior resistance to extinction of the phobic condition was interpreted to be a specific associative effect. In general, the skin conductance acquisition data showed tendencies similar to those during extinction. For finger pulse volume responses, however, there were very weak conditioning effects, and no effect of stimulus.     

‘Preparedness’ and ‘arousability’ as determinants of electrodermal conditioning

‘Arousability’, as defined through spontaneous electrodermal responses, has been empirically linked to anxiety, phobic symptoms and outcome of systematic desensitization. Previous data from our laboratory indicate that ‘preparedness’, as defined through potentially phobic vs. fear-irrelevant or ‘neutral’ conditioned stimuh, is an important determinant of electrodermal conditioning. The present experiment compared groups selected to be high or low in spontaneous responding during differential conditioning to potentially phobic or neutral stimuh. It was found that the effects of these two factors were essentially additive, i.e. conditioning and resistance to extinction were better for phobic stimuli and for high-arousal groups. The high-aroused group with phobic stimuh showed diffuse responding during acquisition, not differentiating between reinforced and unreinforced cues. However, it was the only group that failed to extinguish during 20 trials, which indicates that high arousal gives superior resistance to extinction particularly for phobic stimuli.     

Electrodermal conditioning to stimuli of evolutionary significance: Failure to replicate the preparedness effect

In the present study, an attempt was made to replicate the preparedness effect reported by Öhman, Fredrickson, Hugdahl, & Rimmö (1976). Following Öhman et al. (1976) as closely as possible, a differential conditioning procedure was carried out in which subjects'skin conductance responses (SCRs) were conditioned either to stimuli of evolutionary significance (slides of snakes and spiders) or to evolutionally neutral stimuli (slides of mushrooms and flowers). The experiment consisted of 8 habituation, 12 acquisition, and 20 extinction trials. Electric shock served as an unconditioned stimulus during the acquisition phase. Although SCRs showed significant decreases during habituation and were significantly influenced by the conditioning procedure during acquisition, they were not found to extinguish significantly more slowly in the group that saw slides of snakes and spiders. This result contradicts the earlier results reported by Öman and colleagues. Possible explanations for this failure to replicate their results are discussed.     

Concurrent excitors limit the extinction of conditioned fear in humans

In a human fear conditioning experiment, with on-line expectancy ratings and electrodermal responding as indices of fear, two neutral stimuli (pictures of geometric shapes) were first established as reliable predictors of an electric shock. In the subsequent extinction phase, the two stimuli were repeatedly presented in compound, without the shock. The final test phase consisted of individual stimulus presentations again, which resulted in a strong return of the conditioned responses. This effect was not observed in non-conditioned control stimuli. Hence, behavioral effects of extinction seem highly specific to the stimulus constellation that has gone through the extinction procedure. We argue that pharmacological, behavioral and/or cognitive manipulations that could prevent configural processing of stimulus constellations have direct clinical potential.     

Preparedness and electrodermal fear-conditioning: ontogenetic vs phylogenetic explanations

In a review of existing theories of learning, Seligman (Psychol. Rev. 77, 406-418, 1970) suggested that humans should have an evolutionary derived preparedness to associate fear-relevant (e.g. snakes) events with aversive reinforcers. The preparedness hypothesis has been extensively tested by Ohman and his colleagues. One argument against a non-preparedness explanation for the Ohman findings has been that culturally aversive stimuli, like pictures of guns have not shown the same resistance towards extinction as pictures of snakes. However, the effect of pointing a gun directly towards the S vs pointing it to the side has not been tested. Therefore both slides of guns and snakes, directed both towards and aside from the subject, were used as conditioned stimuli (CSs) in the present study. A second question that has been discussed in the preparedness-literature is the quality of the unconditioned stimulus (UCS), i.e. if only shock can act as UCS for prepared CSs. Thus, both shock and noise UCSs were used in the present study. Skin conductance responses (SCRs) were recorded as dependent measures. The results showed conditioned acquisition, i.e. larger SCRs to CS+ than to CS-, in all groups except for the shock and noise UCS groups with the gun pointing aside as CS+ (where actually larger responses were observed to the CS-, i.e. the gun pointing towards). The extinction data showed significantly larger SCRs to CS+ than to CS- for both snakes and guns directed towards the S. Strongest resistance to extinction was observed for the group with the gun pointed towards as CS+ and with noise as UCS. The gun with noise as UCS pointed towards the S was not different from the snake with shock as UCS. Taken together, the results have shown three things; (a) directing a fear-relevant CS towards the S was a potent manipulation, and especially directing a gun with noise as UCS; (b) shock was overall not superior to noise as UCS, and especially not for snake CSs; (c) a weak form of unique belongingness was demonstrated.     

Emotional conditioning to masked stimuli and modulation of visuospatial attention

Two studies investigated the effects of conditioning to masked stimuli on visuospatial attention. During the conditioning phase, masked snakes and spiders were paired with a burst of white noise, or paired with an innocuous tone, in the conditioned stimulus (CS)+ and CS- conditions, respectively. Attentional allocation to the CSs was then assessed with a visual probe task, in which the CSs were presented unmasked (Experiment 1) or both unmasked and masked (Experiment 2), together with fear-irrelevant control stimuli (flowers and mushrooms). In Experiment 1, participants preferentially allocated attention to CS+ relative to control stimuli. Experiment 2 suggested that this attentional bias depended on the perceived aversiveness of the unconditioned stimulus and did not require conscious recognition of the CSs during both acquisition and expression.     

Resistance to extinction of conditioned electrodermal responses: a study of the incubation fear hypothesis

In the present study we examined Eysenck's incubation hypothesis of fear. Probability of skin conductance response (SCR) was analyzed for a sample of 79 undergraduate women, ranging in age from 18 to 25 years. Different groups of participants were conditioned to two levels of unconditioned stimuli (UCS) intensity and presented to three levels of unreinforced conditioned stimuli (CS) exposures (extinction phase) in a delay differential conditioning paradigm. The CSs were fear-relevant slides (snakes and spiders) and the UCSs were aversive tones. Analysis did not show a clear incubation effect; instead an increased resistance to extinction of SCR probability in association to the high-UCS and the short unreinforced CS presentation was evident. Findings support partially Eysenck's incubation theory of fear/anxiety.     

Return of fear in a human differential conditioning paradigm caused by a stimulus change after extinction

In a human fear conditioning experiment, 32 participants were trained in a differential conditioning procedure with geometrical shapes as CS+ and CS- (four presentations each), and an electric shock as US. Measures of conditioned responding were skin conductance response (SCR) and retrospective US-expectancy ratings. For half of the participants (Generalization Group, GG), the subsequent extinction phase consisted of four nonreinforced presentations of generalization stimuli (GS+ and GS-). Participants from the Extinction control Group received an equal amount of nonreinforced presentations of the CSs. Finally, all participants were tested with the original CSs. The results from both measures clearly show an increase in the size of the discrimination upon the stimulus change after extinction in the GG. Because this pattern is not observed in the Extinction control Group, extinction performance appears to be somehow restricted to the perceptual characteristics of the extinction stimulus. Interestingly, the size of the conditioned SCR discrimination in the GG is not influenced by the stimulus change after acquisition. This observation points to a differential impact of stimulus change after acquisition vs. extinction treatment. The findings are discussed from the theoretical perspective of renewal and the clinical perspective of Return of Fear.     

Conscious knowledge of CS-UCS contingency information affects extinction retrieval of conditioned disgust responses: Findings from an online de novo disgust conditioning task

ObjectiveThe present study aimed to establish and develop an online de novo conditioning paradigm for the measurement of conditioned disgust responses. We further explored the effects of explicit instructions about the CS-UCS contingency on extinction learning and retrieval of conditioned disgust responses.MethodThe study included a sample of 115 healthy participants. Geometric figures served as conditioned stimuli (CS) and disgust-evoking pictures as unconditioned stimuli (UCS). During disgust conditioning, the CS+ was paired with the UCS (66% reinforcement) and the CS- remained unpaired; during extinction and retrieval, no UCS was presented. Half of the participants (n = 54) received instructions prior to the disgust extinction stating that the UCS will not be presented anymore. 1-2 days or 7-8 days later participants performed a retrieval test. CS-UCS contingency, disgust and valence ratings were used as dependent measures.ResultsSuccessful acquisition of conditioned disgust response was observed on the level of CS-UCS contingency, disgust and valence ratings. While some decline in valence and disgust ratings during the extinction stage was observed, contingency instructions did not significantly affect extinction performance. Retrieval one week later revealed that contingency instructions increased the discrimination of the CSs.ConclusionsExtinction of conditioned disgust responses is not affected by explicit knowledge of the CS-UCS contingencies. However, contingency instructions prior to extinction seem to have a detrimental effect on long-term extinction retrieval.     

The effects of language conditioning on various aspects of anxiety

Staats' three-function learning theory provides the basis for investigating the effects of emotionally-relevant self-verbalizations (SV) on the physiological, subjective-affective and behavioural aspects of anxiety. Using aversive electric UCSs and slides of snakes (CS), anxiety was classically conditioned in 88 volunteer Ss. In 20 subsequent language-conditioning trials (without aversive electric UCSs), the same snakes slides were paired with UCS verbalizations having either positive or negative connotative meanings. Half of the Ss were exposed to a living snake prior to language conditioning. The results show a complete extinction of the conditioned anxiety response in groups with positive SV whereas negative SV impeded extinction; the latter effect could only be found in groups without exposure to snakes prior to language conditioning. In general, the affective evaluation of snakes improved in groups with positive SV and deteriorated in groups with negative SV. However, these effects were more pronounced in groups without exposure to snakes. Although the results indicate that Ss in all groups with positive SV exhibited more approach behaviours than Ss with negative SV, this trend was not statistically significant. The possible relevance of the results for a language-conditioning approach to anxiety reduction is briefly discussed.     

Brain asymmetry and human electrodermal conditioning

Two experiments are reviewed that demonstrate effects of brain laterality on human classical conditioning. Pictures of facial emotional expressions were used as conditioned stimuli (CSs) together with shock as unconditioned stimlus (UCS). Bilateral electrodermal responses were recorded as dependent measures. In the first experiment, one group was conditioned to an angry face, and one group to a happy face. During extinction, the face-CSs were presented to the right hemisphere on half of the trials and to the left hemisphere on the other half of the trials. Results showed that the right hemisphere was superior in showing persisting effects of learning, and especially to the angry CS+. In the second experiment, lateralized presentations of the angry and happy faces were made during acquisition, with foveal presentations during extinction. Once again, the angry face elicited greater skin conductance responses (SCRs) during extinction in the group that had this stimulus presented to the right hemisphere during acquisition. It is concluded that emotional conditioning is differentially regulated by the two hemispheres of the brain.     

Backward conditioning as an extinction procedure

In two conditioned suppression experiments, rats received Pavlovian forward defense conditioning in which tonal conditioned stimuli (CSs) terminated with the onset of scrambled grid shock unconditioned stimuli (USs). After this experience, the rats then received a Pavlovian backward conditioning procedure in which the same USs now terminated with the onset of the same CSs. Although the two experiments differed greatly in terms of CS and US parameters, number of forward and backward pairings, and in terms of the general techniques used to establish and measure the Pavlovian conditioned response (CR), the results of both experiments agreed in showing that backward conditioning can indeed weaken a CR based on forward pairings. The results also show that, under some conditions, the backward procedure can be at least as effective in weakening an established CR as the traditional CS-alone extinction procedure; but, under other conditions, the backward procedure is less effective and leads to more spontaneous recovery than the CS-alone procedure.     

Sensitization and aversive conditioning: Effects on the startle reflex and electrodermal responding

Animal data suggest that shock sensitization as well as aversive learning potentiates the acoustic startle reflex. The present experiment tested, whether this shock sensitization also occurs in human subjects and whether it precedes aversive conditioning. Sixty subjects viewed—prior to conditioning—a series of slides of different emotional contents including the to be conditioned stimuli (CSs). Afterwards, the experimenter attached the shock electrodes and initiated shock exposure. Then, subjects were randomly assigned to view a series of two slides, each for eight acquisition trials in which one slide was followed by a shock. Subsequently, extinction trials (12 for each slide) were administered. During preconditioning, acquisition, and extinction, startle probes occurred unpredictably during and between slide viewing. Preconditioning data replicated previous results by Lang and his associates, showing that the startle response magnitude is directly related to the affective valence induced by the slides. Shock exposure strongly facilitated the startle reflex magnitude. This shock sensitization was absent for the skin conductance response. Course of learning also varied for both response systems. The data suggest that startle reflex potentiation indexes the acquisition of an avoidance disposition, which is preceded by a general sensitization of the protective reflexes. Skin conductance learning follows arousal changes and is modulated by cognitive processes.     

The renewal of extinguished conditioned fear with fear-relevant and fear-irrelevant stimuli by a context change after extinction

The acquisition, extinction, and subsequent recovery of conditioned fear can be influenced by the nature of the conditional stimulus (CS) and the context in which the CS is presented. The combined effects of these factors were examined in a differential fear-conditioning procedure with humans. Fear-relevant or fear-irrelevant CSs were followed by a shock unconditional stimulus (US) during acquisition and presented alone during extinction. The CSs were images presented upon different background contexts. Half the participants received the same context during acquisition and extinction and the remaining received different contexts. All participants received test trials in the same context as acquisition. In Experiment 1 (N=64), a renewal of shock expectancy and skin conductance responses was found during test for fear-relevant and fear-irrelevant CSs when extinction was given in a different context. In Experiment 2 (N=72), renewal for fear-relevant stimuli was enhanced when acquisition and test was given in an indoor office context and extinction in an outdoor bush context. The opposite context configuration produced the strongest renewal for fear-irrelevant stimuli. The return of extinguished conditioned fear can occur to fear-relevant stimuli that are commonly associated with clinical fears and its strength may be enhanced when the stimuli are encountered in certain contexts after extinction.     

The effect of hypocapnia on extinction of conditioned fear responses

Conditioning models have been very helpful in the understanding of the etiology and maintenance of anxiety. Such laboratory models, however, leave unexplained why in many cases of naturally occurring anxiety, as in the case of agoraphobia, the fear responses do not extinguish. Literature on experimental anxiety provocation suggests that a systemic alkalosis might play a role in the maintenance of phobic fear. It was hypothesized that a subject in a state of respiratory alkalosis would show delayed extinction to classical conditioned anxiety. In a differential classical conditioning paradigm, consisting of a habituation-, an acquisition-, and an extinction-phase, slides and electric shocks were used as conditioned stimuli (CS) and unconditioned stimuli (US) respectively. The skin conductance response was taken as (U)CR. Subjects were randomly assigned to two groups: hyperventilation or control. It was shown that the extinction was not delayed when subjects were hypocapnic during the extinction. These data support the view that a respiratory alkalosis per se is not a sufficient condition for the maintenance of neurotic fears. The data of the present study are discussed in the context of existing literature on a psychological interpretation of the maintenance of anxiety.     

Brain asymmetry and human electrodermal conditioning

Two experiments are reviewed that demonstrate effects of brain laterality on human classical conditioning. Pictures of facial emotional expressions were used as conditioned stimuli (CSs) together with shock as unconditioned stimulus (UCS). Bilateral electrodermal responses were recorded as dependent measures. In the first experiment, one group was conditioned to an angry face, and one group to a happy face. During extinction, the face-CSs were presented to the right hemisphere on half of the trials and to the left hemisphere on the other half of the trials. Results showed that the right hemisphere was superior in showing persisting effects of learning, and especially to the angry CS+. In the second experiment, lateralized presentations of the angry and happy faces were made during acquisition, with foveal presentations during extinction. Once again, the angry face elicited greater skin conductance responses (SCRs) during extinction in the group that had this stimulus presented to the right hemisphere during acquisition. It is concluded that emotional conditioning is differentially regulated by the two hemispheres of the brain.     

Conditioned nausea after cancer chemotherapy and autonomic nervous system conditionability

There are marked individual differences in conditioned nausea after cancer chemotherapy. To examine if part of this variation is associated with individual differences in autonomic nervous system conditionability, the present study addressed whether patients with conditioned nausea acquired conditioned heart rate and electrodermal responses at a different rate than patients without conditioned nausea. Of 28 relapse-free patients who had completed cisplatinum treatment for testicular cancer between 1981 and 1986, 10 reported persistent conditioned nausea, 8 extinguished conditioned nausea and 10 no conditioned nausea. These three groups were subjected to a differential conditioning paradigm with 8 sec pictorial stimuli (circles and triangles) serving as conditioned stimuli for an unconditioned electric shock while heart rate and electrodermal activity was monitored. There were 4 habituation, 8 acquisition and 8 extinction trials with each of the two cues. Analyses of variance using nausea status as the independent variable and physiological responses as the dependent lended some support to the notion that conditioned heart rate deceleration developed in response to the reinforced compared to the nonreinforced cue during acquisition in the two groups with persistent or extinguished conditioned nausea but not in the group with no conditioned nausea. In addition, patients that displayed good, as compared to poor heart rate conditionability during acquisition, were more likely to have persistent conditioned nausea, whereas those who showed poor heart rate conditioning mostly were those without conditioned nausea. Electrodermal variables revealed no systematic differences between groups. This tentatively supports that individual differences in parasympathetic but not sympathetic nervous system conditionability may be associated with individual differences in conditioned nausea resulting from cancer chemotherapy.     

Electrodermal pavlovian conditioning with prepared and unprepared stimuli

One-trial learning referred to by Guthrie has been suggested to occur in autonomic conditioning, if the conditional stimuli (CSs) are so-called prepared ones. To test this idea, half of 28 subjects were given spider or snake slides as “prepared” CSs, while the remainder were given neutral slides as “unprepared” CSs. A shock was employed as the unconditional stimulus (UCS), with a CS-UCS interval of 8 seconds. Electrodermal activity and probe reaction times were the dependent measures of conditioning, conceived in cognitive, information-processing terms as the learning of the CS/UCS contingency. Evidence for the usual CS/UCS contingency learning emerged in both indicators, and during both acquisition and extinction, but none for one-trial learning, perhaps because the UCS was insufficiently aversive.