Acquisition and severity of driving-related fears
Introduction
The three-pathways theory of fear acquisition proposed by Rachman, 1976, Rachman, 1977developed out of the recognition that classical conditioning theories provide an incomplete account of human fear acquisition and ignore less direct pathways to fear (Withers and Deane, 1995, Rachman, 1991). Rachman (1984), Rachman (1991)noted that fears can be acquired through conditioning as well as other processes, such as vicarious and verbal transmissions and proposed that there were three major associative pathways to the acquisition of fear: (1) classical conditioning experiences, (2) observational experiences and (3) instructional or informational experiences. In addition to this three-pathway proposition, Rachman postulated that direct fear-conditioning would lead to severe fears, while indirect fear-conditioning (i.e. vicarious or instructional pathways) would lead to mild to moderate fears. Rachman also predicted that directly-conditioned fears would be characterised by more elevated physiological and behavioural symptoms than cognitive symptoms, while indirectly-conditioned fears would be characterised by more elevated cognitive symptoms than physiological and behavioural symptoms. Rachman referred to this prediction as the `differential-anxiety-response' hypothesis.
It is evident from existing research that the fear acquisition literature which tests Rachman's predictions has produced inconsistent and variable findings and it has been suggested that the acquisition of nonclinical fears has not been adequately addressed (Öst, 1991). In particular, past research has been criticised for the lack of control over retrospective judgements made as respondents construct reasons for their fears and null results have been explained in terms of potential uncontrolled memory distortions (Merckelbach et al., 1996). A recent study which investigated the potentially confounding role of memory in fear acquisition research with 191 university students failed to support Rachman's predictions (Withers and Deane, 1995). However, direct conditioning ascriptions were endorsed with greater certainty, indicating that direct conditioning events may be more memorable than indirect conditioning events and that memories of onset events may be important (Withers and Deane, 1995). The present study attempted to address this problem by selecting driving-related fears as the phenomenon of interest, as it could reasonably be assumed that motor vehicle accidents (MVAs) may account for at least some proportion of these fears and are likely to be memorable events.
Despite the potential advantages of investigating driving-related fears for testing Rachman's theory, only three studies could be located in this area and all three recruited relatively small nonclinical samples of community volunteers through media advertisements in newspapers or on television. Munjack (1984)interviewed thirty respondents and found that a variety of onset circumstances characterised his sample. Panic attacks were most frequently attributed to the onset of driving-related fears (40%), followed by a collision on the freeway (20%) and other upsetting events directly associated with driving (10%). Therefore, 70% of the sample reported a history of direct conditioning experiences, although indirect pathways to fear such as observation and instruction were not investigated.
Sartory et al. (1992)examined fear onset among sixteen respondents and found that 75% had panic attacks while driving, 31% described one of their parents as being fearful of freeway driving (although two of these also reported panic attacks) and one reported becoming fearful as a result of information regarding fatal accidents on motorways. For most respondents in the study, the cause of their panic attacks was a sudden, unexpected rise in anxiety while driving and the attack was triggered endogenously by worries of having a panic attack while driving, rather than by a MVA. Subsequently, driving was avoided. One subject met DSM-III-R criteria for panic disorder. Finding a high rate of panic attacks as the reported onset event is consistent with the study by Munjack (1984). However, other research has reported that problems after an MVA (such as phobic reactions) were clearly caused by classical conditioning and were a direct result of the accident (Mayou et al., 1991).
Ehlers et al. (1994)used interviews and questionnaire information to investigate driving phobia in fifty-five respondents. Respondents were asked to rate the three most important reasons for their driving phobia from a list of fourteen on a questionnaire. Panic attacks were rated as one of the most important reasons for driving anxiety by 65% of subjects. A traumatic experience such as an accident, dangerous traffic situations or being assaulted while driving was reported by 36%, while 5% rated seeing someone else experience a traumatic event when driving as an onset event. Other reasons which were rated highly were being a generally anxious person (53%) and being generally afraid of high speed (47%). Pathways of fear acquisition were further investigated through an interview, with questions which asked about traumatic accidents, vicarious onset and informational acquisition. Compared with control subjects, phobics were not more likely to have been involved in an accident, not more likely to have had anxious models and not more likely to have been given information regarding the dangers of driving (Ehlers et al., 1994). However, phobics were more anxious about their worst accident as well as reading or hearing about particular dangers associated with driving than control subjects and 15% reported a MVA as the primary reason for their phobia.
These studies have reported a range of pathways consistent with Rachman's theory, although they have not investigated his hypotheses regarding strength of fear and anxiety response patterns (Munjack, 1984; Mayou et al., 1991; Sartory et al., 1992; Ehlers et al., 1994). Furthermore, no research was located which compared people who are fearful because of an MVA with those who are fearful for some other reason. The present study was aimed at addressing both of these issues.
Prior studies have tended to omit investigations of nonassociative pathways through which fears can be acquired in the absence of any previous associative learning experience. The present study aimed to investigate a range of pathways through the use of the Origins Questionnaire (OQ; Menzies and Clarke, 1993), a relatively new instrument developed in an effort to address methodological problems in the assessment of fear acquisition. In particular, the OQ addresses the inconsistent definitions of conditioning events used by different researchers and considers pathways to fear other than conditioning experiences.
From Rachman's predictions, three hypotheses were formulated: (1) respondents will ascribe strong fears to the direct pathway and moderate fears to the indirect (i.e. observational or instructional) pathways, (2) respondents will report higher levels of physiological than cognitive responses for fears ascribed to the direct pathway, but for fears ascribed to either of the indirect pathways, they will report lower levels of physiological than cognitive responses and (3) respondents who report having experienced a MVA will be more likely to ascribe to the direct pathway than those who do not report having experienced a MVA.
Section snippets
Sample and procedure
Of the 190 volunteers who participated in this study, 175 were female (92%) and 15 male. Initial contact with participants was gained through media interest in and coverage of the present study. Two local and two national newspapers published articles about the study and requested interested volunteers to telephone the researcher. Radio coverage about the study was also obtained through national radio stations. This coverage consisted of brief (5–10 min) interviews with the first author who
Results
Based on responses on the OQ, respondents were classified into one of the six fear-onset categories. For the purposes of hypothesis-testing and consistent with prior research (e.g. Öst, 1991; Withers and Deane, 1995), respondents were divided into `direct' (classical) and `indirect' (vicarious and informational) conditioning pathways. To test the prediction relating ascribed pathway to fear severity levels, one-tailed independent t-tests were carried out on the mean pathway-related severity
Discussion
In the present study, no support was found for the predictions derived from Rachman's theory. The low proportion of direct respondents in the present study is inconsistent with the findings of previous nonclinical research on the acquisition of driving-related fears. As previously mentioned, percentages of between 70 and 100% for direct pathway ascriptions have been reported (Munjack, 1984; Sartory et al., 1992; Ehlers et al., 1994). In comparison, only 27% of the present sample attributed
References (35)
- et al.
Anxiety response patterns and etiological factors in dog-fearful and non-fearful subjects
Behaviour Research and Therapy
(1988) - et al.
Clinical characteristics of driving phobia
Journal of Anxiety Disorders
(1994) - et al.
Mode of onset of simple phobia subtypes: further evidence of heterogeneity
Psychiatry Research
(1991) - et al.
Assessment and treatment of motor vehicle accident victims
Cognitive and Behavioral Practice
(1995) - et al.
A brief self-rating scale for PTSD after road vehicle accident
Journal of Anxiety Disorders
(1995) - et al.
The effect of major railway accidents on the psychological health of train drivers. I. Acute psychological responses to accident
Journal of Psychosomatic Research
(1993) - et al.
The effects of road traffic accidents on driving behaviour
Injury
(1991) - et al.
The etiology of fear of heights and its relationship to severity and individual response patterns
Behaviour Research and Therapy
(1993) - et al.
The etiology of phobias: a nonassociative account
Clinical Psychology Review
(1995) - et al.
The etiology of specific phobias: a review
Clinical Psychology Review
(1996)