chapter  7
6 Pages

Individual differences in the acquisition of fears susan mineka

Fear is a highly adaptive emotion that often signals potential or actual danger in humans and many other species. Fear also serves as a central motive state, sometimes motivating escape or avoidance behaviors. Not surprisingly many have argued that fear has been central to mammalian evolution because it is a product of natural selection and therefore shaped and constrained by evolutionary contingencies (e.g., Öhman and Mineka, 2001). Some sources of fear are innate (especially in subhuman animals), but it has long been known that many objects or situations that we fear are based on learning. Indeed, since at least the 1920s it has been known that many fears are based on a fundamental form of learning known as classical or Pavlovian conditioning. In classical conditioning, the conditioned stimulus (CS) is paired with an unconditioned aversive stimulus (US) one or more times. Generally the CS gradually acquires the capacity to elicit a conditioned defensive response that may resemble the unconditioned response, but may also differ from it in significant ways such as being a compensatory response (e.g., Öhman and Mineka, 2001). Another important point about classical conditioning relevant here is that once acquired through conditioning fears are not “forgotten” simply with the passage of time. Instead, for CRs to gradually diminish there must be a number of extinction trials in which the CSs are presented without any USs (Mackintosh, 1974). Direct classical conditioning is not the only pathway to the development of fears and phobias. Indeed, humans and certain other species can readily acquire fears either vicariously simply through observing a conspecific behaving fearfully in the presence of some object or situation or, in humans, through verbal instructions about the danger posed by some object or situation. Mild and transient fears have been conditioned this way in many studies in laboratory settings in both adult humans and young children (see Askew and Field, 2008, for a recent review). Moreover, strong and persistent phobic-like fears have been acquired vicariously in numerous experiments conducted in lab-reared rhesus monkeys by Mineka and colleagues (see Cook and Mineka, 1991, for a review). Indeed lab-reared monkeys who were not initially afraid of snakes quickly acquired an intense phobic-like fear of snakes (as indexed by three different measures) after a total of only 24 minutes of exposure to a wild-reared monkey showing a strong phobic-like fear to a live boa constrictor and toy snakes. Indeed, the level of fear

observed in the lab-reared monkeys following observational conditioning was nearly as intense as that of the model monkeys who had acquired their phobiclike fears in the wild in India several decades earlier. Another indication of the robustness of the vicariously acquired fear was that, in one experiment, monkeys who had acquired the fear vicariously successfully served as models for other unrelated observer monkeys who then also acquired it vicariously. Moreover, there were no signs of diminution of fear of the snakes over a three month follow-up period. When fears are acquired through direct or vicarious conditioning, or through instructional/verbal learning there is a wide variation in how quickly they are acquired. Speed of acquisition is, for example, partly a function of the aversiveness of the US (cf. Mackintosh, 1974) with more intensely aversive USs generally being associated with more rapid and robust conditioning. Another important factor that can contribute to the speed of acquisition is the “belongingness of the CS and US”. That is, some combinations of CSs and USs condition especially well together relative to other combinations of CSs and USs (e.g., Hamm et al., 1989). In the early 1970s, Seligman had proposed the relevance of work on belongingness (much of it done in animals) to the understanding of fears and phobias (1971) in his classic paper on “phobias and preparedness”. Phobias are generally seen as very intense and persistent fears that the person realizes are excessive or unreasonable, and that are triggered by the presence of a specific object or situation. When a person encounters a phobic stimulus, they often show an immediate fear response that may resemble a panic attack except for the presence of a clear external trigger for fear. According to the preparedness theory of fears and phobias, we are evolutionarily prepared to acquire fears and phobias more readily to certain objects or situations that may once have represented a threat during our early evolutionary history (Öhman and Mineka, 2001; Seligman, 1971). A large number of studies conducted by Öhman and his colleagues with human participants and several studies of lab-reared rhesus monkeys conducted by Mineka and colleagues have provided strong support for many aspects of this theory, as reviewed by Öhman and Mineka (2001). In dozens of human studies from Öhman’s laboratory, conditioning to prepared or fear-relevant CSs (e.g., pictures of snakes or spiders) is found to be more robust than when the CSs are unprepared or fear-irrelevant (e.g., flowers or mushrooms). Robustness has most often been demonstrated by participants’ showing greater resistance to extinction with prepared CSs than is seen with unprepared or fear-irrelevant CSs. In addition, however, robustness of prepared fears has also been demonstrated by findings that strong conditioning can occur after only one CS−US pairing using prepared or fear-relevant CSs, but not when using unprepared or fear-irrelevant CSs (Öhman and Mineka, 2001). In Mineka and Cook’s experiments on preparedness, lab-reared rhesus monkeys served as observers who watched model monkeys on spliced videotapes behaving fearfully either with toy snakes or with brightly colored artificial flowers. (A prior study had demonstrated that observer monkeys could acquire the fear vicariously simply through watching a fearful monkey behaving fearfully with snakes on a

videotape.) Only the observers who watched models behaving fearfully with toy snakes acquired a fear of snakes; by contrast, the observers who watched models behaving fearfully with brightly colored flowers acquired no fear of flowers. This is in spite of the fact that the fear performance that the model monkeys showed to toy snakes and to flowers was identical (which was accomplished through video editing). Thus both monkeys and humans seem selectively to associate certain fearrelevant stimuli with threat or danger. So individual differences in who acquires fears depend in part upon which stimuli are paired with threat. Moreover, the lab-reared monkeys had no prior exposure to any of the stimuli involved (e.g., snakes or flowers) before participating in these experiments. Thus the monkey results support the evolutionarily-based preparedness hypothesis even more strongly than do the human experiments. For example, human participants (unlike the lab-reared monkeys) might show superior conditioning in the laboratory to snakes or spiders because of ontogenetic factors, such as preexisting negative associations to snakes or spiders, rather than because of evolutionary factors. The monkey experiments also have the advantage of demonstrating that selective associations occur not only with mild and transient conditioning, as seen in the human experiments, but also with intense and long-lasting phobic-like fears. Nevertheless it should be acknowledged that preparedness theory has generated a good deal of criticism and controversy (e.g., Davey, 1995), which was countered by Öhman and Mineka (2001) in their extensive review of relevant evidence. One common but seriously misconceived assumption about conditioning models of fear acquisition is that people should be able to recall their direct, vicarious, or instructional experiences. There are many reasons to doubt the validity and accuracy of retrospective recall for such situations, and accordingly recently developed non-associative accounts of the acquisition of fears and phobias seem to have little merit (e.g., Mineka and Sutton, 2006). Another common but seriously misconceived assumption about conditioning models is that everyone exposed to the same direct or vicarious conditioning experience should acquire comparable levels of fear. This is also clearly not the case. Instead there is a host of genetic and temperamental, as well as experiential, variables that strongly influence the speed and strength of conditioning in any given individual. Among experiential variables, one is simply the familiarity of the CS to the person prior to a conditioning experience. As illustrated by the well-known and carefully studied phenomenon of latent inhibition, pre-exposure to a CS prior to conditioning reduces the amount of fear that is subsequently conditioned relative to what is observed with a truly novel CS. For example, in an illustrative study, Davey (1989) asserted that children who reported having had first painless dental treatments with a dentist were less likely to develop a dental phobia if they were subsequently traumatized during a dental visit than were children without as many earlier benign experiences with a dentist. Cook and Mineka (1991) also reviewed evidence showing that protective or immunization experiences from pre-exposure to a nonfearful model monkey behaving nonfearfully with snakes further reduced the strength of conditioning when the observer later witnessed a fearful model

behaving fearfully with snakes. Indeed, there was no significant effect of conditioning for 6 out of 8 monkeys that first had the immunization experiences. Egliston and Rapee (2007) found somewhat parallel results regarding protective effects of initial positive modeling by a mother against subsequent maternal fearful modeling in one-two-year-old human children. If the children had first watched their mothers show positive modeling with a fear-relevant stimulus, they were later protected against the effects of fearful modeling by the mother, relative to children without the initial positive maternal modeling. Another important experiential variable during conditioning that strongly affects the amount of fear that is conditioned in an aversive situation is the degree of control the individual has over either the onset or the offset of the US. In traditional classical conditioning experiments, the organism is a passive recipient of the CSs and USs that are delivered because these are all controlled by an experimenter. But in our everyday lives, when conditioning experiences typically occur, the person often has control over some aspect of the situation, such as controlling when the US will terminate (e.g., by escaping from it). Being able to control the offset of the US paired with any CS has a major effect of attenuating the amount of fear that is conditioned. Indeed, in one illustrative experiment with rats, Mineka, Cook, and Miller (1984) found that rats that had experienced tones paired with escapable shocks later showed only approximately half the levels of conditioned fear seen in rats that had experienced tones paired with exactly the same amount and number of shocks that were inescapable/uncontrollable. Thus the dynamics of classical conditioning are dramatically affected by the controllability of the US. Individual differences in what happens following a conditioning experience also affect the amount of fear that is maintained over time because fear memories are somewhat malleable. For example, Rescorla (1974) reported that if rats are first conditioned to show a mild fear (by pairing a CS with a mild US), their fear level will later increase if they are subsequently exposed to more intense USs (not paired with the CS)—a phenomenon known as the “inflation effect”. Similarly, in humans, White and Davey (1989) demonstrated an inflation effect when US re-evaluation occurred following conditioning by delivering more intense USs (alone) than had been used during conditioning. That is, the participants in the inflation group later showed increased levels of fear, even though they had not had any pairings of the CS with the more intense USs. Thus whether or not someone has an inflation experience affects the amount of fear they maintain into the future. In addition to experiential variables on which individuals may differ, there are also temperamental and personality variables that affect the acquisition of fear. For example, prospective studies initiated in the 1980s determined that toddlers with a behaviorally inhibited temperament (i.e., shy, timid, easily distressed) assessed at 21 months were at higher risk for developing multiple specific phobias by age seven or eight than were uninhibited children (32 per cent versus 5 per cent) (Biederman et al., 1990), although it is not yet clear if these phobias were acquired through conditioning. Personality variables such as neuroticism and trait

anxiety also affect the speed and strength of conditioning in laboratory studies (see Oehlberg and Mineka, 2011, for a review). An exciting recent study (Lonsdorf et al., 2009) using sophisticated molecular genetic techniques has also shown that several specific genetic polymorphisms of the 5-HTTLPR and COMT genes have strong effects on either fear acquisition or fear extinction. Normal college students first underwent DNA extraction from blood for genotyping, and then underwent a discriminative fear conditioning procedure using facial stimuli as CS+s and CS−s. Startle potentiation was the primary index of fear. Carriers of one or two short alleles (s or ss) of the 5HTT gene showed significantly stronger fear potentiation then did the l/l homozygous carriers and this pattern persisted in extinction. In contrast, the two polymorphisms of the COMTVal158Met gene had no effect during acquisition. However, the results during extinction were very different. Specifically, those with the homozygous met/met COMT Val158Met polymorphism showed much greater CS+ fear potentiation than those who were COMT Val-allele carriers who showed no fear at all (i.e., no resistance to extinction). The participants who showed the most pronounced startle responding in extinction were those with at least one short s allele of the 5 HTT gene and two COMT met alleles. The authors suggest that such individuals “are likely to expand their sets of fear-and anxiety-evoking stimuli through facilitated fear conditioning and poor extinction” (p. 204). These exciting results obviously need to be replicated before strong conclusions can be drawn, but they may provide a precise mechanism helping to account for why only a subset of people develop fears of many different stimuli, and why such fears may be so persistent in only a subset of them. In conclusion, although fears can be innate, they are frequently acquired through direct or vicarious conditioning, or instructional learning. Importantly, however, not all individuals who have the same learning experiences will acquire or retain the same levels of fears. To the contrary, the acquisition of fears is strongly influenced by a host of individual difference variables. Some of these variables are genetic and temperamental differences over which we have little or no control, but a myriad of other such variables reflect wide individual differences in life experiences that people have had prior to, during, or following conditioning. Due to space limitations, what has been reviewed here is only an illustrative set of examples of such individual differences, but they should be sufficient to convince readers that there is what I once called a “frightful complexity of the origins of fears.”