Fear-then-relief-then argument: How to sell goods using the EDTR technique of social influence

Abstract

In 1999 Davis and Knowles proposed a social influence technique which they named disrupt-then-reframe (DTR). They demonstrated that compliance could be increased by a subtle disruption to the sales pitch, followed immediately by a reframing that provided additional reasons for purchasing the goods. The DTR technique is strictly cognitive in nature: the person, hearing simple argumentation during the short state of her or his cognitive disorganization, becomes more inclined to fulfill the requests made of her or him. In two field studies presented in this article it is shown that a similar effect can be obtained when the fear-then-relief state, which could be viewed as an emotional disruption, is followed by an argument.

Keywords:

• Social influence
• Fear-then-relief technique
• Disrupt-then-reframe technique
• Compliance
• Sales pitch

The psychological literature provides descriptions of various social influence techniques that increase the likelihood of product purchase (e.g., Cialdini 2009; O’Keefe, 2002). These compliance-without-pressure techniques include door-in-the-face (Cialdini et al., 1975), low-ball (Cialdini, Cacioppo, Basset, & Miller, 1978), dialogue involvement (Dolinski, Nawrat, & Rudak, 2001), and that's not all (Burger, 1986).

Another of the compliance procedures has been identified as the disrupt-then-reframe (DTR). In a series of four experiments Davis and Knowles (1999) demonstrated that compliance could be increased by a subtle disruption to the sales request, followed immediately by a reframing that provided additional reasons for purchasing the goods. Each of the experiments followed the same pattern: the experimenters rang the doorbell of the houses and presented their products to potential buyers (sets of Christmas cards, sets of post-it notes, or packets of biscuits), then, having asked if their interlocutors were interested in the prices of the products, they quoted the prices. However, the way the price information was formulated was different depending on the experimental condition. The price was either presented in the traditional way, i.e., “The cards/cookies cost 3 dollars“, or in a way that disarranged to a certain extent the typical sales scheme: “The price of these notes/Christmas cards/cookies is 300 pennies”, which was followed after 2 seconds with an explanation: “That's 3 dollars”. In the strange sales plus argument condition the seller would also add: “It's a bargain”. In the other conditions either the standard quotation of the price alone was given (i.e., in dollars), or it was followed additionally by the remark “It's a bargain”. The percentage of people who decided to buy the products was about two times higher in the disrupt-then-reframe conditions (where the price was first quoted in a strange manner, then repeated in the standard way, and then supplemented with the simple argument in support of the product purchase) than in any of the other conditions. Analogous patterns of results were obtained by Fennis, Das, and Pruyn (2004), who offered their experiment participants the chance to purchase commercial lottery coupons; by Dolinski (2005) who tried to sell bottled soup concentrate from a market stand; and by Kardes, Fennis, Hirt, Tormala, and Bullington (2007), who were selling candles in a supermarket.

Davis and Knowles (1999), trying to explain the mechanism of the phenomenon, mentioned the action identification theory (Vallacher & Wegner; 1985, 1987; Wegner, Vallacher, Macomber, Wood, & Arps, 1984). According to this theory one should take into account that people not only perform various actions, but also think about what they are doing. The identification of the action can be processed on different levels, from low-level characterizations that pertain to specific details of the behavior to high-level qualifications that include the goals and broader implications of the actions. For instance, a man painting a wall can be thinking about the way he is covering the wall with new paint, but he also can be thinking that he is redecorating his daughter's room or that he is tinkering. According to the authors of the action identification theory, people usually tend to identify their actions at a higher (abstract) level (“I’m redecorating a room”, “I’m tinkering”); low-level identification of actions (“I’m putting on a new layer of paint”) occurs in exceptional conditions, such as a situation in which something unexpected happens that disrupts people's control over the action underway. In our example with wall painting, the man would shift to the low-level interpretation of his action if, for instance, the wall was difficult to paint evenly because of stains. Shifting to the lower, matter-of-fact level of specific details of the action allows us to regain the control lost over what we are doing. After regaining control people tend to revert to the more abstract identification of their actions, as this adds a sense of significance to their actions in a broader context.

The dynamics of shifting from one level of identification to another is the key determinant of the disrupt-then-reframe technique's effectiveness. When it comes to the disruption of a typical, everyday action that we would normally identify on a higher level (e.g., a price is given in cents instead of dollars, or the time of a survey, in seconds instead of minutes), our attention is shifted from the higher-level meanings ascribed to the door-to-door salesman's behavior, and indeed the meaning of the entire dyadic encounter, to a more concrete lower-level focus. The shift in the action identification level is the participant's attempt to regain control over what is going on. Sudden clarification of the “odd bit” (e.g., giving the price in dollars) enables the participant to recapture his or her sense of control and consequently return to the higher level of action identification, which is preferred in typical, everyday conditions. The unique state of the participant's mind, resulting from a double shift from one level of action identification to another within a very short time, makes the participant lose his or her normal orientation and disrupts his or her cognitive functions to a certain extent. In this peculiar moment of disorganisation the participant becomes susceptible to simple and explicit argumentation (e.g., “It's a bargain”).

The disrupt-then-reframe technique is strictly cognitive in nature: the participant, hearing simple argumentation during the short state of cognitive disorganization, becomes more inclined to purchase a product. Empirical evidence can be found in the relevant literature demonstrating that compliance can be successfully induced not only during a momentary state of cognitive disorganization, but also under emotional disorganization.

Dolinski and Nawrat (1998) proposed a compliance procedure called fear-then-relief. This procedure is based on the “good cop – bad cop” interrogation scenario that is common in crime literature and films: first one policeman brutally mistreats a suspect, humiliating him or her and threatening him or her with death. Then everything changes when the “bad cop” leaves the room and another policeman, calm and pleasant, comes in. The “good cop” suggests having coffee or tea and a cigarette, and leads a relatively normal conversation. Most often the suspect, previously obstinate in refusing any form of cooperation, starts to reveal everything. It may be assumed that the good cop – bad cop scenario is only one example of a general rule that is not limited to police interrogations.

In an experiment by Dolinski and Nawrat (1998) the participants were jaywalkers. In some cases, when the participant was in the middle of the road, a police whistle was used. The participants reflexively turned their heads toward the sound, but it turned out there were no policemen on the sidewalk behind them. The rest of the participants were allowed to cross the street undisturbed. In the experimental design there was also a third group of participants who did not cross the street but only walked down the sidewalk. All participants were then spoken to by a confederate who asked them to fill in a psychological questionnaire and announced it would take about 10 minutes. It should be noted that the experiment was conducted on a cold autumn day and that it was not possible for participants to fill in the questionnaire later at home; they had to complete it on the spot. The questionnaire the participants were asked to fill in was the Self-Description Inventory (Spielberger, Gorsuch, & Lushene, 1970) which enabled researchers to measure the fear level experienced by the participants at that moment. It turned out that although the level of fear was similar in all of the experimental conditions, the participants who experienced fear-then-relief agreed to fill in the questionnaire more frequently than did participants from the other groups.

In a further experiment Dolinski and Nawrat (1998) demonstrated that this kind of compliance to requests did not result from the fear participants had just experienced. The participants were high school students individually invited to a laboratory for “measurements of various skills and abilities”. Participants were randomly allocated to one of three conditions: group 1 experienced fear, group 2 experienced fear that was subsequently reduced, and group 3 was not subjected to any initial procedure. Participants from groups 1 and 2 were informed that they would take part in a study concerning the effects of punishment on learning. They were told: “Your task will be to learn associations of various words. However, should you make an error while learning, you will be given a mild, not very painful electric shock.” Participants from group 3 were told that the experiment concerned visual-motor coordination: “Your task will be to throw darts at targets at various distances.” Subsequently, in all conditions, the participants were informed that the experiment would begin in a few minutes and asked to wait in the corridor. In the case of group 2 after about 2 minutes an experimenter would come up to the participants to tell them that the professor supervising the laboratory had just decided to postpone the experiment to the following week, so instead of the “electric shock” experiment the students would take part in another study in which they would have to throw darts at various distances. It was also explained that this new study required some preparation, so the students were asked to wait a little longer in the corridor. During the waiting period before the experiment started, each participant was asked by a female student (who pretended to have no role in the study) to take part in a charity action for an orphanage. It turned out that the participants who experienced first fear and then relief more frequently agreed to participate in the charity action than did the participants in the other groups.

In another study the participants were car drivers who had parked their vehicles in a no-parking zone. A card matching the general appearance of a parking ticket was placed by the experimenters either behind the windscreen wiper (where tickets are commonly found) or on a door. The cards placed on the door were shampoo advertisements (no fear), whereas the windshield wiper cards were either parking tickets (fear) or advertisements (fear-then-relief). Drivers who experienced apprehension followed by assuagement were more likely to comply with a request to fill out a questionnaire than those who remained anxious or those never made anxious in the first place. All drivers who agreed to fill out a questionnaire received the PANAS scale (Watson, Clark, & Tellegen, 1988). As could have been expected, the drivers who received a real ticket experienced higher levels of negative emotions than the participants in the other groups. However, there were no differences in the levels of positive emotions. These results show that the increased compliance achieved in “fear-then-relief” conditions cannot be explained by the fact that people relieved from fear experience positive emotions, as these emotions were no stronger in the “fear-then-relief” situation than in the control group.

Thus one may conclude that the results of the aforementioned experiments also rule out the hypothesis that only fear or only an improved mood cause increases in compliance. Apparently the effectiveness of the fear-then-relief technique seems to lie in the sudden change it causes to people's emotional states.

Dolinski and Nawrat (1998) explain their findings by noting that fear alerts the body, focuses our attention on the source of fear (e.g., Tomkins, 1991), and triggers an action program specific for the given type of emotion (Frijda, 1986; Lazarus, 1991). While such a reaction seems perfectly adequate and adaptive for threatening conditions, it stops being adequate when the circumstances suddenly and unexpectedly reverse, as in the “bad cop – good cop” interrogation procedure, or in the studies by Dolinski and Nawrat (1998). In such a situation people may experience a short-lasting state of disorientation, and function automatically and mindlessly.

The similarity between the disrupt-then-reframe and the fear-then-relief techniques is based mainly on the fact that in both cases the participant is dealing with an untypical situation: the routine course of action is disrupted by introducing an “odd” element to disorganize the usual course of interaction, and in both cases his or her functioning is disturbed. In the case of disrupt-then-reframe the sudden change takes place at the cognitive level of functioning, while in the case of fear-then-relief it occurs at the emotional level.

However, despite this essential similarity, there is one ingredient that differentiates the two techniques. To prove effective, apart from making a temporary mess of the cognitive activity of the participant's mind, the disrupt-then-reframe technique also requires an extra argument to make the participant comply with the request. This simple additional argument plays the role of a ready-to-take instruction for what to do next. The effectiveness of the fear-then-relief technique does not require any external indications or arguments to make the participant compliant. The person subjected to the latter technique supplies a heuristic indication for further action from the resources of his or her own biographical memory (e.g., “When you are asked politely to do a small favor, why shouldn’t you agree to it?”).

The question then arises, how would the effectiveness of the fear-then-relief technique change in conditions when—following a sudden change of the emotions experienced—an individual hears some argumentation aimed at making him or her compliant with the request?

On the other hand, why should a verbal argument heard by people in a state of emotional relief induce compliance more effectively than their own memory stored in their cognitive structures—containing memories of their own behavior in similar situations—and the resulting heuristics: “When you are asked politely to do a small favor, why shouldn’t you agree to it”? It can be assumed that such experiential recollection is not always cognitively available, especially under conditions in which people's ability to freely channel their attention resources is limited; they may have problems with the use of information hidden in their memory structures (Smith & Engel, 2011).

There is direct empirical evidence that the cognitive functioning of individuals who are experiencing relief from fear is impaired. Participants in a study by Dolinski, Ciszek, Godlewski, and Zawadzki (2002) who were subjected to a fear-then-relief scenario took longer to find a particular face in a crowd and solved fewer arithmetical equations than did participants in either the fear-only group or the emotionally neutral group. Nawrat and Dolinski (2007), in turn, present evidence from a study conducted by phone that fear-then-relief is effective even in conditions where the final request is completely absurd. The experimenter phoned randomly chosen people, introducing herself as an employee of Polish Telecom. In some cases she informed her interlocutor that the computer had calculated a considerable overdue sum of money to be paid by the participant. Half of the participants were left in this induced emotional state, while the other half were told after a short time that the computer had actually identified another telephone owner with the same name but a different address. Regardless of the type of manipulation, the experimenter then said: “Polish Telecom is presently testing the permeability of the telephone lines. In connection with the introduction of the TELPOCOL system, I would like you to put the receiver of your telephone to your other ear…” After 3 seconds she asked, “Have you done this?” In the control group, where no emotional state was induced, this message was presented right after the experimenter introduced herself as an employee of Polish Telecom. The participants’ confirmation that they had put the receiver to their other ear was treated as compliance. While this reaction occurred sporadically in the control group or the group in which the negative emotions were induced, it was considerably more common in conditions where information justifying negative emotions had been contradicted.

Because the state of relief following the abrupt withdrawal of the fear-causing stimulus is actually associated with cognitive impairment, it can also be assumed that an individual may encounter some problems recalling from the past that, whenever asked politely to do something, he or she usually met the request without incurring any negative consequences. Conversely the verbal argument, which appears exactly at the moment in which the person has to make the decision about how to behave (“agree to comply with the request or not?”), is clearly and unequivocally cognitively available. The attention decrement typical in the state of fear-then-relief will not hinder the use of this argument. It seems worth noting here that in the classical situation when the person is interrogated by the bad cop – good cop duo—which is in fact a real-life prototype for the very technique described in the psychological literature as fear-then-relief—the “good” policeman in fact uses a verbal argument to make the person talk (he says, e.g., “Your only chance is to admit your guilt”, “We know everything about you anyway”, etc.).

Another important issue is the number of arguments that should be used in relation to people subjected to the fear-then-relief procedure with a view to maximizing the likelihood of their compliance with the request. From a certain perspective, one might think that the more verbal arguments we use (especially serious ones), the higher compliance rate we get. From another perspective, however, it must be noted that the particular state of disorientation resulting from experiencing the fear-then-relief sequence is quite short-lasting. In research by Spiewak (2002), fear-then-relief sequence was induced in the laboratory of the Faculty of Psychology, but the request to participants was formulated only after they had left the building. About 2 minutes elapsed between these two events. The FTR technique turned out to be completely ineffective in this case. Although there are so far no clear data on how long the state of confusion resulting from a fresh experience of a fear-relief sequence lasts (most likely its “shelf life” varies and depends on such factors as the strength of the initial fear and the depth of the relief), it can be expected that too many verbal arguments presented sequentially could exceed that time. Thus, in the decision-making moment, the person will have ceased to be cognitively confused. In other words, it is difficult to anticipate a priori what number of sequentially presented verbal arguments would be best. Due to the complexity of this issue we decided not to tackle this problem in the studies presented in this article, limiting ourselves to a single verbal argument just as did Davis and Knowles (1999) in their experiments. It is worth noting that the problem of the optimal number of arguments also concerns the DTR technique described by these researchers.

Thus we assumed that fear-then-relief is essentially similar to the disrupt-then-reframe technique. However, while in the latter technique a strange situational circumstance compels the person to switch to a different action identification level, the subject of a fear-then-relief sequence is forced to change his or her mode of functioning due to the sudden disappearance of the stimulus which justified the emotion that has just ended. Yet in both techniques the person remains for a time in a similar state of cognitive perplexity resulting from the sudden change of situational dynamics.

Because in the fear-then-relief technique the external verbal argumentation would become an additional (aside from the person's own heuristics located in his or her memory) compliance-enhancing factor, we could expect that the condition “fear-then-relief plus extra argument” should induce an even higher compliance rate than obtained so far in fear-then-relief studies. The aforementioned decrement in cognitive functioning post-disruption is responsible for the problems some people can have recalling previous positive experiences in which they fulfilled requests formulated in a polite and pleasant manner. According to the theoretical assumptions presented here, these people will fail to fulfill the requests directed to them in the classic fear-then-relief manner, because they will not be able to use the heuristic “if someone asks you politely, do it”. However, if during the state of sudden and unexpected relief they hear an additional verbal argument encouraging them to comply with the request, they may simply comply.

The aim of the studies presented below was twofold. First, although there is strong empirical evidence that the fear-then-relief technique is a very effective tool of social influence (e.g., Dolinski, 2001, 2007; Dolinski et al., 2002; Nawrat & Dolinski, 2007), it has never been tested in the area of consumer behavior. We intended to examine whether it is possible to increase the likelihood of product purchase using this procedure.

Second, in the two studies presented below we compared the effectiveness of the standard technique with its enhanced version where the person—undergoing a sudden and unexpected state of relief from fear—is provided with a verbal argument with which to comply. We assumed that both techniques would prove effective in the marketing sphere (i.e., the willingness to purchase various goods would be higher in both experimental conditions than in the control group); however, we also expected the fear-then-relief plus argument condition to generate higher compliance rates than the classical fear-then-relief condition.

Study 1

Overview

In the hall of a railway station a confederate approaches a person walking alone and asks him or her if he or she hasn’t lost his or her wallet. After this participant has made sure her or his wallet is still in place and has informed the confederate that it is not lost, she or he is then presented with the Christmas cards and asked to buy them. Under the control group conditions the confederate simply approaches a person walking alone in the hall of the station and asks him or her to buy the Christmas cards. In both the induced fear-and-relief condition, and emotionally neutral conditions, in half of the cases the request was accompanied by the additional argument “It's a bargain”. The confederates were unaware of the hypothesis being tested.

Method and participants

The experiment was carried out in early December, independently by four female confederates aged 22–25. The study took place in the hall of the main railway station in Jelenia Gora, Poland. The participants were people of adult appearance and walking alone in the hall of the station. Men and women were chosen alternately. Every fifth person fulfilling these conditions was approached. The confederate approached the participants from behind, formulating the request according to the prior random assignment of participants to the experimental conditions. Each of the confederates interviewed 72 people (36 women and 36 men), which provides 18 participants for each of the conditions of the experiment.

The confederate followed the participants for several steps and then spoke to them. The content of what she said depended on the condition of the experiment.

In the control condition, she said:

Excuse me [sir/madam], [and when the person turned towards the confederate, she continued:] I am a student of sociology, working for the “Handicap Center”, which is a non-profit organization that has great programs for mentally handicapped children. I would like to show you some cards made by clients of this Center. The cards come six to a package [the confederate presented six pictures, grouped with envelopes in an attractive packet]. This package of cards sells for 8 zloty and 50 grosz.1

In the “argument only” condition, she also added: “It's sublime to help people who are helpless!”

In the fear-then-relief condition the confederate held a leather wallet in her hand. Following the participant, she bent and lowered her hand holding the wallet towards the floor when she was about one meter from the participant. While saying. “Excuse me [sir/madam]”, she simultaneously straightened herself up slowly, as if she had just picked the wallet up from the floor, asking the participant: “Haven’t you lost your wallet?” In this moment the participants impulsively checked their pockets or searched their bags. They also surveyed the wallet held by the confederate (the one chosen for the study was very original and unusual) and then stated that it was not they who had lost it. The confederate then continued:

Oh, I’ll have to take it to the Information Desk then … but as we are already talking … I am a student of sociology, working for the “Handicap Center”, which is a non-profit organization that has great programs for mentally handicapped children. I would like to show you some cards made by clients of this Center. The cards come six to a package [the confederate presented six pictures, grouped with envelopes in an attractive packet]. This package of cards sells for 8 zloty and 50 grosz.

In the fear-then-relief plus argument condition, the confederate also added: “It's sublime to help people who are helpless!”

The confederate waited for a response from the participant. A purchase of any number of packages2 was recorded as being compliant with the request. When participants decided not to buy the cards the confederate thanked them for listening. All money collected was donated to the actual Handicap Center.

Results

Preliminary analyses indicate that the dependent variable (willingness to purchase Christmas cards) is not affected by the confederate making the request (χ²< 1). To verify whether we obtained a “classic” fear-then-relief effect in Study 1 we compared the rate of compliance in the control condition (emotionally neutral, without argument) to the rate of compliance in the fear-then-relief condition without argument. The difference is statistically significant, χ² (1, N = 144) = 6.21, p < .013, ϕ = .21. Moreover, a log-linear analysis shows the main effect of the emotional state variable. In the fear-then-relief condition participants were more likely to comply with the request (31.25%) than in the emotionally neutral condition (9.03 %), χ² (1, N = 288) = 22,11, p < .001, ϕ = .28. In addition, in the fear-then-relief condition, participants complied more often when the additional argument was applied (38.89%) than in situations when it was not (23.61%), χ² (1, N = 144) = 3.91, p < .048, ϕ = .16. However, in the control condition (neutral emotional state), adding the verbal argument to the proposal to buy Christmas cards did not increase the number of buyers, χ² (1, N = 144) = .08, ns.

Figure 1 presents the proportions of participants who decided to purchase cards in each of the four experimental conditions.

Figure 1. Percentage of compliance: Participants who decided to purchase Christmas cards in each of the four experimental conditions in Study 1. N = 72 in each condition.

Discussion

The experiment confirmed the effectiveness of the fear-then-relief technique of social influence. In addition it turned out that people in the fear-then-relief state manifested an especially high level of compliance when the verbal argument to purchase cards was applied. In the control condition (neutral emotional state), however, the verbal argument did not increase the participants’ inclination to purchase cards. Consequently we are not dealing here with a trivial effect by which the verbal argument constitutes a factor that simply increases people's readiness to comply with proposals made to them. On the contrary, the pattern of results obtained clearly shows that the argument has a strong effect, but only on those who are experiencing a sudden and unexpected relief.

One may say, however, that participants in the fear-then-relief condition had good reason to perceive the confederate as a person who wanted to help them. Hence their willingness to fulfill her request could have resulted from the intention to return the favor (Cialdini, 2009). Our next experiment was a conceptual replication of Study 1. This time, however, we decided to manipulate the fear-then-relief state in a different way to exclude an interpretation based on the reciprocity rule. In addition we decided to apply another verbal argument and to make a sales proposal of another type. Since in Study 1 participants were asked to purchase cards in order to help mentally handicapped children, in Study 2 we intended to apply a clearly commercial request. Correspondingly, we used a commercial argument, precisely the same one used by Davis and Knowles (1999).

Study 2

Overview

The general idea of creating a fear-then-relief state was based on Dolinski and Nawrat's (1998) experiment 2. The participants were car drivers who had parked their vehicles in a no-parking zone. A piece of paper was placed behind a windshield wiper. When the drivers returned to their cars and read the pieces of paper, it turned out that the papers were not police tickets but ads for hair shampoo. In a different condition the same pieces of paper were stuck to car doors. When the drivers were about to drive off, they were approached by a 22-year-old male confederate selling a summer screenwash with insect-removing formula. In some conditions the confederate used a verbal argument to encourage drivers to buy the screenwash. The confederate was unaware of the hypothesis being tested.

Method and participants

The experimental design was 2 (participant's gender) × 2 (emotional state: fear-then-relief vs neutral) × 2 (argument: present vs absent). The study was performed in the streets of Warsaw (Poland), in places notorious for illicit parking. Drivers who parked there became unwitting participants in our experiment (204 male and 116 female). A piece of paper, the size and color typical of traffic tickets, was placed behind a windshield wiper (a condition typical for a situation in which a driver is punished) or stuck on a car door with a piece of adhesive tape (a condition not typical for a situation in which a driver is punished with a fine). Our assumption was that a driver who approaches his or her car and sees a paper behind a wiper, unlike someone who sees it on a car door, will experience anxiety, supposing it to be a fine. The card we used in the experiment was an advertisement for Arcoon, an imaginary hair shampoo. On returning to his or her car each driver was allowed to learn the content of the note, and was then approached by a confederate who encouraged him or her to buy the screenwash:

Hi, I am selling a summer screenwash with insect removing formula. Would you like to buy a 1 liter container from me? The price is six zloty.3

In the argument condition the confederate added: “It's a bargain!”

The purchase of any number4 of screenwash containers was recorded as being compliant with the request.

Results

As in Study 1, we began the data analysis by checking whether the “classic” fear-then-relief effect had occurred. It turned out that the difference between the control condition (emotionally neutral state, no argument) and the fear-then-relief condition with no argument was highly statistically significant, χ(1, N = 160) = 10.0311, p < .002, ϕ = .25.

A log-linear analysis also indicated that the participants in the fear-then-relief condition purchased the screenwash more often (33.75%) than participants in the emotionally neutral condition (6.88%), χ² (1, N = 320) = 35.70, p < .001, ϕ = .33. This effect, however, was modified by the interaction of the factors of emotional state and presence/absence of argument. In the fear-then-relief state an argument increased the chances for a successful sale of screenwash (41.25% vs 26.25%), χ² (1, N = 160) = 4.03, p < .0447, ϕ = .16. Figure 2 illustrates the percentage of participants who purchased screenwash in each of the four experimental conditions.

Figure 2. Percentage of people who decided to purchase screenwash in each of the four experimental conditions in Study 2. N = 80 in each condition.

Discussion

The experiment once again demonstrated the effectiveness of the fear-then-relief technique of social influence and the even stronger effectiveness of this technique in situations where this special affective state is accompanied by a verbal argument. Significantly, this time the aforementioned pattern of results was obtained using a quite different manipulation of the participant's emotional state, which excludes an interpretation of the results obtained in terms of the reciprocity rules. In addition a very different request was used in this experiment. Since in Study 1 participants could interpret their own behavior as altruism (they were asked to buy something to help handicapped children), in this study the situation was clearly commercial: drivers were asked to purchase a screenwash.

General Discussion

The experiments presented in this article indicate unequivocally that the fear-then-relief technique is an effective tool of social influence. In Study 1 the participants were asked a question by someone who had just inquired whether they had lost their wallet. However, the results obtained in this study can be interpreted not only as the effect of the relief following a moment of fear but also through the prism of the reciprocity rule. Participants had good reasons to perceive the confederate as a person who wanted to do them a favor and spare them potential problems. Hence their consent to the confederate's request could have resulted from the participants’ desire to return the favor. In the second study, however, this alternative interpretation was ruled out. The confederate initiated the interaction with the participant by addressing him or her with the sales proposal; hence the participant had nothing to reciprocate.

The increased compliance of those who unexpectedly experience relief after fear corresponds with the results of earlier experiments (Dolinski & Nawrat, 1998; Dolinski et al., 2002). However, this time it has also been demonstrated that the technique becomes even more effective when the person experiencing relief from fear hears a straightforward argument for agreeing to fulfill a request or follow a suggestion. The two studies presented in this article shows that the effect is present in the area of consumer behavior.

An additional verbal argument, encouraging the participants to fulfill the request, increased the compliance rate only in the fear-then-relief conditions. This suggests that this additional element has an impact on compliance only when an individual undergoes the specific moment of emotional disorganization, and does not work when the person is in an emotionally neutral state. An analogous pattern of results has been obtained in studies on the disrupt-then-reframe technique. That argument alone, when added to the procedure of the conventional sales condition, did not increase the probability that participants would buy the products. However, the same argument increased the chances for higher sales when the participants were cognitively disorganized. This analogy in the pattern of results obtained in the studies on each of these techniques indicates their remarkable similarity. Emotional and cognitive disruption both seem to be two varieties of the same phenomenon.

In order to demonstrate this in an unambiguous way, however, an additional series of experiments would be indispensable. Yet, most of all, it is worth noting that Davis and Knowles (1999) not only showed that adding a verbal argument when participants find themselves in a short-term state of cognitive disorganization in fact increases their compliance with the request or proposal; the researchers also demonstrated that this effect of increased compliance does not occur if the verbal argument is added before the state of disorientation appears (for example, at the beginning of interaction). The pattern of results obtained by Davis and Knowles is entirely consistent with the assumptions of the DTR technique: the verbal argument works effectively only if the person is cognitively disturbed. In our study we have indeed shown that the additional argument increases compliance when participants experience relief after experiencing unexpected fear, but we have not tested the role such an argument could play if it occurred at a different moment (for example, at the beginning of the interaction or when the person is still experiencing fear). If it turned out that the extra verbal argument increased participants’ compliance when added while they were experiencing the relief from fear, but failed to increase compliance when it occurred at a different moment, this could be a serious indication that the DTR and FTR techniques are based on the same psychological mechanisms.

For the time being, however, we should emphasize above all the significant differences between these two techniques and try to explain their origins. One difference worth pointing out here is that the sequence of the two distinct emotional states is per se a sufficient condition for the fear-then-relief technique to be effective and adding the verbal argument only further enhances the technique's efficiency, whereas in the case of the disrupt-then-reframe technique the verbal argument is an indispensable condition for the technique to prove effective. What could possibly explain this difference? In our opinion, disorganization in the event of sudden relief from experienced fear is much more intense than the cognitive disorganization resulting from a sudden change of the action identification level. The programs triggered by different emotions are more rigidly defined than cognitive ones. These emotional programs, unlike cognitive programs, are also triggered automatically, immediately, and unconditionally—without any possibility for delay. There is evidence that people experiencing the fear-then-relief condition have hardly any possibilities for adequate evaluation of the situation (Dolinski et al., 2002) and most probably for this reason they tend to use the ready and automatic patterns of behavior that they have gained through prior experience (“When you are asked politely to do something, you should agree to do it”). The external verbal argument in support of fulfilling the request, which indicates the proper course of action and eventually makes people compliant, in fact only more convincingly confirms the scheme of action already developed independently, through generalizations learned in their own lifetimes from multiple incidences of similar experiences. The verbal argument then functions in the same way as the heuristic action indication, based on the generalization of previous experience, but it works more decisively. In the cognitive disrupt-then-reframe, the possibilities of the person making an adequate judgment of the situation are probably higher than in the state of emotional disruption. Therefore, in order to increase the person's compliance, it is necessary to provide her or him with an additional verbal argument supporting the decision to fulfill the request.

Notes

¹ Eight zloty and fifty grosz is equivalent to approximately 3 US$.

² Only two participants bought more than one package.

³ Six zloty is equivalent to approximately 2 US$.

⁴ Only four participants bought more than one container.