1. Introduction
Paradoxical interventions are those in which the patient is encouraged to maintain or increase the behavior or thought that s/he wants to remove. The aim of these techniques is to help the client to see the problem from a new perspective, as well as reach a rapid behavioral change [1].
Although paradoxical practices have been applied since ancient Greece, Viktor Frankl appears to be the first to set the bases of paradoxical interventions in psychotherapy [2]. Frankl applied paradoxical techniques to overcome phobias and anticipatory anxiety by encouraging patients to do, or wish to happen, the things that are feared. These directives appear to be especially indicated to treat anxiety difficulties [1,3,4] and high-reactant patients [1,4] since allowing or increasing the problem may be less threatening to high-reactant patients than prohibiting it [1].
Another significant figure in the history of paradoxical interventions was undoubtedly Milton Erickson. Rather than attempting to resolve or eliminate the symptom, he skillfully integrated it into the treatment process. His approach often involved instructing patients to deliberately engage in the symptom at specific times and for a predetermined duration [5]. Notably, Erickson refrained from passing clinical or moral judgment on the symptom and avoided expressing a goal of its removal. Instead, he focused on redefining or reframing the symptom’s meaning. He had an intuitive grasp of the idea that resisting or fighting against symptoms often intensified their disruptive effects.
Other prominent authors who contributed significantly to the expansion of paradoxical interventions were undoubtedly the research group from the Mental Research Institute in Palo Alto, California, which included Paul Watzlawick, Jackson, and Weakland, as well as others. Among their major contributions to communication theories, they discovered that communication is full of paradoxes. For example, “Be spontaneous”. This communicative paradox aptly illustrates the essence of paradoxical interventions: prescribing the patient to do something that is beyond their control [6].
In fact, Watzlawick, Beavin, and Jackson, in their book Pragmatics of Human Communication [7], a seminal work in the study of interpersonal relationships, argue that therapeutic communication must transcend the common, yet ineffective, advice patients often receive. Statements like “Be kind to your siblings” or “Stay out of trouble” may correctly define the desired change, but they cannot be considered therapeutic. These statements assume that the patient can choose to be kind or not, or to stay out of trouble or not. However, in the clinical realm, symptoms are essentially involuntary, making it virtually impossible to stop experiencing them voluntarily.
Due to their disruptive nature, paradoxical interventions have not been free from criticism and have often been considered manipulative and unethical [8], as the patient is asked to increase the problem voluntarily. However, for some authors, these criticisms are based on misunderstandings since paradoxical interventions have proven to be effective in the treatment of psychological symptoms.
Moreover, when discussing the ethics of the intervention, Hunsley [9] emphasizes the importance of distinguishing between compliance-based paradoxical interventions and defiance-based paradoxical interventions. In the former, the patient’s symptoms improve when they follow the therapist’s instructions. In the latter, symptom improvement is expected when the patient disobeys the therapist’s directives. It is in this latter case that psychological practice may be perceived as deceptive. However, in general, and as discussed in this article, when we refer to paradoxical interventions, we are speaking about compliance-based approaches.
For instance, if a patient is instructed to exaggerate their anxious symptoms, the expectation is that doing so will lead to a reduction in their anxiety. According to O’Connell [10], the reason paradoxical interventions are effective is that they transfer control of the situation to the patient. As stated previously, symptoms are beyond the client’s control and arise involuntarily. By deliberately performing these symptoms, the patient learns to manage them and, ultimately, to extinguish them. Gaining control over sensations that were previously beyond your control is a pleasant feeling. As Dunlap [11] noted, if the patient continues accomplishing the same thing, no new problems are created, whereas new “solutions” may introduce additional complications.
According to previous studies, as noted before, paradoxical interventions have had solid evidence supporting their therapeutic effectiveness. Shoham-Salomon and Rosenthal, in their meta-analysis, stated that paradoxical interventions were as effective as the typical treatment modes [12]. Moreover, Hampton concluded that paradoxical interventions were more effective to overcome certain psychopathologies than all comparison groups, including non-paradoxical treatments, placebo, and being included in a waiting list [13]. These results agree with those previously obtained by Hill, who concluded that patients treated by paradoxical interventions were consistently and significantly better than those treated with non-paradoxical interventions [3]. Nevertheless, there are no review studies that address the particularities of its proper execution. Hence, although studies on its therapeutic effectiveness are encouraging, there may be some confusion about when and how these types of techniques should be applied.
One of the main controversies regarding how to use paradoxical interventions is related to the use of a rationale. A rationale is an explanation of the prescription that the therapist provides to the patient, justifying its use. Since these techniques involve voluntarily increasing the behavior or thought one aims to eliminate, justifying this request may seem necessary and should have a significant beneficial effect [14]. However, other authors such as Ott [15] believe that paradoxical interventions should be given without a rationale so that patients do not expect any improvement during their execution. In his view, the correct implementation of the prescription should be free of expectations. Perhaps expecting clinical improvement from voluntarily increasing the problem goes against the paradoxical nature of the treatment.
It is important to clarify that when we refer to ‘the paradoxical prescription’, we are specifically talking about the moment when the therapist explains to the patient what they will need to do or think about between one therapy session and the next. For example, ‘exaggerate your anxiety symptoms when you are about to have a crisis’. The prescription, as we have seen, may or may not be accompanied by a rationale. For instance, ‘this will make the symptoms fall under your control’. The concept of ‘paradoxical intervention’, on the other hand, refers to the entirety of the treatment, which may include several psychotherapy sessions.
Another important aspect of how to employ paradoxical prescriptions is the use of humor. Viktor Frankl stated that the use of humor was a fundamental part of paradoxical interventions [16], as it helped the patient to reduce his/her anxiety when his/her problematic behavior was asked to increase. However, there does not seem to be consensus regarding the use of humor during prescription. Other authors, such as Ascher [17], consider that although it can help to reduce the patient’s reactance, the use of humor is not a requirement for the success of the prescription since the paradoxical directive should be effective on its own. Following this point of view, the use of humor may therefore be a feature of each therapist’s personal style, but not part of the intervention itself. It is also worth noting the potential problem of using humor in a clinical context, where patients are often going through a difficult time in their lives. For instance, Browning and Hull [18] argue that therapists must be very cautious with the use of humor, as empathy is a fundamental part of counseling, and it can be truly difficult to find humor in genuine suffering.
On the other hand, regarding when to use paradoxical interventions, although previous studies indicate particularly high effectiveness in anxious symptoms and highly reactant patients, as previously mentioned, review studies providing more specific information on when it may be advisable to employ these techniques are lacking. For example, whether they should be used for severe or mild symptoms.
For these reasons, the purpose of this study is to elucidate the specific aspects of prescribing paradoxical interventions, including both ‘how’ and ‘when’ it is most appropriate to employ them. To know when to use them, the study will examine (a) for which symptoms paradoxical interventions have been most studied and with what outcomes, and (b) with what type of patient it is most appropriate to use them, including patient reactance and symptom severity. On the other hand, to understand how to apply these techniques, the study will investigate the suitability of using humor and a rationale. Additionally, the influence of change expectancy on the therapeutic success of paradoxical prescriptions will be studied. The change expectancy is the patient’s belief that they will improve, and its level of influence on the effectiveness of paradoxical interventions could provide relevant information about how to use them.
2. Materials and Methods
A scoping review was performed following the preferred reporting items for systematic reviews and meta-analyses extension for scoping reviews (PRISMA-ScR) protocol registered on DOI 10.17605/OSF.IO/924HZ, using the following search strategy and searched in Google Scholar, PubMed, Web of Science, Scopus, PsycArticles, Psicodoc, and PsycINFO databases. A scoping review is a form of knowledge synthesis with the aim of informing practice, in this case, the proper clinical practice of paradoxical interventions, and providing future research directions. We chose the search terms shown below as they correspond to how different psychotherapy models refer to paradoxical interventions:
(a). (“Prescripción paradójica” OR “Intervención paradójica” OR “Instrucción paradójica” OR “Técnica paradójica” OR “Técnica de paradoja cognitiva” OR “Intención paradójica” OR “Prescripción del síntoma”) AND (“Psicoterapia” OR “Intervención psicológica”) NOT (“Trastorno de personalidad” OR “Trastorno de la personalidad”).
(b). (“Paradoxical prescription” OR “Paradoxical intervention” OR “Paradoxical instruction” OR “Paradoxical technique” OR “Cognitive paradox technique” OR “Paradoxical intention” OR “Symptom prescription”) AND (“Psychotherapy” OR “Psychological intervention”) NOT “Personality disorder”.
The article selection process was carried out as shown in Figure 1, following inclusion and exclusion criteria. Studies that address the effectiveness of paradoxical interventions or factors that affect it and are published in scientific journals are included. Theoretical or review articles and studies written in a language other than English and Spanish are excluded. We also excluded articles that address the effectiveness of paradoxical interventions in personality disorders to narrow the search, as they require distinct and complex treatments, which would complicate our analysis of other psychopathologies. We did not select a range of years of publication to address as many studies as possible. In addition, we excluded articles written in a language other than English or Spanish because Spanish is the authors’ mother tongue, and those are the languages we can evaluate properly. We were not restrictive in the inclusion and exclusion criteria to collect as much information as possible. We also decided to include case studies, as they can provide relevant information on the concrete execution of paradoxical prescriptions and their effectiveness in a personalized context. Screening was performed by the first author of the present article, and those studies whose inclusion or exclusion was not entirely clear were pooled to decide their selection among all three authors. Last search was executed on 16 May 2024. Finally, 20 studies were included in the review.
For the data charting process, a dataset was created with the relevant information of the selected articles to compare them systematically. The variables that were analyzed were year and country of publication, type of study, number of subjects, age and gender, symptomatology, paradoxical intervention employed, purpose, instruments used, measure of effectiveness, implications regarding patient’s reactance, implications regarding rationale, implications regarding symptom severity, implications regarding the use of humor, implication regarding change expectancy, main conclusions, and limitations.
We grouped the articles by the type of information they provided to the present study as follows:
-. Articles that provide information on when to use paradoxical interventions.
-. Studies that give information on how to use paradoxical interventions.
3. Results
A flow diagram is shown in Figure 1 explaining the selection of sources of evidence process.
Firstly, an evaluation of the quality of the selected studies was performed. We chose the Downs and Black Quality Checklist [19], a tool designed to assess the quality of both randomized and non-randomized studies of healthcare interventions, to evaluate the quality of clinical trials, and the single case experimental design (SCED) scale [20], a rating tool for evaluating the quality of single-case experimental designs in clinical research, to evaluate the quality of the single case studies included. The analysis was carried out by the first author of the present article and reviewed by the rest of the authors.
Regarding the clinical trials, 13.3% (n = 2) of them obtained a “good” quality rating [14,21], rated with 23 and 20, respectively; 66.6% (n = 10) received an “acceptable” quality rating [15,22,23,24,25,26,27,28,29,30], rated with 15, 18, 17, 18, 19, 19, 15, 17, and 15, respectively; and 20% (n = 3) were classified as “poor” [31,32,33], rated with 13, 13, and 14, respectively. As for the latter, the quality of the studies was mainly compromised by issues with external validity, as the sample was not representative of the general population, and certain shortcomings related to internal validity were also identified. It is worth noting that none of the studies conducted a double-blind procedure.
On the other hand, as regards single case studies, 20% (n = 1) obtained an “excellent” quality rating, rated with 9 [34]; 20% (n = 1) received a “good” quality outcome, rated with 5 [35]; 40% (n = 2) were classified as “acceptable”, both rated with 4 [36,37]; and 20% (n = 1) obtained a “poor” quality rating, rated with 3 [38], since some relevant quality issues were identified as not following an ABA method and the lack of evidence for generalization.
The quality outcomes will be considered during the interpretation of the results. No main conclusions will be drawn from studies classified as poor.
3.1. Characteristics of Selected Studies
As can be seen in Table 1, there are few recently published studies on the effectiveness of paradoxical interventions in psychotherapy. In particular, 75% (n = 15) of the selected studies were published in the 20th century. Furthermore, only four studies have been published in the last 20 years.
Regarding the type of study, 75% (n = 15) of the articles reviewed are clinical trials. Among them, 80% (n = 12) carried out a random assignment of the subjects to the different experimental conditions. In addition, five case studies were included. The selected articles studied, in total, the effectiveness of paradoxical prescriptions or factors that affect it in 745 subjects between the ages of 13 and 85.
The most studied paradoxical intervention is symptom prescription, investigated by 90% (n = 18) of the selected studies. Symptom prescription consists of asking the patient to increase the referred symptom voluntarily. For example, for onset insomnia, the demand would be “try to stay awake as long as possible” [14,15,25,27]. For anxiety attacks, it would consist of encouraging the patient to try to have an anxiety attack right away, voluntarily, or to increase voluntarily the anxiety sensations when you start to feel them [34]. Reframing, which is the paradoxical technique used to give a positive connotation to the problem, has been studied by two articles. Negative self-statements, a mood induction procedure where subjects are asked to read out loud a series of increasingly negative self-referential statements, have been studied by one article. Lastly, Chance and Action was also studied in one work. This technique involves allowing and even increasing the unwanted behavior in specific contexts. For example, in this case, to overcome excessive pornography use, the subject could only consume sexually explicit media when prompted to do so by a card game [37].
Insomnia is the symptom for which the effectiveness of paradoxical prescriptions has been studied the most, investigated by 25% (n = 5) of the selected articles; four of those studied onset insomnia [14,15,25,27], and the other studied intermittent and terminal insomnia [30]. This is followed by social-related anxiety, studied by 15% (n = 3) of the articles reviewed, including speaking anxiety [21,29] and social anxiety [33].
Some type of anxious symptomatology is studied by 70% (n = 14) of the articles. However, we can see a change in trend in terms of the symptomatology studied since all the studies published in the last 15 years (n = 3) evaluate the effectiveness of paradoxical prescriptions in non-anxious disorders, such as depression [28], procrastination [26], or pornography addiction [37]. The total amount of symptoms addressed by the reviewed articles can be seen above in Table 1.
3.2. When to Use Paradoxical Interventions
To determine when it is most appropriate to use paradoxical interventions, we will study which type of symptomatology and which type of patient they should be used with.
3.2.1. Type of Symptomatology
-
(a). Paradoxical interventions vs. no treatment
The effectiveness of paradoxical interventions compared to groups without treatment is assessed by 35% (n = 7) of studies, as shown in Table 2. The ‘Best group’ column refers to the group that received the most effective treatment, according to the results of each study.
Paradoxical interventions were more effective than groups without treatment in 71.4% (n = 5) of studies. It should be noted that the study carried out by Akillas [33] showed that the group receiving both reframing and symptom prescription improved significantly more than the group without treatment, but there were no differences between no treatment and symptom prescription alone. It is believed, according to the authors, that the reframing technique could have worked as a rationale for prescribing the symptom. The role of a rationale will be discussed later in Section 3.3.1.
Paradoxical intervention was not effective in the randomized clinical trial performed by Newton [31], where no significant differences were found between the paradoxical intervention and no-treatment groups in treating generalized anxiety according to their Sarason test anxiety scale results. However, in this case, it is worth noting that statistically significant differences were found between the pre-test and post-test in both the experimental and the no-treatment groups. Both improved their results, but possibly due to factors unrelated to the treatment. Symptom prescription was also not effective in treating snake phobia, according to results obtained by Haygeman [24]. Findings of this study are discussed further in Section 3.2.2.
Paradoxical interventions were significantly more effective than no treatment for treating onset insomnia in all studies (100%, n = 3) where this comparison was evaluated [15,25,27], as well as procrastination [26] and social anxiety [33].
-
(b). Paradoxical interventions vs. non-paradoxical treatment
The effectiveness of paradoxical interventions compared to non-paradoxical treatments is assessed by 45% (n = 9) of studies, as shown below in Table 3.
Paradoxical interventions were found to be as effective as non-paradoxical treatments in 66.6% (n = 6) of studies. Among them, in 66.6% (n = 4), both the paradoxical and non-paradoxical treatments were effective since significant differences between pre-test and post-test were found [14,15,21,29]. In that way, Ascher and Schotte [21] studied symptom prescription effectiveness on treating speaking anxiety compared to a non-paradoxical treatment alone, which consisted of cognitive therapy and anxiety reduction procedures. Even though both the paradoxical and the non-paradoxical interventions were effective, they found that subjects with a recursive anxiety component responded better to treatment when symptom prescription was included, and subjects without recursive anxiety responded better to treatment when symptom prescription was not included. Horvath and Goheen [14] compared the paradoxical intervention to stimulus control treatment; Ott [15] used sleeping feedback as non-paradoxical treatment, and Delvey [29] employed task-motivation directives as the non-paradoxical treatment to overcome speaking anxiety.
In the study performed by Newton [31], as noted previously, significant differences were found between pre- and post-test in both the paradoxical and non-paradoxical treatment, but they could not be considered effective since no significant differences were found between them and the no-treatment group. In addition, in the study performed by Haygeman [24], neither the paradoxical nor the non-paradoxical treatments were effective in treating snake phobia, even though the non-paradoxical treatment was exposure, based on evidence for treating phobias. The authors reported that only three subjects had tried the paradoxical strategy in its entirety and that group members initially were skeptical about whether the intervention was serious.
In addition, paradoxical interventions were more effective than non-paradoxical treatments in 22.2% (n = 2) of reviewed articles. In the study carried out by Ataoglu et al. [22], symptom prescription was found to be significantly more effective in treating conversion disorder than diazepam, which is accepted to be a valuable therapeutic agent in the treatment of conversion disorder.
In the study conducted by Gould [25], three treatments were compared as follows: paradoxical intervention, coping imagery (hereafter CI), and sleep information (hereafter SI), along with a control group on a waiting list. Regarding sleep onset latency, the paradoxical intervention was significantly more effective than SI, but no significant differences were found compared to CI, and neither between SI and the waiting list. For restedness, there were no significant differences among the three treatments, but there were significant differences between all three treatments and the waiting list. Finally, for the total amount of time slept during the night, the paradoxical intervention was significantly more effective than both CI and the waiting list, but no significant differences were found compared to SI. No differences were found between CI and the waiting list. The paradoxical intervention was the only treatment that showed significant differences compared to the waiting list across all variables. Additionally, regarding differences between pre- and post-treatment measures, the paradoxical intervention always showed significant changes for all variables. CI and SI showed significant changes for most variables but not for all, while the waiting list showed no changes in any variable. Therefore, according to the author, the paradoxical intervention was the most effective treatment, although all three treatments could be considered effective compared to the control group on the waiting list.
On the other hand, paradoxical intervention was less effective than non-paradoxical treatment in 11.1% (n = 1) of studies. Tripp [30] compared three non-drug interventions for the treatment of intermittent and terminal insomnia: paradoxical intervention, stimulus control (hereafter SC), and progressive relaxation (pR). Subsequently, PR was added to the other two treatments. The author concluded that the treatment combining SC and PR was the most effective based on participants’ self-reports for intermittent insomnia. Paradoxical intervention was the treatment that showed the least improvement in participants’ self-reports, with subjects even attributing any perceived improvement, when combined with PR, to the latter. It is important to emphasize that, although the objectives were to evaluate the effectiveness of the three interventions for the treatment of intermittent and terminal insomnia, complaints from patients about onset insomnia were also considered. Consequently, the effect of the treatments on this issue was analyzed as well, and the paradoxical intervention appeared to be the least effective according to reported latency. However, no significant data were provided. Results about terminal insomnia were found to be inconclusive.
-
(c). Other studies
The effectiveness of paradoxical interventions without comparing them to other treatments is assessed by 30% (n = 6) of studies, shown below in Table 4.
Paradoxical interventions were effective in 100% (n = 6) of studies. Among them, 83.3% (n = 5) were single case studies. The other was the clinical trial performed by Lanza et al. [28], in which the negative self-statements (NTS) technique was employed to overcome depression in geriatric subjects. Geriatric patients with major depressive disorder reduced significantly their depressed mood according to their scores in the Hamilton Depression Scale. NTS was also applied to healthy young subjects, and no improvement was found.
Among the case studies, 80% (n = 4) of them treated successfully an anxiety-related case, including a child’s performance anxiety [35], an adult man with an anxious-depressed mood [36], recurrent anxiety attacks [34], and adolescent vocational indecision [38]. The remaining study is carried out by Wernik [37], which successfully overcame pornography overuse using the chance and action technique.
3.2.2. Type of Patient
Continuing with the study of when paradoxical interventions should be used, we will next examine with which type of patient it is most advisable to use them, according to the reviewed studies.
-
(a). The role of reactance
The role of reactance when prescribing a paradoxical directive was assessed in 4 studies [14,23,32,36]. Among them, 75% (n = 3) concluded that paradoxical interventions were useful to reduce patients’ reactance.
In the randomized clinical trial performed by Keith [23], the effect of reframing and symptom prescription on patients’ reactance was studied. It was concluded that both high and low reactant subjects saw the counselor as able to help and understanding the issue. However, Keith stated that the high-reactant patients responded better to the paradoxical reframing technique, while the low-reactant patients evaluated the symptom prescription more positively. Moreover, in Horvath and Goheen’s study [14] assessing the effectiveness of symptom prescription in treating onset insomnia, they found that high-reactant patients continued to improve after treatment when the paradoxical intervention was applied and low-reactant patients continued to improve after treatment when stimulus control was employed. In addition, in the article carried out by Avni and Lazar [36], a case of an anxious-depressed adult man with a high reactance level was studied. Paradoxical intervention worked successfully. According to the authors, the suggestion that the subject’s symptoms should worsen gave new meaning to symptomatic improvement, no longer related to the therapist’s success.
Nevertheless, in the clinical trial performed by Gann [32], an increase in reactance levels was found when employing the paradoxical intervention. However, it should be noted that Gann studied reactance to paradoxical directives in healthy populations.
-
(b). Symptom severity
The relationship between symptom severity and paradoxical interventions is pending investigation, as only one study assessed it. Barach evaluated levels of anxiety, depression, multiphobia, and locus of control among insomniacs, and no interactions were found over the course of treatment [27].
3.3. How to Use Paradoxical Interventions
3.3.1. The Role of a Rationale
As noted previously in the Introduction section, a rationale is an explanation of the prescription that the therapist provides to the patient. Regarding its use to accompany the paradoxical prescription, 50% (n = 10) of articles reviewed give a rationale, and 50% (n = 10) do not.
However, only 30% (n = 6) of studies mention whether a rationale should be given. Among them, 50% (n = 3) stated that giving a rationale was beneficial for the success of therapy.
Nevertheless, the only study to assess whether a rationale should be given comparing a group with rationale and a group without rationale was the one carried out by Akillas to overcome social anxiety [33]. One group was treated with symptom prescription alone, another with symptom prescription together with reframing, and a control group without treatment. Reframing, as noted previously, is the paradoxical technique used to give a positive connotation of the problem, changing the patient’s mindset. That is why when it comes to asking to increase the problem voluntarily, reframing could work as a rationale. In the study performed by Akillas, the group treated with symptom prescription and reframing improved significantly while the group treated with symptom prescription alone did not. According to the author, the difference between groups was that in the first one the prescription was presented along with an explanation that reframed the problematic behaviors as legitimate and rational choices, acting as a useful rationale.
3.3.2. The Role of Humor
Only two studies addressed the role of humor when prescribing a paradoxical directive. The one performed by Newton to overcome generalized anxiety disorder assessed the patient’s sense of humor, and it was found that the paradoxical intervention worked best in low sense of humor patients, contrary to the previous hypothesis [31]. In Haygeman’s randomized clinical trial [24], patients were encouraged to make humorous self-statements while completing the prescription. The paradoxical directive was not effective.
3.4. The Importance of Change Expectancy
Only two studies assessed how change expectancy affects paradoxical interventions’ effectiveness. Both concluded that change expectancy does not predict therapy success.
In the randomized clinical trial performed by Delvey to overcome speaking anxiety, therapists with high apparent credibility and therapists with low apparent credibility were assigned to subjects, and no differences were found between groups [29]. In addition, in the study executed by Barach [27], as noted before, there were no differences in improvement between subjects that found the rationale as “highly credible” or “less credible”.
4. Discussion
Interest in the study of paradoxical interventions has been waning over the last 20 years. However, it seems that this lack of interest is not justified by results since these have been promising. This aligns with the fact that the few studies on paradoxical interventions published in the 21st century have shifted the research trend that was common in the previous century. While 81.3% (n = 13) of articles published in the 20th century studied anxiety difficulties, only 25% (n = 1) of those published this century assess anxiety disorders. Instead, depression and modern difficulties, such as procrastination and pornography overuse, have been studied.
Regarding the use of paradoxical interventions, the results provide compelling evidence that anxiety-related concerns are the most studied symptoms. Specifically, sleep difficulties appear to have the strongest clinical evidence. Treating onset insomnia has been the purpose of four clinical trials, and paradoxical interventions were reported to be effective in all of them, according to the authors [14,15,25,27]. However, it is worth noting that onset insomnia worsened when treating intermittent and terminal insomnia in the study conducted by Tripp [30], in which the paradoxical intervention was found to be the least effective treatment for intermittent insomnia as well. However, in the opinion of the authors of this article, based on the logic behind paradoxical interventions described in the Introduction, it is difficult to understand why a paradoxical intervention could be effective in treating intermittent insomnia. This type of insomnia is characterized by frequent awakenings during the night, and given the impossibility of waking up voluntarily in the middle of the night, it cannot be prescribed. Symptom prescription is not possible in this case, and the author, in his work, does not provide a justification for its use.
Furthermore, paradoxical interventions seem to be a useful tool for high-reactant patients. High-reactant patients are those who exhibit strong resistance to perceived threats to their autonomy and independence. This is why allowing or encouraging the problem may be less threatening to high-reactant patients than prohibiting it [1]. Additionally, Avni and Lazar [36], in their case study involving a high-reactance patient, noted that encouraging the patient to allow their symptoms to worsen reframed the meaning of symptomatic improvement, making it less tied to the therapist’s success. These results agree with those obtained by Peluso and Freund [1] and Hill and Norcross [4]. However, we found that the relationship between symptom severity and paradoxical interventions remains unknown.
In addition, Ascher and Schotte [21] found that paradoxical intervention was more effective than cognitive therapy and anxiety reduction techniques alone in cases of speaking anxiety when recursive anxiety was present. The authors define recursive anxiety as the ‘fear of fear’. Specifically, the anxiety that arises in response to the initial symptoms of anxiety themselves. In other words, recursive anxiety is a self-perpetuating cycle where fear of experiencing anxiety exacerbates the anxious response, creating a feedback loop of escalating anxiety.
This concept has notable implications for the current investigation. Onset insomnia may also involve recursive anxiety; in many cases, the individual’s concern over their inability to fall asleep fuels their insomnia, setting off a self-sustaining cycle of worry and wakefulness. Therefore, the effectiveness of paradoxical interventions may derive from their ability to disrupt recursive anxiety. By intentionally amplifying their own sensations or fears, individuals may reduce their anxiety by lessening the fear associated with those experiences. For example, if you want to stay awake all night, which is the paradoxical prescription in this case, the worry about not being able to fall asleep disappears. This could explain why paradoxical interventions have shown significant success in treating onset insomnia [14,15,25,27].
Another possible reason for the effectiveness of paradoxical interventions with recursive anxiety could be that they work by encouraging individuals to face their fears head-on. In therapy, one of the most common and effective ways to treat intense fears, like those seen in panic or phobias, is through exposure, where gradually confronting the feared object or situation helps reduce the anxiety it causes [39]. Similarly, when people experience fear in response to their own anxious sensations, intentionally increasing those sensations may help reduce the fear they feel towards them. This approach aligns with the principles of exposure by fostering tolerance and reducing sensitivity to anxiety. Nonetheless, these ideas remain speculative and would need thorough investigation in future research.
In addition, paradoxical interventions were effective in all the case studies included in the review. This could give us information about the importance of applying paradoxical interventions in a personalized way, attending to the specific needs of each patient.
Regarding how to use paradoxical interventions, there remains considerable ambiguity around their appropriate application.
There does not seem to be consensus about whether the prescription should be accompanied by a rationale since only 50% of studies gave a rationale for the prescription. On the one hand, Horvath and Goheen concluded that a clear and honest rationale appeared to have a significant beneficial effect on treating onset insomnia [14]; this conclusion agrees with the one found by Gould, also for treating onset insomnia [25]. It is noticeable that when it comes to insomnia, 80% (n = 4) of studies gave a rationale of the prescription. On the other hand, Haygeman stated that the counterintuitive nature of the paradoxical rationale may have detracted from expectation for treatment effectiveness, even though the rationale was credible [24], and Barach concluded that improvement was not related to how believable the subjects found the rationale [27]. In addition, in the randomized clinical trial performed by Ott to overcome onset insomnia, patients were told not to expect any improvement during the period they were following the paradoxical directive [15]. Paradoxically, subjects improved significantly. This seems to be the opposite of giving a rationale for the prescription. Further research is needed on when a rationale should be given, including for which disorders and with what type of patients it is most advisable to provide a rationale.
Continuing with the analysis of the appropriate execution of paradoxical interventions, Viktor Frankl [3] stated that humor should be used to reduce the patient’s anxiety when asked to increase the problem voluntarily. However, there is no evidence supporting that statement since the studies that assess the use of humor studied it in a different way than Frankl does, but with contrary conclusions. Newton [31] concluded that the paradoxical intervention worked better in patients with low sense of humor, and Haygeman [24] encouraged subjects to make humorous self-statements, finding that the treatment was not effective and exposing that the directive was not taken seriously. Finally, another significant aspect to consider regarding the execution of any therapy is the influence of patients’ change expectancy. According to the results found by Barach [27] and Delvey [29], paradoxical interventions’ effectiveness appears to be independent from patients’ change expectancy. This is a positive aspect, as it suggests that paradoxical interventions are effective in and of themselves, regardless of, for example, whether the therapist is seen as highly credible or not. In any case, more research is required since only two studies were evaluated.
We cannot overlook that one aspect that has generated some controversy regarding paradoxical interventions is the ethical dimension. As highlighted in the Introduction, opinions on this matter are divided. This controversy is also evident in the results of this research, as there does not appear to be a consensus on whether a rationale should accompany a paradoxical prescription. While a justified prescription might be more honest, some authors, such as Ott [15], prefer a purer application of the paradoxical intervention, leaving it free of expectations. This latter perspective could reasonably be considered somewhat dishonest. However, Hunsley [40] argues that this is not exclusive to paradoxical interventions but applies to psychotherapy in general, as therapists do not always share their theories, hypotheses, and intentions with the patient, as doing so could even be detrimental. Furthermore, it is worth noting that, according to Gerkes and collaborators [41], in clinical practice, information about the expected benefits, common factors, or risks of psychotherapy is rarely disclosed.
This research has had some limitations. Firstly, a sample of 20 studies is small. Also, the variability in inclusion criteria among the studies, such as differences in design, sample, or comparison treatment, hindered the direct comparison of paradoxical interventions, even in cases where they were studied for the same symptoms. This heterogeneity limited the ability to draw consistent conclusions about their relative effectiveness. Moreover, studies with negative or null results may not be published, affecting the representativeness of the review. In addition, the quality of the studies was poor in three (20%) clinical trials [31,32,33]. However, it should be noted that the Downs and Black Quality Checklist gives two points if double blinding is applied. In psychotherapy studies, as stated by Ascher and Schotte [21], it is not possible to apply double blinding since the psychologist must know what therapy is being applied. It is also possible that since they are mostly studies from the last century, their quality is not as refined as current studies, for example, to make the random division of the subjects to the experimental conditions digitally. In any case, no main conclusions have been drawn from poor-quality studies. Moreover, only articles published in scientific journals were included, potentially missing valuable unpublished information such as presentations at conferences. On the other hand, all articles published in a language other than English and Spanish were excluded. Finally, we identified some articles that appeared to be eligible but for which we did not obtain access through the authors’ institutions.
Regarding future research, we consider it essential to continue studying the effectiveness of paradoxical interventions, as past research results suggest that their decline in popularity was more a result of shifting therapeutic trends than of any definitive empirical dismissal, a concept echoed in Thomas Kuhn’s theory of scientific paradigms [42]. Kuhn argues that shifts in scientific focus often stem not from concrete evidence against a technique but from changing conceptual frameworks or “fashions” in scientific thought. This seems applicable to paradoxical interventions, which, despite showing promising results in the 1980s and 1990s, were largely set aside as the therapeutic paradigm shifted.
Moreover, with third-wave therapies, we have seen a resurgence of techniques that had fallen into disuse in the past. One example is mindfulness, a practice rooted in ancient Buddhist tradition but without scientific evidence until relatively recent times. Paradoxical approaches, similarly, could offer innovative ways to address involuntary symptoms, enriching modern therapeutic frameworks.
Revisiting these interventions could unlock valuable therapeutic potential, especially if we deepen our understanding of their appropriate use. Current confusion around how and when to apply paradoxical techniques effectively suggests that we may be neglecting a powerful tool. Systematic investigation could clarify their mechanisms and applications, potentially broadening the therapeutic repertoire and offering new strategies for working with anxious and high-reactant patients.
More specifically, we see potential in the study of anxiety difficulties with a recursive component. In addition, more research is needed on the particularities of its execution, including the study of both how and when paradoxical interventions should be used. For example, studying whether the appropriateness of accompanying a paradoxical prescription with a rationale depends on the type of patient or the type of problem might be more useful than attempting to establish an absolute truth about whether providing an explanation is generally advisable or not.
5. Conclusions
Six conclusions can be drawn from our study: (a) Paradoxical interventions appear to be especially indicated to treat onset insomnia. (b) It seems that paradoxical interventions work better in high-reactant patients. (c) There are indicators suggesting high effectiveness of paradoxical interventions in the treatment of recursive anxiety. Therefore, further research is needed. (d) The relationship between effectiveness and symptom severity requires future research. (e) There is no consensus on whether a paradoxical prescription should be accompanied by a rationale. Further research is needed. (f) There is no evidence supporting the therapeutic effectiveness of using humor during the treatment.
Conceptualization, J.V.V., A.D. and M.M.; methodology, J.V.V., A.D. and M.M.; formal analysis, J.V.V., A.D. and M.M.; investigation, J.V.V.; resources, J.V.V., A.D. and M.M.; data curation, J.V.V.; writing—original draft preparation, J.V.V.; writing—review and editing, J.V.V., A.D. and M.M.; supervision, A.D. and M.M. All authors have read and agreed to the published version of the manuscript.
Not applicable.
Not applicable.
Not applicable.
The authors declare no conflicts of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Enlarge this image.Figure 1. Flowchart of articles identified and used in review.
Characteristics of selected studies.
| Authors (Year) | Type of Study (N) | Symptom | Paradoxical Intervention |
|---|---|---|---|
| Keith (1995) [ | Randomized clinical trial (86) | Oppositional behavior | SP * and reframing |
| Ataoglu et al. (2003) [ | Randomized clinical trial (30) | Conversion disorder | Symptom prescription |
| Horvath and Goheen (1990) [ | Randomized clinical trial (41) | Onset insomnia | Symptom prescription |
| Haygeman (1986) [ | Randomized clinical trial (29) | Phobia | Symptom prescription |
| Gould (1988) [ | Randomized clinical trial (48) | Onset insomnia | Symptom prescription |
| Tripp (1985) [ | Randomized clinical trial (11) | Intermittent insomnia | Symptom prescription |
| Ott (1980) [ | Randomized clinical trial (56) | Onset insomnia | Symptom prescription |
| Barach (1982) [ | Clinical trial (40) | Onset insomnia | Symptom prescription |
| Lanza et al. (2017) [ | Clinical trial (52) | Depression | Negative self-statements |
| Otermin and Hautzinger (2018) [ | Random clinical trial (119) | Procrastination | Symptom prescription |
| Gann (1995) [ | Clinical trial (51) | None | Symptom prescription |
| Ascher and Schotte (1999) [ | Randomized clinical trial (20) | Speaking anxiety | Symptom prescription |
| Newton (1987) [ | Randomized clinical trial (53) | Generalized anxiety disorder | Symptom prescription |
| Akillas (1991) [ | Randomized clinical trial (47) | Social anxiety | SP * and reframing |
| Delvey (1980) [ | Randomized clinical trial (57) | Speaking anxiety | Symptom prescription |
| Dollinger (1985) [ | Case study (1) | Performance anxiety | Symptom prescription |
| Avni and Lazar (1998) [ | Case study (1) | Anxious and depressed | Symptom prescription |
| Gentelet and Widdowson (2016) [ | Case study (1) | Anxiety attacks | Paradoxical alliances |
| López (1983) [ | Case study (1) | Vocational indecision | Symptom prescription |
| Wernik (2012) [ | Case study (1) | Pornography overuse | Chance and action |
* SP stands for symptom prescription.
Studies that assess the effectiveness of paradoxical interventions compared to groups without treatment.
| Authors (Year) | Symptom | Effectiveness | Best Group |
|---|---|---|---|
| Haygeman (1986) [ | Phobia | No significant differences between groups | Equal |
| Gould (1988) [ | Onset insomnia | Significant differences between groups | PI |
| Ott (1980) [ | Onset insomnia | Significant differences between groups | PI |
| Barach (1982) [ | Onset insomnia | Significant differences between groups | PI |
| Otermin and Hautzinger (2018) [ | Procrastination | Significant differences between groups | PI |
| Newton (1987) [ | GAD | No significant differences between groups | Equal |
| Akillas (1991) [ | Social Anxiety | Significant differences between groups | PI * |
* PI stands for paradoxical intervention.
Studies that assess the effectiveness of paradoxical interventions compared to non-paradoxical treatments.
| Authors (Year) | Symptomatology | Effectiveness | Best Group |
| Ataoglu et al. (2003) [ | Conversion disorder | Significant differences between groups | PI |
| Horvath and Goheen (1990) [ | Onset insomnia | Significant differences between pre-test and post-test | Equal |
| Haygeman (1986) [ | Phobia | No significant differences between pre-test and post-test | Equal |
| Gould (1988) [ | Onset insomnia | Significant and no significant interactions found | PI |
| Tripp (1985) [ | Int. insomnia | Better self-report improvement | SC * |
| Ott (1980) [ | Onset insomnia | Significant differences between pre-test and post-test | Equal |
| Ascher and Schotte (1999) [ | Speaking anxiety | Significant differences between pre-test and post-test | Equal |
| Newton (1987) [ | GAD | Significant differences between pre-test and post-test | Equal |
| Delvey (1980) [ | Speaking anxiety | Significant differences between pre-test and post-test | Equal |
* PI was the best group according to the author, but some interactions were found to be significant while others did not. * SC stands for stimulus control.
Studies that assess the effectiveness of paradoxical interventions without comparing them to other treatments.
| Authors (Year) | Type of Study (N) | Symptomatology | Effectiveness |
|---|---|---|---|
| Lanza et al. (2017) [ | Clinical trial (52) | Depression | Significant differences between pre- and post-test |
| Dollinger (1985) [ | Case study (1) | Performance anxiety | Self-report improvement |
| Avni and Lazar (1998) [ | Case study (1) | Anxious and depressed | Self-report improvement |
| Gentelet and Widdowson (2016) [ | Case study (1) | Anxiety attacks | Significant differences between pre- and post-test |
| López (1983) [ | Case study (1) | Vocational Indecision | Self-report improvement |
| Wernik (2012) [ | Case study (1) | Pornography overuse | Self-report improvement |
References
1. Peluso, P.R.; Freund, R. Paradoxical interventions: A meta-Analysis. Psychotherapy; 2023; 60, pp. 283-294. [DOI: https://dx.doi.org/10.1037/pst0000481] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36931813]
2. Frankl, V.E. Paradoxical intention and dereflection. Psychotherapy; 1975; 12, pp. 226-237. [DOI: https://dx.doi.org/10.1037/h0086434]
3. Hill, K.A. Meta-analysis of paradoxical interventions. Psychotherapy; 1987; 24, pp. 266-270. [DOI: https://dx.doi.org/10.1037/h0085714]
4. Hill, C.; Norcross, C. Skills and methods that work in psychotherapy: Observations and conclusions from the special issue. Psychotherapy; 2023; 60, pp. 407-416. [DOI: https://dx.doi.org/10.1037/pst0000487]
5. Haley, J. Strategies of Psychotherapy; 1st ed. Grune & Stratton: New York, NY, USA, 1963.
6. Brown, J.; Slee, P. Paradoxical Strategies: Ethics of Intervention. Prof. Psychol. Res. Pract.; 1986; 17, pp. 487-491. [DOI: https://dx.doi.org/10.1037/0735-7028.17.6.487]
7. Watzlawick, P.; Jackson, D.; Beavin, J. Pragmatics of Human Communication: A Study of Interactional Patterns, Pathologies, and Paradoxes; W.W. Norton: New York, NY, USA, 1967.
8. Cavell, T.A.; Frentz, C.E.; Kelley, M.L. Acceptability of paradoxical interventions: Some nonparadoxical findings. Prof. Psychol. Res. Pract.; 1986; 17, pp. 519-523. [DOI: https://dx.doi.org/10.1037/0735-7028.17.6.519]
9. Hunsley, J. Defiance-based symptom prescription and psychological reactance: A critical evaluation. Prof. Psychol. Res. Pract.; 1997; 28, pp. 36-43. [DOI: https://dx.doi.org/10.1037/0735-7028.28.1.36]
10. O’Connell, D.S. Symptom prescription in psychotherapy. Psychother. Theory Res. Pract.; 1983; 20, pp. 12-20. [DOI: https://dx.doi.org/10.1037/h0088472]
11. Dunlap, K. Habits; Liverwright: New York, NY, USA, 1933.
12. Shoham-Salomon, V.; Rosenthal, R. Paradoxical interventions: A meta-Analysis. J. Consul. Clin. Psychol.; 1987; 55, pp. 22-28. [DOI: https://dx.doi.org/10.1037/0022-006X.55.1.22]
13. Hampton, B.R. The Efficacy of Paradoxical Interventions: A Quantitative Review of the Research Evidence. Ph.D. Thesis; University of Texas: Austin, TX, USA, May 1988.
14. Horvath, A.O.; Goheen, M.D. Factors mediating the success of defiance- and compliance-based interventions. J. Couns. Psychol.; 1990; 37, pp. 363-371. [DOI: https://dx.doi.org/10.1037/0022-0167.37.4.363]
15. Ott, B.D. The Efficacy of Paradoxical Intention in the Treatment of Sleep Onset Insomnia Under Differential Feedback Conditions. Ph.D. Thesis; Hofstra University: New York, NY, USA, 1980.
16. Frankl, V.E. The Unconscious God; 1st ed. Simon and Schuster: New York, NY, USA, 1975.
17. Ascher, L.M. Paradoxical techniques in behavior therapy. Proceedings of the Conference on Behavior Therapy and Behavior Modification; New York, NY, USA, 10–11 November 1977.
18. Browning, S.; Hull, R. Reframing paradox. Prof. Psychol. Res. Pract.; 2021; 52, pp. 360-366. [DOI: https://dx.doi.org/10.1037/pro0000384]
19. Downs, S.H.; Black, N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomized and non-randomized studies of health care interventions. J. Epidemiol. Community Health; 1998; 52, pp. 377-384. [DOI: https://dx.doi.org/10.1136/jech.52.6.377] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/9764259]
20. Tate, R.L.; McDonald, S.; Perdices, M.; Togher, L.; Schultz, R.; Savage, S. Rating the methodological quality of single-subject designs and n-of-1 trials: Introducing the Single-Case Experimental Design (SCED) Scale. Neuropsychol. Rehabil.; 2008; 8, pp. 385-401. [DOI: https://dx.doi.org/10.1080/09602010802009201] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/18576270]
21. Ascher, L.M.; Schotte, D.E. Paradoxical intention and recursive anxiety. J. Behav. Ther. Exp. Psychiatry; 1999; 30, pp. 71-79. [DOI: https://dx.doi.org/10.1016/S0005-7916(99)00009-9]
22. Ataoglu, A.; Özcetin, A.; Icmeli, C.; Özbulut, Ö. Paradoxical therapy in conversion reaction. J. Korean Med. Sci.; 2003; 18, pp. 581-584. [DOI: https://dx.doi.org/10.3346/jkms.2003.18.4.581]
23. Keith, B. Adolescent Client Perception of and Reactions to Reframe and Symptom Prescription Techniques. Ph.D. Thesis; Oklahoma State University: Stillwater, OK, USA, December 1995.
24. Haygeman, J. Paradoxical Intention and Phobia: A Laboratory Analogue. Ph.D. Thesis; Washington State University: Pullman, WA, USA, August 1986.
25. Gould, L.P. A Comparison of Three Cognitive Behavioral Treatments for Insomnia: Paradoxical Intention, Coping Imagery and Sleep Information. Ph.D. Thesis; Vanderbilt University: Nashville, TN, USA, 1988.
26. Otermin-Cristeta, S.; Hautzinger, M. Developing an intervention to overcome procrastination. J. Prev. Interv. Community; 2018; 46, pp. 171-183. [DOI: https://dx.doi.org/10.1080/10852352.2016.1198169]
27. Barach, P.M. Client Personality Variables and Paradoxical Intention Treatment for Insomnia. Ph.D. Thesis; Case Western Reserve University: Cleveland, OH, USA, August 1982.
28. Lanza, C.; Müller, C.; Riepe, M. Positive modo on negative self-statements: Paradoxical intervention in geriatric patients with major depressive disorder. Aging Ment. Health; 2017; 22, pp. 748-754. [DOI: https://dx.doi.org/10.1080/13607863.2017.1306834]
29. Delvey, J. Paradoxical Intention in the Treatment of Speaking Anxiety. Ph.D. Thesis; Temple University: Philadelphia, PA, USA, December 1980.
30. Tripp, E.C. A Comparison of the Effectiveness of Three Behavioral Treatments on Intermittent and Terminal Insomnia. Ph.D. Thesis; Western Michigan University: Kalamazoo, MI, USA, December 1985.
31. Newton, G.R. Effect of Client Sense of Humor and Paradoxical Intervention on Anxiety. Ph.D. Thesis; University of Nebraska: Omaha, NE, USA, August 1987.
32. Gann, M.K. Cognitive Assessment of Reactance Using the Articulated Thoughts in Simulated Situations Paradigm. Master’s Thesis; University of Southern California: Los Angeles, CA, USA, August 1995.
33. Akillas, E. Symptom Prescription with and Without Reframing: Relative Efficacy of Two Paradoxical Directives in the Treatment of Socially Anxious Males. Ph.D. Thesis; Temple University: Philadelphia, PA, USA, January 1991.
34. Gentelet, B.; Widdowson, M. Paradoxical alliances in transacional analysis psychotherapy for anxiety. Trans. Anal. J.; 2017; 46, pp. 182-195. [DOI: https://dx.doi.org/10.1177/0362153716650657]
35. Dollinger, S.J. Effects of a paradoxical intervention on a child’s anxiety about sleep and sports-related performance. Percept. Mot. Skills; 1985; 61, pp. 83-86. [DOI: https://dx.doi.org/10.2466/pms.1985.61.1.83]
36. Avni, A.; Lazar, R. Dynamic Perspectives of paradoxical intervention. Isr. J. Psychiatry Relat. Sci.; 1988; 35, pp. 38-47.
37. Wernik, U. A non-blaming chance and action approach to therapy with sexually explicit media overuse: A case study. Int. J. Ment. Health Addict.; 2012; 10, pp. 770-777. [DOI: https://dx.doi.org/10.1007/s11469-011-9371-0]
38. López, F.G. A Paradoxical approach to vocational indecision. Pers. Guid. J.; 1983; 61, pp. 410-412.
39. Barlow, D.H. Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic; 2nd ed. Guilford Press: New York, NY, USA, 2002.
40. Hunsley, J. Conceptions and misconceptions about the context of paradoxical therapy. Prof. Psychol. Res. Pract.; 1988; 19, pp. 553-559. [DOI: https://dx.doi.org/10.1037/0735-7028.19.5.553]
41. Gerke, L.; Meyrose, A.K.; Nestoriuc, Y. Informed consent for psychotherapy: Ethical illusion or clinical reality? A survey about psychotherapists’ attitudes and practices in Germany. Clin. Psychol. Psychother.; 2023; 30, pp. 166-178. [DOI: https://dx.doi.org/10.1002/cpp.2790]
42. Kuhn, T.S. The Structure of Scientific Revolutions; 1st ed. University of Chicago Press: Chicago, IL, USA, 1962.
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Martín, Martha 2
1 Faculty of Psychology, University of Valencia, 46100 Valencia, Spain;
2 Faculty of Psychology, Universidad Cooperativa de Colombia, Santa Marta 470003, Colombia;