Empirical Investigation of the Verbal Cues Involved in Delivering Experiential Metaphors

Context and Objectives

The use of experiential metaphors has been emphasized in various psychotherapies, particularly acceptance and commitment therapy (ACT). Experiential metaphors create a symbolic context in which clients vividly contact the consequences of their behavior. Although the relevance of delivering metaphors experientially in ACT is recognized, little research has analyzed the specific verbal variables involved in their efficacy.

The purpose of this study was to advance the experimental analysis of components involved in the experiential delivery of metaphors. Two main components were analyzed: (1) the perspective of the protagonist (Self vs. Other) and (2) the inclusion of verbal cues prompting relational elaboration (Elaboration vs. No Elaboration). The study hypothesized that both components would interact positively to increase the metaphor effect, particularly when evaluated through pain tolerance.

Method

Participants

Eighty-four undergraduates (42 women and 42 men; age range 18-40 years; M = 19.64, SD = 2.75) from different academic programs at a university in Bogotá, Colombia participated in the experiment. Inclusion criteria were: (a) being an undergraduate student at the university where the study was conducted (with institutional bioethical approval), and (b) being 18 years or older. Exclusion criteria were: (a) prior experience with the procedures used (cold pressor task or CPT, RFT, and ACT), and (b) having medical conditions incompatible with the study (cardiac and circulatory affections, hypertension, diabetes, epilepsy, chronic pain conditions, or recent wounds). Participants received COP 10,000 (approximately USD 3) as compensation.

Design

A double-blind, randomized, 2 × 2 factorial experimental design was employed that analyzed the effect of two independent variables: (a) presenting the metaphor by asking participants to imagine themselves as protagonists of the story versus imagining a fictional character as the protagonist (Self vs. Other), and (b) including relational cues prompting elaboration of rules derived from the metaphor versus not including such cues (Elaboration vs. No Elaboration). This resulted in four experimental conditions: Condition A (Self & Elaboration), Condition B (Self & No Elaboration), Condition C (Other & Elaboration), and Condition D (Other & No Elaboration).

Materials and Intervention

The experiment used the cold pressor task (CPT) as the experimental task. Participants inserted their right hand into a glass container with circulating water at 3-4 degrees Celsius and were instructed to maintain their hand in the water for as long as possible (maximum 300 seconds). Pain tolerance was measured as the percentage of time tolerating pain at posttest in relation to pretest, where spending the same amount of time corresponds to 100%, spending twice the time corresponds to 200%, and so on.

Metaphors were presented via audio (260 seconds duration) through headphones connected to a tablet. The metaphor used was based on the swamp metaphor, a classic metaphor in ACT that illustrates how advancing toward values involves willingness to experience psychological discomfort.

The four protocols were:

• Protocol A: Metaphor in first person with pauses and relational elaboration cues (Self & Elaboration)
• Protocol B: Metaphor in first person without pauses or elaboration cues (Self & No Elaboration)
• Protocol C: Metaphor in third person with pauses and relational elaboration cues (Other & Elaboration)
• Protocol D: Metaphor in third person without pauses or elaboration cues (Other & No Elaboration)

Instruments

Generalized Pliance Questionnaire (GPQ): A measure of generalized pliance consisting of 18 items rated on a 7-point Likert-type scale. The GPQ obtained an alpha of 0.92 in this study and was used to control for the possible influence of social approval on participants' performance.

Visual Analog Scale (VAS): A 10 cm scale to measure the intensity of pain induced after each exposure to the CPT.

Analysis

Analyses were conducted using Bayesian informative analyses of variance (BAIN ANOVAs) in JASP 0.14.1.0 software. Six alternative hypotheses regarding the increase in pain tolerance were tested to determine which hypothesis had the highest posterior model probability (PMP). The BAIN ANOVA provides Bayes factors (BF) that quantify the degree of support for each hypothesis under consideration, allowing evaluation of multiple alternative hypotheses simultaneously without the need for multiple comparison corrections.

Results

Initial Equivalence Between Groups

Participants were equivalent on generalized pliance (BF₁₀ = 70.426), pretest pain tolerance (BF₁₀ = 9.870), and pretest pain intensity (BF₁₀ = 18.608). Therefore, the experimental conditions were equivalent at baseline on these variables.

Effect of Experimental Protocols

A total of 10 of 20 participants in Condition A showed improvements in pain tolerance greater than 250%, compared to 6 in Condition B, 5 in Condition C, and 4 in Condition D. Descriptive data showed that participants in Condition A presented the highest percentage increase in pain tolerance (M = 268.211, SD = 167.467), followed by Condition C (M = 204.809, SD = 100.193), Condition B (M = 180.857, SD = 73.013), and Condition D (M = 175.407, SD = 75.998).

Bayesian Informative Analysis of Variance

Hypothesis analysis showed that H₁ (Self and Elaboration components show a positive interaction effect but no separate effect) was the hypothesis most supported by the observed data (PMP = 0.545). The second best-supported hypothesis was H₂ (both Self and Elaboration components have positive effects and interact; PMP = 0.223), suggesting that Self and Elaboration positively affected pain tolerance and interacted. The remaining hypotheses obtained PMPs lower than 0.223.

Pain Intensity Data

Regarding pain intensity (pretest-posttest differential), the hypothesis best supported was H₁ (no differences in pain intensity across conditions; PMP = 0.991), indicating that the process of change was not a reduction in pain intensity per se, but rather a reduction of the discriminative functions for avoiding pain and contact with its symbolic and emotional consequences.

Effect Sizes

Effect sizes between conditions ranged from small to moderate: Condition A vs. Condition B (d = 0.73), Condition A vs. Condition C (d = 0.47), Condition A vs. Condition D (d = 0.76).

Discussion and Conclusions

The results demonstrated that delivering experiential metaphors with two specific active components (Self and Elaboration) produces a greater effect on pain tolerance than using only one of these components or neither. The finding of a positive interaction between Self and Elaboration is consistent with the study's initial hypothesis. Presenting the metaphor with the client as protagonist (Self) allows a more intense transformation of functions while establishing contact with the symbolic and emotional consequences. When the metaphor is introduced with a fictional character as protagonist (Other), an additional perspective-taking step is required for the client to contact the personally relevant consequences.

Including relational elaboration cues (Elaboration) allows the client to derive and follow rules related to their own values. Without relational elaboration, the derivation of augmented functions depends on the client's relational framing skills, which could limit efficacy in clients with less fluency in these skills.

The results support the integration of Relational Frame Theory (RFT) conceptualizations into the experiential delivery of metaphors. The study contributes to understanding how therapists can enhance the efficacy of experiential metaphors by systematically manipulating specific verbal components.

Importance and Contribution

This study represents a significant contribution to research on mechanisms of change in metaphor-based psychotherapy. It provides empirical evidence regarding specific variables that increase the efficacy of experiential metaphors in ACT. The findings have direct practical implications for therapist training, suggesting that therapists should be explicitly trained to introduce metaphors with the client as protagonist and to provide verbal cues that promote relational elaboration of derived rules. This is consistent with RFT theoretical texts that provide guidelines on the use of experiential metaphors. Additionally, the study emphasizes the importance of investigating beyond a single experimental task, as generalization to actual clinical interventions requires replication with multiple tasks and clinical populations.

---