The Role of Common Physical Properties and Augmental Functions in Metaphor Effect

Full reference: Sierra, M. A., Ruiz, F. J., Flórez, C. L., Riaño-Hernández, D., & Luciano, C. (2016). The role of common physical properties and augmental functions in metaphor effect. International Journal of Psychology and Psychological Therapy, 16(3), 265-279.

Study type: Laboratory experiment / Analogue study

Background and objectives

Metaphor is a tool widely used in psychotherapy, particularly in Acceptance and Commitment Therapy (ACT). From the perspective of Relational Frame Theory (RFT), metaphors involve relating two distinct relational networks (the metaphor and the individual's experience) through common physical properties. Previous RFT research has identified two key components that facilitate the effectiveness of metaphors: (a) common physical properties between the metaphor and the person's experience, which facilitate the relationship between the two relational networks; and (b) augmental functions (appetitive values specified in the metaphor), which can alter the behavioral functions of the experience. Prior studies (such as Luciano et al., 2014; Páez Blarrina et al., 2008; Gutiérrez et al., 2004) have demonstrated that ACT-based protocols increase pain tolerance without reducing pain perception, suggesting that the mechanism is alteration of discriminative/augmental functions rather than pain reduction. The objective of this study was to experimentally analyze the effect of these two metaphor components on pain tolerance using a cold-pressor task.

Method

Participants

The study included 83 undergraduate students (42 women; age range = 18 to 34 years; M = 19.47, SD = 1.46) from different courses at the university. Participants were randomly allocated to the four experimental conditions. Participants were excluded if they had Raynaud's disease, chronic pain, cardiac conditions, pregnancy, or prior experience with the cold-pressor task. The experimenter was blinded to condition assignment, as participants listened to the protocols through headphones.

Experimental design

The study followed a 2×2 between-subjects factorial design. The independent variables were: (a) the presence or absence of common physical properties between the metaphor and the discomfort experienced in the cold-pressor task (present vs. absent), and (b) the presence or absence of explicit specification of appetitive augmental functions in the metaphor content (present vs. absent). The combination of these two levels of independent variables created four experimental conditions:

• Condition A (Both present): Metaphor that included common physical properties with the cold-pressor discomfort AND specification of augmental functions (values) to tolerate pain (helping others who suffer).
• Condition B (Physical properties only): Metaphor with common physical properties but WITHOUT specification of augmental functions.
• Condition C (Augmental functions only): Metaphor with specification of augmental functions but WITHOUT common physical properties with the pain experience (narrated a non-physical challenging experience).
• Condition D (Control): Neutral metaphor without common physical properties or specified augmental functions.

Random allocation of participants to conditions resulted in approximately 20-21 participants per condition.

Materials and apparatus

A cold-pressor apparatus was used, consisting of a glass container measuring 30×20×20 cm with two interconnected compartments: one for ice and one for water. A digital thermometer adhered to the container maintained the temperature, and two water pumps (300 liters/hour) circulated the water. An ice maker machine kept the container temperature constant. The experimental protocols were metaphorical narratives lasting approximately 4 minutes, presented as audio through headphones. The protocols were adaptations of the swamp metaphor (Hayes et al., 1999; Gutiérrez et al., 2004), with specific manipulations of the independent variables through modulation of references to common physical properties and appetitive values.

Condition A: The metaphor described a person in physical rehabilitation after surgery, with emphasis on appetitive values (helping others who suffer). The narrative referenced the "cold" quality of the swamp and the effort of crossing it, creating common physical properties with the cold-pressor experience.

Condition B: Described the same physical rehabilitation narrative (common properties) but without specification of values or augmental functions.

Condition C: Metaphor about a challenging but non-physical experience, with explicit emphasis on appetitive values (important and meaningful things) without common physical properties.

Condition D: Neutral descriptive narrative of a landscape without metaphorical elements or value references.

Procedure

The procedure was approved by the Center for Psychological Research of the institution. All participants signed informed consent detailing the procedures and participation conditions. Experimental sessions were conducted individually and lasted approximately 30-40 minutes, distributed across four phases:

Phase 1 (Pre-experimental measures, ~15 min): Participants completed baseline questionnaires including the Acceptance and Action Questionnaire-II (AAQ-II), the Cognitive Fusion Questionnaire (CFQ), and the Generalized Pliance Questionnaire (GPQ).

Phase 2 (Pretest cold-pressor): Participants immersed their non-dominant hand to the wrist in a container with circulating water at 4.5-5.5 degrees Celsius (a higher temperature than the typical 1-3 degrees, according to Mitchell, MacDonald, & Brodie, 2004, to facilitate the use of trained strategies). They were instructed to keep their hand in the water as long as possible, with a maximum of 5 minutes. Tolerance time was recorded. Subsequently, participants reported pain intensity on a 10 cm visual analogue scale (VAS).

Phase 3 (Protocols and intervention): Participants were randomly assigned to one of the four experimental conditions and listened to the 4-minute protocol through headphones. To avoid influences from experimenter expectations, the experimenter did not know which protocol each participant was receiving. After listening to the protocol, participants completed a brief comprehension questionnaire with four questions about the story content.

Phase 4 (Posttest cold-pressor): Participants performed the cold-pressor task again under the same conditions as the pretest. Tolerance time and pain intensity report were recorded using the same procedures. Afterwards, participants were fully debriefed about the study objectives.

Measures

Primary dependent variable: Pain tolerance measured as the percentage of time tolerating pain at posttest relative to pretest. Scores were calculated as (posttest time / pretest time) × 100%. A score of 100% indicates no change; a score of 200% indicates that the participant tolerated pain twice as long at posttest as at pretest.

Secondary dependent variable: Differential pain perception, measured as the pre-post difference in reported pain intensity (posttest minus pretest score on the VAS scale).

Data Analysis

Sample size was determined a priori using G*Power 3.1.9.2 (Faul, Erdfelder, Lang, & Füchner, 2007), specifying a medium to large effect size (f² = .32), power of .80, and alpha error probability of .05. The analysis indicated that 75 participants were necessary to detect a medium to large effect through two-way ANOVA. One-way ANOVAs were conducted to explore equivalence of experimental conditions on pretest variables (AAQ-II, CFQ, GPQ, pretest pain tolerance, pretest pain intensity). Statistical analyses were performed using SPSS 19°. Due to the presence of outliers in the percentage of time tolerating pain at posttest, the top and bottom 10% of data in each experimental condition were trimmed (2 participants at the upper and lower limits per condition). Subsequently, two-way ANOVAs were calculated to analyze the differential effect of the protocols. Proportions of total variance attributed to an effect (eta-squared or η²) were reported as effect sizes typical of ANOVA. Eta-squared values of .01, .06, and .14 were interpreted as small, medium, and large effects, respectively. Between-condition effect sizes in Phase 4 were calculated using Cohen's d (Cohen, 1988), which can be interpreted as small (d = .20 to .49), medium (d = .50 to .79), and large (d = .80 or greater).

Results

The one-way ANOVA revealed that there were no significant differences between experimental conditions on AAQ-II, CFQ, GPQ, pretest pain tolerance, and pretest pain intensity (see Table 1), indicating that the experimental conditions were equivalent on pretest variables.

Pain tolerance (primary dependent variable):

The two-way ANOVA showed that both independent variables had statistically significant main effects on pain tolerance. Common physical properties produced a main effect: F(1,79) = 6.135, p = .016, η² = .093. Augmental functions produced a main effect: F(1,79) = 6.024, p = .017, η² = .091. No interaction was found between the two independent variables: F(1,79) = .008, p = .93, η² = .000.

The descriptive data presented in Table 2 and Figure 2 show the percentage of time tolerating pain at posttest relative to pretest for each condition:

• Condition A (common physical properties + augmental functions): M = 202.63, SD = 62.27 (N = 20)
• Condition B (common physical properties only): M = 172.19, SD = 41.54 (N = 20)
• Condition C (augmental functions only): M = 171.91, SD = 53.09 (N = 20)
• Condition D (control, neither): M = 139.21, SD = 46.52 (N = 20)

Cohen's d effect sizes between conditions, relative to Condition D, were:

• Condition A vs. D: d = 1.17 (very large effect)
• Condition B vs. D: d = .75 (medium effect)
• Condition C vs. D: d = .66 (medium effect)

Within-group comparisons revealed:

• Condition A vs. B: d = .59 (medium effect)
• Condition A vs. C: d = .53 (medium effect)
• Condition B vs. C: d = .01 (negligible effect)

These patterns indicate that the effects of common physical properties and augmental functions were additive (summative), with no interaction. The combination of both components (Condition A) produced the largest increases in pain tolerance, while either component alone (Conditions B and C) produced more modest but equivalent effects. Condition D (control) produced the smallest increase in pain tolerance.

Pain perception (secondary dependent variable):

The two-way ANOVA showed that the independent variables did NOT have statistically significant effects on pain perception. Common physical properties: F(1,79) = .876, p = .846, η² = .000 (not significant). Augmental functions: F(1,79) = 1.861, p = .178, η² = .03 (not significant). No interaction: F(1,79) = .038, p = .846, η² = .001.

The descriptive data for pre-post change in pain intensity (Table 3) were:

• Condition A (both present): M = -.94, SD = 2.41
• Condition B (physical properties only): M = -.27, SD = 1.85
• Condition C (augmental functions only): M = -.92, SD = 1.89
• Condition D (control): M = -.34, SD = 1.87

In all cases, scores were close to zero, indicating minimal changes in pain intensity perception between pretest and posttest, regardless of condition.

Discussion and conclusions

The current study analyzed the effect of two metaphor components from the RFT perspective in an analogue study in which participants were exposed twice (pretest and posttest) to a cold-pressor task. Specifically, the two manipulated variables were: (a) the presence of common physical properties between the metaphor and the participant's painful experience, and (b) the specification of appetitive augmental functions to tolerate pain.

The results of the experiment demonstrated a significant effect of both independent variables on pain tolerance. In other words, the inclusion of common physical properties between the metaphor content and the participant's painful experience and the specification of augmental functions to tolerate pain increased the metaphor effect. These variables showed no significant change in pain intensity. This finding suggests that the process of change of the protocols was not necessarily a decrease in pain perception but rather an alteration of discriminative functions for pain avoidance through incorporation of appetitive augmental functions to tolerate pain. The results are consistent with other studies (for example, Gil Luciano, Ruiz, Valdivia-Salas, & Suárez Falcón, in press; Gutiérrez, Luciano, Rodríguez, & Fink, 2004) where ACT-based protocols were found not to reduce pain but to increase pain tolerance.

An interesting finding of this study was that common physical properties had an effect of increasing pain tolerance even when the metaphor did not specify appetitive augmental functions (directed values). It appears that common physical properties might have generated identification with the metaphor and might have provided value to the task, as if the experimental context positioned some general positive meanings that might be functioning as positive abstract reinforcers, such as helping people who suffer from persistent pain or overcoming oneself.

The fact that specified augmental functions (values) also worked when the metaphor did not include common physical properties (Condition C) suggests that providing value-based reasons to tolerate pain can be sufficient to increase tolerance, independent of whether the metaphor shares physical properties with the experience. The additive (non-interactive) effects of both variables suggest that they represent distinct mechanisms of change that operate independently.

Study limitations include: first, the effect of the protocols was tested only in an experimental task and not in real clinical situations where individuals need to tolerate intense pain. Second, only undergraduate students participated, which limits generalization of results. Third, the sample size was relatively small to conduct moderation analyses that would explore whether certain factors (such as levels of psychological flexibility or analogical reasoning abilities) might moderate the observed effects. Fourth, future research could explore whether augmental functions work better in participants with high psychological flexibility, and whether common physical properties show greater effects in participants with low to medium levels of analogical reasoning abilities.

In conclusion, based on the results of this study, ACT therapists working with people in situations of painful rehabilitation after surgery should design metaphors that include common physical properties with the client's painful experience AND specify appetitive augmental functions (valued directions) to maximize the metaphor effect in promoting psychological flexibility and pain tolerance.

Significance and Contribution

This study is important because it provides experimental analysis of the specific mechanisms through which metaphors can promote psychological flexibility from the RFT perspective. The findings directly inform ACT therapists about which components to include when designing clinical metaphors. Although the study was limited to a laboratory experimental context, it provides the empirical foundation for understanding how two key characteristics of metaphors (common physical properties and appetitive values) contribute independently and additively to increased pain tolerance, offering valuable contributions to both basic RFT process research and clinical application of ACT-based interventions.