Printer Friendly

The On-Going Search for Perspective-Taking IRAPs: Exploring the Potential of the Natural Language-IRAP.

Perspective-taking appears to be a key process in the development of the self (Bames-Holmes, Bames-Holmes, Roche, & Smeets, 2001). As well as the mainstream approach to perspective-taking, largely represented by Theory of Mind (ToM; Sodian & Kristen, 2010), behavioral researchers working under the rubric of Relational Frame Theory (RFT) have approached perspective-taking as involving responding in accordance with three deictic relations: the interpersonal (I/you); the spatial (here/there); and the temporal (now/then; see Hayes, Bames-Holmes, & Roche, 2001). The basic idea is that individuals must be able to locate themselves relationally, in space and time, relative to others, in order to learn how to take the perspective of another individual. Imagine a young child who is asked about the color of his brother's jumper and the child replies with the color of his own jumper rather than his brother's. In this case, the child has failed to discriminate self from other, and it is thus difficult to see how this child could take the separate perspective of someone else (for a review of published RFT work on deictic relational responding, see Montoya-Rodriguez, Molina, & McHugh, 2017).

Most RFT studies on deictic relations, or perspective-taking that appears to involve these relations, have employed the Barnes-Holmes (2001) protocol, originally developed to assess and establish these relations in young children. The findings from this body of work may be summarized as follows: (1) there appear to be functional distinctions among the interpersonal, spatial, and temporal relations (e.g., Rendon, Soler, & Cortes, 2012); (2) there appear to be functional distinctions among simple deictic relations (e.g., I versus you), reversed relations (e.g., if I were you and you were me), and double reversed relations (e.g., if I were you and you were me, and if here were there and there were here, see McHugh, Barnes-Holmes, & Barnes-Holmes, 2004); (3) deictic relational responding can be established through training and appears to generalize to other examples of natural language (e.g., Heagle & Rehfeldt, 2006; Weil, Hayes, & Capurro, 2011); (4) interpersonal and simple relations appear to emerge first (e.g., McHugh et al., 2004); and (5) competence in deictic relational responding correlates with cognitive abilities, IQ (Gore, Barnes-Holmes, & Murphy, 2010), and ToM tests (Weil et al., 2011).

Several later studies employed the Barnes-Holmes (2001) protocol for more clinical rather than developmental purposes. For example, Villatte, Monestes, McHugh, Baque, and Loas (2008) reported somewhat greater competence on the protocol for individuals with high versus low social anhedonia. Likewise, Villatte, Monestes, McHugh, Freixa i Baque, and Loas (2010) reported some superiority for a control group versus individuals with a diagnosis of schizophrenia. However, some researchers have raised concerns about using the original protocol to draw conclusions about clinical phenomena (Hussey et al., 2014), which include the protocol's developmental focus; the fact that even typically developing adults show deficits on specific deictic relations (Vitale, Barnes-Holmes, Barnes-Holmes, & Campbell, 2008); and the fact that RFT does not necessarily predict that psychological suffering involves deficits in deictic relational responding.

Researchers have therefore begun to explore other methodologies, such as the Implicit Relational Assessment Procedure (IRAP), to study deictic relational responding (see Golijani-Moghaddam, Hart, & Dawson, 2013; Vahey, Nicholson, & Bames-Holmes, 2015, for recent reviews of the reliability and validity of the IRAP). Two published studies have thus far used the IRAP to investigate deictic relations, particularly in terms of comparing responding to self versus responding to others. Barbero-Rubio, Lopez- Lopez, Luciano, and Eisenbeck (2016) presented participants with what they referred to as a perspective-taking IRAP that contained each participant's own name (self) versus the name of the researcher (other) as label stimuli, and statements describing specific characteristics of the self (e.g., "is in front of the laptop") versus other (e.g., "is standing up") as targets, along with "Yes" and "No" as response options. In order to manipulate perspective-taking, explicit rules were provided prior to each block of trials. In particular, in order to encourage responding from one's own perspective, participants in some blocks were instructed: "For the next block of trials, you have to respond as if you were you and Adrian [researcher] were Adrian." In contrast, in order to encourage responding from the perspective of another, participants in other blocks were instructed: "For the next block of trials, you have to respond as if you were Adrian and Adrian were you." The four trial types were referred to as: 1-1 (participant name-participant characteristics); Other-Other (researcher name-researcher characteristics); I-Other (participant name-researcher characteristics); and Other-I (researcher name-participant characteristics). In general, the latency data reported by Barbero-Rubio et al. (2016) showed significantly more rapid responding on the II trial type, relative to the other three trial-types during self-perspective blocks. The differences between self- and other-perspective blocks for each trial type were also in the predicted direction (i.e., shorter on self-perspective), and these differences were significant in terms of the normalized [D.sub.IRAP]-scores. Overall, the researchers concluded that these significant [D.sub.IRAP].scores indicated that the participants had little flexibility in changing from their own perspective to another perspective.

In a systematic replication of the Barbero-Rubio et al. (2016) study, Kavanagh, Barnes-Holmes, Barnes-Holmes, McEnteggart, and Finn (2018) used a similar IRAP, but the study also included a control IRAP that did not require responding to self versus other. That is, instead of comparing self with other, the control IRAP compared responding between two separate others (i.e., the researcher and a picture of another unknown participant). In Experiment 1, the data from the IRAP showed significantly stronger responding on the 1-1 trial type versus Other-Other, but there was no difference in the control IRAP between Researcher-Researcher and Other-Other. Although a range of methodological differences between the two studies preclude systematic comparisons, both studies did show evidence of differences in responding to self versus other, but no difference in responding to two others in the context of the control IRAP.

One possible concern that could be raised regarding the two studies involving the self- versus other-IRAPs described above is that differences that emerged between responding to self and other within the IRAP could be attributed to factors other than perspective-taking per se. For example, in the study by Kavanagh et al. (2018) a pattern known as the single trial type dominance effect (STTDE) emerged in Experiment 1. That is, the size of the [D.sub.IRAP].score for the 1-1 trial type was significantly larger than for the Other-Other trial type. Although this dominance effect could indicate a history of responding from one's own perspective more frequently than from another perspective, it does not necessarily indicate differences in the relative ability to take the perspective of self versus another (see Kavanagh et al., 2018, for a detailed discussion). To appreciate the argument we are making, other recent research has also reported a STTDE when shapes and colors were presented as categories within the IRAP (Finn, Barnes-Holmes, & McEnteggart, 2018). In particular, larger [D.sub.IRAP].scores were shown on color-color than on shape-shape trial types and this effect appeared to be driven by a higher frequency of the use of color-related words in natural language over the use of shape-related words. It is obvious that no perspective-taking was required when participants were simply asked to categorize colors as colors and shapes as shapes, and thus the same logic could be applied to a single IRAP that requires responding to self versus other (i.e., the effect could be the result of responding to self more frequently than to other in natural language, rather than an ability to perspective-take).

A second potential concern that could be raised regarding the two studies involving the self- versus other-IRAP pertains to the simple target phrases that specified characteristics of self and other (e.g., "is sitting down," "is the participant," "is in front of the computer"). As such, it could be argued that responding on the IRAP simply required deictic relational responding but not perspective-taking. Indeed, perspective-taking would appear to require more complex target statements or relational networks that involve taking the perspective of self versus other. For example, such statements could take the form of "When event X happens, self or other thinks or feels Y." In principle, this sort of complex relational network requires that the participant responds to statements that coordinate with how the self responds to particular events versus how they perceive others will respond to the same events. Such an IRAP, at least in terms of face validity, appears to target perspective-taking, as we generally understand it. Developing such an IRAP would necessarily involve inserting relatively complex statements or networks into the procedure. In doing so, the separation of the stimuli within an IRAP into labels and targets may be problematic because participants may simplify the task by responding to single words or subclauses within the labels and targets in such a way that fails to capture perspective-taking. One way of potentially avoiding this problem would be to employ a natural language format previously reported by Kavanagh, Hussey, McEnteggart, Bames-Holmes, and Bames-Holmes (2016). In that study, the IRAP combined the label and target stimuli to form statements that are more similar to natural language.

Based on this reasoning, the current series of experiments employed two IRAPs to study deictic relations, but using a natural language format, with one IRAP targeting self-perspective and the other targeting other-perspective. The general strategy pursued here was to determine the extent to which the two Natural Language-IRAPs (NL-IRAPs) in each experiment would yield any differences in the direction and/or strength of the individual trial type effects. In addition, correlational analyses were conducted to determine the extent to which the two NL-ERAPs correlated, or failed to correlate, with each other and with a range of self-report measures that were deemed to be relevant to perspective-taking.

Given the exploratory nature of the research, the self-report measures we selected from experiment to experiment were not based on a well-established theoretical rationale. Rather, we simply used measures that had frequently been used in previous studies of perspective-taking within the field of contextual behavioral science. In the later experiments, measures of attachment were included based on comments provided by the first author's doctoral advisory committee, who advised that these measures could be relevant when attempting to analyze perspective-taking with regard to others with whom a participant might have a significant relationship.

Experiments 1 and 2 focused on self versus other when the other was unspecified, whereas Experiments 3-6 focused on self versus other across differences in potentially key aspects of the other, such as similarity with, and difference from, self. In developing the two IRAPs employed in each experiment (i.e., self- versus other-focused), we sought to manipulate only the "self versus other" variable. Thus, any difference that emerged between the two IRAPs could be readily interpreted as based on the self versus other foci. Because Experiment 1 was exploratory, we did not make any formal predictions.

Experiment 1

Method

Participants Fifty-five participants were recruited for Experiment 1: 44 females and 11 males. Participants ranged from 17 to 38 years old (M = 20.89) and were recruited through random convenience sampling from the Ghent University participant pool. Each participant was paid an hourly rate of 10 [euro]. The general strategy for recruiting numbers of participants was guided by the results of a recent metaanalysis of IRAP effects in the clinical domain, indicating that a minimum of 29 is required to achieve a power of 0.8 for firstorder correlations (Vahey et al, 2015). Because participants sometimes failed to reach various performance criteria for the IRAP (details provided subsequently), it was necessary to recruit more than 29 participants in each experiment to yield an adequate dataset for analyses.

Materials and apparatus Experiment 1 comprised two computer-based tasks: an other-focused IRAP and a self-focused IRAP. The experiment also included the Community Assessment of Psychic Experiences (CAPE) and the Acceptance and Action Questionnaire-II (AAQ-II). All materials were presented in Dutch, but are presented in English in this article. The CAPE was employed because perspective-taking has been implicated in psychotic-like experiences (e.g., Savla, Vella, Armstrong, Penn, & Twamley, 2013). The use of the AAQ-II was more exploratory in that we aimed simply to determine if perspective-taking measured with self and other IRAPs correlated with psychological flexibility.

Other-focused IRAP The other-focused IRAP required participants to respond to various statements about other people. These statements were presented in a natural language format, rather than as label and target stimuli, to form a regular sentence or statement. The IRAP presented 16 statements, each of which comprised two parts. The statement referred to an event (deemed positive or negative) and a positive or negative reaction to that event. The sequence in which the event and reaction were presented in each statement could vary, in that the event could be presented before the reaction, or the reaction could be presented before the event. Nevertheless, the 16 statements were divided into four trial types (see Figure 1), based on whether the event and the reaction were both positive, both negative, or a combination of positive and negative (see Table 1). For illustrative purposes, consider the four following statements "People will be proud if they succeed in their exams" (Positive Event-Positive Reaction); "Others are frustrated when they pass an exam" (Positive Event-Negative Reaction); "Others feel lucky if they fail an exam" (Negative Event-Positive Reaction); and "If others fail an exam, then they are disappointed" (Negative Event-Negative Reaction). The response options "Yes" and "No" were presented at the bottom left- and right-hand comers on each trial. The reader should note that the NL-ERAP presented some statements that could be seen as potentially confusing, in terms of how participants should respond, particularly the Positive-Negative and Negative-Positive statements. However, the general structure of the NL-IRAP remained the same as the standard IRAP in that two of the four trial types could be seen as less coherent than the other two.

Self-focused IRAP The self-focused IRAP was similar to the other-focused IRAP, but required participants to respond to various statements about themselves, rather than about others (e.g., "I'm proud when I succeed in my exams"). These 16 statements were also presented in a natural language format and comprised two parts, an event and a reaction (see Table 2). Again, "Yes" and "No" were the response options. The self-focused IRAP comprised the same four trial types as above (see Figure 2).

The Community Assessment of Psychic Experiences (CAPE; Stefanis et al., 2002) The CAPE measures psychotic-like experiences in the general population. The scale consists of 42 symptom items rated along three subscales: positive symptoms (20 items, e.g., "Do you ever feel as if there is a conspiracy against you?"), negative symptoms (14 items, e.g., "Do you ever feel that you experience few or no emotions at important events?") or depressive symptoms (8 items, e.g., "Do you ever feel sad?"). Each item is rated on two 4-point Likert scales from 0 (never) to 3 (nearly always) to indicate (1) the frequency of symptoms and (2) the level of distress associated with each symptom. The CAPE provides overall frequency and distress scores of psychic experiences, and total frequency and distress scores for each of the three subscales. In order to account for partial nonresponses, all scores are weighted for the number of valid answers per subscale (i.e., sum score divided by number of items completed). Overall frequency and distress scores are also weighted. In all cases, higher scores indicate greater frequency or distress regarding symptoms, although there are no clinical cut-offs for this measure.

The Dutch version was completed by participants. The scale has demonstrated adequate reliability: positive dimension alpha = 0.63, negative dimension alpha = 0.64, and depressive dimension alpha = 0.62 (Konings, Hanssen, Van Os, & Krabbendam, 2006).

Acceptance and Action Questionnaire-II (AAQ-II 7-item version; Bernaerts, De Groot, & Kleen, 2012) The AAQ-II measures acceptance, experiential avoidance, and psychological inflexibility. The scale consists of seven items rated on a 7point Likert scale from 0 (never true) to 7 (always true). The AAQ-II yields an overall score with a maximum of 49 indicating low psychological flexibility and a minimum of 7 indicating high psychological flexibility. Although the measure was not designed as a diagnostic tool. Bond et al. (2011) reported that scores [greater than or equal to] 24 correlate with psychological distress. The English version of this scale has been shown to have good internal consistency (alpha = .84; Bond et al., 2011). The Dutch translation, used here, has yielded similar reliability values (alpha = .85; Bernaerts et al., 2012).

Procedure Experiment 1 took place on an individual basis in sound-proof cubicles at the Department of Experimental, Clinical, and Health Psychology of Ghent University. Informed consent was obtained from all participants. Each participant was exposed to the other-focused and selffocused IRAPs, with the order of each counterbalanced across participants. Participants thereafter completed the CAPE, followed by the AAQ-II, always presented in this order.

Other-focused IRAP The other-focused IRAP comprised a maximum of eight pairs of practice blocks, followed by three pairs of test blocks. On each trial, an other-related statement was presented in the middle of the screen (e.g., "People will be proud if they succeed in their exams"), with two response options ("Yes" and "No") at the bottom left and right of the screen. Participants were simply instructed to figure out, based on individual trial feedback, what the task involved. Participants responded on each trial using either the "d" key for the response option on the left or the "k" key for the response option on the right. The locations of the response options "Yes" and "No" alternated from trial to trial in a quasi-random order, such that they did not remain in the same left-right locations for more than three successive trials. The instruction "The previously correct and incorrect answers have been reversed" was presented between blocks of trials.

When participants selected the response option that was deemed correct within that block, an intertrial interval of 400 ms was presented, after which the next trial occurred. When participants selected the response option that was deemed incorrect for that block, the stimuli remained on the screen and a red "X" appeared beneath the statement. Only when the correct response option was selected did the program proceed to the 400 ms intertrial interval (followed by the next trial). This pattern of trial presentations, with corrective feedback, continued until the entire block of 32 trials was presented. Trials were presented in a quasi-random order within each block, with the constraint that each statement was presented twice across the 32 trials. Consistent trial blocks required responding that was deemed to be in accordance with positive events producing positive reactions and negative events producing negative reactions (i.e., Positive Event-Positive Reaction/Yes, Positive Event-Negative Reaction/No, Negative Event-Positive Reaction/No, and Negative Event-Negative Event/Yes). Inconsistent trial blocks required responding that was in accordance with positive events producing negative reactions and negative events producing positive reactions (i.e., Positive Event-Positive Reaction/No, Positive Event-Negative Reaction/Yes, Negative Event-Positive Reaction/Yes, and Negative Event-Negative Reaction/No). The other-focused IRAP always commenced with a consistent block of trials.

When participants completed each block of trials, the IRAP program provided them with feedback on their performance during that block. The feedback consisted of a message informing them how accurately and how quickly they had responded overall during that block. The average speed of responding was calculated from stimulus onset to the first correct response across all 32 trials within the block. Participants were required to achieve a maximum median latency of no more than 5000 ms on each trial type. Although we recognize that 5000 ms is unusually long for the latency criterion in an IRAP (most range between 2000 ms and 3000 ms), our initial pilot work indicated that most participants failed to reach the latency criterion when set at <5000 ms, especially when the criterion was set at the trial type level (the IRAP latency criterion is typically set at the block level). It is also important to remember that the IRAP effect is calculated from the difference in mean latencies between blocks of trials, rather than from the absolute length of the latencies. As such, the IRAP effect, even when the latency criterion is set relatively high at 5000ms, consists of the difference in average latencies between the blocks, which remains relatively short (i.e., no more than a few hundred milliseconds).

Participants were also required to achieve a minimum accuracy of no less than 75%, also set at the trial type level (i.e., no more than two errors were permitted per trial type). If participants achieved both accuracy and latency criteria on any pair of practice blocks, they proceeded to the first pair of test blocks; if they failed on the eighth pair of practice blocks, participation in the experiment was terminated. Although setting the accuracy criterion at 75% is lower than many other studies that have used 80% or more, it is important to note that the accuracy criterion was set at the trial type, rather than the block, level. Setting the accuracy criterion at the trial type level requires a high level of accuracy across all trial types and thus, in a sense, is more stringent than 80% at the block level.

A fixed set of six test blocks was presented with no accuracy or latency criteria required for participants to progress from one block to the next. However, percentage correct and median latency were again presented at the end of each block to encourage participants to maintain criterion-level responding from the practice blocks.

Self-focused IRAP The format of the self-focused IRAP was identical to the other-focused IRAP, but with statements regarding the self, rather than statements regarding others (e.g., "It irritates me if someone I hate wins the lottery"). It was particularly important in this IRAP to ensure that participants were responding to each of the statements from their own perspective. Hence, participants were instructed at the beginning of the IRAP, as follows: "The program will present statements on the screen which refer to you. Please remember that when you see 'I' or 'me' on-screen, this refers to you [the participant]." The designation of consistent and inconsistent blocks was identical to the previous IRAP. Again, all participants were first presented with a consistent block of trials.

Questionnaires Participants completed the two questionnaires in the following sequence: the CAPE followed by the AAQ-II.

Results and Discussion

Questionnaire data A summary of the means and standard deviations for the AAQ-II and the CAPE weighted overall and with subscale scores is provided in Appendix A. The mean AAQ-II score fell below 24, indicating little or no psychological distress. All weighted overall and subscale scores of the CAPE were relatively low.

IRAP data Consistent with standard IRAP practice, mean response latencies for consistent and inconsistent blocks were initially divided according to trial type and calculated for each participant (see Barnes-Holmes, Barnes-Holmes, Stewart, & Boles, 2010). Based on the latency and accuracy criteria, three participants failed to complete the practice blocks (and did not proceed to the test blocks) on the self-focused IRAP, and four failed to complete the practice blocks on the other-focused IRAP. Hence, all seven datasets were excluded from further analyses. For the remaining participants, the same accuracy and latency criteria were applied in the test blocks, except that the criteria now applied across all six test blocks. This meant that no more than eight errors were permitted per trial type across the six test blocks. Using these criteria, five participants failed to complete the self-focused IRAP and five failed the other-focused IRAP. All 10 datasets were excluded from further analyses, leaving the final number of datasets in the analyses at N = 37.

[D.sub.IRAP]-scores Consistent with the majority of published IRAP studies, [D.sub.IRAP]-scores for both IRAPs were calculated for each of the four trial types (see Bames-Holmes et al., 2010), such that positive [D.sub.IRAP]-scores during consistent blocks indicated responding "Yes" more quickly than "No" on Positive Event-Positive Reaction and Negative Event-Negative Reaction trial types, and responding "No" more quickly than "Yes" on Positive Event-Negative Reaction and Negative Event-Positive Reaction trial types. Negative [D.sub.IRAP]-scores indicated the opposite pattern: "No" more quickly than "Yes" on Positive Event-Positive Reaction and Negative Event-Negative Reaction trial types, and responding "Yes" more quickly than "No" on Positive Event-Negative Reaction and Negative Event-Positive Reaction trial types.

The mean [D.sub.IRAP]-scores and standard errors for each trial type for both IRAPs are presented in Figure 3. Positive scores were recorded for all four trial types, with the weakest observed for Negative Event-Positive Reaction. For each of the trial types, the mean [D.sub.IRAP]-scores were greater for the other-focused IRAP, with the largest difference for the Negative Event-Negative Reaction trial type. A 2x4 repeated measures Analysis of Variance (ANOVA) yielded a significant main effect for IRAP type [F(1, 36) = 4.12, p < .05, [[eta].sup.2.sub.[rho]] =.1] and for trial type [F(l, 36) = 10.14, p < .01, [[eta].sup.2.sub.[rho]] =.22] but no significant interaction (p > .3). Post-hoc (Fisher's PLSD) comparisons among the four trial types (see Table 3) indicated that Negative Event-Positive Reaction differed significantly from the other three trial types and Positive Event-Positive Reaction differed from Negative Event-Negative Reaction. In examining Figure 3, it appears that the significant difference between the Positive Event-Positive Reaction and Negative Event-Negative Reaction trial types was driven largely by the other-focused IRAP. On balance, the interaction between IRAP type and trial type was nonsignificant, and thus it would be unwise to draw any strong conclusions at this point. Indeed, the next study was a direct replication conducted primarily to determine if the difference between the IRAPs was robust and if the relatively strong [D.sub.IRAP] effect for the Negative Event-Negative Reaction trial type in the other-focused IRAP would be observed again.

Correlations Given that there was a significant main effect for IRAP type, but no interaction with trial type, a single overall D-score (the mean of the four trial types) was calculated for each IRAP, and then subjected to correlational analyses with the questionnaires (a total of nine correlations for each IRAP). Neither set of correlations proved to be significant (all rs < .32 and all ps > .05). In order to determine if differences in responding to self and others across the two IRAPs correlated with psychological symptoms, we calculated an IRAP difference score by subtracting the other-focused overall D-score from the self-focused overall D-score for each participant, and then subjected this to correlational analyses with the questionnaires (a total of nine correlations, not shown). The frequency of positive psychotic-like symptoms was the only significant correlation [r(37) = 0.34, p = 0.039], but did not remain so after a Bonferroni correction. (1) Finally, the overall D-scores from the two IRAPs were correlated and this single correlation proved to be nonsignificant (r = 0.208 and p = 0.219).

Summary The results from Experiment 1 indicated that the other-focused IRAP produced bias scores that were significantly stronger than the self-focused IRAP. There was no significant interaction between IRAP type and trial type, although visual inspection of Figure 3 indicated a large difference between IRAPs for the Negative Event-Negative Reaction trial type. Only 1 of the 27 correlations proved to be significant, but did not remain so after a Bonferroni correction. Overall therefore, the data indicated that there was a significant difference between the self- and other-focused IRAPs, that the IRAP difference score correlated with psychotic-like symptoms, and that the two IRAPs failed to correlate with each other. This pattern of results suggests that the self- and other-focused IRAPs were tapping into different behavioral repertoires, and in this sense this first experiment could be seen as a success. On balance, given the novelty of the procedure employed in Experiment 1 (i.e., the NL-IRAP), it was deemed important to conduct a direct replication of the study, with additional questionnaires. Because Experiment 2 remained exploratory, we did not make any formal predictions.

Experiment 2

Experiment 2 was a direct replication of Experiment 1, except that the AAQ-II was omitted and replaced with several measures of emotional attachment, relationships, and self-warmth. A measure that was in development at that time, the Psychological Flexibility Index (PFI), was also included in the battery of tests. All questionnaires were completed after the participants had finished the IRAPs.

Method

Participants Fifty-one individuals were recruited for Experiment 2: 37 females, 14 males. Ages ranged from 18 to 49 years (M= 24.5). All participants were recruited through random convenience sampling from the Ghent University participant pool and were paid an hourly rate of 10 [euro].

Materials and apparatus Both IRAPs were identical to Experiment 1. The CAPE was included again. Five additional questionnaires assessed: psychological flexibility (using the Psychological Flexibility Index, PFI); self-warmth (using a Self-Warmth Thermometer); emotional attachments (using the Experiences in Close Relationships-Relationship Structures questionnaire, ECR-RS); and relationships with others (Inclusion of Other in the Self [IOS]; Experiencing of Self Questionnaire [ESQ]). The PFI replaced the AAQ-II as a measure of psychological flexibility because it was being developed by Bond et al. (2011) as an alternative to the AAQ-II. The Self-Warmth Thermometer was included to determine whether performance in the self-IRAP correlated with self-warmth (Vahey, Barnes-Holmes, Barnes-Holmes, & Stewart, 2009). The various attachment questionnaires were included because preexisting difficulties in attachment relationships may manifest in difficulties in perspective-taking with regard to others (Bernstein, Laurent, Nelson, & Laurent, 2015). All materials were presented in Dutch (translated into English when referred to in the text). The CAPE was the only questionnaire with a validated Dutch version. The instructions and items of the remaining measures were created using a backward forward translation procedure (World Health Organization, 2017). There are no clinical cut-offs for any of the measures.

Psychological Flexibility Index (PFI) The PFI is a measure of psychological flexibility currently being developed by Bond et al. (2011). In its current format, the measure includes 82 items. Each item is rated on a 6-point Likert scale from 1 (idisagree strongly) to 6 (agree strongly), with a minimum of 82 and a maximum of 492, generated by reversing relevant items and then summing the scores. Higher scores indicate higher levels of psychological flexibility, with lower scores indicating lower flexibility. At present, there are no reliability data on this measure.

Experiences in Close Relationships-Relationship Structures questionnaire (ECR-RS; Fraley, Heffernan, Vicary, & Brumbaugh, 2011) The ECR-RS assesses attachment patterns in four close relationships (mother, father, romantic partner, and best friend). Each of the four relationships is rated as a separate domain along two subscales: (1) anxious attachment and (2) avoidant attachment. The anxious attachment subscale comprises three items (e.g., "I'm afraid that this person may abandon me") with a maximum possible score of 21 and a minimum of 3. The avoidant attachment subscale comprises six items (e.g., "I don't feel comfortable opening up to this person"), with a maximum possible score of 42 and a minimum of 6. Each item is rated on a 7-point Likert scale from 1 (strongly disagree) to 7 (strongly agree). Higher scores indicate higher levels of avoidant attachment and anxious attachment. According to Fraley et al. (2011), the alpha reliabilities for the four relationship domains in the avoidant subscale are between .81 and .92, with the anxiety subscale between .83 and .87. Test-retest reliability is available for only two domains on each subscale, but is adequate (alpha =.65 for romantic relationships and .80 for parental relationships).

Inclusion of Other in the Self (IOS; Aron, Aron, & Smollan, 1992) The IOS is a measure of closeness in relationships, comprising two sets of seven Venn diagrams. All Venn diagrams contain one circle that represents the self, whereas the other circle represents a "best friend" or "other people generally." As such, each set of Venn diagrams represents the relationship between self and a significant other (best friend) or between self and a nonsignificant other (other people generally). Seven Venn diagrams were presented in each set, with each Venn diagram differing systematically in terms of the extent of the overlap between the two circles. In particular, in the first Venn diagram, the two circles are completely separate, whereas in the seventh Venn diagram, the two circles are almost fully overlapping, with each Venn diagram in-between showing some variation from one extreme to the other. In order to yield one overall score for the relationship between self and best friend, and one overall score for the relationship between self and other people generally, each Venn diagram is allocated a number between 1 and 7, where 1 represented the least overlap, and 7 represented the most. Hence, the maximum score for best friend/other people generally was 7, with the minimum score 1. The IOS has demonstrated adequate reliability (alpha = .93; see Aron et al., 1992).

Experiencing of Self Scale (EOSS; Kanter, Parker, & Kohlenberg, 2001) The EOSS measures the control of others over the experience of the self. It consists of 20 items rated along four subscales (each with five items): casual acquaintances-absent (e.g., "My feelings are influenced by casual acquaintances when I am alone"); casual acquaintances-present (e.g., "My wants are influenced by casual acquaintances when I am with them"); close relationships-absent (e.g., "My attitudes are influenced by close relationships when I am alone"); and close relationships-present (e.g., "My actions are influenced by close relationships when I am with them"). Each item is rated on a 7-point Likert scale from 1 (never true) to 7 (always true). The maximum overall score is 140 and the minimum is 20, with high scores indicating greater control of others over the experience of self. According to Kanter et al. (2001), the scale overall has high internal consistency (alpha = 0.91), with internal consistency in the subscales ranging from alpha .83 to .93.

Self-Warmth Thermometer A feeling-thermometer adapted from Vahey et al. (2009) was used as a measure of subjective self-warmth. The current measure composed an illustrated thermometer with a continuous horizontal scale from 0 (cold), rising in intervals of 10, to 100 (warm). Participants were required to indicate their self-warmth from 0 to 100. Given that this is not a standardized measure, there are no reliability data.

Procedure The two IRAPs were identical to Experiment 1. Participants thereafter completed the CAPE, PFI, IOS, ECRRS, EOSS, and the Self-Warmth Thermometer, always presented in this order. All questionnaires were presented to participants via computer using the program Psychopy (Pierce, 2007).

Results and Discussion

Questionnaire data A summary of the means and standard deviations of each questionnaire and relevant subscales is provided in Appendix B. The mean score on the Self-Warmth Thermometer was around the midway point at 56.65 (/100) and as such was consistent with previous samples (see Vahey et al., 2009). The overall PFI scores were relatively high, indicating high psychological flexibility. The overall CAPE and subscale scores were relatively low, indicating low psychotic-like symptoms. The ECR-RS scores were also relatively low in terms of both attachment-anxiety and attachment-avoidance. The overall EOSS and subscale scores were low, indicating low control by others over the experience of self. The IOS scores for best friend were higher than for other people, suggesting a closer relationship in this regard. Nothing unusual or unexpected, therefore, emerged from the questionnaires, given the use of a nonclinical sample.

IRAP data One participant failed the practice blocks on the self-focused IRAP and another participant failed to complete self-reports, and thus their IRAP data was excluded from further analysis. The final number of participants included in the analyses was N = 49.

[D.sub.IRAP-scores The mean [D.sub.IRAP]-scores and standard errors for each trial type for both IRAPs are presented in Figure 4. Positive scores were recorded for all four trial types, with the weakest observed for Negative Event-Positive Reaction. For three of the four trial types, the mean [D.sub.IRAP]-scores were greater on the other-focused IRAP, with the largest difference between IRAPs for Positive Event-Negative Reaction. A 2x4 mixed repeated measures ANOVA produced a main effect for IRAP type [F(1, 48) = 4.61, p = .04, [[eta].sup.2.sub.[rho]] =.09], and for trial type [F(1,48) = 4.42, p < .01, [[eta].sup.2.sub.[rho]] =.09] but no significant interaction (p > .1). Post-hoc (Fisher's PLSD) comparisons among the four trial types, see Table 4, indicated that Negative Event-Positive Reaction differed significantly, or marginally so, from the other three trial types.

Correlations The overall D-score from each IRAP, as well as the difference scores between the two IRAPs, were subjected to correlational analyses with the questionnaires (i.e., 25 correlations for each IRAP and 25 for the IRAP difference score, not shown). The correlation matrix showed three significant correlations. The other-focused IRAP significantly correlated with the frequency of positive psychotic-like symptoms (r(49) = -0.3, p < 0.04], but did not do so after a Bonferroni correction. The other-focused IRAP also significantly correlated with the ECR-RS attachment-avoidance subscale for romantic partner [r(49) = 0.34, p < 0.02] and for mother [r(49) = 0.31, p < 0.03], but neither remained significant after Bonferroni corrections. There were no significant correlations with the IRAP difference score (all rs <.28 and all ps > .05). Finally, the overall D-scores from the two IRAPs were correlated and this single correlation proved to be significant (r =0.315, p = 0.027).

Summary The results from Experiment 2 again indicated that the other-focused IRAP produced bias scores that were significantly stronger than the self-focused IRAP. Again, there was no significant interaction between IRAP type and trial type. It is interesting that the relatively large difference between the IRAPs for the Negative Event-Negative Reaction trial type that was observed in Experiment 1, was completely absent in Experiment 2, despite the fact that Experiment 2 was a direct replication. The reason for the failure to replicate the effect for this trial type remains unclear at this time. The reader should note that we also failed to replicate this effect across the remaining four studies, and thus it is reasonable to conclude that the effect observed in Experiment 1 was likely due to genuine error variance in the sample. Only 3 of the 75 correlations proved to be significant, but did not remain so after Bonferroni corrections, and thus should be interpreted with extreme caution. At this point, therefore, the main effect for IRAP type had been replicated, in that the other-focused IRAP produced larger positive [D.sub.IRAP] effects relative to the self-focused IRAP. Differences between the two IRAPs for individual trial types did not appear to be particularly important, in that both experiments failed to yield even a trend towards a significant interaction, and the relatively large difference in the IRAPs observed in Experiment 1 for the Negative Event-Negative Reaction trial type was completely absent in Experiment 2. The lack of significant correlations between the IRAPs and the questionnaires in both studies suggests that the measures were not targeting behaviors that overlapped to any great degree. On balance however, the overall D-scores from the two IRAPs did correlate in Experiment 2 (but not in Experiment 1), suggesting some overlap in the two measures.

At this point in the research program, we thought it wise to begin to address the fact that the other-focused IRAP produced stronger [D.sub.IRAP]-scores than the self-focused IRAP. One simple explanation for this difference might be that questions about how other people in general react may tend to produce more stereotypical responding. That is, when the other is unspecified, participants tend to produce responses that reflect some general view of other people, rather than of a particular individual. In contrast, when responding to self, a range of potentially important contextual variables may be involved. Thus, for example, when asked how you as an individual react to positive events, there may be a tendency to confirm that in general you react positively but that you are not the type of person to get overexcited, relative to unspecified others. Likewise, an individual may confirm that they react negatively to negative events, but not to the same degree, in general, as other people do. Insofar as this may be the case, it is possible that an other-focused IRAP that specified a particular other would produce a different result. The next experiment, therefore, was a systematic replication of Experiments 1 and 2, but other was now specified and identified as someone who the participant believed was "the person they are closest to." Identifying this person involved the participant completing a bespoke specified other-focusing task, which will be described below. Conducting a third experiment also allowed us to determine if we would replicate the correlation between the two IRAPs found in Experiment 2, but not in Experiment 1. Finding such a correlation would undermine, to some extent, the argument that the two IRAPs were targeting fundamentally different behavioral repertoires. The research remained relatively exploratory and thus we did not make any formal predictions.

Experiment 3

Method

Participants Forty participants were recruited for Experiment 3, 29 females, 10 males and 1 participant who identified as neither male nor female. Ages ranged from 18 to 45 years (M = 22.67). All participants were recruited through random convenience sampling from the Ghent University participant pool and were paid an hourly rate of 10 [euro].

Materials and apparatus The self-focused IRAP was identical to that used in Experiments 1 and 2. The other-focused IRAP was similar to the previous experiments, except that the phrase "other people" was replaced with the name of a significant other identified by each participant (e.g., "Sarah"). All questionnaires were identical to those used in Experiment 2.

Specified other-focusing task This task was designed to help ensure that all participants, in a broadly similar way, identified a significant other to whom they felt closest. The name of the person identified was then presented in the other-focused IRAP. In particular, participants were asked to write down the name of the individual to whom they felt closest, followed by answers to six further questions about their relationship: (1) length of relationship, rated on Likert scale from 1 (0-1 year) to 4 (7 Of years)-, (2) closeness of relationship (using one set of the IOS Venn diagrams and referring specifically to the name of the significant other); (3) frequency of contact, rated from 1 (several times a day) to 6 (less than once a month)-, (4) level of sharing personal information, rated from 1 ("/ share everything I possibly could") to 5 ("I share nothing about myself with this person"); (5) warmth toward the significant other, on a thermometer from 0 ("very cold toward the person") to 100 ("very warm toward the person")-, and (6) an openended description of the most pleasant event shared with this person. The scores on Questions 1-5 were combined and divided by 5 to provide an overall score for the specified other-focusing task, with higher scores indicating greater closeness in the relationship.

Self-focusing task This task was also developed for the purpose of the current experiment, primarily to control for the extra 5 minutes it took to complete the specified other-focusing task. In this case, participants were simply asked to write down their own first name, followed by answers to five further questions: (1) List your top three hobbies; (2) How frequently do you engage in these hobbies, rated from 1 (several times a day) to 6 (less than once a month)?; (3) What is your favorite food?; 4. Where is your favorite place?; and (5) How warm do you feel toward yourself, from 0 (very cold toward myself) to 100 (very warm toward myself)? Outcomes on this control task were not quantified.

Specified other-focused IRAP The specified other-focused IRAP was identical to the other-focused IRAP used previously, except that each statement included the name of the individual from the specified other-focusing task (e.g., "Getting a fine makes David angry").

Self-focused IRAP The self-focused IRAP was identical to the previous experiments.

Questionnaires All questionnaires included in Experiment 3 were identical to Experiment 2.

Procedure The procedure for Experiment 3 was similar to the procedure from the previous studies, except for the inclusion of the two focusing tasks before each IRAP. Experiment 3 comprised three phases: (1) specified other-focusing task and specified other-focused IRAP; (2) self-focusing task and self-focused IRAP; and (3) Questionnaires (CAPE, PFI, ECRRS, EOS, IOS, and the Self-Warmth Thennometer). The order of the presentation of Phases 1 and 2 was counterbalanced across participants.

Phase 1: Specified other-focusing task and specified other-focused IRAP The researcher guided participants through the seven items of the specified other-focusing task in a semistructured manner. Upon completion of the specified otherfocusing task, participants were instructed as follows: "Before we begin the next task, I just want you to take a moment to think about [name of significant other]. The next phase of the experiment will be focused, in part, on this person." The specified other-focused IRAP was similar to the other-focused IRAP from Experiment 2, except that each statement used the name of the significant other and participants were instructed as follows: "The program will present statements on the screen which refer to [name of significant other]. Please remember that when you see [name of significant other], this refers to [nature of the relationship, e.g., your best friend]."

Phase 2: Self-focusing task and self-focused IRAP The researcher guided participants through the items of the selffocusing task also in a semi-structured manner. Participants were then instructed as follows: "The next task will contain statements about you. Consider the statement, 'If I win the lottery I feel happy.' It is important to remember that the 'I' we are referring to in this statement is you the participant." The procedure for the self-focused IRAP was identical to Experiment 2.

Questionnaires Participants completed the questionnaires in the following sequence: the CAPE, PFI, ECR-RS, EOSS, IOS, and the Self-Warmth Thermometer.

Results and Discussion

Questionnaire data A summary of the means and standard deviations of each questionnaire and relevant subscales is provided in Appendix C. The overall pattern of results from the questionnaires was broadly similar to Experiment 2.

IRAP data One participant failed the practice blocks on the specified other-focused IRAP; one failed to complete the self-focused IRAP because of limited time; three failed to maintain criteria on the self-focused IRAP; and two failed to do so on the specified other-focused IRAP. Thus, the final number of participants included in the analyses was N = 33.

[D.sub.IRAP].scores. The mean [D.sub.IRAP-scores] for each trial type are presented in Figure 5. Positive scores were recorded for all four trial types, with the weakest on Negative EventPositive Reaction. For each of the four trial types, the difference between the two IRAPs appeared relatively modest. A 2x4 mixed repeated measures ANOVA produced a main effect for trial type [F(1, 32) = 9.98, p < .001, [[eta].sup.2.sub.[rho]] =.24], but not for IRAP type (p > .8), nor for the interaction (p > 0.7). Post-hoc (Fisher's PLSD) comparisons indicated that Positive Event-Positive Reaction differed significantly from the other three trial types, and Negative Event-Positive Reaction differed from Positive Event-Negative Reaction and from Negative Event-Negative Reaction (see Table 5).

Correlations The overall D-score from each IRAP, as well as the difference scores between the two IRAPs, were subjected to correlational analyses (not shown) with the questionnaires (i.e., 25 correlations for each IRAP and 25 for the IRAP difference score). A correlation matrix showed four significant correlations. The self-focused IRAP correlated positively with frequency of negative psychotic-like symptoms [r(33) = .38, p < 0.03] and with warmth towards oneself [r(33) = .49, p < 0.01]. The specified other-focused IRAP correlated negatively with avoidant attachment with one's mother [r(33) = -.43,p< 0.01], The IRAP difference score correlated positively with avoidant attachment with one's mother [r(33) = 0.41, p < 0.02]. None of these correlations remained significant after Bonferroni corrections. Finally, the overall D-scores from the two IRAPs were correlated and this single correlation proved to be significant (r = 0.414,p = 0.016).

Summary In contrast to the two previous experiments, the difference for IRAP type was nonsignificant in this third experiment. This suggests that when the other is specified, participants respond in a broadly similar fashion to both self- and other-focused IRAPs. The overall D-scores from the two IRAPs also correlated, suggesting some overlap in the two measures. Only 4 of the 75 correlations proved to be significant but did not remain so after Bonferroni corrections, and thus again should be interpreted with extreme caution. On balance, Experiment 3 differed from the previous two experiments in two ways. Experiment 3 introduced a specified other, but also involved a focusing task designed to help ensure that all participants identified, in a similar manner, with the significant other to whom they felt closest. It is possible that completing this task before the IRAPs, in some undefined way, undermined the differences in IRAP performances previously recorded in Experiments 1 and 2. In Experiment 4, therefore, we sought to replicate Experiment 3, which again involved a specified other, but without exposure to the focusing tasks.

Experiment 4

Method

Participants Thirty-four participants were recruited for Experiment 4: 21 females and 13 males. Ages ranged from 18 to 39 years old (M= 22.64). All participants were recruited through random convenience sampling from the Ghent University participant pool and were paid an hourly rate of 10 [euro].

Materials and apparatus Both IRAPs were identical to Experiment 3.

Procedure The procedure was similar to Experiment 3, except that the self-focusing and specified other-focusing tasks were no longer included. That is, participants were again asked to write down the name of the individual to whom they felt closest, but this was not followed by any questions about the nature of their relationship. Note also that no additional questions preceded the self-focused IRAP.

Results and Discussion

Questionnaire data A summary of the means and standard deviations of each questionnaire and relevant subscales is provided in Appendix D. The overall pattern of results from the questionnaires was broadly similar to Experiment 3.

IRAP data Three participants failed to maintain criteria on the self-focused IRAP and one failed to complete the self-focused IRAP because of time constraints. Thus, the final number of participants included in the analyses was N = 30.

[D.sub.IRAP]-scores The mean [D.sub.IRAP]-scores and standard errors for each trial type for both IRAPs are presented in Figure 6. Positive scores were recorded for all four trial types, with the weakest observed for Negative Event-Positive Reaction. For each of the four trial types, the difference between the two IRAPs appeared relatively modest, except for Positive Event-Negative Reaction. A 2x4 repeated measures ANOVA yielded a significant main effect for trial type [F(1, 29) = 10.99, p < .001, [[eta].sup.2.sub.[rho]] =21] but no main effect for IRAP type (p > .8) or the interaction (p > .4). Post-hoc (Fisher's PLSD) comparisons among the four trial types (see Table 6) indicated that Negative Event-Positive Reaction differed significantly from the other three trial types.

Correlations The overall D-score from each IRAP, as well as the difference scores between the two IRAPs, were subjected to correlational analyses with the questionnaires (i.e., 25 correlations for each IRAP and 25 for the IRAP difference score, not shown). A correlation matrix showed six significant correlations. The self-focused IRAP correlated positively with the EOSS close relationships-absent subscale [r(30) = 0.47, p < 0.01]. The specified other-focused IRAP correlated negatively with the ECR-RS attachment-avoidance subscale for one's mother [r(30) = -0.42, p < 0.05], one's father [r(30) = -.391, p < 0.04], and with attachment-anxiety of best friend [r(30) = -0.38, p < 0.04], and again with the overall PFI score [r(30) = -0.42, p < 0.03]. The difference score between the IRAPs correlated negatively with attachment avoidance with romantic partner [r(30) = -0.38, p < 0.04], None of these correlations remained significant after Bonferroni corrections. Finally, the overall D-scores from the two IRAPs were correlated and this single correlation proved to be nonsignificant (r = 0.284, p = 0.129).

Summary Like Experiment 3, the difference for IRAP type was again nonsignificant suggesting that when the other is specified, participants respond in a broadly similar fashion to both self- and other-focused IRAPs. As a result, the lack of difference between the IRAPs observed in Experiment 3 was unlikely due to the focusing tasks (because in Experiment 4 these tasks were not included). However, the overall D-scores from the two IRAPs did not correlate significantly (in Experiment 3 the correlation was significant), suggesting no overlap in the two measures. Only 6 of the 75 correlations proved to be significant but again these did not remain so after Bonferroni corrections.

At this point in the research, the first two experiments had yielded significant differences between the IRAPs when other was unspecified, but nonsignificant differences were recorded across the third and fourth experiments when other was specified. In these latter experiments, the specified other was identified as the person to whom each participant was closest, and thus the variable of closeness, but not similarity, was manipulated. It is possible, therefore, that the lack of significant differences between the self- and other-IRAPs was not due to closeness but to similarity, if we assume that in general people are close to others who are similar to them. In Experiment 5, therefore, we replicated Experiment 4, but asked participants to identify an individual to whom they were close but who they perceived to be different: would targeting difference in this way produce a significant difference between the two IRAPs?

Experiment 5

Method

Participants Thirty-two participants were recruited for Experiment 5: 24 females, eight males. Ages ranged from 18 to 32 years (M= 20.82). All participants were recruited through random convenience sampling from the Ghent University participant pool and were paid an hourly rate of 10 [euro].

Materials and apparatus The self-focused IRAP was identical to that used in Experiments 1-4. The other-focused IRAP was similar to Experiments 3 and 4, except that the name presented in each of the statements was now of a specified other (e.g., "Sarah") who the participant considered to be different from them. All questionnaires were identical to those used in Experiments 2-4.

Specified other-focused IRAP The specified other-focused IRAP was similar to Experiments 3 and 4, except that each statement now included the name of an individual to whom the participant was close, but who was nonetheless perceived to be different. Participants were instructed as follows "I want you to think of a person in your life that you are close to, but who you also consider to be very different from you. For example, this could be a friend who has very different interests from you."

Self-focused IRAP The self-focused IRAP was identical to all previous experiments.

Questionnaires All questionnaires were identical to Experiments 2-4.

Procedure The procedure was similar to Experiment 4, except that the specified-other IRAP involved the name of an individual to whom the participant was close, but who was perceived to be different.

Results and Discussion

Questionnaire data A summary of the means and standard deviations of each questionnaire and relevant subscales is provided in Appendix E. The overall pattern of results from the questionnaires was broadly similar to previous experiments.

IRAP data One participant failed to complete the specified other-focused IRAP because of limited time and one failed to maintain criteria on the self-focused IRAP. Thus, the final number of participants included in the analyses was N = 30.

[D.sub.IRAP]-scores The mean [D.sub.IRAP]-scores for each trial type are presented in Figure 7. Positive scores were recorded for all four trial types, with the weakest on Negative Event-Positive Reaction. For each of the four trial types, the difference between the two IRAPs appeared relatively modest. A 2x4 mixed repeated measures ANOVA produced a main effect for trial type [F(1, 29) = 5.33, p < .001, [[eta].sup.2.sub.[rho]] =.16], but not for IRAP type (p > .5), or for the interaction p > 0.2). Posthoc (Fisher's PLSD) comparisons indicated that Negative Event-Positive Reaction differed significantly from Positive Event-Positive Reaction and Negative Event-Negative Reaction (see Table 7).

Correlations The overall D-score from each IRAP, as well as the difference scores between the two IRAPs, were subjected to correlational analyses with the questionnaires (i.e., 25 correlations for each IRAP and 25 for the IRAP difference score, not shown). A correlation matrix showed only one significant correlation. The self-focused IRAP correlated negatively with avoidant attachment with one's mother [r(30) = -.38, p < 0.04], but again this was nonsignificant after Bonferroni correction. In addition, the overall D-scores from the two IRAPs were correlated and this single correlation proved to be significant (r = 0.410, p = 0.023).

Summary Like Experiments 3 and 4, the difference for IRAP type was again nonsignificant, suggesting that when the other is known by the participant, but perceived as different from them, participants respond in a broadly similar fashion to both self- and other-focused IRAPs. The overall D-scores from the two IRAPs also correlated, suggesting some overlap in the two measures. Only 1 of the 75 correlations proved to be significant but did not remain so after a Bonferroni correction, and thus once again this result should be interpreted cautiously. In Experiments 3 and 4, the specified other was identified as the person to whom each participant felt closest, but in Experiment 5 specified other was identified as the person to whom each participant felt closest but perceived as different. It was assumed that people would feel positive towards others to whom they are close, and thus the variable of valence (i.e., positivity) was kept constant across these three experiments. It is possible, therefore, that the lack of difference in overall IRAP effects was not due to closeness and similarity, but to positive valence, if we assume that in general people feel positive toward those to whom they are close. In Experiment 6, therefore, we replicated Experiment 5, but asked participants to identify an individual who they knew but disliked: would targeting valence in this way produce a significant difference between the two IRAPs?

Experiment 6

Method

Participants Thirty-four participants were recruited for Experiment 6: 30 females, four males. Ages ranged from 18 to 50 years (M= 21.29). All participants were recruited through random convenience sampling from the Ghent University participant pool and were paid an hourly rate of 10 [euro].

Materials and apparatus The self-focused IRAP was identical to that used in Experiments 1-5. The other-focused IRAP was similar to Experiments 3-5, except that the name presented in each of the statements was now of a specified other (e.g., "Sam") who the participant disliked. All questionnaires were identical to those used in Experiments 2-5.

Specified other-focused IRAP The specified other-focused IRAP was similar to Experiments 3-5, except that each statement now included the name of an individual who was known to the participant, but who they disliked. Participants were instructed as follows "I want you to think of a person in your life who you often interact with, but who you don't like. For example, this could be a classmate, work colleague or family member who you do not get along with."

Self-focused IRAP The self-focused IRAP was identical to all previous experiments.

Questionnaires All questionnaires were identical to Experiments 2-5.

Procedure The procedure was similar to Experiment 5, except that the specified-other IRAP involved the name of an individual to whom the participant was close, but who they disliked.

Results and Discussion

Questionnaire data A summary of the means and standard deviations of each questionnaire and relevant subscales is provided in Appendix F. The overall pattern of results from the questionnaires was broadly similar to previous experiments.

IRAP data One participant failed to complete the specified other-focused IRAP because of limited time and two failed to maintain criteria on the specified other-focused IRAP. One participant failed to maintain criteria on the self-focused IRAP. Thus, the final number of participants included in the analyses was N = 30.

[D.sub.IRAP]-scores The mean [D.sub.IRAP]-scores for each trial type are presented in Figure 8. Positive scores were recorded for seven of the eight trial types, with the only negative score recorded on Negative Event-Positive Reaction of the specified other-focused IRAP. For each of the four trial types, the difference between the two IRAPs appeared relatively modest. A 2x4 mixed repeated measures ANOVA produced a main effect for trial type [F(1, 29) = 7.28, p < .001, [[eta].sup.2.sub.[rho]] =.2], but not for IRAP type (p > .09), or for the interaction (p > 0.46). Post-hoc comparisons (Fisher's PLSD) indicated that Negative Event-Positive Reaction differed significantly from Positive Event-Positive Reaction, Positive Event-Negative Reaction, and Negative Event-Negative Reaction (see Table 8).

Correlations The overall D-score from each IRAP, as well as the difference scores between the two IRAPs, were subjected to correlational analyses with the questionnaires (i.e., 25 correlations for each IRAP and 25 for the IRAP difference score, not shown). The correlation matrix yielded five significant correlations. The self-focused IRAP correlated negatively with avoidant attachment to one's romantic partner [r(30) = -.41, p < 0.03] and the control of a person close to them over the experience of self [>(30) = -.43, p < 0.02], The specified-other IRAP correlated negatively with the control of a person close to them over the experience of self [r(30) = -.39, p = 0.03], closeness to other people [r(30) = -.41, p = 0.03], and with warmth towards oneself [r(30) = -.46, p < 0.01], Once again, none of these correlations remained significant after Bonferroni corrections. Finally, the overall Dscores from the two IRAPs were correlated and this single correlation proved to be nonsignificant (r = 0.103, p = 0.59).

Summary Like Experiments 3-5, the difference for IRAP type was again nonsignificant, suggesting that when the other is negatively valenced, participants again respond in a broadly similar fashion to both self- and other-focused IRAPs. On balance, the overall D-scores from the two IRAPs did not correlate significantly, suggesting limited overlap in the two measures. Once again, only a small number of the 75 correlations proved to be significant, but again these did not remain so after Bonferroni corrections.

General Discussion

The overarching purpose of the current series of experiments was to develop IRAPs that would clearly differentiate responding from the perspective of self versus other. Previous research appeared to demonstrate deictic relational responding using the IRAP (i.e., responding to the characteristics of self versus other within an IRAP), but to date clear evidence of responding from the perspective of self versus from the perspective of another has not been shown. To demonstrate such an effect would seem to require inserting statements into an IRAP that ask participants to confirm and deny how the self versus others would react to similar events. Achieving this objective required that relatively complex statements be employed, and for this reason a natural language version of the IRAP was used. The results from the first two experiments reported here indicated that there were significant differences between the self- versus other-focused IRAPs, when the other remained unspecified. The remaining four experiments, however, indicated that when the other was specified there was limited evidence that performances on the two IRAPs differed significantly. The correlational analyses between the IRAPs and the self-report measures yielded few significant effects (and none at all after Bonferroni corrections). Finally, the correlational analyses between the two IRAPs in each experiment were significant in some cases (Experiments 2, 3, and 5), but not others (Experiments 1, 4, and 6). Overall, the relatively small number of uncorrected correlations between the IRAPs and the explicit measures, and the complete absence of any corrected significant correlations, suggest that such effects should be taken with extreme caution, and at the current time they are, in our view, not worthy of further discussion. The reader is referred to Table 9 for a summary of the main findings arising from the six studies presented here.

Overall, the current findings could be seen as encouraging because each experiment produced performances that would be deemed consistent with the preexperimental histories of the participants. That is, the IRAP effects for the most part, were in the predicted direction (e.g., participants confirmed more rapidly than they denied that when positive events occur, they react positively, but when negative events occur, they react negatively). The results were also encouraging because we observed a possibly important distinction in perspective-taking when it applied to a general unspecified other versus someone well known to the participant (see DeBernardis, Hayes, & Fryling, 2014).

On balance, the results could be seen as somewhat disappointing because we failed to find strong evidence of perspective-taking when other was specified, at least in terms of different performances across the two IRAPs, or in correlations among the IRAPs and the self-report measures. Nonetheless, we think it is important to report the data at this point because they may be instructive for other researchers seeking to develop IRAP-based, or even latency-based, measures of perspective-taking. We do recognize, of course, that it would be possible to analyze the current dataset in almost countless ways, using a variety of statistical techniques, thus yielding, perhaps, less disappointing outcomes. To do so, however, would not link directly back to the conceptual assumptions on which the current research was based.

One unexpected pattern that did emerge, and that was generally consistent across all six experiments, was that the trial type Negative Event-Positive Reaction tended to yield the weakest effect. The extent to which this could be explained in terms of a Single Trial-Type Dominance Effect (STTDE), as reported by Kavanagh et al. (2018; see also Finn et al., 2018), cannot readily be applied here because that effect is specific to the two trial types that require a "Yes" response during history-consistent blocks. The critical issue here is that across all six experiments the effect for the Negative EventPositive Reaction was weaker than for the Positive EventNegative Reaction trial type, and both of these trial types require a "No" response during history-consistent blocks. Why would they differ, given that they both require the same response within blocks? (2) This question seems particularly interesting because a natural language format IRAP was employed here, and thus an explanation that appeals to the functions of separate label and target stimuli within a traditional IRAP cannot be applied (see Kavanagh et al., 2018, for a detailed discussion). Nevertheless, given the consistency in this weakness effect for the Negative Event-Positive trial-type across the six studies, it remains a pattern that should be targeted for systematic analyses in future research.

Irrespective of the explanation for the unexpected differential [D.sub.-IRAP] effects for the two trial types discussed above, overall the two IRAPs did produce history-consistent effects. In this sense, the IRAPs produced predictable outcomes and thus it seems important, going forward, to ask why the IRAPs did not distinguish in a clear and consistent way between the perspective of self versus other. One possibility is that the IRAPs simply tapped into "sense-making." For example, confirming, rather than denying that positive events evoke positive reactions in the self and others makes sense in natural language. The only caveat to this interpretation is that significant differences between the IRAPs were observed in Experiments 1 and 2 when the other was unspecified, but not in Experiments 3-6 when other was specified. Perhaps, therefore, there was some sensitivity to self versus other, but the use of complex statements in the IRAPs potentially undermined or reduced the impact of deictic relational responding per se. That is, in presenting such complex stimuli or networks in the IRAP, participants more or less interpreted the task as a sense-making or problem-solving task, in which the self versus other had little or no impact, particularly when the other was specified. The challenge going forward, therefore, is to develop IRAPs that maintain the deictic functions of self and other, in the context of perspective-taking rather than simple sense-making. One way in which we are currently pursuing this is to revert to the more traditional label-target format within the IRAP, but to present pictures of each participant versus matched pictures of another individual (to maintain the differential functions of self versus other). The general idea, therefore, is to reduce the complexity of the relational networks, while maintaining the distinction between self and other.

In closing, we think it is important to draw attention to a pattern in the data that only came to light during the review process, and that might indicate that the two IRAPs may have been sensitive, to some degree, to self versus others, when other was clearly distinct from self. In particular, the trial type effect for Positive Event-Negative Reaction was significantly different from 0 for the self-focused IRAPs, but not for the other-focused IRAPs, across Experiments 5 and 6. This was not the case in the four remaining experiments, in which other was unspecified or specified as similar to self. In a purely posthoc analysis, we compared the difference between the selfand other-focused IRAPs for the Positive Event-Negative Reaction trial type, (Self; M = .164 SD = .324, Other; M = .032 SD = .333, t(59) = 2.195 p =.032. 95% CI [.024, .54], BF1 = 2.560). We raise this issue here because this pattern emerged across the two experiments in which other was clearly distinct from self. If this finding is indeed robust, it could indicate that participants found it easier to respond "False" when self, rather than a distinct other, was coordinated in some way with a negative reaction. In simple terms, participants showed a bias toward denying that they are negative that was stronger than that bias for different others. Although subtle, pursuing this potentially interesting effect could be useful in future research.

Funding This research was conducted with funding from the XXX.

Compliance with Ethical Standards

Conflict of Interest Deirdre Kavanagh declares that she has no conflict of interest. Adeline Roelandt declares that she has no conflict of interest. Lisa Van Raemdonck declares that she has no conflict of interest. Yvonne Barnes-Holmes declares that she has no conflict of interest. Dermot Barnes-Holmes declares he has no conflict of interest. Ciara McEnteggart declares that she has no conflict of interest.

Ethical approval All procedures in the current study were in accordance with the ethical standards of the institutional research committee, and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants.

Appendix A
Table 10. Experiment 1: Descriptive statistics
for questionnaires

Questionnaire                                      M      SD

Acccptance and Action Questionnaire (AAQ-II)     17.16   7.14
CAPE (weighted scores)
  Overall Frequency                               1.5    .19
  Frequency of Positive Symptoms                 1.32    .23
  Frequency of Negative Symptoms                 1.57    .34
  Frequency of Depressive Symptoms               1.83    .41
  Overall Distress                               2.04    .45
  Distress associated with Positive Symptoms     1.68    .53
  Distress associated with Negative Symptoms     2.13    .86
  Distress associated with Depressive Symptoms   2.33    .62

Note: The maximum weighted score for all CAPE subscales is 4.00.
The CAPE has no formal clinical cut-off. The maximum score for
the AAQ-II is 49 and the measure has a suggested clinical
cut-off of [greater than or equal to] 24.


Appendix B
Table 11. Experiment 2: Descriptive statistics for questionnaires

Questionnaire                                       M       SD

Self-Warmth Thermometer                           56.65    22.02
Psychological Flexibility Index (PFI)             355.31   27.19
CAPE (weighted scores)
  Overall Frequency                                1.72     .32
  Frequency of Positive Symptoms                   1.43     .34
  Frequency of Negative Symptoms                   1.94     .44
  Frequency of Depressive Symptoms                 2.08     .52
  Overall Distress                                 2.15     .54
  Distress associated with Positive Symptoms       1.66     .44
  Distress associated with Negative Symptoms       2.07     .58
  Distress associated with Depressive Symptoms     2.54     .73
ECR-RS
  Attachment-rclatcd avoidance (Mother)           19.82    9.47
  Attachment-rclatcd anxiety (Mother)              5.33    3.51
  Attachment-related avoidance (Father)           23.65    9.55
  Attachment-related anxiety (Father)              6.41    4.61
  Attachment-related avoidance (Partner)          11.45    5.55
  Attachment-related anxiety (Partner)             9.02    4.79
  Attachmcnt-rclatcd avoidance (Best Friend)      14.12    6.33
  Attachment-related anxiety (Best Friend)         7.69     4.6
EOSS
  Overall EOSS                                    72.82    15.97
  Casual Acquaintances-absent                     17.47     5.3
  Casual Acquaintances-present                    23.98    4.54
  Close relationships-absent                        11     6.19
  Close relationships-present                     20.38    6.02
IOS
  Best Friend                                      4.71    1.34
  Other people                                     2.98    1.13

Note: The maximum score is 100 for the Self-Warmth Thermometer.
The maximum score for the PFI is 492. The maximum weighted score
for all CAPE subscales is 4.00. The maximum score for each of
the EC-RS attachment related avoidance subscales is 42 and the
Attachment related anxiety subscale is 21. The maximum overall
EOSS score is 140 with the maximum score for each subscale at 35.
Finally, the maximum score for each of the IOS scales is 7.
None of the scales have formal clinical cut-offs


Appendix C
Table 12. Experiment 3: Descriptive statistics for questionnaires

Questionnaire                                       M       SD

Self-Warmth Thermometer                           53.54    22.77
Psychological Flexibility Index (PFI)             323.42   23.99
Overall Other-Focusing Task score                 24.94    3.18
CAPE (weighted scores)
  Overall Frequency                                1.71     .44
  Frequency of Positive Symptoms                   1.39     .4
  Frequency of Negative Symptoms                   1.93     .54
  Frequency of Depressive Symptoms                 2.09     .62
  Overall Distress                                 2.09     .58
  Distress associated with Positive Symptoms       1.66     .51
  Distress associated with Negative Symptoms       2.06     .55
  Distress associated with Depressive Symptoms     2.35     .93
ECR-RS
  Attachment-related avoidance (Mother)           18.24    9.75
  Attachment-related anxiety (Mother)              4.56    2.75
  Attachment-related avoidance (Father)           22.52    8.64
  Attachment-related anxiety (Father)              5.52    3.89
  Attachment-related avoidance (Partner)          11.61    4.61
  Attachment-related anxiety (Partner)             9.21    5.18
  Attachment-related avoidance (Best Friend)      13.88    4.61
  Attachment-related anxiety (Best Friend)         6.91    4.03
EOSS
  Overall EOSS                                     75.7    13.28
  Casual Acquaintances-absent                     17.33    5.53
  Casual Acquaintances-present                    24.97    4.83
  Close relationships-absent                      10.82    5.57
  Close relationships-present                     22.58    5.04
IOS
  Best Friend                                      5.06    1.14
  Other people                                     3.24     1.0

Note: See note for Appendix B


Appendix D
Table 13. Experiment 4: Descriptive statistics for questionnaires

Questionnaire                                        M        SD

Self-Warmth Thermometer                            49.73    24.41
Psychological Flexibility Index (PFI)             346.73    24.64
CAPE (weighted scores)
  Overall Frequency                                 1.53      .38
  Frequency of Positive Symptoms                    1.33      .35
  Frequency of Negative Symptoms                    1.81      .49
  Frequency of Depressive Symptoms                  1.75      .52
  Overall Distress                                  1.83      .55
  Distress associated with Positive Symptoms        1.48      .65
  Distress associated with Negative Symptoms        1.81      .61
  Distress associated with Depressive Symptoms      2.16      .69
ECR-RS
  Attachment-related avoidance (Mother)            16.57     7.99
  Attachment-related anxiety (Mother)                4.5      2.7
  Attachment-related avoidance (Father)            23.93     10.3
  Attachment-related anxiety (Father)               6.97      5.8
  Attachment-related avoidance (Partner)            11.1     4.71
  Attachment-related anxiety (Partner)              8.23     4.88
  Attachment-related avoidance (Best Friend)       12.83     5.32
  Attachment-related anxiety (Best Friend)           6.1     3.47
EOSS
  Overall EOSS                                     72.37    16.65
  Casual Acquaintances-absent                       16.6     4.97
  Casual Acquaintances-present                     23.33     4.72
  Close relationships-absent                       12.53     6.46
  Close relationships-present                       19.9     6.49
IOS
  Best Friend                                        4.3     1.09
  Other people                                       2.7     1.08

Note: See note for Appendix B


Appendix E
Table 14. Experiment 5: Descriptive statistics for questionnaires

Questionnaire                                       M        SD

Self-Warmth Thermometer                           47.43    27.18
Psychological Flexibility Index (PFI)             358.1    23.93
CAPE (weighted scores)
  Overall Frequency                                1.76      .36
  Frequency of Positive Symptoms                   1.46      .28
  Frequency of Negative Symptoms                   1.97      .53
  Frequency of Depressive Symptoms                 2.13      .59
  Overall Distress                                 2.15      .47
  Distress associated with Positive Symptoms        1.7       .5
  Distress associated with Negative Symptoms       2.45      .77
  Distress associated with Depressive Symptoms     2.01      .56
ECR-RS
  Attachment-related avoidance (Mother)            17.9      7.9
  Attachment-related anxiety (Mother)              6.13      3.5
  Attachment-related avoidance (Father)           24.03     9.26
  Attachment-related anxiety (Father)               8.3     5.37
  Attachment-related avoidance (Partner)          12.03     5.51
  Attachment-related anxiety (Partner)            10.33      5.2
  Attachment-related avoidance (Best Friend)      14.07     6.03
  Attachment-related anxiety (Best Friend)         7.73     4.17
EOSS
  Overall EOSS                                    74.93    14.96
  Casual Acquaintances-absent                     17.27     6.25
  Casual Acquaintances-present                    24.63     4.25
  Close relationships-absent                      12.03     5.73
  Close relationships-present                        21     6.26
IOS
  Best Friend                                      4.43     1.25
  Other people                                      2.7     1.06

Note: See note for Appendix B


Appendix F
Table 15. Experiment 6: Descriptive statistics for questionnaires

Questionnaire                                        M        SD

Self-Warmth Thermometer                            49.91     26.2
Psychological Flexibility Index (PFI)             363.13    25.44
CAPE (weighted scores)
  Overall Frequency                                 1.86      .37
  Frequency of Positive Symptoms                    1.62      .41
  Frequency of Negative Symptoms                    1.98      .45
  Frequency of Depressive Symptoms                  2.23      .49
  Overall Distress                                  2.24      .46
  Distress associated with Positive Symptoms        1.84      .49
  Distress associated with Negative Symptoms        1.99      .52
  Distress associated with Depressive Symptoms      2.72      .57
ECR-RS
  Attachment-related avoidance (Mother)            16.07     8.93
  Attachment-related anxiety (Mother)               4.83     3.13
  Attachment-related avoidance (Father)            23.07    10.70
  Attachment-related anxiety (Father)               6.17     4.69
  Attachment-related avoidance (Partner)           11.07     4.28
  Attachment-related anxiety (Partner)               9.2     4.77
  Attachment-related avoidance (Best Friend)       11.77     4.94
  Attachment-related anxiety (Best Friend)           6.1     4.11
EOSS
  Overall EOSS                                      76.6    15.07
  Casual Acquaintances-absent                      18.23     5.35
  Casual Acquaintances-present                     25.77     3.53
  Close relationships-absent                       11.67     6.42
  Close relationships-present                      20.93     6.38
IOS
  Best Friend                                       5.10     1.40
  Other people                                      3.13     1.22

Note: See note for Appendix B


Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of Personality & Social Psychology, 63, 596-612. https://doi.org/10. 1037/0022-3514.63.4.596.

Barbero-Rubio, A., Lopez-Lopez, J., Luciano, C., & Eisenbeck, N. (2016). Perspective-taking measured by implicit relational assessment procedure (IRAP). The Psychological Record, 66, 1-10. https://doi.org/10.1007/s40732-016-0166-3.

Barnes-Holmes, D., Barnes-Holmes, Y., Stewart, I., & Boles, S. (2010). A sketch of the implicit relational assessment procedure (IRAP) and the relational elaboration and coherence (REC) model. The Psychological Record, 60, 527-542. https://doi.org/10.1007/ BF03395726.

Barnes-Holmes, Y. (2001). Analysing relational frames: Studying language and cognition in young children (Unpublished doctoral thesis). National University of Ireland, Maynooth.

Barnes-Holmes, Y., Bames-Holmes, D., Roche, B., & Smeets, P. (2001). The development of self and perspective-taking: A relational frame analysis. Behavioral Development Bulletin, (7), 42-45. https://doi. org/10.1037/h0100482.

Bernaerts, I., De Groot, F., & Kleen, M. (2012). De AAQ-II, een maat voor experientiele vermijding: normering bij jongeren. Gedragstherapie, 4, 389-399.

Bernstein, R. F., Laurent, S. N., Nelson, B. W., & Laurent, H. K. (2015). Perspective-taking induction mitigates the effects of partner attachment avoidance on self-partner overlap. Personal Relationships, 22, 356-367. https://doi.org/10.1111/pere.12085.

Bond, F., Hayes, S., Baer, R., Carpenter, K., Guenole, N., Orcutt, H., et al. (2011). Preliminary psychometric properties of the Acceptance and Action Questionnaire-II: A revised measure of psychological inflexibility and experiential avoidance. Behavior Therapy, 42, 676-688. https://doi.org/10.1016/j.beth.2011.03.007.

DeBemardis, G. M., Hayes, L. J., & Fryling, M. J. (2014). Perspective taking as a continuum. The Psychological Record, 64, 123-131. https://doi.org/10.1007/s40732-614-0008-0.

Finn, M., Bames-Holmes, D., & McEnteggart, C. (2018). Exploring the single-trial-type-dominance-effect on the IRAP: Developing a differential arbitrarily applicable relational responding effects (DAARRE) model. The Psychological Record, 68, 11-25. https:// doi.org/10.1007/s40732-017-0262-z.

Fraley, R. C., Heffernan, M. E., Vicary, A. M., & Brumbaugh, C. C. (2011). The Experiences in Close Relationships-Relationship Structures questionnaire: A method for assessing attachment orientations across relationships. Psychological Assessment, 23, 615625. https://doi.org/10.1037/a0022898.

Golijani-Moghaddam, N., Hart, A., & Dawson, D. L. (2013). The implicit relational assessment procedure: Emerging reliability and validity data. Journal of Contextual Behavioral Science, 2, 105-119. https://doi.org/10.1016/j.jcbs.2013.05.002.

Gore, N. G., Bames-Holmes, Y., & Murphy, G. (2010). The relationship between intellectual functioning and relational perspective-taking. International Journal of Psychology & Psychological Therapy, 10, 1-17.

Hayes, S. C., Barnes-Holmes, D., & Roche, B. (2001). Relational frame theory: A post-Skinnerian account of human language and cognition. New York: Kluwer Academic.

Heagle, A. I., & Rehfeldt, R. A. (2006). Teaching perspective-taking skills to typically developing children through derived relational responding. Journal of Early Intensive Behavioral Intervention, 3, 1-34. https://doi.org/10.1037/h0100321.

Hussey, I., McEnteggart C., Barnes-Holmes, Y., Kavanagh, D., BarnesHolmes, D., Parting, T., & Lundgren, T. (2014). Flexible perspective-taking: New concepts and a new behavioural measure. Dublin: Paper presented at the ACT-CBS Conference.

Kanter, J. W., Parker, C. R., & Kohlenberg, R. J. (2001). Finding the self: A behavioral measure and its clinical implications. Psychotherapy, 38, 198-211. https://doi.org/10.1037/0033-3204.38-2.198.

Kavanagh, D., Barnes-Holmes, Y., Barnes-Holmes, D., McEnteggart, C., & Finn, M. (2018). Exploring differential trial type effects and the impact of a read-aloud procedure on deictic relational responding on the IRAP. The Psychological Record, 68, 163-176. https://doi.org/ 10.1007/s40732-018-0276-1.

Kavanagh, D., Hussey, I.. McEnteggart, C., Barnes-Holmes, Y., & Barnes-Holmes, D. (2016). Using the IRAP to explore natural language statements. Journal of Contextual Behavioral Science, 5(4), 247-251. https://doi.org/10.1016/j.jcbs.2016.10.001.

Konings, M.. Bak, M., Hanssen, M., Van Os, I. & Krabbendam, L. (2006). Validity and reliability of the CAPE: a self-report instrument for the measurement of psychotic experiences in the general population. Acta Psychiatrica Scandinavica, 114, 55-61. https://doi.org/ 10.1111/j.1600-0447.2005.00741.x.

McHugh, L., Barnes-Holmes, Y., & Barnes-Holmes, D. (2004). Perspective-taking as relational responding: A developmental profile. The Psychological Record, 54, 115-144. https://doi.org/10. 1007/BF03395465.

Montoya-Rodriguez, M. M., Molina, F. J., & McHugh, L. (2017). A review of relational frame theory research into deictic relational responding. The Psychological Record, 67, 569-579. https://doi. oig/10.1007/s40732-016-0216-x.

Peirce, J. W. (2007). PsychoPy: Psychophysics software in Python. Journal of Neuroscience Methods, 162, 8-13. https://doi.org/10. 1016/j .jneumeth.2006.11.017.

Rendon, M. I., Soler, F., & Cortes, M. (2012). Relaciones deicticas simples, toma de perspectiva y competencia social. Suma Psicoldgica, 19, 19-37. https://doi.org/10.1007/s40732-016-0216-x.

Savla, G. N., Vella, L., Armstrong, C. C., Penn, D. L., & Twamley, E. W. (2013). Deficits in domains of social cognition in schizophrenia: A meta-analysis of the empirical evidence. Schizophrenia Bulletin, 39(5), 979-992. https://doi.org/10.1093/schbul/sbs080.

Sodian B., & Kristen S. (2010). Theory of Mind. In B. Glatzeder, V. Goel, & A. Miiller (Eds.), Towards a theory of thinking: On thinking (pp. 189-201). Heidelberg, Germany: Springer.

Stefanis. N. C., Hanssen, M., Smirnis, N. K.. Avramopoulos, D. A., Evdokimidis, I. K., Stefanis, C. N., Verdoux, H., & Van Os, J. (2002). Evidence that three dimensions of psychosis have a distribution in the general population. Psychological Medicine, 32, 347358. https://doi.org/10.1017/S0033291701005141.

Vahey, N., Barnes-Holmes, D., Barnes-Holmes, Y., & Stewart, I. (2009). A first test of the Implicit Relational Assessment Procedure (IRAP) as a measure of self-esteem: Irish prisoner groups and university students. The Psychological Record, 59, 371-388. https://doi.org/ 10.1007/BF03395670.

Vahey, N., Nicholson, E., & Barnes-Holmes, D. (2015). A meta-analysis of criterion effects for the Implicit Relational Assessment Procedure (IRAP) in the clinical domain. Journal of Behavior Therapy & Experimental Psychiatry, 48, 59-65. https://doi.org/10.1016/j. jbtep.2015.01.004.

Villatte, M., Monestes, J.-L., McHugh, L., Freixa i Baque, E., & Loas, G. (2010). Assessing perspective taking in schizophrenia using relational frame theory. The Psychological Record, 60, 413-436. https://doi.org/10.1007/BF03395719.

Villatte, M., Monestes, J.-L., McHugh, L., Freixa, i., Baque, E. F. i., & Loas, G. (2008). Assessing deictic relational responding in social anhedonia: A functional approach to the development of theory of mind impairments. International Journal of Behavioral Consultation & Therapy, 4, 360-373. https://doi.org/10.1037/ H0100867.

Vitale, A., Barnes-Holmes, Y., Barnes-Holmes, D., & Campbell, C. (2008). Facilitating responding in accordance with the relational frame of comparison: Systematic empirical analyses. The Psychological Record, 58, 365-390. https://doi.org/10.1007/ BF03395624.

Weil, T. M., Hayes, S. C., & Capurro, P. (2011). Establishing a deictic relational repertoire in young children. The Psychological Record, 61, 371-390. https://doi.org/10.1007/BF03395767.

World Health Organization. (2017). Process of translation and adaptation of instruments. Retrieved from http://www.who.int/substance_abuse/research tools/translation/en/

(1) Throughout this article, we have adopted the strategy of not correcting for multiple tests of statistical significance, but instead we report actual p values. However, we also report if the values remain significant following correction for multiple tests.

(2) As noted above, this difference across the trial types was not expected and thus any analysis or discussion of this finding must remain entirely post-hoc. We decided to employ both regular r-tests and Bayesian paired f-tests, one for each IRAP, because when the data were collapsed across all six experiments the Ns were relatively large (total N= 209) and thus the likelihood of obtaining significant p values increases dramatically. The results for both tests were significant with extremely strong evidence from the Bayes analyses: self-focused IRAP (Positive Event-Negative Reaction; M = .185, SD = 0.345, Negative Event-Positive Reaction, M = .049, SD = .293, t(208) = 5.202, p < .001, 95% CI [.084, .187], BF1 = 43939); other-focused IRAP (Positive Event-Negative Reaction; M = .231, SD = 0.341, Negative Event-Positive Reaction, M = .074, SD = .329, r(208) = 5.747, p < .001, 95% CI [.103, .210], BF1 =578829).

The data for the current manuscript was collected, and prepared with the support of the FWO Type I Odysseus Programme at Ghent University, Belgium. We are very grateful to the reviewers for their thorough and constructive commentaries.

Deirdre Kavanagh (1) [iD] * Adeline Roelandt (1) * Lisa Van Raemdonck (1) * Yvonne Barnes-Holmes (1) * Dermot Barnes-Holmes (1) * Ciara McEnteggart (1)

[mail] Deirdre Kavanagh

deirdre.kavanagh@ugent.be

Adeline Roelandt

adeline.roelandt@ugent.be

Lisa Van Raemdonck

lisa.vanraemdonck@ugent.be

Yvonne Bames-Holmes

yvonne.bamesholmes@ugent.be

Dermot Bames-Holmes

dermot.bames-holmes@ugent.be

Ciara McEnteggart

ciara.mcenteggart@ugent.be

(1) Department of Experimental, Clinical, and Health Psychology, Ghent University, 9000 Ghent, Belgium

https://doi.org/10.1007/s40732-019-00333-w

Published online: 7 March 2019

Caption: Fig. 1 Examples of the four trial types in the other-focused IRAP: Positive Event-Positive Reaction. Positive Event-Negative Reaction, Negative Event-Positive Reaction, and Negative Event-Negative Reaction. The arrows and words Consistent and Inconsistent were not shown onscreen. Trial type labels denote each of the two parts of the statement but not necessarily the sequence in which they appeared in the statement

Caption: Fig. 2 Examples of the four trial types in the self-focused IRAP: Positive Event-Positive Reaction, Positive Event-Negative Reaction, Negative Event-Positive Reaction, and Negative Event-Negative Reaction. The arrows and words Consistent and Inconsistent were not shown onscreen. Trial type labels denote each of the two parts of the statement, but not necessarily the sequence in which they appeared in the statement

Caption: Fig. 3 Mean [D.sub.IRAP]-scores on the self-focused IRAP trial types and other-focused trial types in Experiment 1. Positive [D.sub.IRAP]-scores indicate history-consistent responding and negative [D.sub.IRAP]-scores indicate history-inconsistent responding, indicates ARAP-scores that are significantly different from zero

Caption: Fig. 4 Mean [D.sub.IRAP]-scores on the self-focused and others-focused IRAP trial types in Experiment 2. Positive [D.sub.IRAP]-scores indicate history-consistent responding and negative [D.sub.IRAP]-scores indicate history-inconsistent responding, indicates [D.sub.IRAP]-scores that are significantly different from zero

Caption: Fig. 5 Mean [D.sub.IRAP]-scores on the self-focused IRAP trial types and specified other-focused trial types in Experiment 3. Positive [D.sub.IRAP]-scores indicate history-consistent responding and negative [D.sub.IRAP]-scores indicate history-inconsistent responding. **Indicates D1raf_scoks that are significantly different from zero

Caption: Fig. 6 Mean [D.sub.IRAP]-scores on the self-focused IRAP trial types and specified other-focused trial types in Experiment 4. Positive [D.sub.IRAP]-scores indicate history-consistent responding negative [D.sub.IRAP]-scores indicate historyin-consistent responding. * Indicates [D.sub.IRAP]-scores that are significantly different from zero

Caption: Fig. 7 Mean [D.sub.IRAP]-scorcs on the self-focused IRAP trial types and specified other-focused trial types in Experiment 5. Positive [D.sub.IRAP]-scores indicate history-consistent responding and negative [D.sub.IRAP]-scores indicate history-inconsistent responding. *Indicates Arapscores that are significantly different from zero

Caption: Fig. 8 Mean [D.sub.IRAP]-scores on the self-focused IRAP trial types and specified other-focused trial types in Experiment 6. Positive [D.sub.IRAP]-scores indicate history-consistent responding and negative [D.sub.IRAP]-scores indicate history-inconsistent responding. ""Indicates [D.sub.IRAP]-scores that are significantly different from zero
Table 1. Natural language
statements from the other-focused
IRAP

Trial type           Stimuli

Positive Event-      If others hear that they look good,
Positive Reaction    their self-esteem enhances.
                     If the enemy of other people dies,
                     they feel liberated.
                     It makes other people happy if
                     they win the lottery.
                     People will be proud if they
                     succeed in their exams.

Positive Event-      People don't like it when they
Negative Reaction    are told they look good.
                     If other people's enemy dies they become angry.
                     Other people are disappointed
                     when they win the lottery
                     Others are frustrated when they pass an exam.

Negative Event-      Other people are happy when a loved one dies.
Positive Reaction    Getting fines makes other people happy.
                     Others feel lucky if they fail an exam.
                     If another person's enemy wins the lottery,
                     they will be happy.

Negative Event-      Other people feel despair when a loved dies.
Negative Reaction    Getting a fine makes other people angry.
                     If others fail an exam then
                     they are disappointed.
                     Other people become bitter
                     if their enemy wins the lottery.

Note: Statements were presented to participants in Dutch.
Trial type labels denote each of the two parts of the
statement, but not necessarily the sequence in which
they appeared in the statement

Table 2. Natural language statements
from the self-focused IRAP

Trial-types          Stimuli

Positive Event-      My self-esteem increases
Positive Reaction    if someone says I look good.
                     I feel liberated if my enemy dies.
                     Winning the lottery makes me happy.
                     I'm proud when I succeed in my exams.

Positive Event-      I feel ugly if someone says I look good.
Negative Reaction    I'm angry if my enemy dies.
                     Winning the lottery disappoints me.
                     It frustrates me if I succeed in my exams.

Negative Event-      I'm happy if a loved one dies.
Positive Reaction    Getting a fine makes me happy.
                     Failing an exam is fantastic.
                     I rejoice if someone I hate wins the lottery.

Negative Event-      If a loved one dies, I'm miserable.
Negative Reaction    Getting a fine makes me angry.
                     Failing an exam is disappointing.
                     It irritates me if someone
                     I hate wins the lottery.

Note: Statements were presented to participants in Dutch.
Trial type labels denote each of the two parts of the
statement, but not necessarily the sequence in which
they appeared in the statement

Table 3. Experiment 1 Fisher's PLSD Comparisons

Trial type                            Mean Difference   P

Positive Event-Positive Reaction/     -0.06             .25
Positive Event-Negative Reaction
Positive Event-Positive Reaction/     0.13              <.01 *
  Negative Event-Positive Reaction
Positive Event-Positive Reaction/     -0.13             <.01 *
  Negative Event-Negative Reaction
Positive Event-Negative Reaction/     0.18              <.001 *
  Negative Event-Positive Reaction
Positive Event-Negative Reaction/     -0.07             .14
  Negative Event-Negative Reaction
Negative Event-Positive Reaction/     -0.25             <.001 *
  Negative Event-Negative Reaction

* Indicates significant p values

Table 4. Experiment 2 Fisher's PLSD Comparisons

Trial type                           Mean Difference      p

Positive Event-Positive Reaction/         -0.03          .51
Positive Event-Negative Reaction
Positive Event-Positive Reaction/         0.09           .05
Negative Event-Positive Reaction
Positive Event-Positive Reaction/         -0.07          .13
Negative Event-Negative Reaction
Positive Event-Negative Reaction/         0.12         <.01 *
Negative Event-Positive Reaction
Positive Event-Negative Reaction/         -0.04          .39
Negative Event-Negative Reaction
Negative Event-Positive Reaction/        -0.159        <.001 *
Negative Event-Negative Reaction

* Indicates significant p values

Table 5. Experiment 3 Fisher's PLSD Comparisons

Trial type                           Mean Difference      P

Positive Event-Positive Reaction/         0.15          <01 *
Positive Event-Negative Reaction
Positive Event-Positive Reaction/         0.25         <0001 *
Negative Event-Positive Reaction
Positive Event-Positive Reaction/          0.1          <04 *
Negative Event-Negative Reaction
Positive Event-Negative Reaction/          0.1          <03 *
Negative Event-Positive Reaction
Positive Event-Negative Reaction/         -0.05          .31
Negative Event-Negative Reaction
Negative Event-Positive Reaction/         -0.15         <01 *
Negative Event-Negative Reaction

* Indicates significant p values

Table 6. Experiment 4 Fisher's PLSD Comparisons

Trial type                           Mean            P
                                     Difference

Positive Event-Positive Reaction/    0.04           .49
Positive Event-Negative Reaction
Positive Event-Positive Reaction/    0.27         <.0001 *
Negative Event-Positive Reaction
Positive Event-Positive Reaction/    -0.005         .96
Negative Event-Negative Reaction
Positive Event-Negative Reaction/    0.23         <.0001 *
Negative Event-Positive Reaction
Positive Event-Negative Reaction/    -0.04          .45
Negative Event-Negative Reaction
Negative Event-Positive Reaction/    -0.28        <0001 *
Negative Event-Negative Reaction

* Indicates significant p values

Table 7. Experiment 5 Fisher's PLSD Comparisons

Trial type                           Mean Difference      P

Positive Event-Positive Reaction/    0.03                .51
Positive Event-Negative Reaction
Positive Event-Positive Reaction/    0.12              0.01 *
Negative Event-Positive Reaction
Positive Event-Positive Reaction/    -0.06               0.2
Negative Event-Negative Reaction
Positive Event-Negative Reaction/    0.09               0.05
Negative Event-Positive Reaction
Positive Event-Negative Reaction/    -0.09              0.05
Negative Event-Negative Reaction
Negative Event-Positive Reaction/    -0.18             <0.01 *
Negative Event-Negative Reaction

* Indicates significant p values

Table 8. Experiment 6 Fisher's PLSD Comparisons

Trial type                           Mean Difference     P

Positive Event-Positive Reaction/    0.07               .18
Positive Event-Negative Reaction
Positive Event-Positive Reaction/    0.21              <.01 *
Negative Event-Positive Reaction
Positive Event-Positive Reaction/    -0.006             .90
Negative Event-Negative Reaction
Positive Event-Negative Reaction/    0.14              <.02 *
Negative Event-Positive Reaction
Positive Event-Negative Reaction/    -0.08              .14
Negative Event-Negative Reaction
Negative Event-Positive Reaction/    -0.21             <.01 *
Negative Event-Negative Reaction

* Indicates significant p values

Table 9. Summary of the main findings for the six experiments

Experiment             Difference in overall   IRAPs
                       D-IRAP score between    Correlated
                       Self and Other IRAPs

Experiment 1           Yes                     No
Self and
Unspecified Other

Experiment 2           Yes                     r (49) = 0.315,
Self and                                       p = 0.027
Unspecified Other

Experiment 3           No                      r (33) = 0.414,
Self and Specified                             p = 0.016
Other (closest to),
with focusing task

Experiment 4           No                      No
Self and Specified
Other (closest to),
without focusing
task

Experiment 5           No                      r(30) = 0.410,
Self and Specified                             p = 0.023
Other (dislike)

Experiment 6           No                      No
Self and Specified
Other
(different from)

Experiment             Correlations between IRAPs        Corrected
                       and self-reports (uncorrected)    Correlations

Experiment 1           IRAP difference score             None
Self and                 with CAPE (positive
Unspecified Other        symptoms/frequency)
                       [K37) = 0.34, p = 0.039]

Experiment 2           Other-focused IRAP with CAPE      None
Self and                 (positive symptoms/frequency)
Unspecified Other      [K49) = -0.3, p < 0.04]
                       Other-focused IRAP with ECR-RS
                         attachment-related avoidance
                         subscale for romantic partner
                       [r(49) = 0.34, p < 0.02]
                       and for mother
                       [K49) = 0.31, p < 0.03]

Experiment 3           Self-focused IRAP with CAPE       None
Self and Specified       (negative symptoms/frequency)
Other (closest to),    [r(33) = 0.38, p < 0.03]
with focusing task     and with Self-warmth
                         Thermometer
                       [r(33) = 0 .49, p < 0.01]
                       Specified other-focused IRAP
                         with ECR-RS attachment-
                         related avoidance subscale
                         for one's mother
                       [r(33) = -0.43, p < 0.01]
                       IRAP difference score with
                         ECR-RS attachment-related
                         avoidance for one's mother
                       [r(33) = 0.41, p < 0.02]

Experiment 4           Self-focused IRAP with EOSS       None
Self and Specified       close relationships-absent
Other (closest to),      subscale
without focusing       [r(30) = 0.47, p < 0.01]
task                   Specified other-focused IRAP
                         with ECR-RS attachment-
                         related avoidance subscale
                         for one's mother
                       [r(30) = -0.42, p< 0.05]
                         one's father
                       [r(30) = -0.391, p < 0.04]
                         with ECR-RS attachment-
                         related anxiety subscale
                         for best friend
                       [r(30) =-0.38, p< 0.04]
                       and with the overall PFI score
                       [r(30) =-0.42, p< 0.03]
                       IRAP difference score with
                         ECR-RS attachment-related
                         avoidance subscale for
                         romantic partner
                       [r(30) =-0.38, p< 0.04]

Experiment 5           Self-focused IRAP with ECR-RS     None
Self and Specified       attachment-related avoidance
Other (dislike)          subscale for one's mother
                       [>(30) = -0.38, p < 0.04]

Experiment 6           Self-focused IRAP with ECR-RS     None
Self and Specified       attachment-related avoidance
Other                    subscale for romantic partner
(different from)       [r(30) = -0.41, p < 0.03]
                       and with the EOSS close
                         relationships-present
                         subscale
                       [r(30) = -0.43, p < 0.02]
                       Specified-other IRAP with the
                         EOSS close relationships-
                         present subscale
                       [r(30) = - 0.39, p = 0.03]
                       IOS for best friend
                       [r(30) = - 0.41, p = 0.03]
                       and with the Self Warmth
                         Thermometer
                       [r(30) = -0.46, p < 0.01]
COPYRIGHT 2019 Springer
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:ORIGINAL ARTICLE
Author:Kavanagh, Deirdre; Roelandt, Adeline; Van Raemdonck, Lisa; Barnes-Holmes, Yvonne; Barnes-Holmes, Der
Publication:The Psychological Record
Date:Jun 1, 2019
Words:16090
Previous Article:Relatedness of Equivalence Class Members: Combined Effects of Nodality and Relational Type.
Next Article:Conditioned Inhibition and its Relationship to Impulsivity: Empirical and Theoretical Considerations.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters