From acceptance to change: The role of acceptance in the effectiveness of the Informative Process Model for conflict resolution

Abstract

The Informative Process Model (IPM) proposes an intervention to facilitate change in conflict‐supporting narratives in protracted conflicts. These narratives develop to help societies cope with conflict; but over time, they turn into barriers for its resolution. The IPM suggests raising awareness of the psychological processes responsible for the development of these narratives and their possibility for change, which may unfreeze conflict attitudes. Previous studies in the context of the Israeli–Palestinian conflict found that the IPM (versus control) increased participants' support for negotiations. In three preregistered studies (combined N = 2,509), we illuminate the importance of feeling that one's conflict‐related attitudes are accepted–that is, acknowledged without judgement–in explaining the effectiveness of the modeland expand the IPM's validity and generalizability: By showing the effectiveness of the IPM compared to an intervention similarly based on exposure to conflict‐related information (Study 1); by showing its effectiveness in unfreezing attitudes when communicating different thematic conflict‐supporting narratives–victimhood and security (Study 2); and by showing its effectiveness when using messages referring to ongoing, not only resolved conflicts, and text‐based, not only visually stimulating, message styles (Study 3). These results contribute to theory and practice on psychological interventions addressing the barrier of conflict‐supporting narratives.

Research in social psychology suggests that one reason that violent conflicts endure, despite their immense costs to individuals and societies, lies in the ‘ethos of conflict’ (EOC): a corpus of societal beliefs and attitudes in protracted conflict, organized as conflict‐supporting narratives that imbue the conflict with meaning, but in doing so also obscure alternative perspectives (Bar‐Tal et al., 2012). A dynamic thus develops whereby narratives formed to help societies with coping at times of conflict, turn over time into barriers for conflict resolution themselves (Bar‐Tal, 2020). The Informative Process Model (IPM) (Rosler et al., 2022) suggests that awareness of the psychological processes responsible for the development of conflict‐supporting narratives can help overcome them. The model is unique in offering a dialectic approach based on a combination of messages of acceptance and change. The messages of acceptance towards conflict‐supporting narratives, as functional and natural psychological mechanisms, reduce resistance to new information, whilst the messages that show the narratives' costs and malleability motivate change.

Following the establishment of the IPM as an effective intervention for unfreezing conflict‐supporting attitudes (Rosler et al., 2022), in the present studies, we had three main objectives. First, to empirically test whether feeling acceptance of one's conflict‐supporting attitudes is indeed one of the underlying psychological mechanisms that drive the effectiveness of our model. Second, to empirically address a concern that in previous work on the IPM, the effects were merely driven by engagement with any type of conflict‐related information, given that the IPM messages were compared to a control group that included non‐conflict related information (ads). Third, we aimed to expand the generalizability and validity of the IPM‐based intervention, examining its effectiveness when using messages addressing two new conflict‐supporting narrative themes and narratives from ongoing – not only resolved – conflict contexts, in text‐based – but also visually stimulating – message styles.

The EOC consists of narratives around eight themes¹ that serve basic needs jeopardized by conflict – such as the need for forming meaning, understanding, predictability and a positive self‐image – enabling individuals and groups to cope with the immense challenges of life under conflict. Given their importance, these narratives are constantly reinforced in society through processes of social influence as individuals seek, for example, assimilation or conformity (Flache et al., 2017). Yet with these reinforcements comes a ‘freezing’ of perspectives: the EOC can be understood as a type of social representation whereby the combination of ideas, thoughts and feelings of many individuals become ‘an object for a social group’ (Wagner et al., 1999, p. 96). Such hardening of collective thinking also limits one's views, repeatedly reaffirming what may be divisive beliefs, and preventing seeing alternative perspectives (Baumeister & Hastings, 1997), even when opportunities for peace arise (Cobb, 2013). These dynamics are particularly detrimental in intractable conflicts whose enduring, existential and all‐encompassing nature (Kreisberg, 2010) increases the psychological need for conflict‐supporting narratives, whilst allowing them to take root in society. Consequently, given their process of social ‘objectification’, they become a barrier to conflict resolution as attempts to change these narratives are met with resistance by society members.

Researchers have proposed various interventions to address this challenge; but their effectiveness is limited due to the strong resistance to change that persists in intractable conflict contexts (Bar‐Tal & Hameiri, 2020). One reason for this may be that most interventions advocate for the use of messages that contradict conflict narratives by trying to correct ‘wrong’ beliefs of society members, eliciting dissonance that motivates attitude change, for example, through peace education (Adwan & Bar‐On, 2003; Bekerman & Zembylas, 2011), mass media (Littman et al., 2023) or intergroup contact (Pettigrew & Tropp, 2006). Yet, research has shown that contradicting one's attitudes can be experienced as threatening, thus counterproductively bolstering existing perspectives (Chen & Chaiken, 1997), or eliciting resistance to the intervention from the get‐go (Berndsen et al., 2018; Ron et al., 2017).

In an attempt to overcome this limitation, scholars looked for more subtle ways of influencing conflict‐related attitudes. One way was through exposing inconsistencies with conflict narratives indirectly, by informing people about particular exemplars from the outgroup (Čehajić‐Clancy & Bilewicz, 2021), or through a different intergroup context or even without a specific context (Halperin et al., 2011), expecting participants to draw conclusions concerning their specific context that were inconsistent with their views (Bar‐Tal & Hameiri, 2020). A second approach tried to avoid this dissonance altogether through exposure to consistent information with one's beliefs as a first step towards changing them; for example, through self‐affirmation (Cohen et al., 2007), affirming conflict‐related group pride (Schori‐Eyal et al., 2015), mirroring conflict narratives in their extreme form leading to paradoxical thinking (Hameiri et al., 2019) or combining narrative elements of different conflict parties (Garagozov & Gadirova, 2019). Whilst consistent, these interventions directly engage individuals with their ingroup conflict narratives, which can be challenging. Similarly, before challenging individuals, the IPM approach tries to raise as little resistance as possible by exposing individuals to information about conflict narratives that are consistent with their beliefs and doing so indirectly through comparisons with other conflictual contexts.

IPM AND THE ACCEPTANCE‐CHANGE DIALECTIC

We draw on literature from both attitude change and clinical psychotherapy, grounding the IPM (Rosler et al., 2022) on the acceptance‐change dialectic, combining sequentially an expression of understanding towards one's extant views with motivators to change them. Such dialectic has been proven particularly non‐threatening and effective in facilitating unfreezing of deep‐seated perspectives (Fielding et al., 2020; Hayes & Strosahl, 2004; Hornsey & Fielding, 2017). Various studies have shown that acknowledgement – whether due to reciprocity, high‐quality listening or other mechanisms – can increase openness towards alternative viewpoints (Cialdini & Goldstein, 2004; Hameiri & Nadler, 2017; Hornsey & Fielding, 2017; Itzchakov et al., 2017; Twali et al., 2020; Xu & Petty, 2022), and in clinical psychotherapy can promote change in dysfunctional patterns by giving patients the sense that their beliefs are valid and normal before challenging them to change (Brodsky & Stanley, 2013; Jacobson et al., 2000; Linehan, 1997).

IPM‐based messages emphasize four elements (Bar‐Tal & Hameiri, 2020; Rosler et al., 2022). First, describing the functional evolvement of conflict‐supporting narratives to fulfil basic human needs (Maslow, 1970). Second, communicating their prevalent and universal nature in intractable conflicts. Third, communicating the role of conflict‐supporting narratives in fueling continued conflict and its related societal costs (Bar‐Tal, 2020) and fourth, communicating the possibility for change. Whilst the first two elements communicate acceptance, the second two elements motivate change.

Previous IPM studies were all conducted in the Israeli–Palestinian context – often referred to as a typical example of a protracted conflict and one in which the EOC is strongly present, influencing its intractability (Bar‐Tal, 1998). In two experimental studies evaluating the IPM, exposure to IPM‐based messages significantly increased support amongst Israeli Jews for Israeli–Palestinian negotiations in comparison to a control group (Rosler et al., 2022). In these studies, Jewish Israelis were exposed to two to four short (40‐s) video‐based messages communicating the IPM principles. Each video began by communicating examples of conflict‐supporting narratives from a distal conflict context and continued by conveying its peaceful resolution. By watching several videos, participants were exposed to examples from different conflicts (e.g., Northern Ireland, France‐Algeria, Spanish‐Basque and Guatemala) exemplifying the universal and malleable nature of conflict‐supporting narratives. Whilst similar to the Israeli narratives, the use of distal conflicts – a unique feature of the IPM intervention compared to other acknowledgement‐based interventions – helped create further distance that reduced threat from the message content, creating an opening for attitude transformation.

Whilst initial evidence is promising, given its novel nature, previous IPM studies suffer from three primary limitations. First, whilst based theoretically on the acceptance‐change dialectic, there is no empirical evaluation of the intervention's influence on feeling accepted. Second, using messages representing only two conflict‐supporting narrative themes out of the eight themes of the EOC limits the generalizability and external validity of the intervention, suggesting that it might be the specific theme rather than the model that is responsible for its effectiveness. Third, the use of strong audiovisuals and examples only from conflicts that ended peacefully, as well as comparison with a control group that is non‐conflict related, may account for the intervention effects, rather than the message contents as the IPM proposes.

ACCEPTANCE

In our current studies, we empirically evaluate the role of acceptance in the acceptance‐change dialectic. Psychological literature typically defines ‘acceptance’ as an action from the perspective of those offering acceptance or as a quality or state from the perspective of those receiving it. Scholars emphasize the function of acceptance in engaging nonjudgmentally with (often negative) experiences of themselves and others (rather than avoiding or trying to control them), which can help individuals accept themselves and can open them to changing their feelings, attitudes and behaviours (Herbert & Forman, 2011; Williams & Lynn, 2010). This rationale is the basis for acceptance‐based psychotherapeutic approaches (e.g., Hayes et al., 2004; Jacobson et al., 2000; Linehan, 1997).

Other scholars focus on the content of acceptance, such as feeling belonging (vs. rejection; Baumeister & Leary, 1995; Kunstman et al., 2013), love and affection (Rohner, 2004) or feeling understood (Livingstone et al., 2019, 2020), with the goal of improving interpersonal relations between those offering and receiving acceptance. However, this is still the result of internal processes. For example, in the context of marital relations, researchers (Jacobson et al., 2000; Jacobson & Christensen, 1996) show acceptance by a spouse reduces external pressure to change, and therefore also the internal rigidity and defensiveness that prevent such change to begin with; whilst in the context of interracial group relations, Kunstman et al. (2013) show that by fulfilling the need for belonging and reducing fear of rejection, outgroup acceptance increases internal motivation to respond without prejudice to the outgroup. Finally, Livingstone et al. (2020) explain the value of feeling understood for improving intergroup relations by quoting Rogers (2011, p. 323): ‘…it is such a relief, such a blessed relaxation of defenses, to find oneself understood’. We find such ‘relaxation of defenses’ particularly important in conflict contexts given the high degree of psychological defences leading to resistance to changing conflict‐supporting narratives.

Importantly, in intractable conflict contexts, mere positive interactions with an adversary – let alone expressions of acceptance – can be difficult to find (Paolini et al., 2018; Ron et al., 2017), making this an impractical approach to attitude changes. Instead, we are interested in how a more generic sense of perceived external acceptance influences one's conflict‐related attitudes. We thus conceptualize acceptance in the context of intractable conflicts as a subjective perception of having one's group attitudes be externally validated without judgement. Acceptance according to our definition does not demand feeling that others agree with you, but rather that they genuinely and non‐judgmentally understand your perspective. Furthermore, it does not require actual acceptance, but its perception in the eyes of the receivers. Finally, it does not need to be a specific person or group offering acceptance, but a general feeling of being externally validated.

We theorize that acceptance, as we define it, has two foundations: similarity and legitimacy. Literature on the similarity‐attraction effect suggests similarity of attitudes satisfies the human drive to be logical and correct and increases certainty (Byrne, 1969; Byrne et al., 1971). Unlike the physical world, however, it is hard to validate attitudes as ‘correct’ through the ‘objective’ reality, leading people to seek such validation through comparison with others' attitudes. Singh et al. (2017) quote Festinger (1950) who noted that ‘an opinion, a belief, an attitude is “correct”, “valid” and “proper” to the extent that it is anchored in a group of people with similar beliefs, opinions, and attitudes’. (p. 272). Thus, feeling attitude acceptance is based on feeling that others have similar perspectives, normalizing one's own. Adding to the literature, we argue that when it comes to group attitudes, such as those developed in intractable conflicts, it is not enough to validate them through comparison with the attitudes of other ingroup members. This would be akin to self‐acceptance, which is already high in contexts of intractable conflicts (Bar‐Tal, 2020) and thus likely to be less effective for eliciting attitude change – rather, validation comes from comparison with other groups' attitudes. In the case of intractable conflicts, this would mean comparison with other groups suffering violent conflicts, which would have similar conflict‐related attitudes.

Yet, similar attitudes can also be framed, for example, as unacceptable or illegitimate, for purposes of condemnation (Guelke, 2004). Thus, our second foundation for acceptance, in addition to similarity, is legitimacy, by which we mean a nonjudgmental validation of conflict attitudes. Dialectical Behavioural Therapy (DBT), which – like our IPM intervention – is also based on the acceptance‐change dialectic, conceptualizes acceptance as validation by the therapist that client perspectives are ‘reasonable in the moment’ and have a well‐founded internal logic (Linehan, 1997, p. 370). Similarly, on the intergroup level, Livingstone et al. (2019) define ‘felt understanding’ through a conceptualization of acceptance that emphasizes its non‐judgmental nature, namely, ‘…the belief that members of an outgroup understand and accept the perspectives of ingroup members… By “accept”, we specifically mean the belief that outgroup members accept ingroup members' perspectives as authentic and subjectively valid…and having a non‐judgmental/non‐dismissive stance towards those perspectives…even if outgroup members disagree with those perspectives’ (p. 634). In sum, a key aspect of external acceptance is having one's attitudes validated and legitimized, given the circumstances within which they were developed.

Based on the above, our hypothesis is that, in the context of intractable conflicts, perceived external acceptance, based on perceived group‐attitude similarity and legitimacy, can unfreeze conflict‐supporting narratives, dialectically increasing openness towards changing them. We theorize that perceived external acceptance of conflict attitudes would relax the (emotional and cognitive) defences in Rogerian terms, increasing unfreezing in turn. As mentioned, such reduction in attitude defensiveness has been effective in other acknowledgement‐based psychological interventions, such as paradoxical thinking (Hameiri et al., 2019), exposing people to messages consistent with their attitudes to change them. When it comes to change in the acceptance‐change dialectic, we argue that the psychological openings that can ‘unfreeze’ strongly held beliefs entail a two‐step process, including first an openness towards new generic information, and then, reconsidering context‐specific conflict‐supporting attitudes. We operationalize this in a mediation model presented in Studies 2 and 3, whilst acknowledging the limitations of mediation models in establishing causality (Rohrer et al., 2022), based on the same sequential logic whereby there is a need first for acceptance that reduces defences, allowing then for consideration of change. We argue that this model offers a valuable theoretical contribution explaining the potential causal relationship between our novel mediator of ‘acceptance’ and our preexisting measures of unfreezing.

THE PRESENT RESEARCH

Replicating previous IPM studies (Rosler et al., 2022), we focus on the Israeli–Palestinian conflict, conducting our pre‐registered experimental studies amongst Israeli Jews. In the current studies, we advance our understanding of the mechanisms explaining the effectiveness of the IPM by evaluating the importance of feelings of acceptance towards one's conflict‐supporting narratives as a mediator. Additionally, we increase the generalizability of the intervention and validate it by controlling for potential confounds. Ensuring the intervention's impact is not due merely to its engagement with conflict‐related information; in Study 1, we compare the impact of the IPM‐based intervention not only with a generic control group but also with another conflict‐focused intervention. In Study 2, with the intention of expanding the generalizability of the IPM, we evaluate two new conflict‐supporting narratives, victimhood and security, and establish whether certain narrative themes are more effective than others for attitude transformation. In Study 3, assessing ongoing not only resolved conflicts, we control for the possibility that exposure to resolved conflicts leads to a change of attitudes due to an increase in beliefs regarding the malleable nature of conflicts and related hopefulness (Cohen‐Chen et al., 2017). Additionally, using text‐based only messages with no visual stimuli or background narration, we control for the potential influence of a well‐produced video clip with strong audiovisuals that can elicit strong emotions compared to other modes of message delivery (see Dawtry et al., 2020) and advance transportation – an immersion in the narrative – which is effective for attitude change (Green, 2021). Finally, we improve the comparability between the intervention and control conditions. Generic commercials – substantially different in style from the intervention – served as control videos in previous IPM, and other, studies (Rosler et al., 2022; see also Bruneau et al., 2022; Hameiri et al., 2018). Therefore, in Study 3, we specially designed commercials to be stylistically identical to our intervention, for example, using the same fonts, letter sizes, colours, number of sentences, balance between quotations and generic statements and background music.

STUDY 1

This preregistered study (https://osf.io/536yh?view_only=true) compares an IPM‐based intervention with another conflict‐related intervention – a “Decisional Balance” (DB) task – assessing their effect on readiness to change attitudes towards intractable conflicts. The DB task asks study participants to write reasons for and against changing attitudes regarding their own conflict, measuring their degree of ambivalence which can relate to different stages of change (from pre‐contemplation to contemplation to action). Studies have proven the DB to be amongst the best predictors of future attitude change (e.g., Velicer et al., 1985). Our original purpose of this preregistered study was to test whether exposure to a DB task would amplify the effect of the IPM‐based message, and we hypothesized that it would. We thus also compared these two conditions with a third condition combining DB and IPM sequentially, and with a control group. In considering the pros and cons of changing conflict attitudes, the DB engages study participants with conflict‐related content, like the IPM, but operationalized differently. Comparing the two will offer additional evidence supporting our argument that the effectiveness of the IPM is not due merely to its conflict‐related content, but rather to its underlying principles.

Additionally, unlike in previous IPM studies, in this study, we conceptualized and operationalized unfreezing through two distinct elements. First, a generic openness to novel information (which we refer to as the first step in ‘unfreezing’) and second, openness to consider change related to one's specific conflict context (in this case, the Israeli–Palestinian conflict) (which we refer to as the second step in ‘unfreezing’). We hypothesized that, compared to only the DB task or exposure to IPM‐based messages, their sequential combination would lead to greater deliberation of new information, reconsideration of currently held beliefs regarding the Israeli–Palestinian conflict, and support for more conciliatory policies (i.e., negotiation) to address the conflict.

Method

Participants

Sensitivity power analysis using G*Power 3.1.9.7 indicated that 75 participants per group, or a sample size of 300 participants, provides 80% power to detect an effect size (f) of 0.162 and 90% power to detect an effect size (f) of 0.188, which are small effect sizes (partial η ² = 0.025–0.035). Therefore, we sought to recruit at least 300 participants. Because we assumed that some would be excluded based on pre‐registered criteria, 410 Israeli Jewish participants were recruited. Advertising for participants was done through the research participation system of Haifa University and through student‐associated social media groups. The study took place between 2nd January and 19th February 2023. Fifty‐five participants were omitted for failing to correctly answer one or more attention check questions embedded in the questionnaire, 31 participants were omitted for completing the study in less than the minimum allowed time (6 min for the control group, 7 min for IPM and DB conditions and 8 min for combined condition), and two were omitted for completing the study over the maximum allotted time (35 min), as we preregistered. The remaining 328 participants (246 women and 81 men) had a mean age of 27.59 (SD = 8.85) and as for political identification, 49.3% identified as rightists, 22.9% as centrists and 27.8% as leftists.

Procedure and materials

In this study, Jewish Israeli participants were randomly assigned to one of four conditions: The IPM‐group (n = 83) received IPM‐based messages (two video clips about intractable conflicts, similar to those used in Rosler et al., 2022; https://youtu.be/E3TRkMJ57EI). The DB‐group (n = 87) was asked to complete a DB task listing possible reasons for first maintaining and then for changing their current beliefs related to the Israeli–Palestinian conflict. Research has shown that weighing such potential benefits and losses associated with attitude change can predict actual change of behaviour in the future (Prochaska et al., 2006). The DB‐IPM group (n = 82) first completed a DB task and then watched IPM‐based messages sequentially. The control group (n = 76) was asked to provide a list of things they did that day so far. After completing the DB or control tasks and/or watching the IPM‐based videos, participants responded to the dependent measures.

The IPM‐based messages included a combination of two video‐clips (in Hebrew, each 40 s long), in randomized order, each following the theoretical underpinnings of the IPM, tailor‐made for this purpose in collaboration with a professional video producer. Each clip exemplifies one conflict‐supporting narrative theme, and uses either the conflict in Northern Ireland, or the French‐Algerian War as examples. Each video clip includes two parts. First, one conflict‐supporting narrative is expressed through a series of quotes. Whilst not mentioning the Israeli–Palestinian conflict explicitly, these quotes lead the viewers to think about it given the similarity across conflict narratives. Only towards the end of this part of the video, the identity of the speaker is revealed to be a fighter from a foreign conflict – either Northern Ireland or the French‐Algerian War. In the second part of the clip, the peaceful resolution of the same conflict is communicated, underscoring the message that conflict is not a necessary reality, rather one can choose otherwise (see videos in the Appendix S1).

Measures

Deliberation of new information was measured using four items (α = .81; based on Rosler et al., 2022) assessing the extent to which the tasks completed conveyed information that was new to study participants (e.g., ‘please rate the extent to which the tasks so far conveyed new information that you were not aware of’) and made participants engage deliberatively with it (e.g., ‘Please rate the extent to which the tasks so far made you think deeply about the conveyed messages’).

Reconsidering attitudes regarding the Israeli–Palestinian conflict were measured (on a ‘slider’ scale ranging from 0 = not at all, I am just as convinced in my views on the topic, to 100 = v ery much so, I am reconsidering my views on the topic) using five items (α = .85). These items assessed the transference of the intervention messages to the Israeli–Palestinian context, through the extent to which the interventions made participants reconsider (and thus ‘unfreeze’) their attitudes on various aspects of the Israeli–Palestinian conflict. In addition to a general item that was included in the previous studies (‘to what extent did the videos make you reconsider your attitudes pertaining to the Israeli–Palestinian conflict’; Rosler et al., 2022), we used four new items. These focused on specific aspects of the conflict (e.g., ‘to what extent did the tasks so far make you reconsider your attitudes pertaining to the cost of continued conflict for Israel’) and the perceptions of the ingroup and outgroup (e.g., ‘to what extent did the tasks so far make you reconsider the way you perceive Israelis/Palestinians in the conflict’).

Support for negotiations was measured using three items (α = .80) assessing participants' support for negotiations towards three different outcomes: achieving peace between Israelis and Palestinians, a long‐term truce between Hamas and Israel, and achieving peace based on the Arab Peace Initiative (Rosler et al., 2022).

Demographic characteristics of the participants were collected in all three studies, including sex, age, place of birth, place of birth of parents, level of education, religious identification and political orientation, assessed on a scale ranging from 1 = strong right to 7 = strong left.²

Results

For means, standard deviations and correlations across all measured variables see Table 1. We ran a series of one‐way ANCOVAs for each of our dependent variables (see means and SDs for each condition in Table 2).³ Deviating from our preregistration, and to be consistent with the following studies, we controlled for political orientation, gender and age. Not controlling for these variables had no effect on the results, except for when evaluating reconsidering attitudes regarding the Israeli–Palestinian conflict (see below). There were no differences between conditions in terms of age (all ps > .165) or gender (p = .770). There were also no differences between conditions in terms of political orientation (all ps > .498).

TABLE 1.

Study 1 means, standard deviations (SD) and Pearson correlation matrix for dependent and control variables.

Variable	M	SD	1	2	3	4	5	6
1. Deliberation	2.84	1.29	–
2. Reconsider attitudes	24.16	20.08	.41**	–
3. Support for negotiations	4.19	1.20	.05	.18**	–
4. Political orientation	3.61	1.48	.04	.19**	.47**	–
5. Gender	1.75	0.43	−.01	−.02	.04	.01	–
6. Age	27.59	8.84	.10	.14*	−.02	−.02	−.13	–

^* p < .05.

^** p < .01.

TABLE 2.

Study 1 means and standard deviations (SD) for dependent variables in each condition.

Conditions	Deliberation		Reconsider attitudes		Support for negotiations
	M	SD	M	SD	M	SD
IPM	3.37a	1.24	28.96a	22.81	4.21a	1.21
DB	2.34b	1.02	22.11b	18.49	4.23a	1.30
DB‐IPM	3.34a	1.14	22.09b	18.45	4.08a	1.20
Control	2.24b	1.26	23.16ab	19.66	4.27a	1.10

Note: Controlling for political orientation, gender and age. Means not sharing the same subscript letter are significantly different from each other at p < .05.

Deliberation of new information

The analysis yielded a significant main effect (F(3,320) = 29.65, p < .001, partial η ² = 0.169) of experimental condition. Paired comparisons revealed that the IPM condition led to greater deliberative processing of new information compared to the control condition and DB condition (ps < .001), but there were no significant differences compared to the DB‐IPM condition (p = .771). The DB‐IPM condition also led to greater deliberation of new information compared to the control and DB conditions (ps < .001). There were no significant differences between the control and the DB condition (p = .654).

Reconsider attitudes regarding the Israeli–Palestinian conflict

The analysis did not yield a significant main effect (F(3, 320) = 2.48, p = .061, partial η ² = 0.023).⁴ Paired comparisons revealed that the IPM condition led to greater reconsideration of attitudes regarding the Israeli–Palestinian conflict compared to the DB (p = .023) and DB‐IPM (p = .016) but not the control (p = .132). There were no other significant differences between conditions (all ps > .390).

Support for negotiations

The analysis did not yield a significant main effect (F(3,320) = .81, p = .490, partial η ² = 0.008). Paired comparisons revealed no significant differences between conditions (all ps > .133).

Discussion

The results of Study 1 show the value of the IPM‐based intervention to advance conflict‐related attitude transformations when compared not only to a generic control group but also when compared to another intervention that exposes study participants to conflict‐related content. The DB task offers a good comparison to the IPM‐based intervention as it similarly engages study participants with content related to conflict attitudes but focuses on internal cognitive argumentation. Unlike the IPM, the DB intervention does not entail any relational experience of messages of acceptance or change communicated by others. The study shows that IPM‐based interventions lead to significantly higher degrees of deliberation and willingness to reconsider conflict‐related attitudes, as compared to the DB task.

Unlike previous IPM studies (Rosler et al., 2022), the present study showed no superior effect of the IPM condition on support for conciliatory policies (i.e., negotiation). We believe this has to do with the deteriorating levels of belief amongst Israeli Jews that a political solution is feasible and that negotiations will lead to successful resolution of the Israeli–Palestinian conflict (Shikaki et al., 2023). Since the previous IPM studies, Israel has suffered unprecedented political instability, and the COVID‐19 pandemic led to reduced global attention towards peacemaking in the region, at the time in which the study was conducted, possibly making peace negotiations seem, at least to some of the study participants, as simply disconnected from reality.

STUDY 2

In this preregistered study (https://osf.io/me2zn?view_only=true), we tested the effectiveness of IPM‐based messages when communicating two new conflict‐supporting narrative themes, victimhood and security, in addition to the previously‐researched theme of delegitimization of the rival. We compared participants' conflict attitudes after being exposed to these three themes with one another and with a control group. Whilst other conflict narratives in the EOC describe either a feature of the in‐group (e.g., ‘patriotic’) or the rival (e.g., ‘delegitimized’), both victimhood and security‐related narratives articulate the interdependent relationship between them. Narratives around victimhood portray the in‐group as the often‐exclusive victim of conflict, thus providing it with moral high ground, a justification for the use of violence, and the legitimacy to seek justice, retribution and external support (Bar‐Tal et al., 2009; Vollhardt, 2020). Security narratives present the threats of violent conflict and articulate the importance of security (both tangible and intangible), providing a mental foundation for the use of violence to protect against such dangers (Bar‐Tal & Jacobson, 1998; Halabi et al., 2021).

Both narrative themes are thus detrimental to conflict resolution, limiting the possibility for conflict transformation, and motivating harmdoing. Following previous studies (Rosler et al., 2022), we hypothesized that the IPM‐based messages would lead once again to greater deliberation of new information and unfreezing of currently held beliefs regarding the Israeli–Palestinian conflict. We also hypothesized that the impact of the intervention would be consistent across thematic messages, and that acceptance would serve as a psychological mechanism, that is, a mediator of the effect of the intervention – exposure to information expressing the functionality and universality of conflict‐supporting narratives as well as motivators to change them – on attitude change.

Method

Participants

A priori power analysis indicated that a sample size of 844 participants would allow us to detect an effect size of partial η ² = 0.02, with 95% power in a one‐way ANOVA with four conditions. In previous studies of IPM‐based interventions, Rosler et al. (2022) found that effect sizes ranged from relatively large (partial η ² = 0.08) to small (partial η ² = 0.01). Therefore, to ensure high power to detect small effects, and considering our resources and the possibility that some participants may later be excluded, we increased the sample size. We therefore recruited 1094 Israeli‐Jews. Out of these respondents, 86 failed to answer the attention verification questions correctly and were taken out of the survey (these were relatively evenly divided across conditions). One thousand and eight participants (M _age = 43.22, SD _age = 16.08; 51.4% women) completed the survey through an online surveying company, Midgam, on 7–8 March 2021. In terms of political orientation, the sample resembled the Israeli‐Jewish population, as 54% were rightists, 24.6% were centrists and the remaining 21.4% were leftists. In exchange for participation, participants received 4.5 ILS (equivalent to $1.4). After watching the video, participants were asked to answer two attention verification questions before continuing to complete the dependent variable measures as detailed below, as well as some additional exploratory measures (for complete materials and data see Appendix S1).

Procedure and materials

Once agreeing to participate in the study, participants were randomly assigned to one of four conditions: IPM‐victimhood (n = 263), IPM‐security (n = 253), IPM‐delegitimization (n = 259) or a control condition (n = 233). In each of the experimental conditions, participants watched a combination of two IPM‐based video clips (each 40 s long), in randomized order, each following the theoretical underpinnings of the IPM and the respective narrative theme, tailor‐made for this purpose in collaboration with a professional video producer. Here, both clips exemplify the same conflict‐supporting narrative theme, but one uses the conflict in Northern Ireland, and the other the French‐Algerian War as examples. In this way, participants were exposed to the consistent nature of conflict narratives across conflict contexts. The structure of the video clips is identical to those used in Study 1 (see example here https://youtu.be/EaN9k4V0824 and all videos in the Appendix S1). In the control condition, participants watched a 1.5‐min video, which contained two generic television commercials unrelated to intergroup relations. After watching the videos, participants responded to a questionnaire.

Measures

We measured deliberation of new information (α = .85), reconsidering attitudes regarding the Israeli–Palestinian conflict (α = .91) and support for negotiations (α = .83) with items almost identical to Study 1. The only difference was that in the first two measures we asked directly about the influence of the ‘videos’ rather than ‘tasks’ on deliberation or reconsidering of conflict attitudes.

Acceptance of the Israeli perspective of conflict was measured using six items⁵ (α = .83) developed for this study. Four items, inspired by research on the value of learning from similar conflict contexts (Lustig, 2003) and of ‘feeling heard’ (Bruneau & Saxe, 2012; Itzchakov et al., 2018), measure the extent to which participants feel that perspectives in the videos resemble perspectives of others and their own perspective on the conflict (i.e., ‘in your opinion, how similar are the perspectives in the video to those of other groups suffering from violent conflict’; 1 = not similar at all, 6 = very similar). Two additional items assess the extent to which the videos made study participants feel like the perspectives of Israelis are understood and thus legitimate. Our first item measuring legitimacy (‘to what extent did watching the videos make you feel like your side of the conflict is understood’) was adapted from items measuring ‘felt understanding’ developed by Livingstone et al. (2019) (‘In general, [outgroup members] have a very good understanding of the views of [ingroup members]’; and ‘[Outgroup members] understand [ingroup] values’). We adapted these to assess acceptance from a generic rather than a specific outgroup. In addition, we also include an item capturing perceived explicit external legitimacy: ‘To what extent did watching the videos make you feel like the perspectives of most Israelis about the Israeli‐Palestinian conflict are legitimate’.⁶

Results

For means, standard deviations and correlations across all measured variables see Table 3. To examine our hypotheses, we ran a series of one‐way ANCOVAs for each of our dependent variables (see means and SDs for each condition in Table 4). Since we found that participants' political orientation, gender and age correlated with the dependent variables (see Table 3), and as pre‐registered, we controlled for these background variables throughout the statistical analysis. Not controlling for these variables had no effect on the results, except for controlling for political orientation when evaluating support for negotiations (see below). There were no differences between conditions in terms of age (all ps > .783) or gender (p = .095). There were also no differences between conditions in terms of political orientation (all ps > .146), except when comparing the victimhood and the delegitimization conditions, such that, compared to the IPM‐victimhood condition, participants in the IPM‐delegitimization condition were more leftists (p = .024).

TABLE 3.

Study 2 means, standard deviations (SD) and Pearson correlation matrix for dependent and control variables.

Variable	M	SD	1	2	3	4	5	6	7
1. Acceptance	3.26	1.15	–
2. Deliberation	2.96	1.32	.47**	–
3. Reconsider attitudes	26.86	23.72	.46**	.49**	–
4. Support for negotiations	4.03	1.43	.30**	.25**	.24**	–
5. Political orientation	3.36	1.44	.20**	.10**	.17***	.50**	–
6. Gender	1.51	0.50	.08**	.10**	.08*	.57	.10**	–
7. Age	43.22	16.08	−.08**	.03	−.16**	−.19**	−.23**	−.04	–

^* p < .05.

^** p < .01.

^*** p < .001.

TABLE 4.

Study 2 means and standard deviations (SD) for dependent variables in each condition.

Conditions	Acceptance		Deliberation		Reconsider attitudes		Support for negotiations
	M	SD	M	SD	M	SD	M	SD
IPM‐victimhood	3.54a	1.07	3.16a	1.31	29.28a	23.19	4.14a	1.46
IPM‐security	3.46a	1.03	3.12a	1.16	29.05a	24.37	4.00ab	1.44
IPM‐delegitimization	3.58a	0.99	3.38b	1.29	29.96a	24.26	4.07ab	1.43
Control	2.30b	1.06	2.09c	1.12	18.28b	20.79	3.91b	1.39

Note: Controlling for political orientation, gender and age. Means not sharing the same subscript letter are significantly different from each other at p < .05.

Acceptance

The analysis yielded a significant main effect (F(3,1001) = 76.46, p < .001, partial η ² = 0.186) of experimental condition. Paired comparisons revealed that all IPM conditions led to a greater sense of acceptance compared to the control condition (all ps < .001). All other comparisons were not significant (all ps > .187).

Deliberation of new information

The analysis yielded a significant main effect (F(3,1000) = 53.32, p < .001, partial η ² = 0.138) of experimental condition. Paired comparisons revealed that all IPM conditions led to greater deliberative processing of new information compared to the control condition (all ps < .001), and the delegitimization condition led to greater deliberation compared to both the security and victimhood conditions (both ps < .05). There were no significant differences between the victimhood and security conditions (p = .755).

Reconsider attitudes regarding the Israeli–Palestinian conflict

The analysis yielded a significant main effect (F(3,991) = 14.24, p < .001, partial η ² = 0.04). Paired comparisons revealed that all IPM experimental conditions led to a greater reconsideration of attitudes regarding the Israeli–Palestinian conflict, compared to the control condition (p < .001). No significant differences were found between the three IPM conditions (all ps > .656).

Support for negotiations

The analysis did not yield a significant main effect (F(3,1001) = 1.61, p = .185, partial η ² = 0.005). Paired comparisons revealed that compared to the control condition, support for negotiations was significantly higher in the IPM‐victimhood condition (p = .035) whilst all other comparisons were not significant (all ps > .150).⁷

Serial multiple mediation model

Given that the predicted effect of our intervention on the outcome variable support for negotiations was not confirmed in the security and delegitimization conditions, we did not pursue the mediation model originally envisioned and pre‐registered (see Appendix S1 for additional mediation analyses). Instead, we tested a serial mediation in which the outcome variable was reconsidering attitudes regarding the Israeli–Palestinian conflict. We then tested whether acceptance and deliberation of new information were mediating predictors of this new outcome variable (see Figure 1).

FIGURE 1.

Serial mediation model with outcome variable reconsider attitudes regarding the Israeli–Palestinian conflict. Results of the serial mediation model assessing the IPM‐based intervention. Coefficients are standardized, and the victimhood, security and delegitimization conditions are differentiated where relevant. *p < .05, ***p < .001.

To examine our serial mediation model, we used Hayes' (2018) PROCESS (model 6) with multi‐categorical IV using PROCESS indicator coding, comparing the control condition to the IPM‐victimhood (coded D1), IPM‐security (coded D2) and IPM‐delegitimization (coded D3). We found all comparisons yielded significant indirect effects (D1: indirect effect = β = .14, b = 3.32, SE = 0.02, 95% Confidence Interval (CI) = [0.10, 0.18]; D2: indirect effect = β = .13, b = 3.10, SE = 0.02, 95% CI = [0.09, 0.17]; D3: indirect effect = β = .14, b = 3.46, SE = 0.02, 95% CI = [0.10, 0.19]), indicating that both acceptance and deliberation of new information serially mediated the effects of the different IPM‐based interventions on reconsidering attitudes regarding the Israeli–Palestinian conflict.

Discussion

The results of Study 2 show the effectiveness of the IPM‐based intervention, not only when using the previously studied EOC narrative theme of delegitimization (Rosler et al., 2022) but also when conveying messages about group victimhood and security threat. They also show that the intervention leads message recipients to feel that their conflict‐related attitudes are accepted, which, in turn, is associated with a process of deliberation, ultimately leading to reconsideration of those same attitudes. Similarly to Study 1, the present study did not lead to significant changes in support for conciliatory policies (i.e., negotiation), except in the IPM‐victimhood condition, and given no changes in the political situation, we decided to leave this measure out of the main variables in Study 3.

STUDY 3

The aims of Study 3 were to substantiate the significance of our new psychological mechanism (feeling ‘acceptance’) as well as to affirm the IPM‐based intervention's validity by addressing potential confounds (for preregistration see https://osf.io/xahvt?view_only=true). To do so, we evaluated the effectiveness of the intervention when using examples of conflict‐supporting narratives from ongoing conflict contexts, not only ones that have already been resolved. In addition, we used text‐based videos rather than visually stimulating ones, and our control video was stylistically identical to the intervention videos. In these ways, we can further isolate the value of the IPM in acknowledging conflict‐supporting narratives through distal contexts regardless of their status of resolution or the videos' audio‐visual style.

Method

Participants

A‐priori power analysis indicated that 254 participants per group, or 762 altogether, will be required to detect a small effect size (partial η ² = 0.02) with .95 power in a one‐way ANOVA with three conditions. Given similar considerations as in Study 2, we once again aimed to recruit approximately 1000 participants. We therefore recruited 1173 Israeli‐Jews (M _age = 41.99, SD _age = 15.21; 50.1% men) who completed the survey through an online surveying company, Midgam, between 22 June and 5 July 2021. In terms of political orientation, the sample resembled the Israeli‐Jewish population, as 61.2% were rightists, 23.1% were centrists and the remaining 15.7% were leftists. In exchange for participation, participants received 4.5 ILS (equivalent to 1.4$). After watching the video, participants were once again asked two attention verification questions before continuing to complete the dependent variable measures as detailed below, as well as some additional exploratory measures (for complete materials and data see Appendix S1). It should be noted that due to unintentional mis‐phrasing of some of our attention verification questions, a substantial number of participants supposedly failed to answer them and were taken out of the survey. This was not evenly divided across conditions, as this mainly affected participants in the IPM‐ongoing condition. Specifically, out of the 1358 participants we originally recruited that completed the study, 352 (61 or 13.3% from the IPM‐resolved, 281 or 63.7% from the IPM‐ongoing and 10 or 2.2% from control) failed to answer the attention verification questions correctly, leaving us with 1006 participants. Thus, we recruited an additional 369 participants only in the IPM‐ongoing condition, out of which 202 or 54.7% failed to answer them correctly, leaving us with an additional 167 participants in this condition, and 1173 participants altogether.

Procedure and materials

Once again, our intervention used short video clips that apply the IPM principles, tailor‐made for this purpose, similar in length and style. In this study both experimental and control videos were re‐designed to include only text‐based messages, identical in contents and background music but without the photos and sounds used in previous studies. Unlike previous studies, the control video was also designed specially to have identical length and style whilst simulating two commercials for common Israeli products: a snack (‘Bamba’) and a coffee brand (‘Turkish coffee’; see links to all videos in Appendix S1). In terms of the narrative themes used, given that Study 2 did not indicate significant differences in results across themes, we chose to replicate the themes from the first IPM studies, delegitimization of the rival and justness of one's goals (Rosler et al., 2022), given that a longer time had passed since these first studies (administered in 2018) and that significant political and contextual changes took place since then, including the COVID‐19 pandemic and several rounds of elections in the Israeli political system.

Once agreeing to participate in the study, participants were randomly assigned to one of three conditions: IPM‐ongoing (n = 327), IPM‐resolved (n = 397) or a control condition (n = 449). In each of the three conditions, participants watched a combination of two video clips, in randomized order. In the two intervention conditions, participants were exposed to two narrative themes from two conflict contexts, in randomized order. The difference between the IPM‐ongoing and IPM‐resolved conditions is in the first part of the video, in which participants were exposed to messages about the natural and universal nature of conflict‐supporting narratives. In the IPM‐ongoing condition, participants watched examples of narratives from the ongoing conflict contexts of Nagorno‐Karabakh and Kashmir whilst the IPM‐resolved condition included narratives from the resolved conflict contexts of Northern Ireland and the French‐Algeria war. The second part of the video, in which participants are exposed to information regarding the possibility of peaceful conflict resolution, was similar across both experimental conditions and exposed study participants to examples of peace agreements reached in both the French‐Algeria and Northern‐Ireland conflict contexts. After watching the videos, participants responded to two attention verification questions – one for each video. Those who responded correctly continued to complete the dependent variables questionnaire, which included the measures detailed below, as well as some additional exploratory measures (see details in our Appendix S1).

Measures

We measured deliberation of new information (α = .86) and reconsidering attitudes regarding the Israeli–Palestinian conflict (α = .93) with identical items as Study 2. Acceptance of the Israeli perspective regarding the conflict was measured using eight items (α = .72). Whilst identical in structure to the previous study, we slightly altered our questions for better comparison with the control group, asking about a sense of similarity with generic groups suffering from violent conflict (rather than the people in the videos) and adding two additional items to strengthen our assessment of the sense of legitimacy (see Appendix S1 for full items).

Results

For means, standard deviations and correlations across all measured variables see Table 5. To examine the hypotheses, we conducted one‐way ANCOVAs comparing the two IPM experimental conditions with a control group and with one another for each of our dependent variables (see means and SDs for each condition in Table 6), and tested for mediation models using Hayes's (2018) PROCESS addon for SPSS. For consistency with Studies 1–2 and per our preregistration, we controlled for age, gender and political orientation throughout the statistical analysis. Not controlling for these variables had no effect on the results. There were no differences between conditions in terms of age (all ps > .693) gender (all ps > .337) or political orientation (all ps > .453). Given the changes in the political context, in this study we did not include support for negotiations in our final analysis, consistent with our pre‐registration.

TABLE 5.

Study 3 means, standard deviations (SD) and Pearson correlation matrix for dependent and control variables.

Variable	M	SD	1	2	3	4	5	6
1. Acceptance	3.36	0.84	–
2. Deliberation	2.62	1.33	.28**	–
3. Reconsider attitudes	24.19	23.82	.26**	.41**	–
4. Political orientation	3.15	1.38	.12**	.10**	.22**	–
5. Gender	–	–	.02	.03	.32	.05	–
6. Age	41.99	15.21	−.13**	−.02	−.18***	−.27**	.01	–

^** p < .01.

^*** p < .001.

TABLE 6.

Study 3 means and standard deviations (SD) for dependent variables across conditions.

Conditions	Acceptance		Deliberation		Reconsider attitudes regarding the Israeli–Palestinian conflict
	M	SD	M	SD	M	SD
IPM‐resolved	3.40a	0.86	3.20a	1.27	28.00a	24.80
IPM‐ongoing	3.43a	0.87	3.06a	1.24	26.01a	23.95
Control	3.27b	0.83	1.80b	.98	19.10b	21.88

Note: Controlling for political orientation, gender and age. Means not sharing the same subscript letter are significantly different from each other at p < .05.

Acceptance

The analysis yielded a significant main effect of the experimental condition (F(2,1161) = 4.59 p = .010, partial η ² = 0.008). Paired comparisons revealed that both IPM conditions led to a greater sense of acceptance compared to the control condition (all ps < .05). No significant differences were found between the two experimental conditions (p = .664).

Deliberation of new information

The analysis yielded a significant main effect of the experimental condition (F(2,1158) = 182.06, p < .001, partial η ² = 0.239). Paired comparisons revealed that both IPM conditions led to greater deliberative processing of new information compared to the control condition (all ps < .001). No significant differences were found between the two experimental conditions (p = .147).

Reconsider attitudes regarding the Israeli–Palestinian conflict

The analysis yielded a significant main effect of the experimental condition (F(2,1159) = 17.37, p < .001, partial η ² = 0.029). Paired comparisons revealed that both IPM conditions led to greater reconsideration of attitudes regarding the Israeli–Palestinian conflict, compared to the control condition (p < .001). No significant differences were found between the two experimental conditions (p = .249).

Serial multiple mediation model

To examine our serial mediation model, we again used Hayes' (2018) PROCESS (model 6) with multi‐categorical IV using PROCESS indicator coding, comparing the control condition to the IPM‐resolved (coded D1) and IPM‐ongoing (coded D2). We found that both comparisons yielded significant indirect effects (D1: indirect effect = β = .01, b = 0.33 SE = 0.01, 95% CI = [0.00, 0.03]; D2: indirect effect: β = .02, b = 41, SE = 0.01, 95% CI = [0.00, 0.03]; see Figure 2), indicating that both IPM‐intervention conditions (resolved conflicts and ongoing conflicts) increased a sense of acceptance, which, in turn, increased deliberative thinking, thus enhancing reconsideration of attitudes regarding the Israeli–Palestinian conflict.

FIGURE 2.

Serial mediation model with outcome variable Reconsider Attitudes Regarding the Israeli–Palestinian Conflict. Results of the serial mediation model assessing the IPM‐based intervention. Coefficients are standardized, and the resolved conflicts and ongoing conflicts conditions are differentiated where relevant. *p < .05; **p < .01; ***p < 0.001.

Discussion

The results of Study 3 indicate that the IPM is an effective intervention not only when using examples of conflict‐supporting narratives from resolved conflicts but also ongoing conflicts, and when conveying text‐based messages, rather than visually stimulating ones. Replicating Study 2, we once again found that the intervention leads message recipients to feel that their conflict‐related attitudes are accepted, leading them to a process of deliberation, which ultimately leads to reconsidering those same attitudes, implying a process of attitude unfreezing.

GENERAL DISCUSSION

The goal of the present research was to understand the mechanisms that underlie the IPM – specifically, feeling acceptance – and expand the generalizability and substantiate the validity of the model's interventions. Our results showed that the intervention increased feelings of acceptance amongst study participants, which, in turn, led to greater deliberation of new information, predicting greater reconsidering of attitudes concerning the Israeli–Palestinian conflict, meaning an ‘unfreezing’ of conflict‐supporting attitudes (see Figures 1 and 2). Studies 1–2 also show that the current political context makes negotiations seem highly unrealistic for most Jewish Israelis, reducing their support for such an improbable policy, hence our focus on unfreezing in broader terms. Our findings empirically substantiate the theoretical basis of the acceptance‐change dialectic that underlies the IPM‐based interventions (Rosler et al., 2022). Whilst this approach is fundamental and empirically validated in clinical psychology relating to individual‐level beliefs (Brodsky & Stanley, 2013; Linehan, 1997), our study provides evidence that the dialectic is also an explanatory mechanism when it comes to changing collective narratives. Our findings also validate that the effectiveness of the IPM intervention is not the result of merely communicating any conflict‐related content (Study 1); show the effectiveness of IPM interventions when using messages of two new narrative themes – victimhood and security (Study 2) – and of narratives from ongoing, not only resolved conflict contexts, and in text‐based, rather than visually stimulating, message styles (Study 3). These findings strengthen the basis for the value of the IPM for attitude transformation in the context of intractable conflicts.

As a new psychological mechanism, the effectiveness of messages that convey acceptance demands further unpacking. We suggest that its impact hangs first on the consistency of its message with individuals' existing beliefs. Studies have already demonstrated the value of messages consistent with one's beliefs as means of changing them (Bar‐Tal et al., 2021), as the acknowledgement of narratives has been shown to reduce identity threat, which can increase openness to new perspectives (Čehajić‐Clancy et al., 2011; Sherman et al., 2017). However, consistency may not be enough to explain the effectiveness of acceptance in mediating the IPM intervention's impact on conflict attitudes. Non‐confrontational interventions such as self‐affirmation (Cohen et al., 2007) are often based on an intrapersonal process that – unless going to their paradoxical form (Bar‐Tal et al., 2021) – may not be as effective for addressing conflict‐supporting narratives as these are already constantly being self‐affirmed within societies in conflict.

Unlike self‐affirmation, acceptance by others implies a relationship between an accepter and an accepted. Therefore, the second key mechanism that helps explain the effectiveness of acceptance is its externalization. Our evidence suggests that mirroring strongly held conflict‐supporting narratives through the experiences of similar others can indirectly elicit a sense that one's own conflict narratives are acknowledged and understood. We use the term ‘similar others’ referring to the similarity of experience, rather than categorical similarity (such as of identity markers), as evidence suggests this can be uniquely important for emotional wellbeing in stressful situations (Thoits, 2021) and for improving intergroup relations (Shnabel et al., 2013). Such similarity is also a key condition for stimulating comparative learning (Holyoak & Thagard, 1997): meaning, the possibility of peace in a distal (but similar) conflict context might indicate the same possibility in one's own conflict context, increasing openness to change (see also Shulman et al., 2020). Whilst more research is needed, we could even explore the possibility that the IPM intervention is indirectly recategorizing message recipients into a common identity of ‘peoples in violent conflicts’, leading to a normalization of the in‐group experience of conflict‐narratives, combined with a willingness to learn from others, given the potential for a common fate (Gaertner & Dovidio, 2000).

We further understand the importance of this externalization in relation to research on the capacity to change attitudes through acceptance‐oriented listening and feeling heard which has been shown to increase attitude ambivalence. By reducing social anxiety and defensive information processing and increasing the capacity to tolerate inconsistencies in one's beliefs, quality listening opens space for contemplation and ultimately moderates attitudes (Bruneau & Saxe, 2012; Gutenbrunner & Wagner, 2016; Livingstone et al., 2019, 2020). Much of the research in these areas has focused on the importance of feeling heard by the adversary in conflict. Yet, encounters of this quality between conflict parties can be particularly challenging to implement in intractable conflict contexts. Whilst IPM‐based interventions are not listening or perspective‐giving exercises, our findings extend these theories by showing that even a general sense of having one's perspective echoed through similar others and thus indirectly acknowledged and normalized can have a meaningful influence on unfreezing conflict‐supporting attitudes.

Finally, our findings contribute to extant research on the effectiveness of media‐based interventions in peace promotion (for a recent review, see Littman et al., 2023), and its potential scalability, which is especially important in societal‐level conflicts. Media‐based interventions also have the potential for creating transportational experiences of immersion in a story that increase the likelihood of attitude change due, amongst other things, to reduced counterarguing and character identification (Green, 2021). Whilst lack of longitudinal data is a general limitation of psychological interventions (Paluck et al., 2021), we would argue that media‐based interventions are also particularly well suited for repetitive exposure in campaign‐like structures, which has been found to have enduring effects even in intractable conflicts (Hameiri et al., 2014, 2016, 2018; see also Bruneau et al., 2022).

Limitations and areas for future research

The current research has several limitations. First, like past studies that examined many other interventions, the present research focused on short‐term changes, thus we are unable to say much regarding the sustainability of the observed effects (Paluck et al., 2021). Second, our effect sizes are relatively small. However, we would argue that in difficult conflict contexts – as the events starting on 7th October 2023 in the region exemplify – even small effect sizes could be meaningful when reaching many people, which is possible with media‐based interventions (Littman et al., 2023). Third, all IPM studies have thus far focused on examining the effectiveness of the intervention amongst the Israeli Jewish side of the Israeli–Palestinian conflict; further research is necessary to evaluate its effectiveness in other conflict contexts, as well as the Palestinian side of the Israeli–Palestinian conflict. This is especially important given the unequal power relations between Israelis and Palestinians and research indicating advantaged and disadvantaged groups have different needs and need different stimuli for attitude change (Bruneau & Saxe, 2012; Shnabel & Nadler, 2008). Fourth, whilst our results indicate that acceptance mediates the relationship between the intervention and unfreezing of attitudes, we recognize the limitation of establishing causality from such mediation models (see Rohrer et al., 2022) and the need for complementary research in this regard, that can establish the sequential causal relationship we suggest in our mediation models, for example, through manipulating the extent to which the messages convey acceptance. Finally, whilst in Study 1 we compared the IPM intervention with a DB‐based task – both exposing study participants to conflict‐related content – in general, the control conditions across studies, do not include exposure to new conflict‐related information, slightly reducing their comparability with the IPM. Future studies should ensure increased deliberation is due to the IPM foundations and not mere exposure to any new information.

This study opens new avenues of research within and beyond the IPM. In particular, on the notion of acceptance, we welcome further research on the importance of the identity of the messenger of acceptance, amongst other things. For example, whether and how does interaction with members of societies from different conflict contexts lead to attitude changes regarding one's own conflict (see e.g., Darby, 2008; Guelke, 2004; Lustig, 2003; Mitchell, 2021a, 2021b) and about the intersubjectivity of the researchers as invisible but omniscient participants in the studies – potentially offering reason and certainty in the midst of the chaotic reality.

CONCLUSION

In conclusion, the present studies aimed at advancing our understanding of processes of attitude changes towards peaceful conflict resolution. We looked to the IPM as a means of overcoming the barrier of conflict‐supporting narratives in intractable conflicts. By extending the generalizability of IPM‐based interventions, our findings further substantiate the effectiveness of acknowledgement‐based approaches for attitude change and the acceptance‐change dialectic. In revealing the importance of acceptance in driving the effects of IPM‐based interventions, we further our understanding of the underlying mechanisms that explain the interaction between acknowledgement of conflict attitudes and their unfreezing. We thus provide firmer grounds for policy makers and practitioners to implement the IPM principles in peace processes, for example, through outreach strategies communicating peace processes to the public, or within elite negotiations. This can take the form of a better‐sequenced process or more effective confidence‐building measures; the intentional use of comparative case studies, emphasizing the universal nature of various conflict dynamics; as well as the possible use of short video messages as a scalable intervention on a societal level. The implications of our contributions can also go well beyond the IPM‐based intervention, shedding new light on other socio‐psychological interventions to change conflict attitudes. Researchers and practitioners may want to consider ways of incorporating messages that convey an experience of external acceptance towards one's beliefs, and only then motivate their change, preempting dissonance with acknowledgement.

Whilst writing this paper, the world witnessed an unprecedented attack by Hamas on Israel on October 7, 2023, leading to an equally unprecedented escalation in the region since decades. In parallel to the physical violence, we witness both sides draw upon their polarized conflict‐supporting narratives, in an effort to make sense of the events. The path towards any kind of reconciliation seems highly improbable. Yet, in the words of Maimonides, hope lies in the ‘belief in plausibility of the possible, as opposed to the necessity of the probable’ (Ganz et al., 2023, p. 173). In our conflict‐ridden world, where conflicts persist and many peace agreements still culminate in failure (Quinn & Echavarría Alvarez, 2021), this study sheds a hopeful light on our capacity to better transform societies and build peace.

AUTHOR CONTRIBUTIONS

Inbal Ben‐Ezer: Conceptualization; data curation; formal analysis; methodology; writing – original review and editing; writing – original draft. Nimrod Rosler: Conceptualization; writing – review and editing; supervision. Keren Sharvit: Conceptualization; writing – review and editing. Ori Wiener‐Blotner: Conceptualization; writing – review and editing; data curation. Daniel Bar‐Tal: Conceptualization; writing – review and editing. Meytal Nasie: Conceptualization; writing – review and editing. Boaz Hameiri: Conceptualization; data curation; formal analysis; methodology; writing – review and editing; supervision.

CONFLICT OF INTEREST STATEMENT

The authors have no conflict of interests to declare that are relevant to the content of this article.

Supporting information

Appendix S1.

Ben‐Ezer, I. , Rosler, N. , Sharvit, K. , Wiener‐Blotner, O. , Bar‐Tal, D. , Nasie, M. , & Hameiri, B. (2025). From acceptance to change: The role of acceptance in the effectiveness of the Informative Process Model for conflict resolution. British Journal of Social Psychology, 64, e12802. 10.1111/bjso.12802

DATA AVAILABILITY STATEMENT

Anonymized pre‐registration for study 1 is available at: https://osf.io/536yh?view_only=true. Anonymized pre‐registration for study 2 is available at: https://osf.io/me2zn?view_only=true Anonymized pre‐registration for study 3 is available at: https://osf.io/xahvt?view_only=true. Additional supplementary materials for studies 2‐3 are available at: https://osf.io/qkgtm/?view_only=None.