Empathy as a foot in the inter-ethnic door : Facilitation of positive bilateral perception among Jewish and Arab students in Israel *

A heavy load of blame, hostility, and violence typifies the relationships between Jews and Arabs in Israel for nearly two centuries in a row. The present study is an exploratory experimental attempt, within the framework of Functional Measurement, to examine the viability of empathy as a means to facilitate a positive shift in functional cognitive schemata of helping. The participants, 143 Jewish and Arab (57 and 86, respectively) female students, were sampled from an Israeli academic college, where the Jews-Arabs ratio is nearly 50-50. Each Participant met individually 4-8 times, once a week, with a same sex and same age experimenter (another student). Each experimental session included a conversation where the experimenter approached the participant empathetically for approximately 15-20 minutes and asked her repeatedly to imagine a series of meetings between two female students (Jews and Arabs in all combinations) where one such protagonist attempts to receive form the other (Jew or Arab) lesson notes. This manipulation was arranged within the framework of a bi-factorial model, that isi.e., the level of needs of both protagonists (3 x 3). In each such encounter the participant was asked to estimate the likelihood that the imagined request will be fulfilled. The results show that serial exposure to empathy affected positively mainly Jewish students. The relevance of the findings to the notion of 'functional social cognition' and to the body of knowledge which deals with majority-minority relations is pointed at.


Para citar este artículo:
Yehuda, L., & Wolf, Y. (2016).Empathy as a foot in the inter-ethnic door: Facilitation of positive bilateral perception among Jewish and Arab students in Israel.Universitas Psychologica,15 (3).http://dx.doi.org/10.11144/Javeriana.upsy15-3.efidJewish-Arab relations in Israel have a history of heavy frictions which was ignited more than a century ago.This long term tension is reflected in many aspects of everyday life, in terms of latent and overt inter-party hostility.The present study does not pretend to offer an ultimate solution to the problem of Jewish-Arab relations.However, the hypothesis that some positive change can be facilitated following empathetic communication between individuals from both parties is offered here.
There is a vast array of literature which advocates the potential of empathy as a facilitator of coping with social problems (e.g., Berger, 1984;Billikopf, 2004;Buie, 1981;Cialdini, Brown, Lewis, & Luce, 1997;Graziano, Gaertner, Flores Neimann, & Sneider, 2007;Rogers, 1975).According to Kohut (1981) empathy connects human beings more than love.Along the process of empathetic communication, the empathy giver collects psychological information on the receiver, forms an image of the latter's internal experience, and develops a sort of objective evaluation of his or her mental state.It facilitates in the receiver of empathy responses or actions which can be related to as coping behavior.Berger (1984) postulates that empathetic understanding facilitates catharsis of hard and painful feelings.According to Warner (1997), prolonged empathetic relations can contribute to the development of an ability to accept the other's experience, even when it does not go along with the beholder's experience.Premises which relate to the nature of this sort of social communication can be found in quite a few recent publications (e.g., Angus & Hardtke, 2006;Castonguay, 2005;Castonguay & Hill, 2006;Goldfried, 2003;Hardtke & Angus, 2004;Pachankis & Goldfried, 2007;Winslade & Monk, 2000).
Based on a merger of Anderson's theory of functional cognition and Wolf's hypothesis of moral/judgmental modularity, it is hypothesized that empathy will increase the relative importance assigned to the needs of individuals from the other party (Anderson, 1996(Anderson, , 2008(Anderson, , 2013;;Wolf, 2001Wolf, , 2002)).In order to exemplify a possible breakthrough, the present explorative experimental venture offers an overarching (Staats, 1999;Anderson, 2013) between two bodies of knowledge -empathy and helping.That is in a way which unifies circumstantial and primordial accounts for the related phenomenon.This work was conducted within the framework of the Information Integration Theory (Anderson, 1991a(Anderson, , 1991b(Anderson, , 1996(Anderson, , 2008(Anderson, , 2013) ) and its methodological counterpart -Functional Measurement (Anderson, 2001;henceforth FM).
Anderson postulates that any individual's decision and behavior is a function of a personally meaningful purpose (helping in the present context).According to another postulate, information processing and action is based on integration of perceived facts which are relevant to the timely purpose.Such bits of information are channeled spontaneously to the related cognitive categories; in the present context the level of need of each of the two role partners (a person who asks for help and the one who is being asked to grant help).
The present study attempts to shed some preliminary light on the functional way in which inter-ethnic (Jewish-Arabs in the present context) helping requests are coded in the cognitive system of the beholders.Such coding is defined here as functional cognitive helping schema (Laskov-Peled & Wolf, 2010).Formula 1 is deliberated to represent, in terms of IIT, the way in which information regarding the needs of two personae, that is, protagonists A and B (a Jew and an Arab, or any other combination of two protagonists from these primordial origins), is integrated algebraically (Å) into individuals' (Jews and Arabs) cognitive system, as reflected in terms of perceived likelihood of helping (PLH).This formula is used in the present study as a means to reflect Jewish and Arab students' functional cognitive schemata of helping in terms of requests to receive lesson summaries.
(1) It is hypothesized that repeated exposures to empathetic Jew-Arab and Arab-Jew approach in an academic context will facilitate a shift toward helping schemata which represent assignment of greater weight (importance) to the needs of those who belong to the other party.

Participants and experimenters
The participants, 143 female students from the Western Galilee Academic College in Israel, 57 Jews and 86 non Jews -53 Islamic Arabs, 28 Christian Arabs, and five Druze (for the sake of writing simplicity, all minorities, that is, non-Jews, will henceforth be called Arabs) were sampled incidentally.All are 1 st and 2 nd year 1 st degree students; all reside in Western Galilee.The participants' age mode (nearly 2/3 of them) is 23-25, the range being 18-27.All of them come from middle to lower class families and are not married.A team of 12 confederates/ experimenters, with similar age range and socioeconomic background, half Jews and half Arabs, chosen on the basis of peers' recommendations, participated in a mini (two days) guidance course, destined to teach the basics of empathetic communication.They served as applicants of FM and providers of empathy reflections to the study participants, as described below.They received students' payment per hours of working.

Design
The present attempt to achieve reflections of bilateral helping schemata (see Formula 1) of Jewish and Arab students was conducted within the framework of FM, before and after exposure to empathy referred personally to Jews or Arabs by a Jewish or an Arab confederate (all combinations).More specifically, the measure was a product of an FM procedure which provided raw data (likelihood estimation, as detailed below) for the computation of the relative importance assigned by the participant to the needs of the two protagonists that she should have imagined to herself.The model was bi-factorial ( 3 x 3 ) -the (level of) needs of a protagonist who asks for help multiplied by the (level of) needs of the protagonist to whom the helping request is referred (hence receiver and giver, R and G, respectively).These two protagonists are imagined by the participant, whose task is to rate the likelihood that G will help R.There were two variations of this procedure.In one variation R belonged to the participant's own group.This variation is labeled here as 'stereotypic', presumably due to the stereotypes facilitation quality of an estimation of the other party intentions.In the other variation, labeled as 'responsive', R belonged to the other group.
The response to be recorded in each of the nine cells of the model in each of the two variation was the participant's estimation of the likelihood that G will help R, given a specific combination (one out of nine) of the needs of G and R. Estimates of the relative importance (RI) assigned to the needs of both protagonists are extracted from the entire set of nine estimations, using an algorithm offered by FM (Anderson, 1982, pp. 266-270).Of special significance, in terms of empathy theory, is the relative importance assigned to the needs of R (hence RIr); it should be noted that the importance assigned to the needs of G is as follows: RIg = 1 -RIr.The comparison of this statistic, before and after exposure to empathy, served as a measure of the impact of empathy as a function of a series of exposures to this type of social communication.
The present focus on relatively long term changes in the relative assignment of importance to the needs of R as a function of an exposure to empathy is due to the following analysis.While even one session (i.e., 1 st meeting) support for the Empathy => RIr hypothesis should be theoretically meaningful, the importance of a possible exemplification of longitudinal effect (of several consecutive exposures to empathy) is much greater, conceptually and in applied terms; simply because a relatively longitudinal exposure to empathy should reinforce resistance to extinction.The choice to use a time gap of one week between the repeated empathy => FM was based on a compromise between the need to allow sufficient extinction time and the fact that students, like anyone in the Western society, live in periods of weeks.
The participant was asked repeatedly nine times to imagine request making meetings between two protagonists (female students), a Jew and an Arab, where one of them asks for the other's help (lectures records).Eventually, in each of the nine episodes the specific combination between a specific level of the needs of R and G was noted very plainly and the participant was asked to repeat it in her own terms.Then she was asked to make a spontaneous (prompt) likelihood estimation that G will respond positively to R's request.Table 1 illustrates how a clear conclusion with regard to either perspective (G and R) and empathy effects can be made, based on a comparison between two RIr statistics extracted from two sets of nine likelihood-of-helping ratings.This statistic, compared to the equal weighting before empathy, represents an assignment of two thirds of the relative weight to the needs of R,.Moreover, a comparison of the arithmetic means 'before' and 'after' would provide another support for the conclusion with regard to the positive effect of empathy: The mean 'before' is 50 while the mean 'after' is 60.A similar twofold effect is expected in a comparison between these two FM statistics which will be extracted from likelihood estimation of the same person in two perspectives -G and R, simply G < R.
Recalling that in Israel practically Jews are in-group and Arabs are out-group, a complete design has to include all possible cross-comparisons of these two parties as experimenters-participants (independent conjoint samples) and protagonists (dependent samples).In terms of mundane validity, the Western Galilee Academic College, where the students are Jews and Arabs (nearly half and half), is about to be an ideal arena for such a study.Since the vast majority of the students in this college are Female, the participants were sampled from this population.

Procedure
Each experimenter met individually 1-8 times (depending on each participant's ability to complete the seven-meeting schedule) about once a week, with 8-12 Jewish and Arab students.In each session there was a 15-20 minute open conversation where the experimenter did her best to empathize with the participant.Empathetic reflections to the participant were delivered by the experimenter during the conversation whenever it was possible (a post-experimental count revealed a range of 9-14 moments of reflection in each conversation).
In each such moment, the empathy provider followed five (quite fast) phases, as follows: 1. Figure out whether or not the conversation partner is attending to you, and if positive try to empathize with her; 2. Focus on verbal and bodily reflections of her state of mind and emotions; 3. Do your best to compare these impressions to your own past experience; 4. Exclude your own relevant memories from the illustration that you form with regard to the partner's state of mind and emotions, and relate only to what belongs to her; 5. On the basis of the above phases, form a brief question-like statement with regard to the partner's state of mind and/or emotional state, and refer it to her as gently as possible.
In the 1 st session, the FM procedure, deliberated to produce stereotypic and responsive functional coding of helping in terms of Formula 2, was conducted, before (to provide a baseline) and after the empathetic conversation.In the following sessions, the FM procedure (a request to imagine serially nine episodes) was conducted only after the conversation, without measurement 'before'.This core series (i.e., 1 st + 2 nd +…8 th sessions) was replicated in a variety of conditions which were included in the complete design.
Each participant (Jewish or Arabic) was exposed to two experimental conditions, in each of which she was asked nine (3 x 3) times to imagine to herself an encounter of two (female) students, one of whom is asking for lectures notes.The operative decision to focus on lecture notes requests was based on the results of a preexperimental questioning of 12 Jewish and Arab students (half and half), who stated very clearly that note -request is the most salient sort of helping requests among students in that specific college.In one of the experimental conditions, the participant was asked to imagine that R belongs to her ethno-social-religious group (ingroup or out-group) and in the other condition she was asked to imagine that R belongs to the other group.

Results and discussion
The quasi longitudinal design of this study requires comparisons between the following two helping request episodes: 1.The participant, either in-group or out-group, responds to requests made by protagonists who belong to her group; 2. The participant, either in-group or out-group, responds to requests made by protagonists who belong to the other group.Due to unexpected personal limitations there was drop-out along the weekly sessions, due to insufficient calibration, as depicted in Figure 1.

Figure 1 Number of Participants in Each Session
Source: own work As it can be seen in Figure 1, there was a sizeable dropout along the range of eight weekly sessions.To maximize the viability of the results, two types of data arrangement were conductedparametric and non-parametric.The parametric one was destined to enable a test of the overall three-way design (Origin x Perspective x Empathy).Only the participants in the 1st to 4th sessions were included in this (F) test in order to include a sufficient number of participants in the longitudinal (repeated exposure to empathy) part of the design.Another, non-parametric, test attempted to cover the entire range of sessions.
The three factors which relate to real life personae and their experiences along the study are empathy giver (henceforth EG), empathy receiver (henceforth ER), and session (1-4); the three of them make a three factorial model -EG x ER x Session (2 x 2 x 4, respectively).Following the substantive distinction made in this composition between the participants' perception of in-group and out-group, this model was tested in the two following conditions of G (protagonists whose help is requested): 1. G belongs to the participant's (ER) group (responsive); 2. G belongs to the other group (stereotypic).These two sets of analysis are presented below.

Responses to requests from the same-group protagonists
Table 2 presents, within the frame of the three-factorial design -EG x ER x Session -grouped likelihood ratings in conditions of requests from protagonists who belong to the participants' group (responsive).The table shows that Jews' estimation of the probability of help provided by Jews increased as a function of repeated exposure to empathy, while Arabs did not.Nevertheless, a Three-way ANOVA reveals, under the conservative criterion of p < 0.01, a significant ER x Session interaction, F (3, 234) = 4.76, and a significant main effect of ER, F (3, 234) = 4.04.In order to get a clearer descriptive exposition of these effects, especially the interaction, Table 3 presents the picture shown in Table 2 beyond EG, which can be summarized as an indication that only Jewish ER assume an increased help of G who belongs to the ER's group.

Table 2 Mean Likelihood Ratings in Conditions of Requests by Protagonists from the Participant's Same Group in Terms of the Model -EG x ER x Session
*EG = empathy giver; ER = empathy receiver; NJ = non Jew Source: own work

Table 3 Mean Likelihood Ratings in Terms of the Model -ER x Session
Source: own work While being nearly illegitimate in conventional experimental modeling, a quick look at the entire array of data in nonparametric terms might provide some (though quite marginal) support for the main line findings.A contingency c 2 was made between Jews who received empathy from Arabs and Arabs who received empathy from Jews.This test compared the relative number of the participants from both parties who were affected positively or negatively by empathy in all eight sessions (a 2 x 2 matrix).An inverse trend was found -positive shift in Jews and negative in Arabs, c 2 (233) = 6.31, p < 0.05.In the same line is the positive shift in Jews who received empathy from Jews and the negative shift in Non-Jews who received empathy from Jews, c 2 (233) = 6.91, p < 0.05.Other nonparametric effects were not found whatsoever.Overall, the findings thus far signify a support for the hypothesis of this study regarding a positive longitudinal effect of empathy on the way helping episodes are perceived by in-group (i.e., Jews) members of the Israeli society.4 shows that the estimation of the probability of help by Jews in conditions of a request from a member of the other group is greater than that of Arabs (the respective means are 42.47 and 31.18),F (1, 66) = 8.62, p < 0.01 .No other effect of F test or of c 2 was found whatsoever.These analyses, in terms of two three-way ANOVAs and non-parametric tests, seem to point to the same tendency of Jewish students to perceive positively help requests between members of their own sociopsychological (in) group.Such a view was not found in the members of the other (out) group.

General discussion
The findings, based on Israeli female students' estimations of helping likelihood, are in line with the general hypothesis that empathy can facilitate positive shifts in the context of inter ethnic-religious tension and friction.However, only (female) Jewish students' (in-group in Israel), perception of help request episodes was found to be affected by repeated empathetic communication referred to them by classmates from both parties.No such shift was observed in Arab (out-group in Israel) students; there are even some indications of an inverse (i.e., negative) shift in those students' likelihood estimations following empathy.When the target protagonists belonged to the other group, no positive shift was found in the responses of students from both parties.However, in such conditions the Jews' helping estimations were much higher than the estimations of the Arab students.
The findings seem to support the offer to overarch three different bodies of psychological knowledge -functional cognition, empathy, and helping.The viability of studying empathy in terms of functional cognition was exemplified by Shimkin, Idisis, and Wolf (in press), in a study which examined functional cognitive coding/schemata of three aspects of empathy by psychotherapists and patients.Support for the assumption that empathy has positive social consequences can be found in the literature (Angus & Hardtke, 2006;Castonguay, 2005;Goldfried, 2003;Hardtke & Angus, 2004;Kohut, 1981;Ladany, Walker, Pate-Carolan, & Gray Evans, 2008;Pachankis & Goldfried, 2007;Winslade & Monk, 2000).Dovidio, Gaertner, Flores Neimann, & Sneider (2001) found, as in the present study, that in-group white students tended to behave empathetically and helpfully toward black students who did not emphasized their ethnic identity and identify with the majority group (see also Kaiser & Prat-Hyatt, 2009;Schmitt & Branscombe, 2001;Sellers & Shelton, 2003).This might means that in-group => out-group empathy and help might occur in conditions of perceived similarity between these groups.
This conclusion is in line with Mollov and Barhoum (1998), who found greater Jews-Arabs cooperation and personal relations following a four-year dialogue between Jewish students from Bar-Ilan University and Arabic students from Hebron, where similar elements of Judaism and Islam were discussed.The present findings, which show a positive shift in Jewish students' perceptions of inter ethnic help, might be related to the very fact that the Arabs volitionally joined an Israeli-Jewish academic institute, recalling that they could have alternatively joined an Israeli Arabic institute.
Three sorts of validity of this work can be noted.The quite informative results, which speak in terms of Anderson's (1996Anderson's ( , 2008Anderson's ( , 2013) ) functional cognitive theory, seem to point to a convincing level of construct validity.In terms of internal validity, the viable findings seems to gain support by repeated statements of the participants, which were documented throughout the study.Most statements reflect sincere involvement in the to-be-imagined situation.For instance, a typical statement of Arab participants was as follows: "Based on my experience, it is not likely that Jewish students will give me lectures' summaries."Nevertheless, Jewish students statements reflected focus on the immediate needs of the two protagonists; for instance: "My needs count for me more than others'".The mundane validity of the experimental task is beyond any question; all over the academic world lectures summaries are well requested and are a source for tensionloaded interpersonal, and, possibly, inter-group dynamics.
As about external validity, on the one hand, the campus of the Western Galilee College (where the Jews-Arabs ratio is nearly half and half) can be conceived as an in vivo micro-cosmos of Jewish-Arab relations.The findings, however, require much purification, especially in terms of the type of helping request, situation and context, ethno-religious-political attitudes of the participants, education, age, etc.
A few prospective notes can be made.Firstly, a shift from perception of helping, as observed in the present study, to helping in vivo is called for, as implied from the findings of Graziano et al. (2007), which show that prosocial and empathetic dispositions correlate positively with helping.Possibly, the presumed suspicious approach of Arabs, as hinted by the present findings, should be moderated using warming up activities aside of empathy, as can be learned from the work of Mollov and Barhoum (1998).In such conditions, if empathy will be shown to affect helping in the desired way, applied possibilities seem to offer themselves.
A tempting reality to begin with is, as in the present study, academic institutions where in-group and out-group students participate in the same programs and courses.There, volunteers from both parties can participate in accredited empathy courses (workshops), in which they will be encouraged to relate empathetically to students from the other party.Simultaneously, opportunities for inter-group and inter-personal encounters should be formed, including dyadic participation in course assignment and common social activities.Assuming that such students will take key social, political, and professional positions, the effect of a stone thrown to a pool of water can be expected.
Preferably, Arabic lecturers should be part of the support of the system for such a venture.Wolf and Laskov-Peled (2010) report on a nationwide Israeli project deliberated to identify empathetic educational figures and form a positive atmosphere around them, assuming a negative correlation between empathy and socially negative types of communication at school.In this line, and assuming a positive correlation between empathy and constructive in-group/outgroup communication, it is recommended that an empathetic approach and empathy techniques should be taught in teachers' education schools.
Finally, the present work might overlay a novel way, programmatic, and methodological, for the study of reconciliation, beginning with the relations between Jews and Arabs in Israel.Discussions on this issue can be found in the literature (e.g., Bar-Tal, 1998, 2000;Deutsch, 1957;Hayner, 1996;Lederach, 1997;Van der Merwe, 1999).
following a hypothetic exposure of some participant to empathetic conversation (let us define the relative weight assigned in the condition before empathy as RIrb and the weight 'after' as RWra).The calculation of these statistics is based on two terms which represent the independent weights assigned to the needs of R and the needs of G before and after the conversation (henceforth Rb-Gb and Ra-Ga, respectively).Rb is the difference between the means of the two extreme rows = 70 -30 = 40; Gb is the difference between the means of the two extreme columns = 70 -30 = 40.RIrb = R/(R + G) = 40/(40 + 40) = 0.50.RIg (the relative weight assigned to G) = 1 -RIr = 0.50.Such a finding would represent an assignment of equal weighting to both components of the assumed helping code (schema) of the 'demo' participant before his or her exposure to empathy.Nonetheless, this participant's relative weighting after empathy is much different: RIra = 40/(40 + 20) = 0.67.

Table 1
Two Core FM Design, Before and After an Empathetic Conversation, with Demo Likelihood Estimations * M, S, and L represent Much, Some, and Little, respectively Source: own work Table 1 exemplifies how FM can reflect a positive shift in RIr

Table 4
Mean Likelihood Ratings in Conditions of Requests by Protagonists not from the Participant's Group in Terms of the Model -EG x ER x Session Source: own work Table