Effects of dispositional approach/avoidance and attitudes on
interpersonal distance.
Research report
of
Lea Hildebrandt
6240917
For the 2
ndresearch project within the masters’ program
Brain and Cognitive Sciences,
Cognitive Neuroscience Track,
University of Amsterdam, Netherlands
At the
Max Planck Institute for Human Cognitive and Brain Sciences
Social Neuroscience Department
Leipzig, Germany
Supervisor:
Cade McCall, PhD,
Max Planck Institute for Human Cognitive and Brain Sciences,
Social Neuroscience Department,
Leipzig, Germany.
Co-Assessor:
Dr. Nexhmedin Morina,
University of Amsterdam,
Clinical Psychology Group,
Amsterdam, Netherlands.
Abstract
The aim of this study was to investigate whether dispositional approach and avoidance tendencies are embodied in interpersonal distance during social interactions. Particularly, it was expected that this relationship would be moderated by situational factors, such as attitudes towards the interaction partner. The results indeed indicate an interaction between avoidance tendencies and perceptions of fairness. Specifically, the higher the dispositional avoidance tendency, the more participants
Effects of dispositional approach/avoidance and attitudes on interpersonal distance
Proxemics, the “study of how man unconsciously structures microspace” (Hall, 1963, p. 1003), deals with the question which factors influence interpersonal distance1the distance that is kept between two people in a social interaction. It has been widely accepted that this distance varies, depending on rather stable cultural norms (Hall, 1966) but also on dynamic situational factors. The nature of the relationship with the interaction partner, for example, is related to the proximity to that person: the closer the relationship, the closer the interaction distance.
This idea is central in the proxemic framework of E.T. Hall. He emphasized a division of personal space into four distance zones around one’s body: the intimate, personal, social, and public distances. Intimate distance is the space up to 46cm, which is reserved for intimate partners and close friends. Family members and acquaintances are allowed in the personal distance, which ranges from 46cm to 1.2m. The space beyond the ‘arm-length’ of 1.2m and up to 3.6m is used for more formal encounters. Finally, the distance greater than 3.6m is the public distance that is usually kept between a public speaker and an anonymous audience (Hall, 1963).
Support for this proposition has been found using both simulation techniques (Little, 1965; Mehrabian, 1968) and field observations (Heshka & Nelson, 1972; Willis, 1966). In general, in these studies, it has been found that the distance between friends is smallest, increasing between acquaintances, and greatest between strangers.
Attitudes
Besides aspects of a relationship such as trust, one factor that could be related to the intimacy of the relationship is the valence of attitudes held toward the interaction partner. In general, attitude has been defined as “the degree of liking, positive evaluation, and/or preference of one person for another” (Mehrabian, 1969). It has been shown that familiarity, similarity, and attraction – factors high in close relationships – are all related to liking (Moreland & Zajonc, 1982).
The assumption that attitudes can affect spatial approach and avoidanceis not only reflected in other definitions, which emphasize a behavioral readiness to respond (Allport, 1935; Droba, 1933), but was also tested empirically by Chen and Bargh (1999). In their study, participants had to either push or pull a lever in response to valenced stimuli. Participants were faster at pushing the lever in response to negative stimuli than to a positive one, and they were also faster at pulling it when positive stimuli were presented, compared to negative ones. Thus, positive attitudes facilitate approach and negative attitudes facilitate avoidance.
1 Often, personal space and interpersonal distance are used interchangeably, as will be the case in this report.
Recently, two studies provide evidence that these behavioral predispositions influence interpersonal distance. Dotsch and Wigboldus (2008) found that implicit attitudes, in this case prejudice, predicted physical approach of (virtual) Moroccan and White agents. Along the same line, McCall, Blascovich, Young, and Persky (2009) showed that interpersonal distance (in an immersive virtual environment) to a black agent accounted for 50% of subsequent aggressive shooting behavior towards this agent. These findings support the thesis that attitudes are reflected in proxemic
behavior.
More generally, earlier findings have shown that positive attitudes are related to closer interaction distances. Byrne, Ervin, and Lamberth (1970) matched female-male pairs for a date based on similarity or dissimilarity of attitudes. The authors found that subsequently, the participants chose closer standing distances to a similar than a dissimilar person. Furthermore, attraction also
influenced the proximity in this study. Similarly, Gifford (1982) showed that attraction and cooperation result in a smaller interpersonal distance.
To summarize, it can be concluded that positive attitudes lead to greater interpersonal proximity and negative attitudes lead to less interpersonal proximity, a pattern we expected to be replicated in the current study.
Personality
Besides situational aspects, it has also been suggested that factors that are more stable across situations have an effect on interpersonal distance as well. Hall (1959), for example, developed his framework based on observed cross-cultural differences in the use of interpersonal space. Furthermore, it has often been assumed that personal space size also depends on intra-individual differences, such as personality traits. Although some traits – especially those indicating one’s own control and confidence in the interaction: locus of control (Altman & Vinsel, 1977) and self-esteem (Frankel & Barrett, 1971) – have been found to predict interpersonal distance, the findings on more fundamental dispositions are inconsistent(see Aiello, 1987). Interestingly,
dispositional approach and avoidance tendencies, a fundamental division in personality psychology, have not been studied as a predictor for actual social approach and avoidance.
Approach and avoidance. The distinction between approach of positive and avoidance of
negative outcomes is widely used in psychological research. It is the basis of a number of theories on, among others, motivation (Elliot, Murayama, & Pekrun, 2011), self-regulation (Higgins, 1997), and reinforcement learning (Azrin & Holz, 1966). What is more, Gray’s influential biopsychological theory of personality (1982, 1990) proposes that approach and avoidance tendencies relate to the basic components of personality. He suggested that two separate neurological systems exist, the
behavioral activation system (BAS) and the behavioral inhibition system (BIS), that respond to cues of
reward (incl. non-punishment and escape from punishment) and punishment (and non-reward as well as novelty), respectively. The BAS would regulate approach of goals, whereas the BIS would regulate avoidance of negative outcomes. Individual differences in the responsiveness of these two systems would therefore lead to differences in sensitivity to reward and punishment cues, which would, in turn, manifest in dispositional differences in approach and avoidance motivation.
Moreover, high BAS activity is associated with extraversion and positive affect, and high BIS activity with introversion and negative affect (Carver & White, 1994).
The link between dispositional approach and avoidance and proxemic behavior is less clear but might be very similar to the relationship between attitudes and proximity. Interestingly, Elliot and Thrash (2010) define approach and avoidance, in the context of personality traits. They argue that positively evaluated stimuli are inherently associated with an approach orientation (literally or figuratively) toward the stimuli, whereas negatively evaluated stimuli are inherently associated with an avoidance orientation (literally or figuratively) away from the stimuli. [… Approach and avoidance] produce immediate affective, cognitive, and behavioral inclinations” (p. 866-878). Therefore, it can be assumed that approach and avoidance personality differences result in behavioral predispositions toward attitude objects, which should be reflected in unconscious spatial behavior.
Although BIS and BAS are assumed to be separate, orthogonal dimensions the relationship between them is often treated as being negatively correlated. It is therefore often not clear if effects found are due to approach or avoidance tendency, or both. Therefore, no concrete hypotheses about the interaction between these two dimensions can be formulated; however, it is generally expected that the higher the approach tendency, the higher the social approach would be. In contrast, higher avoidance tendency would be related to less social approach.
The primary aim of this report, however, is to shed light on a possible interaction between dispositional approach and avoidance, and situational factors, such as attitudes. This interaction is expected based on two assumptions:
First, the lack of consistent findings relating personality traits to proxemic behavior (Aiello, 1987) has been suggested to be due to contextual variability , or “a lack of congruence between the measured personality dimension and the situation in which the distance is measured” (Karabenick & Meisels, 1972). Specifically, it has been argued that personality traits could have an effect if these traits are relevant to a certain situation. Karabenick and Meisels (1972), for example, have found that trait anxiety affects interpersonal distance only if the evaluative nature of the situation is made salient. According to these authors, “operations that increase the avoidance tendency and/or
decrease the approach tendency would be expected to lead to greater distance while increasing the approach and/or decreasing the avoidance tendency would lead to smaller distances” (p. 285). In the context of the present study, it can therefore be expected that manipulations that activate approach and avoidance tendencies will, in turn, manifest in social distance.
Second, this activation of approach and avoidance tendencies is due to individual differences in sensitivity to reward and punishment (Gray (1982). A positive interaction could be rewarding and a negative interaction might have a punishing, or at least non-rewarding, character. Given these assumptions, a positive interaction with a fair and cooperative person should activate the BAS, especially in those individuals that have a highly responsive BAS. In contrast, a negative interaction with an unfair person should activate the BIS, especially in individuals that have a highly responsive BIS.. In other words, individuals with a more responsive BAS should approach a positively evaluated interaction partner while individuals with a more responsive BIS should avoid a negatively evaluated interaction partner. To test these predictions, we staged a social interaction between participants and two confederates, one of whom behaved in a fair fashion and the other of whom did not. We then tracked their movement during a social interaction by using a tracking system that allowed for very precise and reliable position and orientation measures while permitting free, naturalistic movements.
Methods Participants
Eighty participants took part in the study, 42 of whom were female. The age of the
participants ranged from 19 to 39 (M = 26.36, SD = 4.159). The data of 6 participants were excluded from the analysis due to technical problems leading to either abortion of the session (2 cases) or bad quality tracking data (3), or the suspicion of being deceived (1). The study was conducted at the Max Planck Institute for Human Cognitive and Brain Sciences (MPI) in Leipzig, Germany and was approved by the ethics committee of the institute. The participants were recruited by telephone from the Institute’s database, accounting for gender of the participant and the confederates. Gender of the two confederates was always the same. The aim of this was to control for gender effects by having approximately the same number of each of the four gender combinations. Each session lasted approximately 90 minutes. The tasks reported here were parts of a larger study.
Procedure
Participants completed a series of tasks with two confederates whom they were led to believe were fellow experimental participants. These confederates were student research assistant from the MPI who were instructed to maintain a neutral facial expression and to avoid initiating small talk and eye contact in order to minimize sympathy effects. Furthermore, one of the
confederates always arrived before and the other shortly after the participant in the lobby of the MPI to reduce suspicion. Before beginning the experiment, the three alleged participants were informed about the purpose of the study – to investigate strategic thinking in game situations – and gave their informed consent before being led to the lab (see figure 1 for schematic layout of lab). After having been equipped with headphones necessary for the physical tracking (see below), the participants first completed an economic game on their assigned computers in which fairness of the two
confederates was manipulated. Subsequently, an ostensible memory task that required participants to walk around the room was carried out. After a short post-questionnaire, the participants were probed for suspicion, debriefed and paid.
Pre-questionnaires
The participants were asked to fill in an online questionnaire prior to participation. This questionnaire was made available by personalized email invitations that linked to LimeSurvey, an open source online survey application. The questionnaire comprised a set of demographic questions and the German version of the behavioral inhibition and activation scales (BIS/BAS, Carver & White, 1994, Strobel, Beauducel, Debener, & Brocke, 2001) as measures of dispositional approach and avoidance motivations.
Figure 1. Schematic overview of the lab.
Table for instructions Participants‘ table Columns Confederates‘ tables Pictures for memory game Door
Position Tracking
Participants were tracked using a Worldviz Precision Position Tracker (PPT-H) system. During the whole session, the participants as well as the two confederates wore headphones with infrared LED markers attached to the left and right side (see figure 2). The positions of the LED markers were tracked by means of 8 optical sensors/cameras with a precision and accuracy of 0.25mm per 3m3 volume. The data were recorded at
approximately 15 Hz. This set-up allowed for recording of the position and yaw of the participants’ heads over the course of the experiment.
Economic Game
A sequential iterated variant of the Prisoners’ Dilemma was used to manipulate perceived fairness, which was adapted from
Singer et al. (2006). Beforehand, the participant and the two confederates had been asked to draw one of three envelopes, all of which contained a sheet with the letter B. These envelopes were used to give the participants the impression that the role of player B was randomly allotted to them. The two confederates, however, had been informed in advance which letter (A or C) they were supposed to pretend to have drawn.
At the beginning of every round, all of the three players received 10 coins. For the first play, Player B (the participant) could either keep the 10 coins or give them to one of the other players. However, the recipient was chosen randomly such that Player B could not know beforehand which of the other players would receive the money. If chosen to give the money away, the money was tripled and allocated to one of the two other players. The second move consisted of the chosen player sending back an amount between 0 and 10 coins, which was also tripled. This move took place irrespective of whether the participant in move 1 had sent money or not. Although the participants thought to be playing against the two confederates, the responses of the two players were actually programmed computer responses. This way, the responses were controlled and it was ensured that one of the two confederates seemed to be a fair, reciprocating player whereas the other was perceived to be unfair and gave little or no money back. The game consisted of 4 training trials and 18 rounds, whereby every round where the participant had sent no coins was repeated at the end, up until a maximum of 8 repeated trials. At the end of the session, all coins earned were translated into Eurocent with an exchange rate of 1:1, and paid out to the participants in addition to the
Figure 2. Representation of the tracking system.
standard reimbursement. The Prisoners’ Dilemma was programmed and presented with Vizard (WorldViz).
Memory Game/Proxemic Task
During the memory game, the three players were asked to walk around a room with pictures attached to the walls for 5 minutes and to find and memorize pairs of matching pictures. Twenty pairs of neutral pictures obtained from the International Affective Picture System (IAPS, Lang, Bradley, & Cuthbert, 2008) were attached to the walls and ceiling of the room at a height of approximately 1.50m. The pictures were printed on A4 sheets and covered by a liftable white thick sheet of paper so that participants had to walk over to the pictures and lift the paper before seeing which picture was underneath. Those white sheets had a number between 1 and 40 on them. The two confederates were instructed to follow a fixed walking pattern designed to appear as though they were also completing the task: After having made a clockwise round through the room, looking at every picture beginning at desk C, confederate C then started at picture number 20 and repeatedly followed the pattern +1, +2, +3, +7, +13 until the time was up. For example, (s)he would go to
pictures 20, 21, 23, 30, 3, 4, 6, 9 and so on. Confederate A would follow the same pattern, only counter-clockwise, hence subtracting the numbers of the pattern and starting at picture number 1. Importantly, as can be seen in figure 1, the layout of the room was symmetrical in order to account for environmental effects. After the time had elapsed, the three players went back to their desks to note down the pairs they had found by drawing lines between matching numbers on a graphical overview of the room.
Data Analysis Preprocessing
The raw tracking data were divided into epochs corresponding to the different stages of the experiment. The data of the memory epoch were then visualized using Vizard, and two raters independently corrected samples where LED locations were incorrectly recorded. For this end, instances where the tracked light was jumping from one location to another faster than one would be able to move (threshold: 20 cm per sample) or where the distance between the two lights per head deviated from their fixed distance (A = 0.31m, B = 0.32m, C = 0.32m) by more than 3 cm had been flagged using algorithms written with Matlab. The two raters then corrected those instances by either selecting the location of one of the other five lights (in the case of switched recorded locations of LEDs) or deleting that light from the sample. The former option was used when, for example, two lights were recorded as having switched positions and the actual position could clearly be identified. Otherwise, the light was deleted. The inter-rater agreement for all samples was satisfying (94.4% agreement, Cohen’s kappa κ = 0.8028). Samples where the two raters disagreed were revised by a
third independent rater, who made a final judgment, choosing between the options of the two corrected samples. Often, these samples of disagreement were beginnings or endings of periods of corrections, where, for example, one light had been deleted but the onset of this deletion slightly differed between raters. Therefore, no crucial differences between corrections could be detected.
Subsequently, the dependent variables for the analysis were calculated using Matlab. Minimum distance between the participant (player B) and both of the confederates (A and C) was calculated for every session. Notably, to avoid confounding influences of player height, this distance was calculated in the horizontal plane only. Next, these values were coded for fairness of the confederates, leading to measures for minimum distance towards the fair and the unfair confederate, which will be denoted as MDUF and MDF in the remaining part of this report.
Furthermore, every participant’s scores of the pre-questionnaire’s BIS and BAS scales were calculated. This was done by recoding the responses of reverse items and averaging the responses over all items per subscale according to instructions (Strobel et al., 2001) using SPSS.
Consequently, minimum distances were analyzed using hierarchical multiple linear regression analysis with BIS score, BAS score (both 1st level) and the interaction between BIS and BAS (2nd level) as predictors. The within-subjects variable was accounted for by calculating the difference between the distance to fair and unfair confederate per subject, as suggested by Judd, Kenny, and McClelland (2001). For this end, the difference towards the fair confederate was subtracted from the one towards unfair (MDUF – MDF), as this would lead to a positive difference with results in line with
expectations (greater distance to unfair confederate). The resulting difference score will be denoted as Ddif in the following section. For the regression analyses, all variables were standardized.
Importantly, this analysis does not allow testing for main effects, due to the fact that MDUF and MDF
are combined and cannot be compared. Moreover, testing for main effects of BIS and BAS would require the average, not the difference, of MDUF and MDF. Therefore, the effects of BIS and BAS
tested here will be the effects on the difference between MDUF and MDF, which are thus
interactions. Hence, separate analysis will be used to test for main effects: a paired-samples t-test comparing MDUF – MDF and a linear regression of average minimum distance (over both liking
conditions) on BIS and BAS scores. Any significant effects will be adequately further analyzed, including conducting separate multiple linear regressions of MDF and MDUF on the relevant
Manipulation Check
Paired-samples t-tests of the
post-questionnaire manipulation checks revealed highly significant effects of fairness on ratings of annoyance (t(72) = 14,181; p < 0.001), attractiveness (t(72) = 2.675; p = 0.009), enjoyableness (t(72) = 8.230; p < 0.001), fairness (t(72) = 18.265; p < 0.001), and liking (t(72) = 5.917; p < 0.001). The fair confederate was rated as being more attractive, enjoyable, fair, likable, and less annoying (reversed response options) than the unfair confederate (see table 1). It can thus be concluded that the manipulation worked very well.
Reliability BIS and BAS scales
The BIS and BAS scales showed satisfying internal consistency with Cronbach’s alphas of αBIS =
0.825 and αBAS = 0.767. Mean Standard deviation annoying_fair ,8845 ,18755 annoying_unfair ,3844 ,26773 attractive_fair ,5164 ,23982 attractive_unfair ,4258 ,24092 enjoyable_fair ,7783 ,17264 enjoyable_unfair ,5274 ,19425 fair_fair ,8776 ,14865 fair_unfair ,3469 ,19865 like_fair ,6126 ,16007 like_unfair ,4634 ,16959
Figure 3. Bar chart of the means of the manipulation check questionnaire.
Table 1. Means and standard deviations of manipulation check questionnaire. Note: N = 73
Test of assumptions of multiple linear regression analysis
First of all, Kolmogorov-Smirnov tests resulted in non-significant test statistics for all variables (pBIS = 0.514, pBAS = 0.356, pMDfair = 0.713, pMDunfair = 0.406), which means that normal
distributions of all variables can be concluded. Second, a very low correlation between BIS and BAS (r = 0.06) indicates that
multicollinearity does not exist. Lack of fit tests indicated that the
relationship between the
independent variables (BIS and BAS) and the combined dependent variable (Ddif) is linear: BIS: F(24,48)
= 0.618, p = 0.898; BAS: F(20,52) = 0.545, p = 0.931. Visual inspection of figure 4 led to the conclusion that the assumption of uniform variance, or
homogeneity of errors, is met. This is based on the fact that the distribution is even and not skewed. Furthermore, a Durbin-Watson (d = 1.905) indicated independence of errors of the variables BIS, BAS, and DDif. Finally, Shapiro-Wilk tests confirmed normality of errors (W(79) = 0.971, p = 0.090).
In conclusion, all the assumptions of a multiple regression analysis are satisfied and the analysis seems to be valid to conduct for this dataset.
Results Main effects of fairness
The paired-samples t-test comparing distance towards unfair (M = 0.5281, SD = 0.1441) and fair (M = 0.5444, SD = 0.1337) confederates revealed no significant differences (t(73) = 0.692, p = 0.491).
Main effects of BIS & BAS
The regression model including BIS and BAS scores as predictors of the average minimum distance did not fit the data (R2 = 0.009, R2adj = -0.019, F(71,2) = 0.304, p = 0.739). Both BIS and BAS
were found to be unrelated to the variance of the general minimum distance (BIS: β = -0.092, t(72) = -0.777, p = 0.440; BAS: β = -0.003, t(72) = -0.021, p = 0.983).
Figure 4 . Regression of studentized residuals on standard predicted values to test for uniform variance.
Interactions/Hierarchical regression
First level. The regression of the distance difference on BIS and BAS scores revealed a
significant effect of BIS (see table 2, model 1, for statistics). This finding implies that the difference between the distances toward fair and unfair confederates depends on BIS score, and it therefore indicates an interaction effect. However, model fit is poor, as indicated by the corresponding R2 in table 2 .
The effect of BAS on the Ddif was non-significant (see table 2, model 1). Hence, an interaction
between fairness and BAS scores cannot be concluded.
Second level. Similarly, adding the interaction between BIS and BAS to the model to test for
the three-way interaction does not significantly increase the variance explained.
Coefficients Model fit
Model Regression coefficient t p R2 Adjusted R2 R2 1 BIS -0.227 -1.999 0.049* 0.085 0.059 0.085 BAS -0.170 -1.498 0.139 2 BIS -0.248 -2.093 0.038* 0.093 0.055 0.008 BAS -0.183 -1.591 0.117 BIS x BAS -0.094 -0.797 0.428
Table 2. Regression coefficients and model fit for the hierarchical regression of Ddif on BIS, BAS, and BIS x BAS.
Note. * significant at a level of significance of p = 0.05.
Further analysis
To investigate the nature of the interaction effect found between BIS scores and fairness, two separate regression analyses were carried out for MDUF and MDF. The regression with MDF as
dependent variable revealed that BIS scores were not predictive of the variance in MDF: R2 = 0.013,
F(72, 1) = 0.966, p = 0.329. In contrast, regressing MDUF on BIS resulted in a marginally significant
effect: R2 = 0.052, F(72,1) = 3.946, p = 0.051. Notably, the regression coefficient is negative (β =
-0.228), indicating that with increasing BIS scores, the distance towards the unfair confederate
Figure 5. Regression of minimum distance on BIS scores with lines indicating the outlier thresholds (+/- 2.5).
Figure 6. Regression of minimum distance on BAS scores with lines indicating the outlier thresholds (+/- 2.5).
Outlier removal
To control for possible confounding influences of outliers, data of five participants were removed for further analysis. Three of these had deviating distances of more than +/- 2.5 standard deviation from the mean of the MDUF, whereas two had been identified similarly as univariate
outliers of MDF. No multivariate outliers were identified (according to Mahalanobis’ probability with
a significance level of p = 0.001).
A subsequent repeated analysis resulted in similar findings as the above described analysis with outliers included. No main effect of liking was found: t(68) = 0.886, p = 0.379. Similarly, neither BIS nor BAS predicted average minimum distance (BIS: β = -0.004, t(68) = 0.036, p = 0.971; BAS: β = 0.071, t(68) = 574, p = 0.568). As can be seen in table 3, again, only the interaction between fairness and BIS scores reaches significance.
Coefficients Model fit
Model Regression coefficient t p R2 Adjusted R2 R2 1 BIS -0.191 -2.042 0.045* 0.080 0.052 0.080 BAS -0.109 -1.180 0.242 2 BIS -0.197 -2.049 0.046* 0.081 0.038 0.001 BAS -0.113 -1.200 0.234 BIS x BAS -0.026 -0.270 0.788
Table 3. Regression coefficients and model fit for the hierarchical regression of Ddif on BIS, BAS, and BIS x BAS with
outliers removed
Note. * significant at a level of significance of p = 0.05.
Discussion
This study was designed to investigate the effects of and the interaction between personality and situational factors on proxemic behavior. Particularly, it was expected that approach and
avoidance dispositions would be embodied in physical approach of interaction partners, and that this relationship would be moderated by the valence of attitude towards the interaction partner.
Surprisingly, none of the expected results were found. Most unexpected was the non-significant simple comparing distance toward fair and unfair confederate. In line with a great amount on
literature, it was expected that positive attitudes would lead to a decrease in interpersonal distance. This effect could not be replicated here.
Moreover, it was expected that high dispositional approach would be related to small interpersonal distances and dispositional avoidance, in contrast, to greater distances kept with an interaction partner. However, neither approach nor avoidance could be shown to have a direct effect on proxemic behavior; moreover, no interaction between approach and avoidance was found.
The interactions between avoidance x fairness and approach x fairness did also not result in the expected findings. It was hypothesized that with increasing BIS scores, the distance towards an unfair interaction partner would increase, and that increasing BAS scores would lead to greater approach of the fair person. In contrast, an interaction between BIS scores and fairness was found that was in the opposite direction as expected: The higher the avoidance, the closer the participants came to the unfair interaction partner.
Based on the results, we failed to find an influence of either perceived fairness or approach and avoidance personality traits on interpersonal distance. Furthermore, the higher an individual scores on the avoidance (BIS) scale, the closer he or she would come to an unfair person. In this study, we only found evidence for an influence of the behavioral inhibition system on interpersonal distance when moderated by situational factors. However, there are a few aspects that indicate that this result should be treated with caution.
First, even though the interaction between approach disposition and fairness of confederates did not reach significance, visualization of the effect (see figure 6) shows a trend in the same
direction as the interaction between avoidance and fairness. Thus increasing approach tendency may be related to higher physical approach of the negatively evaluated interaction partner and avoidance of the positively evaluated one. This, especially the failure to find an opposing and significant effect of BAS and fairness, suggests that high avoidance and approach tendencies do not necessarily lead to opposite behavioral responses. However, this type of pattern could also be due to other factors. For example, similar trends for BIS and BAS might result from the fact that individuals giving extreme responses on questionnaires might also approach a negatively evaluated person more.
Second, having found no relationship between personality and behavior in this study, the question remains whether this relationship simply does not exist or the translation into quantifiable measurements has not been successful yet. Although widely used, the BIS/BAS scales developed by Carver and White (Carver & White, 1994) might not be adequate measures of approach and
avoidance dispositions. Critics (e.g. Cogswell, Alloy, van Dulmen, & Fresco, 2006; Heubeck, Wilkinson, & Cologon, 1998; Smillie & Jackson, 2005) have addressed the lack of evidence for the BIS/BAS scales
measuring the construct’s suggested by Gray. A number of more recent attempts to quantify these constructs exist; especially promising seems the scale of Elliot (2010) who proposed general approach and avoidance constructs underlying extroversion-introversion, positive-negative affect, and behavioral inhibition and activation. However, as this study showed and had been suggested by Karabenick and Meisels (1972), instead of trying to isolate more fundamental constructs, it might be more fruitful to concentrate on finding moderators and mediators to clarify the conditions and processes that might link personality to behavior.
Furthermore, it could be argued that the manipulation induced different affective states, not only positive and negative attitudes towards the confederates. Participants could react differently on the reciprocity of one confederate or the unfairness of the other in the Prisoner’s Dilemma. For example, it could lead to a positive feeling based on the cooperation of the one confederate, but it could also result in aggression or a feeling of foolishness due to the refusal of cooperation of the other, which could, in turn, lead to different proxemic behaviors. Therefore, in a further study, it would be advisable to measure affective response to the manipulation and include this in the analysis.
Aggression, for example, has already been suggested by Hall (1966) to result in smaller interpersonal distances, which is contrary to the expectation that negative attitudes lead to greater distance. Therefore, aggression could explain the results found here, if high avoidance tendency is related to higher aggression. Behavioral inhibition has been related to negative affect (Gray,
1982;Elliot & Thrash, 2010). However, in the literature, aggression, or anger, has rather been related to behavioral activation (Harmon-Jones, 2003; Verona, Sadeh, & Curtin, 2009). For example, it has been shown that left frontal brain asymmetry is related to both anger and high scores on the BAS scale (Harmon-Jones & Siegelman, 2001). The relationship between BAS sensitivity and anger might be due to a higher frustration of individuals with a high reward expectation (Corr, 2002). It could be possible that the different traits are predictive of different forms of aggression: approach tendency with an overt, physical aggression and avoidance with a more covert form. As indicated in the study by McCall and colleagues (2009), proxemics can be seen as a measure of implicit attitudes and thus could possibly also be a sign of covert aggression. Therefore, further research on the role of
aggression in proxemics could lead to an explanation for the results found here.
Finally, although carefully designed to meet all requirements of a naturalistic but controllable interaction, the memory task itself might have led to confounding effects. The task demands,
associated with looking at the pictures, might have overridden any implicit attitudes or the typical behavior of approach and avoidance orientation. The pictures were attached to the walls with a distance of only around 20cm, and it happened frequently that two players were looking at the same
or neighboring pictures. These task-related encounters could have easily affected minimum distance measures, as these detect the one-time closest approach only. Thus, I would suggest using a
different, less restricted, more naturalistic task in further research. Moreover, the use of minimum distance as a measure of proxemics might confound some interesting findings. I would suggest using a more differentiated measure, which incorporates minimum distance, mean distance, and possibly even orientation (on which side of the body is an interaction partner located). Dividing the space around the participant into different zones for the analysis could be a way to include more of the information about the interaction. Even though (one-sample) minimum distance might not differ, the amount of time spent in different proximities and/or spatial orientations could still differ.
In conclusion, this study found no evidence for the influence of approach or avoidance traits or of perceptions of fairness on social spatial behavior. However, having found a significant
interaction between BIS and fairness supports the proposition that the effect of personality traits on proxemics should be studied together with moderating situational factors (Karabenick & Meisels, 1972). However, surprisingly, the effect found is not in the direction hypothesized, which calls for research on mediating factors to explain this relationship, for example aggression could be a candidate. To summarize, I would suggest investigating the relationship between behavioral
inhibition and approach of a negatively evaluated person more closely by using different measures of this avoidance tendency and an even more naturalistic proxemic task. Furthermore, I would include another questionnaire after the manipulation, measuring the affective state of the participants.
R
eferencesAiello, J. R. (1987). Human spatial behavior. In D. Stokols & I. Altman (Eds.), Handbook of
Environmental Psychology (1st ed., pp. 389-504). New York: John Wiley & Sons.
Allport, G. W. (1935). Attitudes. In C. Murchison (Ed.), Handbook of Social Psychology (pp. 798-844). Worchester, MA: Clark University Press.
Altman, I., & Vinsel, A. M. (1977). Personal space: An analysis of E. T. Hall’s proxemic framework. In I. Altman & J. Wohlwill (Eds.), Human Behavior and Environment (Vol. 2.). New York: Plenum. Azrin, N. H., & Holz, W. C. (1966). Punishment. In W. K. Honig (Ed.), Operant behavior: areas of
research and application (pp. 380-447). New York: Appleton-Century-Crofts.
Byrne, D., Ervin, C. R., & Lamberth, J. (1970). Continuity between the experimental study of attraction and real-life computer dating. Journal of Personality and Social Psychology, 16(1), 157-165. Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective
responses to impending reward and punishment: The BIS/BAS scales. Journal of Personality and
Social Psychology, 67(2), 319-333.
Chen, M., & Bargh, J. A. (1999). Consequences of Automatic Evaluation: Immediate Behavioral Predispositions to Approach or Avoid the Stimulus. Personality and Social Psychology Bulletin,
25(2), 215-224.
Cogswell, A., Alloy, L. B., van Dulmen, M. H. M., & Fresco, D. M. (2006). A psychometric evaluation of behavioral inhibition and approach self-report measures. Personality and Individual Differences,
40(8), 1649-1658.
Cook, M. (1970). Experiments on Orientation and Proxemics. Human Relations, 23(1), 61-76. Corr, P. J. (2002). Gray’s reinforcement sensitivity theory and frustrative nonreward: a theoretical
note on expectancies in reactions to rewarding stimuli. Personality and Individual Differences,
32, 1247-1253.
Dotsch, R., & Wigboldus, D. H. J. (2008). Virtual prejudice. Journal of Experimental Social Psychology,
44(4), 1194-1198.
Droba, D. D. (1933). The Nature of Attitude. Journal of Social Psychology, 4, 444-462. Elliot, A. J., Murayama, K., & Pekrun, R. (2011). A 3 × 2 achievement goal model. Journal of
Educational Psychology, 103(3), 632- 648.
Elliot, A. J., & Thrash, T. M. (2010). Approach and avoidance temperament as basic dimensions of personality. Journal of personality, 78(3), 865-906.
Frankel, A. S., & Barrett, J. (1971). Variations in personal space as a function of authoritarianism, self-esteem, and racial characteristics of a stimulus situation. Journal of Consulting and Clinical
Psychology1, 37, 95-98.
Gifford, R. (1982). Projected Interpersonal Distance and Orientation Choices : Personality , Sex , and Social Situation. Social Psychology Quarterly, 45, 145-152.
Gray, J. A. (1982). The Neuropsychology of Anxiety. New York: Oxford University Press.
Gray, J. A. (1990). Brain systems that mediate both emotion and cognition. Cognition & Emotion,
4(3), 269-288.
Hall, E. T. (1959). The Silent Language. New York: Doubleday.
Hall, E. T. (1963). A System for the Notation of Proxemic Behavior. American anthropologist, 65(5), 1003–1026. Wiley Online Library.
Hall, E. T. (1966). The Hidden Dimension. (Anchor Books Doubleday, Ed.)The Hidden Dimension (Vol. 6, pp. 94–94). New York: Doubleday.
Harmon-Jones, E. (2003). Anger and the behavioral approach system. Personality and Individual
Differences, 35(5), 995-1005.
Harmon-Jones, E., & Siegelman, J. (2001). State anger and prefrontal brain activity: evidence that insult-related relative left prefrontal activity is associated with experienced anger and aggression. Journal of Personality and Social Psychology, 80, 797-803.
Heshka, S., & Nelson, Y. (1972). Interpersonal Speaking Distance as a Function of Age , Sex , and Relationship. Social Influence, 35, 491-498.
Heubeck, B. G., Wilkinson, R. B., & Cologon, J. (1998). A second look at Carver and White’s (1994) BIS/BAS scales. Personality and Individual Differences, 25(4), 785-800.
Higgins, E. T. (1997). Beyond pleasure and pain. The American Psychologist, 52(12), 1280-1300. Judd, C. M., Kenny, D. A., & McClelland, G. H. (2001). Estimating and testing mediation and
moderation in within-subject designs. Psychological Methods, 6(2), 115-134.
Karabenick, S. A., & Meisels, M. (1972). Effects of performance evaluation on interpersonal distance.
Journal of personality, 40(2), 275-286.
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2008). International affective picture system (IAPS):
Affective ratings of pictures and instruction manual. Technical Report A-8. Gainesville, FL.
Little, K. B. (1965). Personal space. Journal of Experimental Social Psychology, 1, 237-247. McCall, C., Blascovich, J., Young, A., & Persky, S. (2009). Proxemic behaviors as predictors of
aggression towards Black (but not White) males in an immersive virtual environment. Social
Influence, 4(2), 138-154.
Mehrabian, A. (1968). Relationship of attitude to seated posture, orientation, and distance. Journal
of Personality and Social Psychology, 10, 26-30.
Mehrabian, A. (1969). Significance of posture and position in the communication of attitude and status relationships. Psychological Bulletin, 71, 359-372.
Moreland, R. L., & Zajonc, R. B. (1982). Exposure effects in perception: Familiarity, similarity, and attraction. Journal of Experimental Social Psychology, 18, 395-415.
Singer, T., Seymour, B., O’Doherty, J. P., Stephan, K. E., Dolan, R. J., & Frith, C. D. (2006). Empathic neural responses are modulated by the perceived fairness of others. Nature, 439(7075), 466-9. Smillie, L. D., & Jackson, C. J. (2005). The appetitive motivation scale and other BAS measures in the
prediction of Approach and Active Avoidance. Personality and Individual Differences, 38(4), 981-994.
Strobel, A., Beauducel, A., Debener, S., & Brocke, B. (2001). Eine deutschsprachige Version des BIS/BAS-Fragebogens von Carver und White. Zeitschrift für Differentielle und Diagnostische
Psychologie, 22(3), 216-227.
Strube, M. J., & Werner, C. (1982). Interpersonal distance and personal space: A conceptual and methodological note. Journal of Nonverbal Behavior, 6.
Verona, E., Sadeh, N., & Curtin, J. J. (2009). Stress-induced asymmetric frontal brain activity and aggression risk. Journal of abnormal psychology, 118(1), 131-45.
Willis, F. N. J. (1966). Initial speaking distance as a function of the speakers’relationship.
