It is Written in Your Eyes: Hostile Attributions and Self-Directed Gaze Perception in Incarcerated Violent Adolescent Male Offenders

Abstract

To date, we have frugal knowledge about the hostile attribution bias (HAB) and the biased gaze perception in violent adolescent offenders. This however is a major contributing factor in understanding delinquent behavior. Using a computer-based approach, presenting faces modulated in gaze direction (0°, 2°, 4°, 6°, 8°) and valence (angry, fearful, happy, neutral), the present study examined the impact of HAB of the feeling of being stared at in a sample of 27 adolescent offenders (aged 17-24 years). The study was conducted institution-intern in the Department for Social Therapy of a German correctional facility. Results showed that in comparison with faces with negative expressions, happy faces were more likely to be perceived as self-directed. Interestingly, emotion showed significant influence of the gaze perception in only two viewing angles (2° and 6°), revealing the role of the facial expression in highly ambiguous conditions. Furthermore, hostility did not modulate the relationship between the self-referential gaze perception and (negative) facial expression. Possible frameworks and limitations of the study are discussed.

Keywords

aggressive behavior hostile attribution bias emotion self-directed gaze perception juvenile delinquency

The phenomenon of juvenile delinquency and violent behavior has received an enormous amount of scientific attention in the recent times. Discovering the factors associated with the manifestation and maintenance of aggressive behavior is an important step in understanding youth crime. Deficient social information processing may at least partly explain socially inappropriate behavioral reactions in aggression-prone individuals. The hostile attribution bias (HAB; e.g., Nasby, Hayden, & DePaulo, 1980) is a fundamental cognitive process found to reinforce aggressive behavior. The HAB refers to the individual’s tendency to perceive other people’s provocative, yet ambiguous, actions as hostile and to overattribute malicious intent to other people’s behavior (e.g., Crick & Dodge, 1994; Dodge, 1980; Dodge & Coie, 1987; Dodge & Newman, 1981; Dodge, Price, Newman, & Bacharowski, 1990; Orbio de Castro, Veerman, Koops, Bosch, & Monshouwer, 2002; Van Oostrum & Horvath, 1997).

HAB is usually assessed with vignettes describing (if written) or showing (if videotaped) different social situations, for example, being hit by a ball in the back. The protagonist in the vignettes acts in an unclear or ambiguous way which has a negative outcome for the subject of the story, in this case for the participants in the study. When asked to attribute intent to other’s behavior, aggressive subjects were significantly more likely to interpret the peer’s intention as hostile, whereas nonaggressive controls tended to attribute a benign intent (for overview, see Orbio de Castro et al., 2002). In addition, antisocial-aggressive individuals were most likely to solve social problems in the vignettes by defining problems in hostile ways, adopting hostile goals, seeking few additional facts, generating few alternative solutions, anticipating few consequences for aggression, and choosing few “best” and “second best” solutions that were rated as “effective” (Slaby & Guerra, 1988).

During social interactions, facial expressions represent one of the most important information sources in the ongoing stream of various social cues. When involved in a social interaction, individuals have to deal with facial expressions with diverse emotional valences and then use these subtle cues to evaluate other people’s intentions or current emotional state. An efficient processing of such social information provides the base for successful social interactions and suitable social behavior. However, due to its complex and dynamic nature, facial affect has a high potential for ambiguity and tends to support the biases of the beholder. Schönenberg and Jusyte (2014) suggested that the impaired ability to decode important social signals such as facial expressions might be a probable reason for individuals to interpret another’s behavior inaccurately and to attribute hostile intention. In fact, aggressive behavior was shown to be linked to deficits in the perception of unambiguous facial affect (Marsh & Blair, 2008). However, the possible connection between HAB and the processing of facial expression has been only partially investigated. Initial studies with normative adolescent and adult populations showed that the aggression trait is moderately associated with hostile perception of others (Burt, Mikolajewski, & Larson, 2009) and that subjects with high scores in hostility perceived happy and neutral faces as less friendly (Knyazev, Bocharov, & Slobodskoj-Plusnin, 2009). Furthermore, an investigation into a sample of institutionalized boys suggested that a generalized attributional bias to infer hostility (negative-dominant affect) from various classes of unambiguous social stimuli becomes more pronounced as aggressiveness increases (Nasby et al., 1980).

Schönenberg and Jusyte (2014) found a hostile response bias to emotionally ambiguous faces in a population of incarcerated antisocial violent offenders. They used a morph task, in which affective pictures (angry, happy, and fearful) were subsequently morphed with each other to create three continuous dimensions (happy–fearful, happy–angry, and fearful–angry). In this way, the authors created ambiguous facial stimuli that contained different proportions of angry, happy, and fearful expressions. Results suggested that aggression is associated with a strong preference to interpret ambiguous stimuli containing proportions of an angry expression as hostile, while there was no evidence for a generally biased interpretation of distress cues under conditions of uncertainty. Aggressive individuals, as compared with controls, not only (mis)interpreted ambiguous facial cues as hostile but also showed a strong tendency to systematically overrate the perceived intensity of anger. This may suggest that antisocial individuals felt more confident when making their hostile responses, whereas control participants were less convinced that they responded appropriately under maximal uncertainty conditions (Schönenberg & Jusyte, 2014).

Although a biased face perception could modulate an individual’s reaction to ambiguous stimuli, there is one more factor, which could be essential for the display of socially inappropriate or aggressive behavior. This is the tendency to perceive and interpret social stimuli as being directed to oneself. An example of an interpersonal factor, indicating whether someone refers to us or not, is the gaze direction. Gaze direction carries substantial information for the social interaction revealing one’s disinterest or attention towards another individual (Itier & Batty, 2009). Notably, human beings perceive a remarkable range of gaze directions as self-directed (Gamer & Hecht, 2007). The perception of gazes is influenced by several factors, including the emotional expression. In particular, individuals tended to feel more likely to be looked at by happy rather than by negative or neutral facial expressions (Lobmaier & Perrett, 2011; Lobmaier, Tiddeman, & Perrett, 2008). This phenomenon is known as self-referential positivity bias, a mechanism that serves to increase one’s self-esteem and to reinforce a positive view of the world.

Gaze direction provides socially relevant information that reflects other people’s intentions. Thus, the abnormal processing of gaze cues in social interactions may induce irritation, annoyance, or feelings of being provoked and, thereby, trigger an aggressive response. Therefore, the tendency to misinterpret gaze directions may at least partly underlie aggressive–impulsive behavior in susceptible individuals.

To our knowledge, there are no studies investigating this phenomenon in a population of violent offenders. As a result, the present study sought to investigate whether highly hostile violent juvenile offenders (aged 17-24 years) do feel more looked at. Individuals convicted for violent crimes were recruited from the Department for Social Therapy (DST) in a German correctional facility (Jugendstrafanstalt Berlin). Experimental, as well as clinical, assessments were conducted in designated rooms of the facility by trained psychologists from our research group. First, we address the question of whether incarcerated juvenile violent offenders show the same self-referential positivity bias for preferable processing of happy faces as compared with self-directed. Second, we investigate whether emotional expressions influence the general interpretation of gaze direction. Considering the findings in the sphere of social cognition and emotion processing in violent offenders, we hypothesized a positive relation between a self-directed perception of gaze directions and hostility, especially in the presence of negative facial expressions.

Method

Sampling and Procedure

The initial sample consisted of 45 male adolescent violent offenders recruited face-to-face in the DST of a German correctional facility (Jugendstrafanstalt Berlin). All of them were incarcerated for a violent or sexually motivated offense and were currently undergoing social therapy.

Experimental as well as clinical assessments were conducted in designated rooms of the DST facility by psychologists from our research group trained to administer and interpret the applied tests and familiar with the technical implementation of the experimental task. To ensure confidentiality of the data collection, prison officers were not present in the assessment room. Participants were initially introduced to the experimental concept. They provided a written informed consent for their participation in the experiment and were then instructed of the experimental task and procedure. The complete experiment consisted of 288 trials in random order (four expressed emotions, nine viewing angles, four actors, two repetitions), and its duration was approximately 15 min. To ensure accurate understanding of the experimental task, participants were informed of the procedure and then completed a short training session (eight trials). Experimental and clinical assessments were performed individually. The entire set of task and tools took approximately 45 min per participant to complete (including introduction and debriefing). After completion of the study, all participants received monetary compensation for their participation. Remuneration was provided to appropriately compensate subjects for their time, efforts, and inconvenience of participation. Even if subjects decided to withdraw from the study, they were compensated. To avoid undue influence and profit-related participation, the level of compensation was set to match subjects’ economic status, resources, and minimal hourly payment rate in the state of Berlin (10 Euro).

Due to the treatment concept of the DST, participants with active schizophrenia, psychotic symptoms, hallucinations, paranoid delusions, contemporary use of psychotropic medication, mental retardation, and/or poor German language skills were not included in the initial sample. In addition, participants who performed at a very low level in the experimental task (see “Stimuli and Experimental Task” and “Statistical Analyses” sections) were also excluded from the analyses (n = 18). The final sample consisted of 27 participants.

Diagnostic Measures for Hostility

To ensure a maximum comprehensive assessment of hostility-prone cognitions, we used a test battery consisting of three different measures for hostility. First, hostility as an immanent part of aggressive behavior was assessed with the Hostility Scale of the German version (Herzberg, 2003) of the Buss–Perry Aggression Questionnaire (BPAQ; Buss & Perry, 1992). The Hostility Scale consists of eight items referring to diverse self-reported hostility-based beliefs, such as “I know that my friend does badmouth me behind my back.” Items are to be answered with a Likert-type scale ranging from 1 = strongly disagree to 5 = strongly agree. In the present sample, somewhat low internal consistency was observed (Cronbach’s α = .61).

Personality Style and Disorder Inventory (Persönlichkeits-Stil- und Störungs-Inventar [PSSI]; Kuhl & Kazén, 2009) was employed to assess hostility as a personality style. PSSI is a self-report questionnaire assessing personality styles, which correspond to the personality disorders described in the psychiatric diagnostic manuals Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) but in fact represent nonpathological expressions of those disorders (Kuhl & Kazén, 2009). The Paranoid Style Scale measures the nonpathological dimensions of the paranoid personality disorder and shows a close semantic relation with the construct hostility. It is characterized by a basic distrust, inability to rely on other people or hesitancy to do so, constant doubt of other’s loyalty, permanent feeling of being exploited or discriminated against others, and interpreting the other’s behavior as deliberately, demeaning, derogatory, or threatening (Kuhl & Kazén, 2009). The scale consists of 10 items such as “A certain distrust of others is often appropriate.” Items are to be answered with a Likert-type scale ranging from 0 = strongly disagree to 3 = strongly agree. In the present sample, somewhat low internal consistency was observed (Cronbach’s α = .61).

Hostility as information processing pattern and attribution of intent was assessed using five hypothetical vignettes originally designed to assess hostile attributions in both institutional and noninstitutional social settings (see appendix). The generation and validation of the presented vignettes are described in detail elsewhere (Serin, 1991). In short, all vignettes represent a hypothetical scenario in which protagonists display an ambiguous action, creating a negative outcome for the subject in the story. Participants are asked to imagine themselves as subjects in those scenarios and then explain why the protagonists acted the way they did and how the participants perceive the actors’ behavior.

Participants responded in writing to series of closed and open-ended questions, identical for each vignette. We used the same question set developed by Serin (1991) and used in the following studies (James & Seager, 2006; Seager, 2005; Vitale, Newman, Serin, & Bolt, 2005) but we submitted the questions with a dichotomous answering format: “Do you think their actions were accidental?” “Do you think their actions were intentional?” “Do you think their actions were done out of disrespect?” “Do you think they thought their actions were right?” Hostility-based answers were scored as 1 and benign answers with 0. In addition, we included one more closed question, explicitly asking whether participants judge the protagonist’s behavior as hostile or benign, and two open-ended questions to give participants the opportunity to write their own interpretation of the situation (“What do you think is the most plausible explanation of this situation?” and “How would you describe the protagonist in the story?”). Answers were scored with 1, when participants provided a hostile interpretation of the situation (“he intended to harm me,” “he wanted to provoke me,” “they wanted to show who has the power and who has to subordinate,” etc.) or described individuals in the hypothetical stories as “malicious,” “aggressive,” towards them, “provocative,” “bad,” “hostile,” and so on. Answers were scored with 0 when whey provided a benign attribution of intent (“he did it by accident,” “he did not mean ill,” “they are just doing their job”) or described the protagonists as “distracted,” “forgetful,” “unintentional,” and so on. Thus, hostile attribution scores for each vignette ranged from 0 to 7 and from 0 to 35 for the whole instrument.

Regarding the vignette tool, Cronbach’s alpha was computed once for each vignette separately and once for the whole instrument. In the present sample, we found a good internal consistency for the first (Cronbach’s α = .84) and the second (Cronbach’s α = .88) vignette as well for the entire tool (Cronbach’s α = .82), an acceptable internal consistency for the fourth (Cronbach’s α = .78) and the fifth (Cronbach’s α = .73) vignette and a low internal consistency for the third one (Cronbach’s α = .62).

Hostility as personality style (PSSI) and hostility as a part of the aggressive behavior (AQ) were strongly positively correlated, r(27) = .59, p = .002. Hostility measured by the vignettes method was not correlated with the other two measures (ps > .05).

Tests to see whether the data met the assumption of collinearity indicated that multicollinearity was not a concern (AQ hostility score: Tolerance = .77, Variance Inflation Factor (VIF) = 1.29; PSSI hostility score: Tolerance = .77, VIF = 1.29).

Stimuli and Experimental Task

The development and validation of the stimuli are already precisely defined elsewhere (Lobmaier & Perrett, 2011; Lobmaier et al., 2008). The acquired images represented faces of two male and two female actors varying in their emotional expression (angry, fearful, neutral, or happy) and viewing angles (direct gaze direction 0° and rotated gaze directions to 2°, 4°, 6°, and 8° to the left and right). Gaze and head orientation were overlapping. Lobmaier and colleagues (Lobmaier & Perrett, 2011; Lobmaier et al., 2008) validated the set of stimuli showing correct recognition and classification of the emotional expressions. A series of example images for each emotional expression in different viewing angles is shown in Figure 1.

Figure 1.

Example images for each emotional expression in four different viewing angles.

Using a forced-choice, yes-or-no task, participants decided whether a presented face was looking directly at them or not. Each stimulus was presented for 300 ms on the screen and participants submitted their answer by pressing the j key for yes, looking at me and n key for no, not looking at me. All stimuli were presented in random order. Participants received no feedback on whether their decision was right or wrong. After participants had answered, the next trial appeared automatically on the screen.

Statistical Analyses

First, we separately calculated the means for both looking-at-me and not-looking-at-me answers for each emotional expression and viewing angle for each participant, and then transformed the means in percentages. A higher mean indicates that a person feels strongly looked at, while a lower mean reflects a weaker self-referential perception in the particular gaze direction. Corresponding gaze directions to the left and to the right were merged together because prior analyses showed no significant difference in the effects between left and right direction (Lobmaier & Perrett, 2011; Lobmaier et al., 2008).

Second, to examine the hypothesis of a positive self-referential bias, looking-at-me answers (dependent variable) were subjected to a two-way repeated measures analysis of variance (ANOVA) with the factors emotion (neutral, angry, fearful, happy) and gaze direction (0°, 2°, 4°, 6°, 8°) while controlling for the effects of age, educational level, and delict. Then we added the individual score of self-reported hostility as a covariate to estimate the potential influence of hostility in gaze perception. The effects of age, educational level, and delict were also controlled for in the multivariable model.

In a third step, the effects of hostility on response latencies were analyzed repeating the same statistical steps. First, we examined the looking-at-me answers with a two-way repeated measures ANOVA with the factors emotion (neutral, angry, fearful, and happy) and gaze direction (0°, 2°, 4°, 6°, 8°). Then we added the self-reported hostility scores as covariate to assess the effect of hostility on the response times.

We created an additional exclusion criteria based on the level of performance in the experiment and the frequency of yes/no responses. Participants who completed less than 100% of all trials (288 in total) and exhibited a response bias by answering with a yes or no in more than 80% of all trials were excluded from the final analyses (n = 18).

Partial eta square ( $η_{p}^{2}$ ) is reported to indicate the effect size. In case of violating the sphericity assumption, we used the Greenhouse-Geisser correction. Significant main effects were followed by Bonferroni corrected pairwise comparisons. Post hoc analyses of significant interactions included separate one-way ANOVAs for repeated measures. Statistical analyses were calculated using SPSS version 22. The significance level for all tests was p < .05.

Since Shapiro–Wilk test is more appropriate for small sample sizes (<50), we used it as numerical means of assessing normality. A Shapiro–Wilk’s test (p > .05; Shapiro & Wilk, 1965) and a visual inspection of their histograms, normal Q-Q plots, and box plots showed that the proportions of looking-at-me-answers were approximately normally distributed among all four emotional expressions and five viewing angles.

Results

Demographic Data

The subjects ranged in age from 17 to 24 years (M = 20.33 years, SD = 1.81 years). The majority were Germans (32.3%), followed by Turkish (16.1%), Bosnian (9.7%), Lebanese, Polish, and double citizenship holders (each 3.2%). Regarding the convicted offense, the majority of the inmates were incarcerated for robbery and extortion (35.5%), followed by nonsexual violent offenses such as assault (29.0%), (attempted) homicide (6.5%), sexual violent offenses (6.5%), sexual abuse of children or juveniles (3.2%), fraud (3.2%), and other offenses (3.2%). Additional offenses, committed parallel with the current crime, were registered for 32.3% of the participants. A total of 51.6% of the participants had already committed at least one additional prior crime which was taken into account for the current incarceration.

Testing the Social Cognition

First, proportions of looking-at-me answers as well as reaction times (RTs) were analyzed. To replicate the findings of Lobmaier and colleagues (Lobmaier & Perrett, 2011), a repeated measures ANOVA was conducted on the influence of two independent variables (emotion and gaze direction) on the proportions of looking-at-me-answers. Emotion included four levels (neutral, angry, happy, and fearful) and gaze direction consisted of five levels (0°, 2°, 4°, 6°, 8°). All effects were significant at the .05 significance level. The main effect of emotion yielded an F ratio of F(1, 3) = 3.903, p = .012, $η_{p}^{2}$ = .131, indicating a significant difference between the four emotional expressions. Happy faces evoked a stronger feeling of being looked at (M = .59, SD = .02), followed by neutral (M = .55, SD = .02), angry (M = .53, SD = .03), and fearful faces (M = .52, SD = .03). Post hoc pairwise comparisons (Bonferroni corrected) of all emotions showed that happy faces evoked significantly more looking-at-me-answers than angry faces (p = .007) and fearful faces (p = .018). Neutral faces did not differ (p = 1).

The main effect for the factor gaze direction yielded an F ratio of F(1, 4) = 48.048, p > .001, $η_{p}^{2}$ = .649. Except the 0° (direct) gaze which did not differ from the 2° averted gaze (p = 1), all pairwise comparisons (Bonferroni corrected) between the different angles were statistically significant (all ps < .001). This is not unexpected as it reflects more looking-at-me-answers by less averted gazes (a gaze position which confront the observer straight on) than by larger viewing angles. Direct gazes 0° evoked the strongest feeling of being looked at (M = .69, SD = .03), followed by 2° averted gazes (M = .69, SD = .02), 4° (M = .57, SD = .03), 6° (M = .45, SD = .03), and 8° (M = .34, SD = .03).

The interaction effect between both the factors emotion and gaze direction was as well significant—F(1, 12) = 2.094, p = .047, $η_{p}^{2}$ = .075. To further investigate the significant interaction, separate one-way ANOVA with the factor emotion was conducted for each viewing angle. We found significant effects for emotion only in the conditions of 2°—F(1, 3) = 5.602, p = .002, $η_{p}^{2}$ = .177—and 6°—F(1, 3) = 4.483, p = .007, $η_{p}^{2}$ = .147—while there were no significant effects in the remaining viewing angles (all ps > .050). Moreover, we found that the effect of the facial expressions of the self-referential perception of averted gazes seems to be specific for happy faces: more than other emotional expressions participants tend to perceive averted gaze of happy faces in all gaze directions as more self-referential (see Table 1).

Table 1.

Descriptive Statistics (Means and Standard Deviations) for the Self-Referential Gaze Perception for Every Emotion in Every Viewing Angle.

Gaze	Happy	Neutral	Angry	Fearful
0°	.741 ± .036	.657 ±. 040	.681 ± .048	.718 ± .038
2°	.762 ± .028	.699 ± .037	.683 ± .036	.618 ± .037
4°	.625 ± .034	.581 ± .043	.542 ± .042	.558 ± .039
6°	.516 ± .038	.477 ± .037	.440 ± .043	.396 ± .045
8°	.331 ± .037	.361 ± .033	.331 ± .042	.352 ± .043

Note. The cells show the mean values and standard deviations (±) of the self-referential perception of the gaze direction for the respective viewing angle and the respective emotion.

Next, to analyze the relationship between hostility and the self-referential gaze perception, we separately entered the three different hostility variables (AQ, PSSI, and the hypothetical vignettes) as covariates in the multivariable analysis while controlling for the effects of age, educational level, and offense. Both the factors, emotion and gaze perception, their interaction, as well as all interactions comprising the factor hostility failed to reach statistical significance (all ps > .050).

For the RTs, we calculated a two-way ANOVA with the factors emotion (happy, angry, fearful, neutral) and viewing angle (0°, 2°, 4°, 6°, 8°) as within-participant factors, revealing a nearly significant effect of emotion—F(1, 3) = 2.609, p = .057, $η_{p}^{2}$ = .091—and significant effect of viewing angle—F(1, 4) = 2.893, p = .026, $η_{p}^{2}$ = .100—while there were no effects for the interaction between both the factors (all ps = .412). RTs were shortest for happy faces (1,079.072 ms), followed by neutral faces (1,085.304 ms), angry faces (1,111.683 ms), and fearful faces (1,170.797 ms). Contrary to the expectations, RTs did not increase with the increment of the viewing angle but showed a dynamic fluctuation: RTs were shortest for stimuli facing the observer in the 0° condition (1,065.337 ms), followed by 6° (1,073.738 ms), 2° (1,099.366 ms), 8° (1,124.059 ms), and 4° (1,196.069 ms), again indicating a high ambiguity in the middle angles conditions. All other interactions, after including the different hostility scores, were not significant (all ps > .050).

Discussion

In the present study, we investigated both the self-directed perception of eye gaze and facial affect processing in incarcerated juvenile violent offenders with high levels of hostility. Hostility was assessed in three different ways: as part of the aggressive behavior, as a personality trait, and as attribution and interpretation bias in ambiguous social settings.

Consistent with prior studies (e.g., Lobmaier et al., 2008), our data highlighted the importance of the emotional facial expression as a key factor which determines the frequency of perceived gaze as self-directed. In comparison to faces with negative expressions such as fear and anger, happy faces were more likely to be perceived as self-directed. In addition, angry faces evoked a stronger feeling of being looked at than fearful faces. Moreover, juvenile violent offenders also exhibited an increased sensitivity to perceive happy faces fastest (as reflected by the RTs), followed by neutral, angry, and fearful faces. Regarding the viewing angle, we found that participants responded with more looking-at-me-answers by less averted gazes than by larger viewing angles. Interestingly, emotion showed a significant influence of the gaze perception in only two viewing angles (2° and 6°) while there were no significant effects in the remaining gaze directions. This suggests that the emotional face expression plays an essential role of gaze processing, especially in highly ambiguous conditions (angles). Although it might be easy to determine a direct gaze (0°) as self-directed and strongly averted gaze (8°) as turned aside, the middle viewing angles, such as 2° and 6°, seemed to evoke a high uncertainty in deciding whether a face is looking at the observer or not. In such ambiguous conditions, participants not only tended to perceive gazes as self-directed but were also faster when judging perceived direction (as reflected by the RTs). Again, the effect of the facial expressions of the self-referential perception of averted gazes was shown to be specific for happy faces: compared with other emotional expressions participants tended to perceive averted gaze of happy faces in all gaze directions as more self-referred.

Hostility, Affect Recognition, and Self-Directed Perception

The present study is among the first to investigate the impact of hostility of gaze perception. Our findings demonstrate that hostility does not modulate the relationship between the self-referential perception of viewing angles and (negative) facial expression processing in juvenile violent offenders. Contrary to expectations, participants with a high level of hostility neither exhibit a preferable processing of angry faces nor tended to perceive angry faces as looking at the observer significantly more often than faces with neutral or positive expressions. In highly ambiguous conditions, such as 2°- and 6°-averted gaze directions, participants tended to perceive happy faces as self-directed. However, the prioritized perception of happy faces could also be an indicator of a hostile reaction to provocation in social situations. In fact, aggressive behavior and hostility are both partly driven by the belief that oneself has actually been threatened or provoked (Dodge, 2006). Usually, hostile response bias to emotional faces is shown to occur as a reaction to stimuli containing proportions of an angry expression (Schönenberg & Jusyte, 2014). We suppose that happy faces possess the same potential to be perceived as provoking as negative facial expressions do. In real life, smiling faces not only represent a positive and, thus, nonthreatening approach to another person. They could also indicate an insulting or offensive attitude and, therefore, be perceived as disturbing or unpleasant and cause feelings of annoyance, distress, or irritation. Smiling faces might serve as potent signals of threat in hostility-prone individuals. Future studies, however, are needed to investigate these interesting assumptions and provide additional evidence of how people perceive happy expressions as well as what kind of attributions smiling faces can evoke.

Furthermore, the fact that there is no correlation between hostility-prone attributions and a preferable processing of angry face gazes as self-directed could be due to the impaired recognition of unambiguous facial affect that is found among antisocial populations (e.g., Marsh & Blair, 2008). Logically, before perceiving the gaze of a face as self-directed, one has to first be able to detect, identify, and correctly categorize the onset of the facial affect. The assumption of an increased sensitivity to threatening or provoking faces in aggressive offenders can be considered counterintuitive given the fact that there is no consistent evidence showing prioritized detection of angry facial cues in those with antisocial tendencies. In fact, some recent findings provide grounds to assume that aggressive individuals show reduced, rather than increased, sensitivity to facial signals of anger (Best, Williams, & Coccaro, 2002; Fairchild, Stobbe, Van Goozen, Calder, & Goodyer, 2010; Fairchild et al., 2009; Schönenberg, Louis, Mayer, & Jusyte, 2013). Such deficits in facial affect recognition can at least partly explain the lack of a consistent link between hostility and biased perception. Reduced sensitivity to angry expression might block the continuity of the perception–interpretation process and, thus, interrupt the potential manifestation of a biased interpretation and attribution of intent.

In sum, evidence from previous studies suggests that aggressive individuals tend to habitually attribute hostile intent to others in social situations with high ambiguity (Dodge, 2006). In addition, this characteristic pattern is not only driven by concrete actions of a social partner that might be misinterpreted as deliberate provocation but also emerges in the processing of facial cues (Schönenberg & Jusyte, 2014). However, this deficit processing does not seem to take place on a self-referential basis. Our data support past evidence suggesting that antisocial individuals show reduced sensitivity to negative (angry) facial expressions (perceptual bias), but we could not confirm the hypothesized link between hostility and self-referential gaze perception (cognitive interpretation bias).

Limitations of the Study

The study has several methodological and sample-related limitations. First, we used two self-report measures for assessing hostility (AQ und PSSI). Those are prone to recall error and social desirability. This is especially prominent with samples of offenders, who are characterized by low educational levels, drug abuse, and mental health problems. We tried to counteract this issue by including semiprojective measures for hostility such as the vignette tool to neutralize the biased self-reports.

Although we assessed hostility in three different ways (as part of the aggressive behavior, as a personality trait, and as hostile attributional bias), we did not find evidence for the hypothesized link between the tendency to perceive gazes of angry faces as self-related and the level of self-reported hostility. However, the lack of such evidence might be due to some features of the instruments assessing hostility. In the present study, the Hostility Scale has even lower internal consistency (α < .7) than in the original adaptation (Cronbach’s α = .75; Herzberg, 2003). The same limitation applies to PSSI as well as one of the vignettes. The low internal consistency casts doubt on the reliability of these measures.

A methodological issue also concerns the vignettes tool used in this study. Although numerous studies have shown a significant association between aggressive behavior and hostile attribution of intent to peers, as assessed with vignettes, findings are very divergent and effect sizes differed considerably between studies. A recent meta-analysis (Orobio de Castro, Veerman, Koops, Bosch, & Monshouwer, 2002) uncovered the role of different moderators such as participants’ characteristics (kind of behavior problems, sociometric status, intelligence, age, and gender) and methodological tasks’ characteristics (personal involvement, stimulus presentation, context, number of stimuli, response format, setting, scoring, and reliability) accounting for differences between effect sizes in the studies. Larger effects were associated with more severe aggressive behavior, rejection by peers, participants’ age, and absence of control for intelligence. Audio presentation of stimuli was associated with larger effect sizes than were video and picture presentation. Staging of actual social interactions was associated with the largest effects (Orobio de Castro et al., 2002). All of those methodological features of a vignette-based assessment tool might be important moderators affecting the role that hostile attribution of intent plays in social information processing. Considering the limitations of the hostility measures, used in this study, it seems necessary for future studies to increase assessment validity to allow for conclusions to be drown about the relationship between HAB and social perception. This issue could be addressed by including additional assessments of hostility, for example, interview-based diagnostics.

Another limitation refers to the strictly controlled experimental setting, based on the presentation of virtual faces simulating a real social interaction. The use of artificial schematic-looking faces as stimuli neglects the fact that in real-world social situations individuals simultaneously process a variety of social stimuli, including facial expressions, current behavior, body language, social context, and so on. A static and artificially created experimental setting, focusing on only one of all possible characteristics individuals use to interpret another’s intentions, could have decreased the potential effects of hostility of the gaze perception. However, there is some evidence suggesting that the effects, obtained in experimental settings with virtual observer, show an essential overlap with the perception of real persons and, thus, can be generalized to social interaction with real people, confirming the ecological validity of virtual presentation (Harbort, Witthöft, Spiegel, Nick, & Hecht, 2013). It remains a task for future research to examine the link between hostility and self-referred gaze perception in a dynamic real-interaction based framework and verify the transferability from an experimental to a real-life social setting.

The present study examined direct, immediate reactions, using a primary spontaneous response to briefly presented stimuli (300 ms). In reality, secondary judgments take place in the interpersonal context. While interpreting the viewing direction or the emotional facial expression of a person, a secondary review process occurs, which evaluates the primary assessment and strengthens it. This meta-assessment may relate to various factors (e.g., the social context in which the interpersonal interaction takes place) and consider diverse elements of the environment (e.g., distance between the interaction partners) before concluding whether the individual is being looked at or not. These secondary assessments can be understood as controlling or verification processes, since it questions and challenges the accuracy of the spontaneous perception and accordingly confirms or rejects the primary appraisals. Since the consideration of potentially relevant factors warrants a longer time, it could not take place in the present study. It cannot be excluded that additional verification processes, which may also be hostility based, could occur in a real context and influence the processing of social perception.

The study has also a couple of sample-related limitations. The sample was very highly specified and included multiple selection criteria: age (juvenile and adolescent), offense of conviction (violent and sexual crimes only), current incarceration, and current psychotherapeutic treatment. Strictly including only participants who fulfilled all criteria, naturally reduced the sample size, which may have resulted in the overestimation of effect sizes and limited reproducibility of the results. Furthermore, subjects with active schizophrenia, contemporary use of psychotropic medication, mental retardation, and poor language skills, were all excluded due to the intervention concept of the DST. In practice, those characteristics are typical in samples of offenders. Excluding those subjects may therefore have limited the generalizability of the results to other samples of offenders. In addition, we studied male adolescent offenders only, and thus, study findings may not be applicable to nonincarcerated samples and samples with different age groups or gender. In addition, due to the lack of a control group, the possibility that some other third variable (e.g., impulsivity, aggressiveness, psychopathy) may have mediated the relationship between social perception and information-processing bias could not be ruled out entirely. However, we decided to use only a prison sample to ensure similar exposure to potentially hostile situations for all participants. If we had used some nonincarcerated participants, noise could have been introduced into the results because of current environmental differences, and thus, differences between violent and nonviolent groups may have arisen from differences in current environment (institutional–noninstitutional). Although the findings of the study cannot be generalized and are not necessarily applicable to all, they must be appreciated in their specific nature and importance in understanding the unique experiences of the individuals of interest.

However, our data replicated prior findings using a normative (noncriminal) sample (Lobmaier & Perrett, 2011; Lobmaier et al., 2008) and confirmed the evidence for the positive self-referential bias. Future studies should conduct an experiment, including a control group to register possible differences in the social perception and hostility bias between incarcerated and normative participants.

In conclusion, our findings replicate previous evidence, showing a relation between facial expression and the self-directed perception of gaze particularly in the presence of positive facial expressions. Attempts to find a link between hostility-prone attributions of intent and impaired gaze perception, based on preferred perception of angry faces, failed, suggesting that social processing does not take place in a self-directed level. Further investigations of abnormal perception and the role that hostility might play in the processing of gaze cues represent a promising approach to broaden the understanding of juvenile delinquency and youth violence.

Footnotes

Appendix

Vignettes used to assess the hostile attributional bias (HAB):

Acknowledgements

The authors thank the supervisor Mr. Klaus-Peter Dahle for his academic guidance.

Authors’ Note

This article is derived from Zhana Karadenizova’s PhD dissertation, currently being completed at the Institute of Forensic Psychiatry, Charité–Universitätsmedizin Berlin, and the Freie Universität Berlin, Germany, with data collection in 2014 and 2015.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Best

Williams

Coccaro

(2002). Evidence for a dysfunctional prefrontal circuit in patients with an impulsive aggressive disorder. Proceedings of the National Academy of Sciences of the United States of America, 99, 8448-8453. doi:10.1073/pnas.112604099

Burt

Mikolajewski

Larson

(2009). Do aggression and rule-breaking have different interpersonal correlates? A study of antisocial behavior subtypes, negative affect, and hostile perceptions of others. Aggressive Behavior, 35, 453-461. doi:10.1002/ab.20324

Buss

A. H.

Perry

(1992). The Aggression Questionnaire. Journal of Personality and Social Psychology, 63, 452-459. doi:10.1037/0022-3514.63.3.452

Crick

N. R.

Dodge

K. A.

(1994). A review and reformulation of social information-processing mechanisms in children’s social adjustment. Psychological Bulletin, 115, 74-101. doi:10.1037/0033-2909.115.1.74

Dodge

K. A.

(1980). Social cognition and children’s aggressive behavior. Child Development, 51, 162-170. doi:10.2307/1129603

Dodge

K. A.

(2006). Translational science in action: Hostile attributional style and the development of aggressive behavior problems. Development and Psychopathology, 18, 791-814. doi:10.1017/S0954579406060391

Dodge

K. A.

Coie

J. D.

(1987). Social-information-processing factors in reactive and proactive aggression in children’s peer groups. Journal of Personality and Social Psychology, 53, 1146-1158.

Dodge

K. A.

Newman

J. P.

(1981). Biased decision-making processes in aggressive boys. Journal of Abnormal Psychology, 90, 375-379. doi:10.1037/0021-843X.90.4.375

10.

Dodge

K. A.

Price

J. M.

Newman

J. P.

Bachorowski

J. A.

(1990). Hostile attributional biases in severely aggressive adolescents. Journal of Abnormal Psychology, 99, 385-392.

11.

Fairchild

Stobbe

Van Goozen

S. H.

Calder

A. J.

Goodyer

I. M.

(2010). Facial expression recognition, fear conditioning, and startle modulation in female subjects with conduct disorder. Biological Psychiatry, 68, 272-279. doi:10.1016/j.biopsych.2010.02.019

12.

Fairchild

Van Goozen

S. H.

Stollery

Aitken

M. R.

Savage

Moore

S. C.

Goodyer

I. M.

(2009). Decision making and executive function in male adolescents with early-onset or adolescence-onset conduct disorder and control subjects. Biological Psychiatry, 66, 162-168. doi:10.1016/j.biopsych.2009.02.024

13.

Gamer

Hecht

(2007). Are you looking at me? Measuring the cone of gaze. Journal of Experimental Psychology: Human Perception and Performance, 33, 705-715. doi:10.1037/0096-1523.33.3.705

14.

Harbort

Witthöft

Spiegel

Nick

Hecht

(2013). The widening of the gaze cone in patients with social anxiety disorder and its normalization after CBT. Behaviour Research and Therapy, 51, 359-367. doi:10.1016/j.brat.2013.03.009

15.

Herzberg

P. Y.

(2003). Faktorstruktur, Gütekriterien und Konstruktvalidität der deutschen Übersetzung des Aggressionsfragebogens von Buss und Perry [Psychometric evaluation and validity of the German translation of the Aggression Questionnaire by Buss and Perry]. Zeitschrift für Differentielle und Diagnostische Psychologie, 24, 311-323. doi:10.1024/0170-1789.24.4.311

16.

Itier

R. J.

Batty

(2009). Neural bases of eye and gaze processing: The core of social cognition. Neuroscience & Biobehavioral Reviews, 33, 843-863. doi:10.1016/j.neubiorev.2009.02.004

17.

James

Seager

J. A.

(2006). Impulsivity and schemas for a hostile world: Postdictors of violent behaviour. International Journal of Offender Therapy and Comparative Criminology, 50, 47-56. doi:10.1177/0306624X05279804

18.

Knyazev

G. G.

Bocharov

A. V.

Slobodskoj-Plusnin

J. Y.

(2009). Hostility- and gender-related differences in oscillatory responses to emotional facial expressions. Aggressive Behavior, 35, 502-513. doi:10.1002/ab.20318

19.

Kuhl

Kazén

(2009). Persönlichkeits-Stil- und Störungs-Inventar PSSI [Personality Style and Disorder Inventory] (2nd ed.). Göttingen, Germany: Hogrefe.

20.

Lobmaier

J. S.

Perrett

D. I.

(2011). The world smiles at me: Self-referential positivity bias when interpreting direction of attention. Cognition and Emotion, 25, 334-341. doi:10.1080/02699931003794557

21.

Lobmaier

J. S.

Tiddeman

B. P.

Perrett

D. I.

(2008). Emotional expression modulates perceived gaze direction. Emotion, 8, 573-577. doi:10.1037/1528-3542.8.4.573

22.

Marsh

A. A.

Blair

(2008). Deficits in facial affect recognition among antisocial populations: A meta-analysis. Neuroscience & Biobehavioral Reviews, 32, 454-465. doi:10.1016/j.neubiorev.2007.08.003

23.

Nasby

Hayden

De Paulo

B. M.

(1980). Attributional bias among aggressive boys to interpret unambiguous social stimuli as displays of hostility. Journal of Abnormal Psychology, 89, 459-468. doi:10.1037/0021-843X.89.3.459

24.

Orobio

Castro

Veerman

J. W.

Koops

Bosch

J. D.

Monshouwer

H. J.

(2002). Hostile attribution of intent and aggressive behavior: A meta-analysis. Child Development, 73, 916-934. doi:10.1111/1467-8624.00447

25.

Schönenberg

Jusyte

(2014). Investigation of the hostile attribution bias toward ambiguous facial cues in antisocial violent offenders. European Archives of Psychiatry and Clinical Neuroscience, 264, 61-69. doi:10.1007/s00406-013-0440-1

26.

Schönenberg

Louis

Mayer

Jusyte

(2013). Impaired identification of threat-related social information in male delinquents with antisocial personality disorder. Journal of Personality Disorders, 27, 496-505. doi:10.1521/pedi_2013_27_100

27.

Seager

(2005). Violent men: The importance of impulsivity and cognitive schema. Criminal Justice and Behavior, 32, 26-49. doi:10.1177/0093854804270625

28.

Serin

(1991). Psychopathy and violence in criminals. Journal of Interpersonal Violence, 6, 423-431. doi:10.1177/088626091006004002

29.

Shapiro

S. S.

Wilk

M. B.

(1965). An analysis of variance test for normality (complete samples). Biometrica, 52, 591-611.

30.

Slaby

R. G.

Guerra

N. G.

(1988). Cognitive mediators of aggression in adolescent offenders: 1. Assessment. Developmental Psychology, 24, 580-588.

31.

Van Oostrum

Horvath

(1997). The effects of hostile attribution on adolescents’ aggressive responses to social situations. Canadian Journal of School Psychology, 13, 48-59. doi:10.1177/082957359701300105

32.

Vitale

J. E.

Newman

J. P.

Serin

R. C.

Bolt

D. M.

(2005). Hostile attributions in incarcerated adult male offenders: An exploration of diverse pathways. Aggressive Behavior, 31, 99-115. doi:10.1002/ab.20050