There is a growing body of evidence that psychotic symptoms

Transcription

1 ORIGINAL ARTICLE Impact of Adverse Life Events on Individuals With Low and High Schizotypy in a Nonpatient Sample Krisztina Kocsis-Bogár, MA,*Þ Mónika Miklósi, MA,Þþ and Dóra Perczel Forintos, PhD*Þ Abstract: The aims of this study were to gain a better understanding of adverse life events connected with the development of schizotypal personality traits and, also, to examine whether subclinical schizotypy has a relationship with vulnerability to traumatic intrusions and avoidance. In a cross-sectional design, 198 undergraduate students completed the Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE), the Impact of Event Scale (IES), and Paykel s Life Events Scale, together with other relevant scales. The number of adverse life events was significantly related to overall schizotypy measured by O-LIFE scores and positive schizotypy measured by the Unusual Experiences (UnEx) subscale. The subjective severity of life events was significantly related to Cognitive Disorganization (CogDis). Measures of positive schizotypy (UnEx and CogDis) were significantly related to the scores on the IES and on the intrusion and avoidance subscales, too. Adverse life events are associated with schizotypal personality traits, which contribute to a tendency for traumatic intrusions, even in a nonpatient sample. Key Words: Schizotypy, O-LIFE, life events, traumatic intrusions, Impact of Event Scale (J Nerv Ment Dis 2013;201: 208Y215) There is a growing body of evidence that psychotic symptoms form a continuum with certain experiences, beliefs, and appraisals of the healthy personality. According to the dimensional model of schizophrenia (Mason and Claridge, 2006), schizotypy is a fully dimensional trait that varies throughout the healthy population, patients with schizophrenia spectrum disorders, and patients with schizophrenia. This fully dimensional approach has been repeatedly found to be a comprehensive model to explain a range of similarities between patients with schizophrenia and healthy subjects scoring high on schizotypy scales. Schizotypy is also regarded as a multifactorial trait, its main factors being positive schizotypy (Unusual Experiences [UnEx], strange beliefs, and magical thinking), negative schizotypy (anhedonia), and disorganized schizotypy, together with social anxiety and nonconformity (Vollema and van den Bosch, 1995). This multifactoriality is reflected in the construction of the Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE; Mason et al., 1995). Positive and negative schizotypy and Cognitive Disorganization (CogDis) have been found to be continuous variables in a small sample of patients with schizophrenia and matched healthy controls, whereas impulsive nonconformity has not. In the study by Cochrane et al. (2010), the UnEx subscale of the O-LIFE was proved to be the only subscale that mirrors the positive symptoms of patients with schizophrenia on the healthy population, but the same was not true *Department of Clinical Psychology, Semmelweis University, Budapest, Hungary; School of PhD Studies, Semmelweis University, Budapest Hungary; and Heim Pál Children s Hospital, Budapest, Hungary. Send reprint requests to Krisztina Kocsis-Bogár, MA, Department of Clinical Psychology, Semmelweis University, 1083 Budapest, Tómj utca kocsis.bogar.krisztina@gmail.com; klinpszicho@med.semmelweis-univ.hu. Copyright * 2013 by Lippincott Williams & Wilkins ISSN: /13/20103Y0208 DOI: /NMD.0b013e cea for the introvertive anhedonia, CogDis, or impulsive nonconformity subscales in this study (Cochrane et al., 2010). On the one hand, a high level of schizotypy, similar to schizophrenia, is associated with a range of cognitive deficits (poorer cognitive inhibition, as measured by negative priming affect; Beech et al., 1989; Peters et al., 1994; Steel et al., 1996), a reduced distractor cueing effect, and, consequently, a reduced ability to process contextual information (Steel et al., 2002). Schizotypal characteristics have been recognized to be risk factors of developing schizophrenia later on (Miller et al., 2002). In the study of Mason et al. (2004), schizotypal personality traits were the most reliable scale-based predictors of developing a psychotic disorder later on in a sample of young people seeking help at a psychological assessment service. On the other hand, connections with creativity and imagination, in other words, certain protective aspects of schizotypy, especially of positive schizotypy, have also been recognized. Green and Williams (1999) found a significant positive correlation between schizotypy and divergent thinking. In Stoneham and Coughtrey s (2009) study, subjects with high schizotypy were found to use significantly more creative strategies in a group problem-solving situation. There is evidence that a high level of magical thinking together with a low level of CogDis can enhance subjective well-beingvfor these people judge their UnEx to be more pleasant and positive than those with high CogDis. It seems that organized positive schizotypy allows people to construct an adaptive mental framework to interpret their experiences (Schofield and Claridge, 2007). At the same time, individuals with high schizotypy tend to relive their past experiences and prelive their future with a lot more sensual (especially olfactory) details, and this is accompanied by a deeper subjective feeling (Winfield and Kamboj, 2010). This latter finding leads further to the question of reliving traumatic experiences and adverse life events. The leading symptom of posttraumatic stress disorder (PTSD) is the repeated involuntary intrusion of distressing trauma memories. According to the cognitive model of PTSD, patients do not remember or think of what has happened to them; they actually relive it. Sensory features of the intrusive memories are very strong, and time perspective is lost. The patients impression is as if they are experiencing the adverse event right at the moment and the emotions accompanying the event are exactly the same as these were at the original event. (Ehlers et al., 2004) This way, sensory impressions and emotions are dissociated from their source. Two important factors are to be mentioned in maintaining the symptoms. First of all, trauma memories are not elaborated and integrated in their context in time and place. Integration is thought to be impeded partly by the dissociation mentioned above (Ehlers and Clark, 2000). Secondly, avoidance of stimuli connected with the traumatic event and the effort to suppress conscious thoughts of the traumatic event also block the process of integration and maintain or even increase the regularity or the severity of intrusions (Davies and Clark, 1998; Ehlers and Clark, 2000). Traumatic experiences, especially various forms of childhood abuse, have also been widely recognized as a risk factor of developing psychotic symptoms (Bebbington et al., 2004; Hardy et al., 2005; Read, 1997; Spauwen et al., 2006) and a high level of schizotypy (Startup, 1999; Steel et al., 2009). According to the longitudinal study by Spauwen et al. (2006), traumatic life events increase the expression of psychotic symptoms in youngsters both high and low on psychosis The Journal of Nervous and Mental Disease & Volume 201, Number 3, March 2013

2 The Journal of Nervous and Mental Disease & Volume 201, Number 3, March 2013 Schizotypy and Adverse Life Events proneness. Steel et al. (2009) found that individuals who were categorized in the psychosis-prone group on the basis of their relatively high score on the Schizotypal Personality Scale were more than six times more likely to have experienced physical abuse and four times more likely to have been sexually abused than those not in this group. The often cited 1989 study of Romme and Esher shed light on the fact that even the adverse events of later life (e.g., bereavement) increase the likelihood of hallucinationsveven in nonpatientsvand further research (Morrison and Petersen, 2003) gave more support to this statement. Steel et al. (2005) proposed an information-processing account according to which individuals with high schizotypy, similar to patients with schizophrenia, are less able to make use of contextual information surrounding target stimuli (demonstrated by the flanking letter test; Steel et al., 2002) and therefore have an information-processing style in which contextual integration is weaker. This way, adverse life events that may generate traumatic intrusions are more likely to affect individuals with high schizotypy (Steel et al., 2005), which have been a proved risk factor of developing hallucinations, as stated above. At the same time, Bebbington et al. (2004) suggest that people who develop a schizophreniform disorder later in life tend to be survivors of bullying or different forms of abuse more often than healthy controls. Because of their poorer social skills, they might seem more vulnerable, easier to intimidate, and less able to defend themselves. At the same time, because of their social and cognitive deficits, they are less able to seek social support, build further confiding relationships, and mentally organize traumatic material after going through an adverse life event. This way, people with more schizotypal traits are expected to have experienced more adverse life events, which they found more unpleasant and more difficult to cope with than healthy controls. The relationship between childhood trauma and unusual perceptual experiences or hallucinations has been repeatedly interpreted as a consequence of intrusive memories of the stressful events (Hardy et al., 2005; Steel et al., 2009). It has been proposed by several studies that poor source monitoring of individuals who have schizophrenia and individuals with high schizotypy may lead the observers to think certain experiences of schizotypal personalities are schizophrenia symptoms (hallucinations) instead of actual memories of past events (Steel et al., 2009). Relying on the analogue design, it has been argued that high positive schizotypy means vulnerability for traumatic intrusions. Holmes and Steel (2004) found that the UnEx scale of the O-LIFE was the only independent predictor of intrusions of the stressful film. They also showed that subjects scoring high on the UnEx scale reported more personal experiences of traumatic events. It is argued by the authors that high-scoring individuals with positive schizotypy are less able to integrate information within a temporal context (as shown in Steel et al., 2002) and therefore store information in a way that allows it to be triggered involuntarily. Thus, they may become overwhelmed by the content of their consciousness and become highly aroused by trying to find the meaning of their experiences but being unable to connect these to their personal life events. The more traumatic experiences they go through, the more vulnerable they become to intrusions. Steel et al. (2008) replicated these findings in a sample of 45 survivors of road traffic accidents and gained some more empirical support for the connection of positive schizotypy measured by the O-LIFE and posttraumatic symptoms. The results of the study by Marzillier and Steel (2007) in a sample of 50 patients who experienced a traumatic event seeking psychological assistance extend previous knowledge to clinical population. They found individuals with high schizotypy to be especially vulnerable to traumatic intrusions and other posttraumatic stress symptoms, such as hyperviligance and avoidance. In regard to connections of schizotypy and mood disorders, it has been found that depressive and anxious symptoms can be more closely associated with positive than negative schizotypy. Lewandowski et al. (2006) argue that, in addition to positive schizotypy, anxiety and depression involve deregulation of negative affect, whereas negative schizotypy involves a diminution of affective response. Similarly, Marzillier and Steel (2007) found positive schizotypy to be positively connected with low positive-self and positive-others, high negativeself and negative-others beliefs, and therefore higher levels of anxiety and depression in their clinical sample. In summary, it has been proposed that stressful life events increase the likelihood of developing schizophreniform pathology, and there is some evidence that this connection can be found regarding schizotypal personality traits too. At the same time, the cognitive dysfunctions and difficulties of social contacts associated with schizotypal personality make it even harder for these people to work through the effects of adverse life events and make them even more vulnerable for posttraumatic symptoms. Therefore, we intended to test the following hypotheses: 1. Our first hypothesis was that schizotypy, especially positive schizotypy measured by UnEx, is associated with high rates of traumatic intrusions, more so than negative schizotypy measured by introvertive anhedonia. Our intention was to replicate the findings of Holmes and Steel (2004) and Steel et al. (2008), with a more personal material on a large, nonclinical sample. 2. Our second hypothesis was that a larger number of adverse life events that were rated as severe by individuals is connected with schizotypal personality traits, especially positive schizotypy measured by UnEx. 3. On the basis of the findings by Lewandowski et al. (2006) and Marzillier and Steel (2007), we hypothesized that self-reported depressive and anxious thoughts and feelings correlate positively with positive schizotypy but not with negative schizotypy. In the study of posttraumatic intrusions in nonclinical samples, we had three choices, all of which have been used in this field: a) the analogue design (e.g., Holmes and Steel, 2004), b) selecting survivors of a traumatic event concerning a large population (e.g., earthquake, flood; e.g., Steel et al., 2008), or c) asking people to think of a certain traumatic event of their lives that they selected themselves (e.g. Merkelbach and Giesbrecht, 2006; Startup, 1999) or that was determined specifically by the researchers. Following one of the first two would have meant more conceptual clarity of trauma and traumatic intrusion as described in the ICD-10 (World Health Organization, 1992) and the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) (American Psychiatric Association, 2000), whereas adopting design number 3 meant the advantage of reaching a larger sample and working with more personally significant material. Studies relying on the same design refer to a wide range of adverse life events as traumatic, even though these do not necessarily fulfill the exact ICD-10 and DSM-IV criteria of a trauma (e.g.,bereavement or emotional abuse [Morrison and Petersen, 2003]), bullying [Campbell and Morrison, 2007], something terrible happening to others [Spauwen et al., 2006]), and find that these actually function in people s lives in the same way as the traumatic events that fulfill the taxonomic criteria. METHODS Ethics Ethical approval for this study was obtained from the Regional and Institutional Scientific and Research Ethics Committee (Regionális * 2013 Lippincott Williams & Wilkins 209

3 Kocsis-Bogár et al. The Journal of Nervous and Mental Disease & Volume 201, Number 3, March 2013 és Intézményi Tudományos és Kutatásetikai Bizottság) of Semmelweis University, Budapest. The subjects were briefly informed both verbally and in writing on a short information sheet about the purpose of this study and were ensured that their participation was fully voluntary and could be stopped any time. A contact person was named in case psychological and/or medical support was needed. Participants Four hundred booklets of questionnaires were distributed among first and second year undergraduate students of Semmelweis University, University of Szeged, and Budapest Technical University. The students were asked to stay behind after lectures to fill in the questionnaires or were reached at their home and asked to return the completed questionnaires into a closed box at an easily accessible place. The students received no university credit or financial reward for their participation. Finally, 222 booklets were found to be right for statistical analysis. The subjects were asked on an information sheet about their age and sex; habits of consuming alcohol and drugs; and previous psychological and/or psychiatric treatment, including medication. All together, 24 subjects reporting previous psychiatric diagnosis, medication, and/or extensive use of alcohol and/or drugs were excluded from further analysis, so 198 subjects remained. Questionnaires Paykel s Life Events Scale (Short Version) Paykel s Life Events Scale (short version; Paykel, 1991; Hungarian translation, Tringer and Veér, 1977) is a 28-item scale that lists different adverse life events, for example, death of a family member, death of a close friend, suicide of somebody close, or being a survivor of violence. Subjects are asked to mark whether they have experienced any of these and then to rate the subjective severity of these events, ranging from 0 to 5, with 0 indicating not upsetting at all and 5 indicating very upsetting (Paykel, 1983). At the end of the list, subjects can report any further events missing from the list that they experienced and found stressful. (Perczel Forintos et al., 2005) In the study of a healthy sample and different patient populations (Nagy, 2004), the mean number of life events for people with schizophrenia was 5.33, which was very similar to the mean of people with depression (5.36), depression and suicide attempt (5.24), and alcohol dependence (5.10). The mean number of adverse life events in the healthy sample was Impact of Event Scale The 15-item Impact of Event Scale (IES; Horowitz et al., 1979; Hungarian translation, Annus et al., 2005), which measures subjective stress caused by adverse life events, contains two factors: intrusions (seven items) and avoidance (eight items). The subjects were asked to think about a severe traumatic life event and rate on a 4-point frequency scale how often in the last week they had experiences described by the items (0, not at all; 1, rarely; 3, sometimes; 5, often; Perczel Forintos et al., 2005; Sundin and Horowitz, 2002). In a meta-analysis of 18 different studies, internal consistencies were found to be high for both subscales, with a mean alpha of 0.86 (range, 0.72Y0.92) for Intrusions and a mean alpha of 0.82 (range, 0.65Y0.90) for avoidance (Sundin and Horowitz, 2003). Scores on the IES have a possible range of 0 to 75, with higher scores indicating more intrusions and avoidance in relation to a past adverse life event. In our sample, very good internal consistencies were found for IES total (Cronbach s > = 0.88) and for the intrusions subscale (Cronbach s > = 0.89); and good internal consistencies, for avoidance (Cronbach s > = 0.756; Table 1). The O-LIFE The 104-item O-LIFE (Mason et al., 1995) has four subscales: UnEx (30 items), measuring positive schizotypy; CogDis, measuring attention and concentration deficits, poor decision making, and social anxiety (24 items); introvertive anhedonia, measuring negative schizotypy (27 items); and impulsive nonconformity, measuring impulsive, antisocial, and eccentric forms of behavior (23 items; Mason and Claridge, 2006). Subjects are asked to decide whether they agree with each statement of the questionnaire, and yes answers on positive items and no answers on negative items equally score 1. No answers on positive items and yes answers on negative items score 0. Scores on the O-LIFE have a possible range of 0 to 104, with higher scores indicating more schizotypal traits. At the examination of psychometric properties of the O-LIFE, Mason et al. (1995) found good to very good internal consistencies for all four subscales, Cronbach s alphas ranging from 0.77 to The basis of the Hungarian translation was the appendix of the article by Mason and Claridge (2006), where all 104 items were listed. It was translated into Hungarian by the first and the second author and then translated back into English by a Hungarian native speaker living in an English-speaking environment. TABLE 1. Descriptive Statistics and Internal Consistencies of the Main Variables for the Whole Sample Minimum Values Maximum Values Mean (SD) Cronbach s > Age (N = 198), yrs (1.95) No. stressful life events (n = 197) (2.24) (median, 3; interquartile range, 2Y5) Severity of stressful life events (n = 195) (1.11) STAI-T (n = 190) (9.59) BDI (n = 190) (5.98) IES intrusions (N = 198) (5.84) IES avoidance (N = 198) (6.16) IES total (N = 198) (10.99) O-LIFE UnEx (N = 198) (5.15) O-LIFE introvertive anhedonia (N = 198) (3.68) O-LIFE CogDis (N = 198) (4.92) O-LIFE impulsive nonconformity (N = 198) (3.40) O-LIFE total (N = 198) (12.30) BDI indicates Beck Depression InventoryYshortened version * 2013 Lippincott Williams & Wilkins

4 The Journal of Nervous and Mental Disease & Volume 201, Number 3, March 2013 Schizotypy and Adverse Life Events In our sample, the internal consistency of the O-LIFE was very good (Cronbach s > = 0.829), and for the subscales, it was acceptable to good (Cronbach s alphas ranging from to 0.777). The exact results are reported in Table 1. Spielberger Trait-State Anxiety Inventory, Trait Subscale The Spielberger Trait-State Anxiety Inventory, trait subscale (STAI-T; Spielberger, 1970; Hungarian translation, Sipos and Sipos, 1978), is a widely used 20-item scale for measuring personal tendency for anxiety. Every item is rated on a 4-point Likert scale, with 1 indicating almost never; 2,somewhat;3,often; and 4, almost always. Eight of the 20 items are reversed, and, in these cases, scores are also reversed, with 1 indicating almost always; 2,often; 3,somewhat; and 4, almost never (Kopp and Fórizs, 1993). Total scores on the STAI-T have a range of 20 to 80, with higher scores indicating more trait anxiety (Perczel Forintos et al., 2005). In a few recent studies in which the STAI-T was applied to nonpsychiatric samples, the internal consistency was found to be very good (Cronbach s > = 0.86, Dunn et al., 2012, and Cronbach s > = 0.89, Lander et al., 2012). In our sample, the internal consistency of this scale was very good (Cronbach s > = 0.873; Table 1). Beck Depression InventoryYShortened Version The nine-item Beck Depression Inventory (BDI)Yshortened version (Beck and Beck, 1972; Hungarian shortened version, Kopp and Fórizs, 1993) has been designed to measure cognitive, emotional vegetative, and behavioral signs of depression (Kopp and Fórizs, 1993). It is not the exact translation of the shortened version of the BDI by Beck and Beck (1972), but it is very similar to that. Each item is rated from 1 to 4, with 1 indicating that the symptom is not experienced and 4 indicating that the symptom is permanently experienced. Total scores on the BDIYshortened version have a possible range of 9 to 54, with higher frequencies indicating more severe depressive symptoms. Scores of the BDIYshortened version can be converted to the original BDI (Beck et al., 1961) scores using the following template: ð BDIjshortened version scoresj9þ 2 Rózsa et al. (2001), in their study of the psychometric properties of the BDIYshortened version, reported that it had a strong correlation (r = 0.92, n = 101, p G ) with the original BDI (Beck et al., 1961). In their sample, the BDIYshortened version had very good internal consistency (Cronbach s > = 0.83). In our sample, the internal consistency of the BDIYshortened version was proved to be a little lower but still good (Cronbach s > = 0.715; Table 1). Statistical Analysis The data were processed by SPSS 15.0 for Windows (SPSS Inc, 2006). Cronbach s alphas were used to describe the internal consistencies of the applied scales. Descriptive statistics were reported. Sex differences were tested using independent-samples t-tests and Mann-Whitney s U-test. Bivariate relationships of the study variables were tested using Pearson s correlation coefficients. Multivariate relationships between schizotypy, adverse life events, and posttraumatic symptoms were assessed using multiple regression analyses. When the assumption of normality was violated, Box-Cox transformation was used. Levels of significance were set at > = RESULTS Statistical analysis was performed on the basis of the data of the 198 subjects, 114 (57.6%) of which were women. The mean (SD) age was (1.96; range, 18Y31) years. Descriptive statistics of the main variables can be seen in Table 1. The men and the women did not differ significantly with regard to age (t[196] = 0.97, p = 0.336, Cohen s d = 0.139), self-reported depressive symptoms (t[188] = j0.117, p = 0.907, Cohen s d = j0.017), STAI-T score (t[188] = j1.913, p =0.057,Cohen sd = j0.279), IES total score (t[196] = j0.967, p = 0.335, Cohen s d = j0.138), traumarelated intrusions (t[196] = j1.650, p = 0.101, Cohen s d = j0.236), avoidance (t[196] = j0.167, p = 0.868, Cohen s d = j0.024), O-LIFE total score (t[196] = j1.138, p = 0.256, Cohen s d = j0.125), and introvertive anhedonia (t[196] = j0.146, p = 0.884, Cohen s d = j0.02), but the women reported significantly more stressful life events (Mann-Whitney s U-test = ; z = j2.184; p = 0.029; mean rank difference, 17.71) and rated these to be more distressing (t[196] = j2.376, p =0.018,Cohen sd = j0.339) than did the men and scored higher on UnEx (t[196] = j2.059, p = 0.041, Cohen s d = j0.294) and CogDis (t[196] = j2.042, p =0.043,Cohen sd = j0.291). On the other hand, the men scored significantly higher on impulsive nonconformity (t[196] = 2.243, p = 0.026, Cohen s d =0.32). Correlations of the main variables can be seen in Table 2. The O-LIFE CogDis subscale had strong correlations with the STAI-T (0.751, p G 0.001) and the BDI (0.601, p G 0.001). O-LIFE total was also strongly correlated with anxiety (0.568, p G 0.001) and the BDI (0.520, p G 0.001). The STAI-T had a strong relationship with the BDI (0.654, p G 0.001). The STAI-T and the BDI had significant correlations of medium effect size with O-LIFE introvertive anhedonia (0.416, p G 0.001, and 0.425, p G 0.001, respectively), and the STAI-T had a medium effect size correlation with the whole IES scale (0.378, p G 0.001) and both subscales: IES intrusions (0.349, p G 0.001) and IES avoidance (0.343, p G 0.001). Medium effect size correlations were found between O-LIFE total and IES total (r =0.401, p G 0.001). Correlations of O-LIFE CogDis with the IES (r = 0.359, TABLE 2. Correlations of the Main Variables Age IES Intrusions IES Avoidance IES Total STAI-T (n = 190) BDI (n = 190) Age 1 j0.147* j0.066 j0.115 j0.06 j0.05 O-LIFE UnEx j ** 0.282** 0.308** 0.189** 0.230** O-LIFE IntrAnhed * 0.273** 0.229** 0.416** 0.425** O-LIFE CogDis j ** 0.323** 0.359** 0.751** 0.601** O-LIFE ImpNonc j * 0.145* 0.164* 0.162* O-LIFE total j ** 0.386** 0.401** 0.568** 0.520** STAIjT j ** 0.343** 0.378** V 0.654** BDI j ** 0.287** 0.289** 0.654** V N = 198. ImpNonc indicates impulsive nonconformity; IntrAnhed, introvertive anhedonia. *p G **p G * 2013 Lippincott Williams & Wilkins 211

5 Kocsis-Bogár et al. The Journal of Nervous and Mental Disease & Volume 201, Number 3, March 2013 TABLE 3. Results of Linear Regression Analyses Dependent variable: STAI-T (n = 189) Independent Variable A t p Constant G0.001 UnEx j0.097 j CogDis G0.001 Introvertive anhedonia Impulsive nonconformity j0.042 j R 2 = 0.562, F = , df = 188, p G Dependent variable: BDI (n = 190) Constant UnEx CogDis G0.001 Introvertive anhedonia Impulsive nonconformity j0.070 j R 2 = 0.331, F = , df = 189, p G Dependent variables: the STAI-T and the BDIYshortened version. p G 0.001), IES Intrusions (r = 0.334, p G 0.001), and IES Avoidance (r = 0.323, p G 0.001) were also in the medium range. There was a medium effect size correlation between UnEx and IES total (r = 0.308, p G 0.001). Small effect size correlations were found between UnEx and IES Intrusions (r = 0.282, p G 0.001) and IES Avoidance (r = 0.282, p G 0.001) and between introvertive anhedonia and IES Avoidance (r = 0.273, p G 0.001) and IES total (r = 0.229, p G 0.001). As was expected, impulsive nonconformity had no significant correlation with IES total or either of the IES subscales. The BDI scores were significantly related to IES total (r = 0.298, p G 0.001) and both subscales (intrusions: r = 0.258, p G 0.001; avoidance: r = 0.287, p G 0.001); however, the effect sizes were small. The correlation of small effect size was the lowest, in the case of UnEx and the STAI-T (0.189, p G 0.001) and the BDI (0.230, p G 0.001). Age had no significant relationship with any of the variables. Regression analysis of the connection of anxiety and depression with the O-LIFE and the subscales revealed that the STAI-T (R 2 = 0.562, F = , df = 188, p G 0.001) and BDI (R 2 = 0.331, F = , df = 189, p G 0.001) scores both had a significant relationship with CogDis (for STAI-T: A = 0.742, t = , p G 0.001, and for BDI: A = 0.498, t = 6.927, p G 0.001). There was also a significant connection between the BDI and introvertive anhedonia (A = 0.139, t = 2.091, p = 0.03; Table 3). Regression analyses of the O-LIFE and the subscales relating to the IES and the subscales showed that UnExf (A = 0.196, t = 2.688, p = 0.008) and CogDis (A = 0.239, t = 3.039, p = 0.003) were significantly related to IES total (R 2 = 0.17, F = 9.943, df = 197, p G 0.001). The same result emerged with IES Intrusions as dependent (R 2 = 0.14, F = 7.891, df = 197, p G 0.001; UnEx: A = 0.177, t = 2.377, p = 0.018; CogDis: A = 0.264, t = 3.296, p G 0.001). These two schizotypy (UnEx: A = 0.183, t = 2.480, p = 0.014; CogDis: A = 0.176, t = 2.222, p = 0.027) scales together with introvertive anhedonia (A = 0.163, t = 2.218, p = 0.028) had a significant relationship, with IES Avoidance as dependent (R 2 = 0.16, F = 8.981, df = 197, p G 0.001). There was no significant change in these results even after corrections for sex and age (Table 4). Regression analyses of number and severity of adverse life events relating to the O-LIFE and its subscales showed that, in the case of O-LIFE total as dependent (R 2 = 0.06, F = 6.457, df = 194, p = 0.002), subjective severity of life events was a significant predictor (A = 0.205, t = 2.862, p = 0.005), and for CogDis as dependent (R 2 = 0.06, F = 6.181, df = 194, p G 0.001), subjective severity was also significant (A = 0.201, t = 2.807, p = 0.006). For UnEx as dependent (R 2 = 0.08, F = 7.908, df = 194, p G 0.001), the number of adverse life events (A = 0.251, t = 3.524, p G 0.001) was significant. According to our results, the number and the severity of life events did not seem to have a significant relationship with introvertive anhedonia (R 2 = 0.002, F = 0.163, df = 194, p = 0.850) and impulsive nonconformity (R 2 = 0.019, F = 1.895, df = 194, p = 0.153; Table 5). DISCUSSION Our intention was to gain a better understanding of the interconnections of stressful life events, schizotypal personality traits, and posttraumatic symptoms such as intrusion and avoidance. In our bivariate analyses, we mainly got the results expected based on the literature in this field. We found that O-LIFE CogDis had a strong connection with trait anxiety and depression scores, and this was supported by our regression analyses as well, which is not a surprise because it is meant to be a somewhat related phenomenon to trait anxiety and depression. Total schizotypy scores were also highly correlated with trait anxiety and depression. A similarly strong connection appeared between trait anxiety and depression scores. The STAI-T and the BDI were both moderately correlated with introvertive anhedonia, which is contrary to our expectations, on the basis of Lewandowski et al. (2006). Still, it might be understandable because introvertive anhedonia, together with emotional numbing, contains items measuring lack of joy for life. Trait anxiety was moderately correlated with the IES total and subscale; this result can be understood on the basis of the fact that posttraumatic symptoms are considered to be symptoms of an anxiety disorder. As expected, on the basis of previous results (e.g., Marzillier and Steel, 2007), total schizotypy scores were moderately correlated with posttraumatic symptoms, but of the four subscales, somewhat surprisingly, CogDis had the strongest positive connection to the IES total and subscales. UnEx had a moderate positive correlation with IES TABLE 4. Results of Linear Regression Analyses Dependent variable: IES total Independent Variable A t p Constant UnEx CogDis Introvertive anhedonia Impulsive nonconformity R 2 = 0.17, F = 9.943, df = 197, p G Dependent variable: IES intrusions Constant UnEx CogDis G0.001 Introvertive anhedonia j0.004 j Impulsive nonconformity R 2 = 0.14, F = 7.891, df = 197, p G Dependent variable: IES avoidance Constant UnEx CogDis Introvertive anhedonia Impulsive nonconformity R 2 = 0.16, F = 8.981, df = 197, p G Dependent variables: the IES and the subscales * 2013 Lippincott Williams & Wilkins

6 The Journal of Nervous and Mental Disease & Volume 201, Number 3, March 2013 Schizotypy and Adverse Life Events TABLE 5. Results of Regression Analyses Dependent variable: O-LIFE total Independent Variable A t p Constant p G No. life events p = Severity of life events p = R 2 = 0.06, F = 6.457, df = 194, p = Dependent variable: O-LIFE UnEx Constant G0.001 No. life events G0.001 Severity of life events R 2 = 0.08, F = 7.908, df = 194, p G Dependent variable: O-LIFE CogDis Constant G0.001 No. life events Severity of life events R 2 = 0.06, F = 6.181, df = 194, p = Dependent variable: O-LIFE introvertive anhedonia R 2 = 0.002, F = 0.163, df = 194, p = Dependent variable: O-LIFE impulsive nonconformity R 2 = 0.019, F = , df = 194, p = Dependent variables: the O-LIFE and the subscales. total only; correlation with the two subscales remained in the small range. Introvertive anhedonia had a slight positive correlation with IES total and IES Avoidance, which is not surprising, considering emotional numbing being a common factor of both. Contrary to our expectations, the positive schizotypy scores represented by UnEx were only slightly correlated with depression and anxiety scores. The sex differences in our sample also reflected previous results. Similar to the findings in Mason et al. (1995), it was found that the women scored significantly higher on the scales resembling positive symptoms, UnEx and CogDis, whereas the men scored higher on impulsive nonconformity. 1. In our first hypothesis, we expected high positive schizotypy to be strongly associated with posttraumatic symptoms measured by the IES, especially intrusions. Linear regression analysis supported this hypothesis; UnEx scores had a significant positive connection with IES total and Intrusions as well. At the same time, CogDis was another significant factor in both cases. In the case of avoidance, there were three significant factors: UnEx, CogDis, and introvertive anhedonia. The role of this latter subscale is very much in line with the suggestion of Lewandowski et al. (2006) regarding the decrease of emotional response in negative schizotypy measured by introvertive anhedonia. This way, emotional numbing in posttraumatic stress and the diminution of emotional reactions seem to be quite similar phenomena. On the basis of our data, we must attribute more importance to CogDis in posttraumatic symptoms than do most studies in the field. The results of the bivariate analyses, as well as the linear regression analyses, show CogDis to have almost as strong a connection with posttraumatic symptoms as UnEx. These findings are very much in line with the conclusions of Schofield and Claridge (2007), who found that higher levels of cognitive organization predicted a more positive attitude toward one s odd or paranormal experiences and therefore less distress because of them. The authors argue that high magical thinking together with high organization can be adaptive because it can provide a better imaginative framework to place odd experiences. On the basis of our results, we can suppose the same for coping with the impact of adverse events, especially considering the fact that we did not find a significant correlation between UnEx and trait anxiety, but there was a strong significant connection between trait anxiety and CogDis. 2. Our second hypothesis was that the number and perceived severity of stressful life events are significantly associated with the level of schizotypy, especially positive schizotypy. Our hypothesis was supported by the fact that the subjective severity of life events was significantly connected to CogDis and the number of stressful life events was positively connected to O-LIFE UnEx. The former can also be explained by the reasoning that high mental organization means a higher level of psychological functioning and a better cognitive framework to interpret the stressful experiences. The lack of this must make processing of one s adverse life events especially painful, so that those events seem even more severe to these individuals than the ones with a higher level of cognitive organization. In regard to the latter result, similar to several studies (e.g., Bebbington et al., 2004, Campbell and Morrison, 2007), one interpretation can be that the higher number of adverse life events (e.g., bullying) increases the likelihood of schizotypal characteristics, especially unusual and psychoticlike experiences, but a possible explanation might also be that people with schizotypal characteristics attract unpleasant life events. This question can be further investigated in longitudinal studies. The finding that aspects of stressful life events were found to be associated with different types of schizotypy is in accordance with the results of several studies in the field (Bebbington et al., 2004; Startup, 1999; Steel et al., 2009) that adverse life events tend to be linked to schizotypal characteristics. 3. In regard to our third hypothesis, in line with the previous findings of Lewandowski et al. (2006) and Marzillier and Steel (2007), higher schizotypy scores (O-LIFE total) were correlated with higher levels of depression and anxiety, and this relationship was especially strong in the case of CogDis. However, the correlations between the UnEx subscale and the STAI-T and the BDI were relatively low. Contrary to our expectations, depressive and anxious symptoms were positively correlated with introvertive anhedonia. One explanation for this can be that, unlike Marzillier and Steel (2007), we used a healthy sample in which schizotypal traits and depressive and anxious symptoms were all subclinical. CONCLUSIONS It can be concluded that in thinking about schizotypy, we must consider its multifactoriality, which is expressed in the different factors of the different schizotypy questionnaires, for example, in the O-LIFE. Apart from the positive-negative dichotomy, it seems to be important to differentiate between organized and disorganized (positive) schizotypy, which can determine the level of psychological functioning, especially cognitive processing and, as a consequence of this, the subjective evaluation of life events for the individuals. This way, the level of organization or disorganization can determine the person s system of beliefs about the self and the world, which is closely connected with his/her mental health and well-being. Further investigations can be conducted to gain more understanding of the role CogDis plays in the expression and functioning of schizotypal personality traits on the subclinical and the clinical level. At the same time, it is important to conclude that the larger number of traumatic life events (even in the everyday sense of traumatic ) is a factor that, apart from other important determinants such as genetics and biological abnormalities, is associated with schizotypal personality traits, which, on the long-term, can make people more vulnerable to triggers of schizophreniform symptoms. * 2013 Lippincott Williams & Wilkins 213

7 Kocsis-Bogár et al. The Journal of Nervous and Mental Disease & Volume 201, Number 3, March 2013 On the whole, even in a nonpatient sample, we can suppose a tendency for a vicious circle. It is possible that the more adverse life events people experience, the more they tend to develop certain schizotypal characteristics. The more distressing they find these experiences, the more they tend to become disorganized. At the same time, our results lead us to the conclusion that the more disorganized people are, the more difficult they find it to cope with distressing experiences. LIMITATIONS As a limitation of our study, it must be mentioned that the IES was used in a nonclinical sample in a retrospective way and that the types of traumatic events recalled differed a lot. There was also a remarkable difference in the length of time elapsed since the traumatic event. It must be mentioned as a limitation that our cross-sectional design did not allow establishing causal relationships. Further research is needed to study the relationships between schizotypy, adverse life events, and posttraumatic symptoms in a longitudinal design. Stressful life events were assessed by Paykel s Life Events Scale, and we gained no information about the exact time, length, and circumstances of the adverse life events marked by the individuals. Despite the limitations, our results highlight the connections between schizotypal traits, adverse life events, and posttraumatic symptoms. ACKNOWLEDGMENTS The authors thank Professor Erzsébet Nagy, Professor György M. Nagy, Dóra Károly, and their colleagues for the collection of data and Anna Újfaludi for her help with translation. DISCLOSURES The authors declare no conflict of interest. REFERENCES American Psychiatric Association (2000) Diagnostic and statistical manual of mental disorders (4th edyrev). Washington, DC: American Psychiatric Association. Annus R, Perczel Forintos D, Kiss Zs (2005) IES. In Perczel Forintos D, Kiss ZS, Ajtay GY (eds) Kérdjjívek, becsljskálák a klinikai pszicholgiában. Budapest, Hungary: OPNI. Bebbington PE, Bhugra D, Brugha T, Singleton N, Farrell M, Jenkins R, Lewis G, Meltzer H (2004) Psychosis, victimisation and childhood disadvantage: Evidence from second British National Survey of Psychiatric Morbidity. Br J Psychiatry. 185:220Y226. Beck AT, Beck RW (1972) Shortened version of BDI. Postgrad Med. 52:81Y85. Beck AT, Ward CH, MendelsonM, MockJ, Erbaugh EJ (1961) An inventory for measuring depression. Arch Gen Psychiatry. 4:561Y571. Beech A, Powell T, McWilliam J, Claridge G (1989) Evidence of reduced cognitive inhibition in schizophrenia. Br J Clin Psychol. 28:109Y116. Campbell M, Morrison AP (2007) The relationship between bullying, psychoticlike experiences and appraisal in year olds. Behav Res Ther. 45:1579Y1591. Cochrane M, Petch I, Pickering AD (2010) Do measures of schizotypal personality provide non-clinical analogues of schizophrenic symptomatology? Psychiatry Res. 176:150Y154. Davies MI, Clark DM (1998) Thought suppression produces a rebound effect with analogue post-traumatic intrusions. Behav Res Ther. 36:571Y582. Dunn LB, Bradley EA, Cooper AB, Dodd M, Lee K, West C, Paul SM, Wara W, Swift P, Miaskowski C (2012) Trajectories of anxiety in oncology patients and family caregivers during and after radiation therapy. Eur J Oncol Nurs. 16:1Y9. Ehlers A, Clark D (2000) A cognitive model of posttraumatic stress disorder. Behav Res Ther. 38:319Y345. Ehlers A, Hackmann A, Michael T (2004) Intrusive re-experiencing in post-traumatic stress disorder: Phenomenology, theory, and therapy. Memory. 12:403Y415. Green MJ, Williams LM (1999) Schizotypy and creativity as effects of reduced cognitive inhibition. Pers Individ Dif. 27:263Y276. Hardy A, Fowler D, Freeman D, Smith D, Steel C, Evans J, Garety P, Kuipers E, Bebbington P, Dunn G (2005) Trauma and hallucinatory experiences in psychosis. J Nerv Ment Dis. 193:501Y507. Holmes EA, Steel C (2004) Schizotypy: A vulnerability factor for traumatic intrusions. J Nerv Ment Dis. 192:28Y34. Horowitz M, Wilner N, Alvarez, W (1979) Impact of Event Scale: A measure of subjective stress. Psychosom Med. 41:209Y218. Kopp M, Fórizs N (1993) A szorongás kognitív viselkedésterápiája. Budapest, Hungary: Végeken Kiadó. Lander GC, Lutz-Zois CJ, Rye MS, Goodnight JA (2012) The differential association between alexithymia and primary versus secondary psychopathy. Pers Individ Dif. 52:45Y50. Lewandowski KE, Barrantes-Vidal N, Nelson-Gray RO, Clancy C, Kepley HO, Kwapil TR (2006) Anxiety and depression symptoms in psychometrically identified schizotypy. Schizophr Res. 83:225Y235. Marzillier S, Steel C (2007) Positive schizotypy and trauma-related intrusions. J Nerv Ment Dis. 195:60Y64. Mason O, Claridge G (2006) The Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE): Further description and extended norms. Schizophr Res. 82:203Y211. Mason O, Claridge G, Jackson M (1995) New scales for the assessment of schizotypy. Pers Individ Dif. 18:7Y13. Mason O, Startup M, Halpin S, Schall U, Conrad A, Carr V (2004) Risk factors for transition to first episode psychosis among individuals with at risk mental states. Schizophr Res. 71:137Y227. Merkelbach H, Giesbrecht T (2006) Subclinical dissociation, schizotypy and traumatic stress. Pers Individ Dif. 40:365Y374. Miller P, Byrne M, Hodges A, Lawrie S, Owens DGC, Johnstone EC (2002) Schizotypal components at people in high risk of developing schizophrenia: Early findings from the Edinburgh High-Risk Study. Br J Psychiaty. 180:179Y184. Morrison AP, Petersen T (2003) Trauma, metacognition and predisposition to hallucinations in non-patients. Behav Cogn Psychother. 31:235Y246. Nagy J (2004) The analysis of life events in Hungary. Lege Artis Medicinae. 14:690Y692. Paykel SE (1983) Methodological aspects of measuring life events. J Psychosom Res. 27:341Y352. Paykel ES (1991) Stress and life events. In Davidson, L, Linolla, M (eds), Risk factors for youth suicide. New York, NY: Hemisphere. Perczel Forintos D, Kiss ZS, Ajtay GY (2005) Kérdjívek, becsljskálák a klinikai pszichológiában. Budapest, Hungary: OPNI. Peters ER, Pickering AD, Hemsley DR (1994) Cognitive inhibition and positive symptomatology in schizotypy. Br J Psychol. 33:33Y48. Read J (1997) Child abuse and psychosis: A literature review and implications for professional practice. Prof Psychol Res Pract. 28:448Y456. Romme M, Escher AD (1989) Hearing voices. Schizophr Bull. 15:209Y216. Rózsa S, Szádóczky E, Füredi J (2001) A Beck Depresszió Kérdjív rjvidített változatának jellemzji a hazai mintán. Psychiatr Hung. 16:384Y402. Schofield K, Claridge G (2007) Paranormal experiences and mental health: Schizotypy as an underlying factor, Pers Individ Dif. 43:1908Y1916. Sipos K, Sipos M (1978) The developement and validation of the hungarian form of the STAI. Cross-Cultural Anxiety. Washington-London: Hemisphere Publishing Corporation. Spauwen J, Krabbendam L, Lieb R, Wittchen H, van Os J (2006) Impact of psychological trauma on the trajectory of psychosis proneness Br J Psychiatry. 188:527Y533. Spielberger CD, Gorsuch RL, Lushene RE (1970) Manual for the State-Trait Anxiety Inventory. Palo Alto, California: Consulting Psychologist Press. SPSS Inc (2006) SPSS [computer program]: Version 15.0 for Windows. Chicago: SPSS, Inc. Startup M (1999) Schizotypy, dissociative experiences and childhood abuse: Relationships among self-report measures. Br J Clin Psychol. 38:333Y344. Steel C, Fowler D, Holmes EA (2005) Trauma-related intrusions and psychosis: An information processing account. Behav Cogn Psychother. 33:139Y152. Steel C, Hemsley DR, Jones S (1996) Cognitive inhibition and schizotypy as measured by the Oxford-Liverpool Inventory of Feelings and Experiences. Pers Individ Dif. 20:769Y773. Steel C, Hemsley DR, Pickering AD (2002) Distractor cueing effects on choice reaction time and their relationship with schizotypal personality. Br J Clin Psychol. 41:143Y * 2013 Lippincott Williams & Wilkins

8 The Journal of Nervous and Mental Disease & Volume 201, Number 3, March 2013 Schizotypy and Adverse Life Events Steel C, Mahmood M, Holmes EA (2008) Positive schizotypy and trait dissociation a vulnerability factors for post-traumatic stress. Br J Clin Psychol. 47:245Y249. Steel C, Marzilier S, Fearon P, Ruddle A (2009) Childhood abuse and schizotypal personality. Soc Psychiatry Psychiatr Epidemiol. 44:917Y923. Stoneham ACS, Coughtrey AE (2009) The role of schizotypy and creativity in a group problem-solving task. Pers Individ Dif. 46:827Y831. Sundin E, Horowitz MJ (2002) Impact of Life Event Scale: Psychometric properties. Br J Psychiatry. 180:205Y209. Sundin E, Horowitz MJ (2003) Horowitz s Impact of Event Scale Evaluation of 20 years of use. Psychosom Med. 65:870Y876. Tringer, L, Veér, A (1977) Egyes élethelyzetek stressz-hatásainak elemzése. Ideggyogy Szemle. 30:37Y32. Vollema MG, van den Bosch RJ (1995) The multidimensionality of schizotypy. Schizophr Bull. 21:19Y31. Winfield H, Kamboj SK (2010) Schizotypy and mental time travel. Conscious Cogn. 19:321Y327. WorldHealthOrganization(1992)The ICD-10. Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. Geneva, Switzerland: World Health Organization. Hungarian translation: BNO-10 Zsebkönyv (2004) Budapest, Animula Kiadó. * 2013 Lippincott Williams & Wilkins 215