* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Three Dimensions of Oppositionality in Autism
Mental disorder wikipedia , lookup
Residential treatment center wikipedia , lookup
Symptoms of victimization wikipedia , lookup
Dyadic developmental psychotherapy wikipedia , lookup
Causes of mental disorders wikipedia , lookup
Geraldine Dawson wikipedia , lookup
Abnormal psychology wikipedia , lookup
Psychopathy Checklist wikipedia , lookup
Antisocial personality disorder wikipedia , lookup
Discrete trial training wikipedia , lookup
Schizoaffective disorder wikipedia , lookup
Narcissistic personality disorder wikipedia , lookup
Empathizing–systemizing theory wikipedia , lookup
Sluggish cognitive tempo wikipedia , lookup
Dissociative identity disorder wikipedia , lookup
Factitious disorder imposed on another wikipedia , lookup
Diagnostic and Statistical Manual of Mental Disorders wikipedia , lookup
Conduct disorder wikipedia , lookup
Autism and working memory wikipedia , lookup
Depression in childhood and adolescence wikipedia , lookup
Spectrum disorder wikipedia , lookup
Epidemiology of autism wikipedia , lookup
J Abnorm Child Psychol DOI 10.1007/s10802-013-9778-0 Three Dimensions of Oppositionality in Autism Spectrum Disorder William Mandy & Laura Roughan & David Skuse # Springer Science+Business Media New York 2013 Abstract In autism spectrum disorder (ASD), symptoms of oppositional defiant disorder (ODD) are common but poorly understood. DSM-5 has adopted a tripartite model of ODD, parsing its features into ‘angry and irritable symptoms’ (AIS), ‘argumentative and defiant behavior’ (ADB) and ‘vindictiveness’. This was based on findings in non-autistic populations that each of these dimensions of oppositionality has a distinct constellation of associations with internalising and externalising psychopathology. We applied the tripartite DSM-5 ODD model to ASD to test its generalisability beyond non-ASD populations; and to elucidate the nature of ODD symptoms in ASD. Participants were 216 verbally-fluent young people (mean age=9.6 years, range 3.0 to 16.2 years, 82 % male) with ASD. Cross-sectional parent-and teacherreport data were analysed using bootstrap multiple regression to test the following predictions, derived from studies of nonASD young people: (1) AIS will be the main predictor of internalising problems; (2) ADB will be the main predictor of ADHD symptoms; (3) all ODD traits will independently predict conduct disorder symptoms; (4) vindictiveness will be the main predictor of aggressive conduct problems. Our findings using both parent and teacher data were consistent with the non-ASD ODD literature. AIS were associated with internalising but not externalising problems; ADB and vindictiveness were associated with externalising but not internalising problems; and vindictiveness was the main predictor of aggression. The DSM-5 tripartite model of ODD appears to be W. Mandy (*) Research Department of Clinical, Educational and Health Psychology, UCL, Gower Street, WC1N 6BT London, UK e-mail: [email protected] L. Roughan Department of Child and Adolescent Mental Health, Great Ormond Street Hospital for Children, London, UK D. Skuse Behavioural and Brain Science Unit, UCL Institute of Child Health, Guilford Street, London, UK generalisable to ASD: for people with an autistic disorder, AIS, ADB and vindictive dimensions of oppositionality have distinct associations with concurrent psychopathology, suggesting the need to assess them as separate constructs. Keywords Autism spectrum disorder (ASD) . Oppositional defiant disorder (ODD) symptoms . Conduct disorder symptoms . Angry and irritable symptoms . Argumentative and defiant behavior . Vindictiveness . Diagnostic and Statistical Manual—Fifth Edition (DSM-5) Whilst the core and definitive features of autism spectrum disorder (ASD) are deficits in social reciprocity, communication and flexibility, most people with ASD experience a constellation of additional difficulties, including emotional and behavioural disturbance (Gadow et al. 2008b). This is reflected in the frequency with which non-autistic psychiatric diagnoses are applicable to people with ASD. The only populationrepresentative study to investigate psychiatric comorbidity in ASD found that 70 % of their sample of 12-year-olds met criteria for at least one additional mental disorder, and 41 % met criteria for two (Simonoff et al. 2008). This is consistent with studies of clinically-ascertained samples which have applied DSM-IV diagnoses and found similarly elevated psychiatric comorbidity (e.g. Gadow et al. 2004; Gadow et al. 2005). High rates of co-occurring psychiatric symptomatology persist into adolescence (Simonoff et al. 2013), and have also been observed in adult ASD samples (Lugnegård et al. 2011). In addition to its high prevalence, a striking feature of the non-autistic psychopathology that accompanies ASD is its variety. Simonoff and colleagues (2008) discovered an array of internalising and externalising problems in their community ASD sample. The most common diagnoses were of anxiety disorders (48 % had at least one), followed by attention deficit hyperactivity disorder (ADHD; 28 %) and oppositional defiant disorder (ODD; 28 %). A similar picture has emerged from the study of clinically ascertained ASD samples, in which high rates of internalising difficulties (van Steensel et al. 2011), ODD and ADHD (Gadow et al. 2005) have been found. J Abnorm Child Psychol Relatively little is known about the processes that underpin the development and maintenance of behavioural and emotional disturbances in people with ASD, and this limits quality of clinical care. One approach to increasing knowledge in this area, which we take in the current study, is to test whether discoveries about the development of psychopathology in non-autistic populations apply to people with ASD. We argue that such an approach offers the promise of ‘quick wins’ in improving clinical practice, by identifying insights and clinical technologies based on non-autistic populations that could usefully be applied to ASD. Furthermore, testing the applicability to ASD of general models of psychopathology can also highlight areas of divergence, where superficially similar behavioural or emotional difficulties turn out to have distinct causes and treatment needs in ASD and nonASD groups (Gadow et al. 2004; Simonoff et al. 2013). In the current study we test whether recent advances in understanding ODD, achieved through the study of nonautistic people, can be usefully applied to ASD. We have focused on ODD as it is one of the most commonly diagnosed comorbidities in ASD (Simonoff et al. 2008), and because it is a key construct within developmental psychopathology, operating as a risk factor for a wide variety of internalising and externalising difficulties (Stringaris and Goodman 2009a). As such, increasing understanding of ODD in ASD may offer insights into mechanisms underpinning the broad range of psychopathology found in autistic populations. ODD is characterised by ‘a pattern of negativistic, defiant, disobedient or hostile behaviour towards authority figures’ (American Psychiatric Association 2000, p.100), and represents the extreme of a trait distribution that extends throughout the population (Sterba et al. 2010). ODD and ODD traits (‘oppositionality’) are a risk factor for a spectrum of internalising and externalising problems, and cannot simply be understood as early and/or mild manifestations of disruptive behaviour. This multifinality, and some factor analytic work, has led to the hypothesis that oppositionality is a multidimensional construct that should be parsed into separate factors, each with partially distinct associations with different types of maladaption (Krieger et al. 2013; Stringaris and Goodman 2009a; 2009b). The DSM-5 separates the features of ODD into ‘angry and irritable symptoms’ (AIS), ‘argumentative and defiant behaviour’ (ADB) and ‘vindictiveness’ (American Psychiatric Association 2013).1 This partitioning of ODD symptomatology is based on studies of non-autistic community and clinical samples that suggest each of these symptom dimensions has The DSM-5 constructs of ‘angry/irritable symptoms’, ‘argumentative disruptive behavior’ and ‘vindictiveness’ are sometimes referred to as ‘irritable’, ‘headstrong’ and ‘hurtful’ dimensions of oppositionality, notably by Stringaris and Goodman (2009a; 2009b). In the interests of clarity, we have chosen to use only the DSM-5 nomenclature when reviewing the literature. 1 distinct associations with other forms of psychopathology. AIS, but not ADB and vindictiveness, independently predict concurrent (Stringaris and Goodman 2009a) and future (Leibenluft 2011; Rowe et al. 2010; Stringaris and Goodman 2009b) mood and anxiety disorders. Of the three ODD dimensions, ADB behaviour is the best predictor of ADHD (Stringaris and Goodman 2009a; 2009b). All three factors are associated with conduct disorder symptoms, although cross-sectional and longitudinal associations between AIS and conduct problems have been observed to be weak (Stringaris et al. 2012), or non-existent when controlling for other ODD traits (Stringaris and Goodman 2009b). Vindictiveness is especially associated with aggressive symptoms of CD and callous-unemotional behaviour (Stringaris and Goodman 2009a; 2009b). We investigated whether this specific pattern of associations holds in an ASD sample, in order to test the validity of a three factor model of ODD traits in ASD. Specifically we tested the following hypotheses in a sample of children and young people with ASD: (1) AIS, but not ADB or vindictiveness, will be uniquely associated with internalising problems; (2) ADB will be particularly associated with ADHD symptoms; (3) AIS, ADB and vindictiveness will all be associated with CD symptoms; (4) Vindictiveness will be particularly associated with aggressive CD symptoms. Some phenomena of ODD in ASD have been found to be informant-specific, meaning that they vary depending on whether a parent or teacher provided data (Gadow et al. 2008a). On this basis we tested our hypotheses using information collected from both parents and teachers. We argue that our test of a tripartite model of ODD could offer insights into the variety of emotional and behavioural difficulties that co-occur with ASD. Furthermore, by testing whether findings on ODD from non-autistic samples are relevant to people with ASD, we can contribute to debates about whether ODD behaviours in ASD are manifestations of true comorbidity, or if instead they are a phenocopy, representing an autistic epiphenomenon, with only superficial resemblance to true ODD symptoms. Methods Participants Participants were 216 children and adolescents with ASD (82 % male) who received a full ASD and comorbidity assessment at a specialist social communication disorders service in London, UK. Information about their age, IQ, ASD symptoms and comorbidities are presented in Table 1. IQ data were only available for a subsection of the sample (verbal IQ available for 72 %, performance IQ for 56 %). This was not because a decision whether or not to test IQ was dependent on a child’s particular presenting problems; it occurred because IQ testing J Abnorm Child Psychol Table 1 Characteristics of the sample (n=216) Age, IQ and autistic symptomatology Mean (SD) [Range] Age in years 9.1 (2.9) [3.1 to 16.2] 3Di Reciprocal Social Interaction 15.1 (3.4) [9.0 to 24.2] 3Di Communication 13.9 (3.4) [6.3 to 22.2] 3Di Repetitive 4.3 (2.4) [0.0 to 11.5] Verbal IQ1 93.6 (19.9) [40 to 153] Performance IQ2 94.5 [57 to 143] Comorbidity Proportion in SDQ ‘definite problems’ range SDQ Total problems 77.8 % SDQ Hyperactivity 68.1 % SDQ Conduct problems 56.5 % SDQ Emotional problems 49.1 % SDQ Peer problems 80.1 % 3Di developmental, dimensional and diagnostic interview; SDQ strengths and difficulties questionnaire; 1 Verbal IQ n=156; 2 Performance IQ n=122 only became a routine part of the clinic’s assessment process after the start of the current study’s sampling period. There were no significant differences in level of ASD, CD or ODD symptomatology between participants with and without IQ data. Of the sample 81 (37.5 %) received a diagnosis of DSM-IV autistic disorder, 62 (28.7 %) were diagnosed with Asperger’s disorder and 73 (33.8 %) met criteria for pervasive developmental disorder—not otherwise specified. This is a predominantly high-functioning sample, with all participants demonstrating fluent language and attending mainstream school at the time of assessment, as these are referral criteria of the clinic in which this study was based. Nevertheless, autistic symptomatology and comorbidity was substantial and severe, equivalent to that found in a community-ascertained UK sample (Baird et al. 2006; Simonoff et al. 2013). Procedure All young people were assessed using the Dimensional, Developmental and Diagnostic Interview (3Di; Skuse et al. 2004), administered by an experienced child and adolescent psychiatrist or clinical psychologist. The 3Di is a parent-report interview that uses computerised algorithms to generate severity scores for each element of the autism triad, and also collects information from parents and teachers on symptoms of ODD and CD. In addition, the Strengths and Difficulties Questionnaire (Goodman 1997) was administered to parents and teachers. ASD diagnosis was based on information from the 3Di, plus structured reports from the child’s school or nursery and direct observation of the child in clinic, which was performed using the Autism Diagnostic Observation Schedule (ADOS; Lord et al. 2000) in most (n=167, 77 %) cases. This retrospective chart review study used nonidentifiable data collected in the course of routine clinical assessment. Ethical permission was provided after review by the research and development office of the hospital in which this study was based, and appropriate measures were taken to ensure young person and parent confidentiality. Measures The Developmental, Dimensional and Diagnostic Interview (3Di)—The 3Di is a validated parent report interview designed for the assessment of ASD and related psychopathology (Skuse et al. 2004). It covers developmental history and a range of past and current symptoms. A subset of 122 questions are used in an ASD diagnostic algorithm that generates dimensional scores for social reciprocity, social communication and repetitive, stereotyped behaviour (RSB). The test-retest and interrater reliability of the 3Di’s ASD algorithm is strong, with all intraclass correlations in the psychometric evaluation study exceeding 0.85 (Skuse et al. 2004). In terms of criterion validity, the 3Di shows excellent agreement with clinician ASD diagnosis (sensitivity=1; specificity=0.97). Also there is a high level of agreement between the 3Di and Autism Diagnostic Interview-Revised algorithms on whether a case is above the clinical threshold for social reciprocity (86 %), communication (97 %) and RSB (76 %) (Skuse et al. 2004). In addition to its ASD algorithm, the 3Di measures DSM-IVdefined symptoms of ODD and CD, by both parent and teacher report, with each relevant item directly corresponding to a DSM-IV diagnostic criterion. Inter-rater and test-retest reliability for diagnoses based on these sections of the 3Di is excellent, with greater than 95 % agreement between and within clinicians in the original validation study. Criterion validity, assessed by comparing 3Di and clinician diagnoses, is also excellent for both ODD (91 % agreement) and CD (93 % agreement). In the 3Di, parents report on the presence of each symptom of ODD, with answers coded as ‘not present’, ‘possibly present’, or ‘definitely present’. The three parent-report ODD subscores were calculated on the basis of guidelines proposed for DSM-5. To generate an AIS score, three items measuring frequent temper loss, touchiness and anger/resentfulness were summed (Cronbach’s α=0.72 in the current sample). ADB scores were summed from the four items measuring arguing with adults, refusal to comply, deliberately annoying others and blaming others for mistakes (α=0.77). The vindictiveness score was based on response to the 3Di item measuring spiteful or vindictive behaviour. Teacher ODD scores were calculated in the same way, with one exception. Teachers were not asked about the ‘touchiness’ symptom of ODD, so the teacher-report AIS score was calculated from two items concerning temper tantrums and anger/resentfulness (Kendall’s ΤB =0.66, p<0.001). The teacher ADB scale showed good reliability (α=0.86). J Abnorm Child Psychol Conduct disorder total scores were summed from 3Di items measuring DSM-IV defined symptoms. By both parent and teacher report, each symptom was marked as present or absent within the previous 12 months. We summed these items to derive total CD symptom scores from parent (α=0.79) and teacher (α=0.85) data. Also, we used all CD items to derive distinct subscales for aggressive and rule-breaking CD symptoms. Based on previous factor analytic work (Tackett et al. 2005), and DSM-IV, the aggressive CD scale by parent report was summed from items measuring the following symptoms of CD: bullying, fighting, use of weapons, cruelty to people, cruelty to animals, sexual assault and stealing while confronting a victim (α=0.60 in parent data). The rule-breaking CD scale from parent report was summed from items measuring fire setting, other destruction of property, stealing without confrontation, breaking into property, often staying out beyond curfew, running away, truanting and lying (α=0.71). The teacher-rated conduct scales were identical in composition with one exception: the rule-breaking scale did not include items measuring breaking into property, staying out beyond curfew and running away from home. Teachers were not asked to rate the presence of these symptoms, as it was thought that they would not have the requisite knowledge of the child’s behaviour to provide a valid answer. Internal consistency was acceptable for both teacher-report aggressive (α=0.79) and rule-breaking (α=0.67) CD scales. The Strengths and Difficulties Questionnaire (SDQ)—The SDQ comprises 25 items in five subscales: conduct problems, emotional problems, hyperactivity, peer problems and prosocial behaviour (Goodman 1997). Rated by parents and by teachers, it has acceptable reliability and validity when assessing adaptation and psychopathology in children and adolescents both as a screening instrument (Goodman 2001) and as a dimensional measure for quantifying the degree of an individual’s strengths and difficulties (Goodman and Goodman 2009). The SDQ appears to offer a valid measure of psychopathology in autistic populations (Simonoff et al. 2013). In the current study published thresholds for ‘definite’ caseness were used to derive the comorbidity estimates presented in Table 1 (Goodman 1997). In addition, SDQ ‘emotional problems’ and ‘hyperactivity’ scales were used as dimensional measures of internalising difficulties and ADHD symptoms respectively. The SDQ emotional problems scale comprises items measuring anxiety, low mood and somatising. It has been shown to correlate strongly with the Child Behaviour Checklist’s internalising scale (r=0.74; Goodman and Scott 1999). The SDQ hyperactivity scale measures ADHD symptomatology, comprising items on inattention, hyperactivity and impulsivity. It correlates strongly with the Child Behaviour Checklist scale measuring inattention/hyperactivity (r=0.71) and with clinician ADHD diagnosis (area under the curve=0.86; Goodman and Scott 1999). Analyses STATA version 12 (StataCorp 2011) was used. The study hypotheses were tested using hierarchical multiple regression. Some of our outcome variables deviated from normality so we used bootstrap regression based on 1,000 random samples to ensure robustness of our parameter estimates. This allowed empirically-based 95 % confidence intervals and p values to be derived, avoiding any normality-based assumption associated with the t-distribution used in standard linear regression. Each regression model had a first block containing age, sex, 3Di reciprocal social interaction, 3Di communication and 3Di repetitive, stereotyped behaviour. These were entered as control variables. The second block comprised AIS, ADB and vindictiveness scores. ADHD symptoms are associated with ODD and CD (Burke et al. 2002). Therefore we added the SDQ hyperactivity subscale as an additional control variable to parent and teacher models regressed on total CD, aggressive CD and rule-breaking CD symptoms. When more than one of the ODD dimensions was significant in the final model, we used the STATA ‘lincom’ command to execute a z-test, which tested the null hypothesis that the difference between their coefficients was not different from zero. This enabled us to test if one dimension of oppositionality made a greater contribution to the model than another. Results As Table 2 shows, in this sample, symptoms of ODD were common. Eighty-seven percent (n=187) of the sample had at least one definitely present ODD symptom by parent report, Table 2 Oppositional defiant disorder symptoms by parent and teacher report Parent report Number with symptom definitely present (%) N=216 Teacher report Number with symptom definitely present (%) Angry and irritable symptoms Temper loss 139 (64 %) 78 (36 %) Angry and resentful 140 (65 %) 75 (35 %) Touchiness 113 (52 %) – Argumentative and defiant behaviour Argumentative 112 (52 %) 77 (36 %) Refusal to do as told 113 (52 %) 87 (40 %) Blames others for mistakes 121 (56 %) 65 (30 %) Deliberately annoys others 75 (35 %) 60 (28 %) 49 (23 %) 40 (19 %) Vindictiveness Spiteful or vindictive J Abnorm Child Psychol and 55 % (n=119) had at least four. By teacher report, 56 % (n=122) participants had at least one definite ODD symptom, and 31 % (n=66) had four or more. Forty-seven percent (n=102) of participants had at least one definite CD symptom by parent report, and 19 % had three or more. The most common parent-reported CD symptoms were lying (27 %) and fighting (25 %). Sexual assault, burglary and robbery involving confrontation were rarest, each reported by parents in one percent or less of the sample. By teacher report, 32 % of children had at least one symptom of conduct disorder, and 13 % had three or more. Most commonly observed by teachers were lying (16 %) and fighting (13 %). Sexual assault, animal cruelty and fire starting were rare, reported by teachers in one percent or less of participants. Correlations between the three dimensions of oppositionality and the other study variables are presented for parent and teacher data in Table 3. Associations between ODD traits and 3Di scores were small (<±0.14) and, with one exception, non-significant. In the parent data AIS correlated with ADB (r(214)=0.70, p<0.001) and vindictiveness (rs(214)=0.44, p<0.001). ADB and vindictiveness were also associated (rs(214)=0.58, p<0.001). By teacher report AIS was associated with ADB (r(214)=0.73, p<0.001) and vindictiveness (rs(213)=0.57, p<0.001). There was also a strong correlation between ADB and vindictiveness (rs(215)=0.66, p<0.001). Tables 4 and 5 show standardised coefficients (β’s) and their 95 % confidence intervals from the bootstrap hierarchical regressions for parent and teacher data respectively. Each regression controlled for age, sex, and 3Di ASD algorithm scores by including these variables in block 1. The regressions on total CD, aggressive CD and rule-breaking CD symptoms also included the SDQ hyperactivity scale in block 1, to control for ADHD symptoms. For all regressions, the second block comprised the three ODD dimensions, allowing us to assess their individual, independent contributions to each model, above and beyond the control variables. Hypothesis one—AIS, but not ADB and Vindictiveness, Will Predict Internalising Problems Parent and teacher models offered support for hypothesis one, as in each model AIS was the only ODD dimension that independently predicted internalising problems, as measured by the SDQ emotional problems subscale. In the parent data, female sex and repetitive stereotyped behaviour were also independently associated with internalising. Hypothesis two—ADB Will be Particularly Associated with ADHD Symptoms In the parent-report model both AIS and vindictive behaviour were independently associated with ADHD symptoms, as measured by the SDQ hyperactivity subscale. Although β was larger for ADB than for vindictiveness, a z-test revealed that this difference was not significant (p=0.292). By teacher report ADB was the only ODD dimension that predicted ADHD symptoms. Hypothesis Three—AIS, ADB and Vindictive Dimensions Will all be Associated with Overall CD Symptoms The parent model shows that ADB and vindictiveness were each independently associated with the overall CD symptom score. Vindictiveness was a significantly stronger predictor of CD symptoms than ADB (p=0.044). In the teacher model it was also the case that ADB and vindictiveness, but not AIS, independently predicted CD symptoms. Whilst the coefficient for vindictiveness was bigger than that for ADB in the teacher model, our z-test showed that this difference did not reach significance (p=0.125). Hypothesis Four—Vindictiveness Will be Particularly Associated with Aggressive CD Symptoms The parent and teacher models predicting aggressive symptoms of conduct disorder support hypothesis four. In the parent and teacher models vindictiveness was the only significant predictor of aggression, and this was a substantial effect (βs>0.45). We also ran models predicting the rulebreaking symptoms of CD using both parent and teacher data. They both showed ADB and vindictiveness to be independently associated with rule-breaking CD symptoms. In parent (p=0.421) and teacher (p=0.301) models, coefficients for these dimensions of oppositionality were not significantly different in size. Discussion In a sample of young people with ASD, we investigated the DSM-5 model of oppositional defiant disorder (ODD), which parses features of oppositionality into ‘angry and irritable symptoms’ (AIS), ‘argumentative and defiant behaviour’ (ADB) and ‘vindictiveness’. To assess the validity of this division we tested whether these three dimensions of oppositionality had distinct associations with other measures of concurrent psychopathology. Specifically we tested the following hypotheses, which were based on findings from non-autistic young people: (1) AIS, but not ADB or vindictiveness, will be associated with internalising problems; (2) ADB traits will be particularly associated with ADHD symptoms; (3) AIS, ADB and vindictive traits will all be associated with conduct disorder (CD) symptoms; (4) Vindictiveness will be particularly associated with aggressive CD symptoms. 0.07 –0.09 0.04 –0.06 0.54*** 0.46*** 0.42*** 0.36*** 0.37*** Parent rule-breaking CD symptoms Age 0.54*** 0.57*** 0.59*** 0.14* Teacher rule-breaking CD symptoms Teacher SDQ hyperactivity (ADHD symptoms) Teacher SDQ emotional problems (internalising) Teacher total conduct disorder symptoms Teacher aggressive CD symptoms 3Di Repetitive stereotyped behaviour –0.10 0.32*** 3Di Communication 3Di Social Sex (male=0, female=1) –0.10 0.01 0.02 0.14* Vindictiveness1 Teacher report 0.49*** 0.39*** 0.44*** 0.28*** 0.36*** 0.05 0.60*** 0.39*** 0.54*** 0.21** 0.37*** 0.06 –0.07 –0.11 –0.03 –0.02 0.13 –0.16* –0.10 Argumentative disruptive behaviour –0.00 Angry irritable symptoms 0.57*** 0.62*** 0.62*** 0.22** 0.26*** 0.10 –0.13* –0.12 –0.14* 0.07 Vindictiveness1 0.03 [−0.12 to 0.18] 0.13* [0.00 to 0.27] – 0.11 [−0.03 to 0.25] 0.08 [−0.06 to 0.23] –0.02 [−0.13 to 0.10] – 3Di social Reciprocity 3Di communication 3Di repetitive Stereotyped Behaviour SDQ hyperactivity (ADHD symptoms) 0.22*** Overall R2 of model 0.20*** 0.10*** 0.04 [−0.11 to 0.19] -0.08 [−0.27 to 0.12] 0.37*** [0.18 to 0.56] 0.44*** 0.31*** 0.45*** [0.31 to 0.59] 0.23** [0.09 to 0.36] 0.04 [−0.08 to 0.16] 0.06 [−0.03 to 0.14] –0.02 [−0.15 to 0.11] 0.00 [−0.11 to 0.11] 0.05 [−0.06 to 0.16] –0.01 [−0.12 to 0.09] 0.03 [−0.06 to 0.11] Total conduct disorder symptoms 0.40*** 0.29*** 0.47*** [0.32 to 0.63] 0.14 [−0.02 to 0.30] 0.07 [−0.06 to 0.20] 0.07 [−0.03 to 0.17] –0.00 [−0.13 to 0.12] 0.00 [−0.13 to 0.13] 0.02 [−0.11 to 0.14] –0.04 [−0.15 to 0.08] –0.04 [−0.13 to 0.06] Aggressive conduct disorder symptoms 0.34*** 0.28*** 0.35*** [0.21 to 0.49] 0.26*** [0.13 to 0.39] 0.01 [−0.10 to 0.11] 0.03 [−0.05 to 0.12] –0.03 [−0.18 to 0.12] 0.00 [−0.12 to 0.13] 0.08 [−0.04 to 0.19] 0.01 [−0.11 to 0.13] 0.08 [−0.02 to 0.17] Rule-breaking conduct disorder symptoms SDQ strengths and difficulties questionnaire; 3Di developmental, dimensional and diagnostic interview; ADHD attention deficit hyperactivity disorder; ODD oppositional defiant disorder *p<0.05, **p<0.01, ***p<0.001 0.15* [0.03 to 0.27] 0.19*** Δ R2 from addition of ODD traits 0.28** [0.11 to 0.44] Argumentative defiant behaviour Vindictiveness 0.11 [−0.08 to 0.29] Angry irritable mood Block two – ODD variables 0.20** [0.07 to 0.32] –0.07 [−0.20 to 0.07] 0.05 [−0.09 to 0.19] 0.07 [−0.06 to 0.20] –0.12 [−0.25 to 0.01] Sex (male=0, female-1) SDQ emotional problems (Internalising) Age Block one—control variables SDQ hyperactivity (ADHD symptoms) Table 4 Bootstrap regression models testing the study hypotheses in parent-report data, showing standardised betas and their 95 % confidence intervals 3Di developmental, dimensional and diagnostic interview; SDQ strengths and difficulties questionnaire; ADHD attention deficit hyperactivity disorder; 1 All coefficients for Vindictiveness are Spearman’s ρ *p<0.05, **p<0.01, ***p<0.001 0.24*** 0.36*** 0.49*** 0.40*** 0.32*** Parent SDQ hyperactivity (ADHD symptoms) Parent SDQ emotional problems (internalising) Parent total conduct disorder symptoms Parent aggressive CD symptoms 0.04 0.02 3Di Repetitive stereotyped behaviour 3Di Communication 3Di Social 0.24*** 0.02 0.27*** 0.02 Argumentative disruptive behaviour Sex (male=0, female=1) Angry irritable symptoms Age Parent report Table 3 Correlations between traits of oppositional defiant disorder and study variables, in parent and teacher data J Abnorm Child Psychol SDQ strengths and difficulties questionnaire; 3Di developmental, dimensional and diagnostic interview; ADHD attention deficit hyperactivity disorder; ODD oppositional defiant disorder *p<0.05, **p<0.01, ***p<0.001 0.35*** 0.43*** 13*** 0.08*** 0.45*** 0.26*** 0.37*** 0.26** [0.08 to 0.43] Overall R2 of model 0.26*** 0.44*** [0.24 to 0.64] 0.13*** 0.35*** 0.09 [−0.10 to 0.27] −0.03 [−0.20 to 0.14] Vindictiveness Δ R2 from addition of ODD traits 0.44*** [0.26 to 0.61] 0.01 [−0.16 to 0.18] 0.41*** [0.20 to 0.62] 0.21* [0.04 to 0.39] −0.05 [−0.27 to 0.17] 0.27** [0.08 to 0.46] Argumentative defiant behaviour 0.01 [−0.15 to 0.17] −0.01 [−0.15 to 0.13] 0.26** [0.06 to 0.45] 0.15 [−0.04 to 0.34] Angry irritable mood Block two—ODD variables 0.05 [−0.11 to 0.21] 0.01 [−0.11 to 0.13] −0.02 [−0.09 to 0.12] 0.06 [−0.05 to 0.18] 0.04 [−0.07 to 0.14] 0.04 [−0.09 to 0.17] 0.04 [−0.11 to 0.18] − −0.03 [−0.16 to 0.10] − 3Di repetitive Stereotyped Behaviour SDQ hyperactivity (ADHD symptoms) 0.08 [−0.05 to 0.21] 0.03 [−0.11 to 0.16] 0.06 [−0.07 to 0.19] 0.05 [−0.09 to 0.20] 0.09 [−0.06 to 0.25] 0.05 [−0.08 to 0.18] −0.06 [−0.20 to 0.08] 0.03 [−0.11 to 0.18] −0.07 [−0.23 to 0.08] 0.02 [−0.12 to 0.17] 3Di social Reciprocity 3Di communication −0.14 [−0.32 to 0.04] 0.03 [−0.09 to 0.14] 0.06 [−0.04 to 0.16] 0.03 [−0.06 to 0.12] 0.03 [−0.06 to 0.13] 0.00 [−0.10 to 0.10] Block one—control variables 0.04 [−0.04 to 0.12] 0.21** [0.08 to 0.35] 0.07 [−0.06 to 0.20] –0.14* [−0.27 to −0.01] −0.30*** [−0.43 to −0.15] Age Sex (male=0, female-1) Aggressive conduct disorder symptoms Total conduct disorder symptoms SDQ emotional problems (Internalising) SDQ hyperactivity (ADHD symptoms) Table 5 Bootstrap regression models testing the study hypotheses in teacher-report data, showing standardised betas and their 95 % confidence intervals Rule-breaking conduct disorder symptoms J Abnorm Child Psychol Our findings were mostly supportive of the study hypotheses, thus providing evidence for the validity of the DSM-5 tripartite ODD model in ASD. As in studies of non-ASD samples (Stringaris and Goodman 2009a), we found that AIS was the only ODD dimension uniquely associated with internalising problems. This was the case in both parentand teacher-report data. There was support for the second hypothesis, that ADB would have the strongest association with ADHD, as ADB was the only variable associated with ADHD symptoms in both parent and teacher models. In line with hypotheses three and four, ADB and vindictiveness were both associated with overall CD symptoms; and by both parent and teacher report, vindictiveness was the sole predictor of aggressive CD behaviours. Nevertheless, in partial contradiction of our third hypothesis, there was no relationship between AIS and CD symptoms. This contrasts with cross-sectional (but not longitudinal) analyses in nonASD samples, which have found a modest but significant relationship between AIS and CD (Stringaris and Goodman 2009a; 2009b). As such, our findings with respect to AIS and CD symptoms may signify an area of divergence between ASD and non-ASD populations. Theorists of comorbidity have debated the meaning of the high rates of diagnosable additional mental disorders in ASD (e.g. Gadow et al. 2004; Lecavalier et al. 2009). Is this true comorbidity, in which the psychiatric symptoms seen in ASD are genuine manifestations of DSM-IV mental disorders? Or does it actually reflect the presence of phenocopies which are epiphenomena of ASD? Our findings speak to this debate, offering support for the idea that the high rates of diagnosable ODD in ASD may reflect authentic comorbidity. There were many more similarities than differences between our findings and those reported for non-autistic samples (e.g., Stringaris and Goodman 2009a; 2009b), suggesting that the three dimensions of ODD configure similarly in autistic and non-autistic populations, both in their relationships with each other and in their associations with external validators. This is compatible with the possibility that they share a similar underlying structure and aetiology. Furthermore, the associations we observed between autistic symptomatology and ODD traits were small and mostly non-significant. This does not fit with the idea that the ODD symptoms are merely epiphenomena of ASD. If that was the case, we would expect oppositionality to intensify as autistic symptomatology increased. Other investigations of ODD in ASD (Guttmann-Steinmetz et al. 2009; Gadow et al. 2008a) also found more similarities than differences between the presentation of ODD in autistic and non-autistic samples. It will be important in future to compare the neuropsychological underpinnings of ODD traits in autistic and non-autistic samples to further investigate whether manifest oppositionality in these two groups signifies the same underlying disorder. Also, the use of confirmatory factor analysis to compare the structure of ODD symptoms in people with and without ASD would J Abnorm Child Psychol provide additional information about the applicability of the DSM-5 tripartite model to autistic populations. To our knowledge, this is the first study to measure vindictiveness (i.e., vindictive and spiteful behaviour) in people with ASD. Our findings suggest that this construct should receive more attention in future. In both parent and teacher data, around a fifth of our participants were described as definitely showing vindictiveness; and this characteristic emerged as the most consistent and strongest predictor of conduct disorder symptoms. In particular, it was the only variable independently associated with aggressive behaviour. Evolutionary psychologists have shown some interest in vindictiveness and spite, conceptualising them as motivating adaptive social behaviours that enforce moral codes through punishing unfair acts (Tooby and Cosmides 2008). In our clinical experience, many children with ASD have a heightened and somewhat rigid sense of fairness, and can have difficulty accepting what they perceive to be an injustice. It will be interesting to test to what extent reported acts of vindictiveness and spite in our sample arise from this characteristic; and by extension what role moral inflexibility plays in aggressive behaviour in ASD. Our findings highlight the importance of assessing AIS in young people with ASD. Symptoms of AIS were common in our clinical ASD sample and were associated with internalising difficulties. People with ASD often struggle to understand and articulate their feelings, which can make their anxiety and low mood hard to detect (Hill et al. 2004). By contrast symptoms of anger and irritability are usually directly observable and obvious. The presence of anger and irritability in children with ASD should alert clinicians to the need for careful assessment of anxiety and mood problems. Why was AIS so common in our autistic sample? Leibenluft (2011) has proposed a model of irritability in which rigidity, dysregulated attention and poor emotional processing play a role in both increasing the likelihood of experiencing frustration and decreasing the capacity to manage frustration. We note that rigidity (Ozonoff et al. 2004), attention problems (Gadow et al. 2005) and emotion processing deficits (Hobson et al. 1988) are all common in people with ASD, and this may go some way to explaining the high rates of AIS in our sample. It will be interesting to test Leibenluft’s model of irritability in ASD by examining whether variability in rigidity, attention problems and emotional processing predicts elevated AIS in people with an autistic disorder. The association between AIS and internalising difficulties that we observed is consistent with the new DSM-5 diagnosis of temper deregulation disorder with dysphoria (TDDD), which is characterised by irritability and negative affect. Simonoff and colleagues (2012) recently examined ‘severe mood problems’ (a construct very similar to TDDD) in their clinically ascertained ASD cohort, and also observed an association between irritability and low mood in ASD. Given our findings and those of Simonoff and colleagues (2012), it is likely that TDDD as described in DSM-5 will be a valuable construct for characterising the nature of mood disturbance in ASD. Studies that apply formal criteria for TDDD in ASD are required. Our cross-sectional analyses raise the question of whether the adoption of the three dimension model of oppostionality could improve the prognostic accuracy of ODD traits found in people with ASD. This has been the case in non-autistic populations (Stringaris and Goodman 2009b). Follow-up studies are required to test whether longitudinal findings in nonASD populations about AIS, ADB and vindictiveness apply to people with ASD. A study of 135 adults with ASD who had been diagnosed in childhood discovered that all psychopathology that emerged in adulthood involved disordered affect, with depression and obsessive compulsive disorder being the most common problems (Hutton et al. 2008). Likewise, comparison of cross-sectional child and adult studies suggests that the transition to adulthood in ASD is associated with an increased risk of depression and obsessive compulsive disorder, as these are diagnosed more often in adult compared to childhood autistic samples. In non-autistic people, AIS in childhood is a risk factor for the subsequent development of affective and anxiety disorders (Burke 2012; Stringaris et al. 2012). Furthermore, temper outbursts, a key symptom of AIS, are highly prevalent in obsessive compulsive disorder (OCD; Krebs et al. 2012). We hypothesise that children with ASD who score high on AIS will be at the greatest risk of developing mood and anxiety disorders, including OCD, in adulthood. If longitudinal investigations found this to be the case, it would promote the identification of high-risk groups amongst young people with ASD, and would suggest measures prophylactic against the development of comorbidity in adolescence and adulthood. Our findings are relevant to the development of treatments for young people with ASD. ODD symptoms are common in this population (Simonoff et al. 2008), and the risk they pose for later aggression is amplified by co-occurring autistic social impairments (Mandy et al. 2013). As such, there is a pressing need for ODD treatments of proven effectiveness for people with ASD. Our observations of commonalities between oppositionality in people with and without ASD suggest the value of testing existing evidence-based treatments for ODD in ASD samples. Interventions that treat conduct problems by teaching parents to change their responses to their child’s problem behaviours (‘contingency management’) are amongst the most effective in child psychology (Scott 2008). Such treatments should be evaluated experimentally in young people with ASD and ODD, with due consideration of whether autistic difficulties, such as rigidity and empathy deficits, moderate treatment response. Furthermore, our findings suggest a need for treatments for AIS in ASD. Currently there is no basis for evidence-based J Abnorm Child Psychol practice in this area, although promising interventions have been manualised (e.g., Attwood 2007) making them amenable for evaluation using randomised controlled trials. The study we report had several strengths, including the use of multiple informants, standardised measures administered by experienced clinicians and robust statistical analyses to test focused hypotheses that were grounded in previous literature. Nevertheless, its findings should be considered in relation to the following limitations. Our data were cross-sectional and so are not well suited to testing causal hypotheses. Also, the sample comprised individuals with fluent language who were in mainstream education, and so is not representative of the full population of people with ASD. As such, whether or not our findings can be generalised to people with low-functioning ASD, who comprise up to 50 % of those with an ASD diagnosis (Baird et al. 2006), remains an open question for further empirical investigation. Our dimensional measures of conduct problems were based on DSM-IV symptom checklists. This allowed us to compare our results to relevant studies of nonautistic people, and generally ensured our measures had content validity, in that they covered the full range of ODD and CD symptoms. Nevertheless, our study would have benefitted from the use of fine-grained measures with more items. This is especially true for our teacher data, and for the vindictiveness construct, which was measured by a single item in the current study, reflecting the fact that only one ODD diagnostic criterion in DSM-IV (and indeed DSM-5) covers vindictiveness. Furthermore, our measure of internalising problems did not distinguish low mood and anxiety. It will be useful in future to measure these constructs independently, to test whether they are differentially associated with irritability. Conclusions Our findings highlight the importance of measuring ODD traits in the clinical assessment of ASD. There was considerable variability amongst our participants in terms of ODD symptom severity, and this variability contained important clinical information about an individual’s tendency to have difficulties with ADHD symptoms, internalising and conduct. Furthermore, despite their high intercorrelations, AIS, ADB and vindictiveness dimensions of oppositionality had distinct associations with concurrent psychopathology, suggesting the need to assess these as separate constructs. AIS in ASD is part of a picture of comorbidity involving anxiety and low mood. By contrast ADB and vindictiveness associate with externalising problems, such as ADHD and disruptive behaviour. Acknowledgements The authors wish to thank Richard Warrington whose 3Di software made this study possible, as well as the anonymous reviewers for their helpful suggestions and comments. References American Psychiatric Association. (2000). Diagnostic and statistical manual (4th ed., text rev.). Washington: Author. American Psychiatric Association. (2013). Diagnostic and statistical manual (5th ed.). Washington: Author. Attwood, T. (2007). Exploring feelings: Cognitive behaviour therapy to manage anger. Arlington: Future Horizons Inc. Baird, G., Simonoff, E., Pickles, A., Chandler, S., Loucas, T., Meldrum, D., & Charman, T. (2006). Prevalence of disorders of the autism spectrum in a population cohort of children in south Thames: The special needs and autism project (SNAP). Lancet, 368, 210–5. Burke, J. D. (2012). An affective dimension within oppositional defiant disorder symptoms among boys: personality and psychopathology outcomes into early adulthood. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 53, 1176–1183. Burke, J., Loeber, R., & Birmaher, B. (2002). Oppositional defiant disorder and conduct disorder: A review of the past 10 years, part II. Journal of the American Academy of Child and Adolescent Psychiatry, 41, 1275–93. Gadow, K. D., DeVincent, C. J., & Drabick, D. A. (2008a). Oppositional defiant disorder as a clinical phenotype in autism spectrum disorder. Journal of Autism and Developmental Disorders, 38, 1302–10. Gadow, K. D., DeVincent, C. J., Pomeroy, J., & Azizian, A. (2005). Comparison of DSM-IV symptoms in elementary school-age children with PDD versus clinic and community samples. Autism: the international journal of research and practice, 9, 392–415. Gadow, K. D., DeVincent, C. J., Pomeroy, J., & Azizian, A. (2004). Psychiatric symptoms in preschool children with PDD and clinic and comparison samples. Journal of Autism and Developmental Disorders, 34, 379–393. Gadow, K. D., DeVincent, C., & Schneider, J. (2008b). Predictors of psychiatric symptoms in children with an autism spectrum disorder. Journal of Autism and Developmental Disorders, 38, 1710–1720. Goodman, A., & Goodman, R. (2009). Strengths and difficulties questionnaire as a dimensional measure of child mental health. Journal of the American Academy of Child and Adolescent Psychiatry, 48, 400–403. Goodman, R. (1997). The strengths and difficulties questionnaire: A research note. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 38, 581–586. Goodman, R. (2001). Psychometric properties of the strengths and difficulties questionnaire. Journal of the American Academy of Child and Adolescent Psychiatry, 40, 1337–1345. Goodman, R., & Scott, S. (1999). Comparing the strengths and difficulties questionnaire and the child behavior checklist: Is small beautiful? Journal of Abnormal Child Psychology, 27, 17–24. Guttmann-Steinmetz, S., Gadow, K. D., & DeVincent, C. J. (2009). Oppositional defiant and conduct disorder behaviors in boys with autism spectrum disorder with and without attention-deficit hyperactivity disorder versus several comparison samples. Journal of Autism and Developmental Disorders, 39, 976–985. Hill, E., Berthoz, S., & Frith, U. (2004). Brief report: cognitive processing of own emotions in individuals with autistic spectrum disorder and in their relatives. Journal of Autism and Developmental Disorders, 34, 229–235. Hobson, R. P., Ouston, J., & Lee, A. (1988). Emotion recognition in autism: Coordinating faces and voices. Psychological Medicine, 18, 911–923. Hutton, J., Goode, S., Murphy, M., Le Couteur, A., & Rutter, M. (2008). New-onset psychiatric disorders in individuals with autism. Autism: the international journal of research and practice, 12, 373–390. J Abnorm Child Psychol Krebs, G., Bolhuis, K., Heyman, I., Mataix-Cols, D., Turner, C., & Stringaris, A. (2012). Temper outbursts in paediatric obsessivecompulsive disorder and their association with depressed mood and treatment outcome. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 54, 313–22. Krieger, F. V., Polanczyk, V. G., Robert, G., Rohde, L. A., GraeffMartins, A. S., Salum, G., & Stringaris, A. (2013). Dimensions of oppositionality in a Brazilian community sample: testing the DSM-5 proposal and etiological links. Journal of the American Academy of Child and Adolescent Psychiatry, 52, 389–400. Lecavalier, L., Gadow, K. D., DeVincent, C. J., & Edwards, M. C. (2009). Validation of DSM-IV model of psychiatric syndromes in children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 39, 278–289. Leibenluft, E. (2011). Severe mood dysregulation, irritability, and the diagnostic boundaries of bipolar disorder in youths. The American Journal of Psychiatry, 168, 129–142. Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Jr., Leventhal, B. L., DiLavore, P. C., & Rutter, M. (2000). The autism diagnostic observation schedule-generic: A standard measure of social and communication deficits associated with the spectrum of autism. Journal of Autism and Developmental Disorders, 24, 659–85. Lugnegård, T., Hallerbäck, M. U., & Gillberg, C. (2011). Psychiatric comorbidity in young adults with a clinical diagnosis of asperger syndrome. Research in Developmental Disabilities, 32, 1910–1917. Mandy, W., Skuse, D., Steer, C., St Pourcain, B., & Oliver, B. (2013). Oppositionality and socio-emotional competence: Interacting risk factors in the development of childhood conduct disorder symptoms. Journal of the American Academy of Child and Adolescent Psychiatry, 52, 718–27. Ozonoff, S., Cook, I., Coon, H., Dawson, G., Joseph, R. M., Klin, A., & Wrathall, D. (2004). Performance on Cambridge neuropsychological test automated battery subtests sensitive to frontal lobe function in people with autistic disorder: Evidence from the collaborative programs of excellence in autism network. Journal of Autism and Developmental Disorders, 34, 139–150. Rowe, R., Costello, E. J., Angold, A., Copeland, W. E., & Maughan, B. (2010). Developmental pathways in oppositional defiant disorder and conduct disorder. Journal of Abnormal Psychology, 119, 726– 738. Scott, S. (2008). Parenting programmes. In M. Rutter, D. Bishop, D. Pine, S. Scott, J. Stevenson, E. Taylor, & A. Thapar (Eds.), Rutter’s child and adolescent psychiatry (5th ed.). Oxford: Blackwell. Simonoff, E., Jones, C. R. G., Baird, G., Pickles, A., Happé, F., & Charman, T. (2013). The persistence and stability of psychiatric problems in adolescents with autism spectrum disorders. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 54, 186–194. Simonoff, E., Jones, C. R. G., Pickles, A., Happé, F., Baird, G., & Charman, T. (2012). Severe mood problems in adolescents with autism spectrum disorder. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 53, 1157–1166. Simonoff, E., Pickles, A., Charman, T., Chandler, S., Loucas, T., & Baird, G. (2008). Psychiatric disorders in children with autism spectrum disorders: prevalence, comorbidity, and associated factors in a population-derived sample. Journal of the American Academy of Child and Adolescent Psychiatry, 47, 921–929. Skuse, D., Warrington, R., Bishop, D., Chowdhury, U., Lau, J., Mandy, W., & Place, M. (2004). The developmental, dimensional and diagnostic interview (3di): A novel computerized assessment for autism spectrum disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 548–558. StataCorp. (2011). Stata statistical software: Release 12. College Station, TX: StataCorp LP. Sterba, S. K., Copeland, W., Egger, H. L., Jane Costello, E., Erkanli, A., & Angold, A. (2010). Longitudinal dimensionality of adolescent psychopathology: Testing the differentiation hypothesis. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 51, 871–884. Stringaris, A., & Goodman, R. (2009a). Three dimensions of oppositionality in youth. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 50, 216–223. Stringaris, A., & Goodman, R. (2009b). Longitudinal outcome of youth oppositionality: Irritable, headstrong, and hurtful behaviors have distinctive predictions. Journal of the American Academy of Child and Adolescent Psychiatry, 48, 404–412. Stringaris, A., Zavos, H., Leibenluft, E., Maughan, B., & Eley, T. C. (2012). Adolescent irritability: Phenotypic associations and genetic links with depressed mood. The American Journal of Psychiatry, 169, 47–54. Tackett, J., Krueger, R., Iacono, W., & McGue, M. (2005). Symptombased subfactors of DSM-defined conduct disorder: Evidence for etiologic distinctions. Journal of Abnormal Psychology, 114, 483–7. Tooby, J., & Cosmides, L. (2008). The evolutionary psychology of the emotions and their relationship to internal regulatory variables. In M. Lewis, J. Haviland-Jones, & L. Barrett (Eds.), Handbook of emotions (3rd ed., pp. 114–137). New York: Guilford. van Steensel, F. J. A., Bögels, S. M., & Perrin, S. (2011). Anxiety disorders in children and adolescents with autistic spectrum disorders: A meta-analysis. Clinical Child and Family Psychology Review, 14, 302–317.