Using generalized additive (mixed) models to analyze single case designs

This article shows how to apply generalized additive models and generalized additive mixed models to single-case design data. These models excel at detecting the functional form between two variables (often called trend), that is, whether trend exists, and if it does, what its shape is (e.g., linear and nonlinear). In many respects, however, these models are also an ideal vehicle for analyzing single-case designs because they can consider level, trend, variability, overlap, immediacy of effect, and phase consistency that single-case design researchers examine when interpreting a functional relation. We show how these models can be implemented in a wide variety of ways to test whether treatment is effective, whether cases differ from each other, whether treatment effects vary over cases, and whether trend varies over cases. We illustrate diagnostic statistics and graphs, and we discuss overdispersion of data in detail, with examples of quasibinomial models for overdispersed data, including how to compute dispersion and quasi-AIC fit indices in generalized additive models. We show how generalized additive mixed models can be used to estimate autoregressive models and random effects and discuss the limitations of the mixed models compared to generalized additive models. We provide extensive annotated syntax for doing all these analyses in the free computer program R.     

The Evolution of Mindfulness‐based Cognitive Therapy

Metcalf and Dimidjian (this issue) present a timely review of the evolving evidence base and mechanisms of mindfulness‐based cognitive therapy (MBCT). The present commentary extends the discussion on the current evidence base for MBCT based on findings from recent meta‐analytic reviews in this field which attest to the promising outcome for mindfulness‐based therapies, particularly for depressed populations. However, the specific effects of MBCT as applied to anxiety, health and developmental populations is still very much in its infancy. The second objective of this commentary extends discussion on the transdiagnostic applications of MBCT versus traditional cognitive behavioural therapy (CBT). It is recommended that with the continuing expansion of MBCT, the effects of this therapeutic approach needs to be evaluated against other empirically supported therapies, including traditional CBT.     

Individual differences in 10‐month‐olds' performance on the A‐not‐B task

This study used the classical A‐not‐B task (Piaget, 1954 ) to explore individual differences in cognitive flexibility in 10‐month‐old infants by: (1) examining how differences in search performance during A trials relate to search performance during B trials; (2) studying the relation between temperamental dimensions and A‐not‐B performance; and (3) investigating differences in search performance between looking and reaching responses within the same task. Forty infants were tested on a fixed‐design‐version of the A‐not‐B task, not allowing for training or individual adjustment, but instead eliciting additional search behaviors than the common correct responses in A trials and perseverative errors in B trials. Infants were also rated by their parents on the temperamental scales Activity level and Attention span. The main findings were: (1) performance on A trials affected B trial performance, with infants being more correct on A trials having more incorrect and less ‘no search’ responses on B trials; (2) activity level, but not attention span, was related to performance on the A‐not‐B task, with infants performing better on A trials having a lower activity level; and (3) there were a few differences in performance with regard to modality, indicating that responding correctly by looking may be less cognitively demanding than doing so by reaching. This study demonstrated that 10‐month‐olds show a wide variation of search behaviors on this A‐not‐B task, resulting in individual differences in performance. These differences are suggested to reflect variation in temperamental activity level as well as maturity of short term/working memory, inhibition and cognitive flexibility.     

A qualitative study of citizens' experience of participating in health counseling

Individual health counseling is a form of intervention designed to minimize the effects of chronic health disease and to offer a path towards good health practices. The aim of the present study was to explore the experiences of those persons who participated in health counseling in order to assess the psychosocial significance of the counseling upon their health behavior. In addition the study was concerned with the factors which underlay peoples' decision to sign up for health counseling. The research involved 11 semi‐structured interviews with individuals who had participated in a municipality based health counseling program. Data was analyzed using Malterud's systemic text condensation and a theoretical framework around Bandura's social cognitive theory. Analysis revealed that an approach tailored to each individual with minor short‐term goals accompanied by feeling supported by the counselor produced the greatest impact on behavior. Receiving feedback from the counselor and feeling positive about the relationship were seen as essential. These aspects were also crucial in the decision to undertake health counseling. The study indicates that whether individual's sign up on their own initiative or conversely are invited to join the program has no influence upon their motivation to change their behavior. Overall the respondents demonstrated improvement in their behavior and attitudes towards their health. However, the study also indicates that this form of intervention is less or even ineffective without the support of the individual's immediate family.     

Rural Neighborhoods and Child Aggression

Structural equation modeling with latent variables was used to evaluate the direct and mediated effects of a neighborhood risk factor (negative teen behaviors) on the parent-report aggressive behavior of 213 students in grades 3 through 5 attending a school in a low-income, rural community. Contagion and social control hypotheses were examined as well as hypotheses about whether the neighborhood served as a microsystem or exosystem for rural pre-adolescents. Analyses took into account the clustering of students and ordinal nature of the data. Findings suggest that rural neighborhoods may operate as both a microsystem and exosystem for children, with direct contagion effects on their aggressive behaviors as well as indirect social control effects through parenting practices. Direct effects on aggression were also found for parenting practices and child reports of friends’ negative behaviors. Pre-adolescence may be a transitional stage, when influences of the neighborhood on child behavior begin to compete with influences of caregivers. Findings can inform the timing and targets of violence prevention in rural communities.     

The Dark Triad,happiness and subjective well-being

Narcissism can be expressed in grandiose or vulnerable forms. We examined whether positive psychological states (defined by the Oxford Happiness Inventory (OHI) and the Diener Satisfaction With Life (SWL) scales) assisted differentiation relative to general personality traits and the “the Dark Triad” (psychopathy, narcissism, and Machiavellianism, measured by the D12 and Short Dark Triad (SD3) indices) for 840 persons primarily from the UK, USA and Canada. The best fitting structural equation model comprised two latent variables, one of positive mood (comprising total scores on the OHI and SWL scales), and another forming a “dark dyad” of Machiavellianism and psychopathy (predicted by low agreeableness and lower positive mood), with narcissism regarded as a separate construct correlated with the dark dyad. Latent positive mood was primarily predicted by higher emotional stability and extraversion. Narcissism was predicted by lower emotional stability, lower agreeableness, and higher extraversion. Latent profile analysis identified four groups in the data: “unhappy but not narcissistic”, “vulnerable narcissism”, “happy non-narcissism” and “grandiose narcissism”. Our results suggest more problematic narcissism can be identified by reference to measures indexing positive mood states and general personality traits.     

Theory and teen dating violence victimization: Considering adolescent development

Teen dating violence is an important public health problem, with implications for the future health and well-being of adolescents. However, most work on teen dating violence has developed separately from literature on normative adolescent romantic relationships and development; understanding teen dating violence within the framework of adolescent psychosocial development may provide new areas for research. Thus, the present paper summarizes five theories of adolescent development that are relevant to the study of teen dating violence victimization, as well as empirical literature that demonstrates support for key theoretical tenets in research examining adolescent romantic relationships. We also present questions for future dating violence study that arise from these key theoretical tenets and past empirical research. Researchers interested in dating violence victimization can use the presented theories to guide new directions in research inquiry, so that findings are situated within the broader field of adolescent development.     

Quantitative Standards for Absolute Linguistic Universals

Absolute linguistic universals are often justified by cross‐linguistic analysis: If all observed languages exhibit a property, the property is taken to be a likely universal, perhaps specified in the cognitive or linguistic systems of language learners and users. In many cases, these patterns are then taken to motivate linguistic theory. Here, we show that cross‐linguistic analysis will very rarely be able to statistically justify absolute, inviolable patterns in language. We formalize two statistical methods—frequentist and Bayesian—and show that in both it is possible to find strict linguistic universals, but that the numbers of independent languages necessary to do so is generally unachievable. This suggests that methods other than typological statistics are necessary to establish absolute properties of human language, and thus that many of the purported universals in linguistics have not received sufficient empirical justification.     

DEFINING “RELIGION” AS NATURAL: A CRITICAL INVITATION TO ROBERT McCAULEY

Previous critics have argued that Robert McCauley defines religion and science selectively and arbitrarily, cutting them to fit his model in Why Religion Is Natural and Science Is Not. McCauley has responded that final definitions are “overrated” and that artificial distinctions can serve an important role in naturalistic investigation. I agree with this position but argue that a genealogy of the category of religion is crucial to the methodology that McCauley describes. Since the inherent ambiguity of religion will undermine any essential claims about its cognitive naturalness, I invite McCauley to consider how his research might investigate scientific and religious cognition in new terms.     

EXPLANATORY MODESTY

Although I certainly have differences with some of my commentators, I am grateful for the time, effort, and attention that each has devoted to my book, Why Religion Is Natural and Science Is Not. They have helpfully pointed out features of my positions that need clarification and elaboration. I am also grateful to the editor of Zygon, Willem Drees, for this opportunity to undertake that task here.