Dynamic Structure in Diagnostic Structured Interviewing: A Comparative Test of Accuracy and Efficiency

The incorporation of Bayesian logic into diagnostic interviewing may assist with empirically based diagnostic assessment strategies in practice settings, balancing cost effectiveness, administration demands, and accuracy, yet few demonstrations of such a system have been undertaken in the context of mental health diagnosis. The present study represented an initial feasibility demonstration of whether a simplified Bayesian approach offered comparative advantages in interview accuracy and efficiency against a standard assessment procedure. Two different diagnostic algorithms were compared targeting three selected diagnoses: generalized anxiety disorder (GAD), major depressive disorder (MDD), and social phobia (SP). The first algorithm was from a standard semi-structured diagnostic interview, and the second was from a dynamic system using diagnostic base rate information to select interview content. The dynamic algorithm reduced administration time and uniformly matched or improved accuracy over standard procedures. Preparation of this article was supported in part by National Institute of Mental Health Grant R03 MH60134, an award from the University of Hawai‘i Research Council, and awards from the Hawaii Departments of Health and Education to the first author.     

Some Factors Explaining the Reliability of a Structured Interview System at a Work Site

Recent meta-analytic research has demonstrated structured interviewing to hold acceptable validity and reliability. While the emphasis has been on refining psychometric properties, there is sufficient evidence to suggest a discrepancy between the manner in which interviewing systems should be used and how they are actually used. The present study examined the use of a commercially available structured interviewing system based on past behaviour. 112 candidates were interviewed on two separate occasions by 28 interviewers. Inter-rater reliability was 0.55. The system required the derivation of a consensus score which was found not to differ significantly from the arithmetical mean of the original scores suggesting the process was not undertaken as required. Follow up discussions with interviewers reported three main areas of misunderstanding; lack of role clarity, different interpretations of job specification and inconsistent use of the rating system. Data also suggested interviewers were inexperienced. A critical finding involved female interviewers making assumptions about female applicant's motivations and suitability to the position. This finding was explained by employing ‘person-in-job’ prototypes. The data support the conclusion that although structured interviews may contain appropriate psychometric properties, the application of the system is critical.     

Selection of gender‐incongruent applicants: No gender bias with structured interviews

Research suggests that the use of structured interviews can reduce gender bias in hiring. However, studies have been limited to gender‐neutral professions or laboratory simulations. The current study evaluated two components of the structured interview in a field sample of 691 applicants interviewing for a male‐dominated position of public transit operator. Multilevel modeling results show no significant differences in ratings across applicant gender with this highly structured interview. This relation was found in individual interviewer ratings and consensus panel ratings, as well as irrespective of interviewer gender and interviewer participation in comprehensive versus minimal training in structured interviewing. This study provided a conservative test in a male‐dominated profession to further validate the value of the structured interview for promoting equal hiring practices.     

The interviewing computer: A technology for gathering comprehensive treatment information

The interviewing computer is a practical device for gathering comprehensive information on mental health patients. The complex decisions of a treatment science will require a large body of patient facts and an automated interviewing system to elicit such facts, covering the patient’s intake assessment, therapeutic progress, and follow-up stages of treatment. Evidence suggests that the computer interview may equal or excel the interview accomplishments of some clinicians who are unknowledgeable about asking certain clinical questions, forget to ask these questions, or fail to record crucial patient facts. The documentation of each patient’s treatment episode is essential for a treatment science, and the computer interview may represent a technology for achieving this information goal.     

Wright,Okasha and Chandler on transmission failure

Crispin Wright has given an explanation of how a first time warrant can fall short of transmitting across a known entailment. Formal epistemologists have struggled to turn Wright’s informal explanation into cogent Bayesian reasoning. In this paper, I analyse two Bayesian models of Wright’s account respectively proposed by Samir Okasha and Jake Chandler. I argue that both formalizations are unsatisfactory for different reasons, and I lay down a third Bayesian model that appears to me to capture the valid kernel of Wright’s explanation. After this, I consider a recent development in Wright’s account of transmission failure. Wright suggests that his condition sufficient for transmission failure of first time warrant also suffices for transmission failure of supplementary warrant. I propose an interpretation of Wright’s suggestion that shields it from objections. I then lay down a fourth Bayesian framework that provides a simplified model of the unified explanation of transmission failure envisaged by Wright.     

Entailment with Near Surety of Scaled Assertions of High Conditional Probability

An assertion of high conditional probability or, more briefly, an HCP assertion is a statement of the type: The conditional probability of B given A is close to one. The goal of this paper is to construct logics of HCP assertions whose conclusions are highly likely to be correct rather than certain to be correct. Such logics would allow useful conclusions to be drawn when the premises are not strong enough to allow conclusions to be reached with certainty. This goal is achieved by taking Adams" (1966) logic, changing its intended application from conditionals to HCP assertions, and then weakening its criterion for entailment. According to the weakened entailment criterion, called the Criterion of Near Surety and which may be loosely interpreted as a Bayesian criterion, a conclusion is entailed if and only if nearly every model of the premises is a model of the conclusion. The resulting logic, called NSL, is nonmonotonic. Entailment in this logic, although not as strict as entailment in Adams" logic, is more strict than entailment in the propositional logic of material conditionals. Next, NSL was modified by requiring that each HCP assertion be scaled; this means that to each HCP assertion was associated a bound on the deviation from 1 of the conditional probability that is the subject of the assertion. Scaling of HCP assertions is useful for breaking entailment deadlocks. For example, it it is known that the conditional probabilities of C given A and of ¬ C given B are both close to one but the bound on the former"s deviation from 1 is much smaller than the latter"s, then it may be concluded that in all likelihood the conditional probability of C given A B is close to one. The resulting logic, called NSL-S, is also nonmonotonic. Despite great differences in their definitions of entailment, entailment in NSL is equivalent to Lehmann and Magidor"s rational closure and, disregarding minor differences concerning which premise sets are considered consistent, entailment in NSL-S is equivalent to entailment in Goldszmidt and Pearl"s System-Z ⁺. Bacchus, Grove, Halpern, and Koller proposed two methods of developing a predicate calculus based on the Criterion of Near Surety. In their random-structures method, which assumed a prior distribution similar to that of NSL, it appears possible to define an entailment relation equivalent to that of NSL. In their random-worlds method, which assumed a prior distribution dramatically different from that of NSL, it is known that the entailment relation is different from that of NSL.     

Systems for multivariate monitoring of behavioral status over time

Decision-theoretic criteria are presented for optimizing the information gathered from a series of interviews over time. It is shown that the optimum interviewing strategy depends strongly on assumptions about the covariation of behavior over time. Standard interviewing strategies, including the major-problem/target-complaints approach, are optimal only under extreme assumptions about behavior. An interviewing strategy based on dynamic programming is presented that will provide optimal information return from a series of interviews under assumptions that are realistic for mental health applications. A system using this approach can tailor its interviewing strategy to adapt to differences in interview content, item importance, and individual response patterns, selecting the optimally informative questions to ask each subject at each point in time. Simulation results show that this approach achieves a 34% reduction in the false negatives obtained with the major-problem/target-complaints method, and, depending on the acceptable error rate, a reduction of 47 % or more in the questions that are needed in standard interviewing.     

Popular or unpopular? Therapists' use of structured interviews and their estimation of patient acceptance

An accurate diagnosis is an important precondition for effective psychotherapeutic treatment. The use of structured interviews provides the gold standard for reliable diagnosis. Suppiger et al. (2009) showed that structured interviews have a high acceptance among patients. On a scale from 0 (not at all satisfied) to 100 (totally satisfied) patients rated overall satisfaction with a structured interview at M = 86.55. Nevertheless, therapists rarely seem to use structured interviews in clinical practice. The aim of this study was to assess how frequently therapists use structured interviews in daily practice. Secondly, we hypothesized that therapists underestimate patient acceptance of structured interviews. As a third goal, we explored further reasons why therapists choose not to use structured interviews. We conducted an online survey of 1,927 psychiatrists and psychotherapists in Switzerland and asked them how frequently they used structured interviews and how they estimated patient satisfaction with these interviews. Furthermore, we asked therapists why they chose to use or not use structured interviews. Therapists reported using structured interviews on average with about 15% of their patients. Furthermore, therapists estimated significantly lower patient acceptance than patients themselves indicated (M_therapist = 49.41, M_patient = 86.55). Our data suggest lack of familiarity with these instruments as well as an overestimation of the utility of open clinical interviews as further reasons for not using structured interviews.     

Development of Standard Anchors for Scoring Generic Past-Behaviour Questions in Structured Interviews

A structured interviewing system was developed which included an item bank of pre-written past-behaviour questions. Scoring anchors for each of 168 questions in the item bank were created from analysis of protocols from individual assessments. A concurrent validation study was conducted, focusing on the job of highway patrol sergeant. Job knowledge experts used a job analysis tool to select a subset of twelve questions from the item bank. Panel interviews were conducted with 32 highway patrol sergeants as subjects. Average inter-rater reliability was 0.65 and interview validity was 0.42 for group consensus scores and 0.44 for panel average scores. Discussion focused on practical issues affecting the interviewing system.     

Probabilities on Sentences in an Expressive Logic

Automated reasoning about uncertain knowledge has many applications. One difficulty when developing such systems is the lack of a completely satisfactory integration of logic and probability. We address this problem directly. Expressive languages like higher-order logic are ideally suited for representing and reasoning about structured knowledge. Uncertain knowledge can be modeled by using graded probabilities rather than binary truth values. The main technical problem studied in this paper is the following: Given a set of sentences, each having some probability of being true, what probability should be ascribed to other (query) sentences? A natural wish-list, among others, is that the probability distribution (i) is consistent with the knowledge base, (ii) allows for a consistent inference procedure and in particular (iii) reduces to deductive logic in the limit of probabilities being 0 and 1, (iv) allows (Bayesian) inductive reasoning and (v) learning in the limit and in particular (vi) allows confirmation of universally quantified hypotheses/sentences. We translate this wish-list into technical requirements for a prior probability and show that probabilities satisfying all our criteria exist. We also give explicit constructions and several general characterizations of probabilities that satisfy some or all of the criteria and various (counter)examples. We also derive necessary and sufficient conditions for extending beliefs about finitely many sentences to suitable probabilities over all sentences, and in particular least dogmatic or least biased ones. We conclude with a brief outlook on how the developed theory might be used and approximated in autonomous reasoning agents. Our theory is a step towards a globally consistent and empirically satisfactory unification of probability and logic.     

Bayesian generalized structured component analysis

Generalized structured component analysis (GSCA) is a component-based approach to structural equation modelling, which adopts components of observed variables as proxies for latent variables and examines directional relationships among latent and observed variables. GSCA has been extended to deal with a wider range of data types, including discrete, multilevel or intensive longitudinal data, as well as to accommodate a greater variety of complex analyses such as latent moderation analysis, the capturing of cluster-level heterogeneity, and regularized analysis. To date, however, there has been no attempt to generalize the scope of GSCA into the Bayesian framework. In this paper, a novel extension of GSCA, called BGSCA, is proposed that estimates parameters within the Bayesian framework. BGSCA can be more attractive than the original GSCA for various reasons. For example, it can infer the probability distributions of random parameters, account for error variances in the measurement model, provide additional fit measures for model assessment and comparison from the Bayesian perspectives, and incorporate external information on parameters, which may be obtainable from past research, expert opinions, subjective beliefs or knowledge on the parameters. We utilize a Markov chain Monte Carlo method, the Gibbs sampler, to update the posterior distributions for the parameters of BGSCA. We conduct a simulation study to evaluate the performance of BGSCA. We also apply BGSCA to real data to demonstrate its empirical usefulness.     

The Counselling Interview Rating Form: a teaching and evaluation tool for counsellor education

The Counselling Interview Rating Form (CIRF) was developed to evaluate trainee counsellors' use of microskills and other aspects of the counselling session. The CIRF provides a structured format for evaluation that enhances the teaching and learning of interviewing skills by supervisors and trainee counsellors respectively. Uses of the CIRF in educational and clinical settings are described.     

Deontic action logic, atomic boolean algebras and fault-tolerance

We introduce a deontic action logic and its axiomatization. This logic has some useful properties (soundness, completeness, compactness and decidability), extending the properties usually associated with such logics. Though the propositional version of the logic is quite expressive, we augment it with temporal operators, and we outline an axiomatic system for this more expressive framework. An important characteristic of this deontic action logic is that we use boolean combinators on actions, and, because of finiteness restrictions, the generated boolean algebra is atomic, which is a crucial point in proving the completeness of the axiomatic system. As our main goal is to use this logic for reasoning about fault-tolerant systems, we provide a complete example of a simple application, with an attempt at formalization of some concepts usually associated with fault-tolerance.     

The Bayesian New Statistics: Hypothesis testing,estimation, meta-analysis,and power analysis from a Bayesian perspective

In the practice of data analysis, there is a conceptual distinction between hypothesis testing, on the one hand, and estimation with quantified uncertainty on the other. Among frequentists in psychology, a shift of emphasis from hypothesis testing to estimation has been dubbed “the New Statistics” (Cumming 2014). A second conceptual distinction is between frequentist methods and Bayesian methods. Our main goal in this article is to explain how Bayesian methods achieve the goals of the New Statistics better than frequentist methods. The article reviews frequentist and Bayesian approaches to hypothesis testing and to estimation with confidence or credible intervals. The article also describes Bayesian approaches to meta-analysis, randomized controlled trials, and power analysis.     

Natural deduction in connectionist systems

The relation between logic and thought has long been controversial, but has recently influenced theorizing about the nature of mental processes in cognitive science. One prominent tradition argues that to explain the systematicity of thought we must posit syntactically structured representations inside the cognitive system which can be operated upon by structure sensitive rules similar to those employed in systems of natural deduction. I have argued elsewhere that the systematicity of human thought might better be explained as resulting from the fact that we have learned natural languages which are themselves syntactically structured. According to this view, symbols of natural language are external to the cognitive processing system and what the cognitive system must learn to do is produce and comprehend such symbols. In this paper I pursue that idea by arguing that ability in natural deduction itself may rely on pattern recognition abilities that enable us to operate on external symbols rather than encodings of rules that might be applied to internal representations. To support this suggestion, I present a series of experiments with connectionist networks that have been trained to construct simple natural deductions in sentential logic. These networks not only succeed in reconstructing the derivations on which they have been trained, but in constructing new derivations that are only similar to the ones on which they have been trained.     

Action understanding as inverse planning

Humans are adept at inferring the mental states underlying other agents’ actions, such as goals, beliefs, desires, emotions and other thoughts. We propose a computational framework based on Bayesian inverse planning for modeling human action understanding. The framework represents an intuitive theory of intentional agents’ behavior based on the principle of rationality: the expectation that agents will plan approximately rationally to achieve their goals, given their beliefs about the world. The mental states that caused an agent’s behavior are inferred by inverting this model of rational planning using Bayesian inference, integrating the likelihood of the observed actions with the prior over mental states. This approach formalizes in precise probabilistic terms the essence of previous qualitative approaches to action understanding based on an “intentional stance” [Dennett, D. C. (1987). The intentional stance. Cambridge, MA: MIT Press] or a “teleological stance” [Gergely, G., Nádasdy, Z., Csibra, G., & Biró, S. (1995). Taking the intentional stance at 12 months of age. Cognition, 56, 165-193]. In three psychophysical experiments using animated stimuli of agents moving in simple mazes, we assess how well different inverse planning models based on different goal priors can predict human goal inferences. The results provide quantitative evidence for an approximately rational inference mechanism in human goal inference within our simplified stimulus paradigm, and for the flexible nature of goal representations that human observers can adopt. We discuss the implications of our experimental results for human action understanding in real-world contexts, and suggest how our framework might be extended to capture other kinds of mental state inferences, such as inferences about beliefs, or inferring whether an entity is an intentional agent.