Assessing the distinguishability of models and the informativeness of data

A difficulty in the development and testing of psychological models is that they are typically evaluated solely on their ability to fit experimental data, with little consideration given to their ability to fit other possible data patterns. By examining how well model A fits data generated by model B, and vice versa (a technique that we call landscaping), much safer inferences can be made about the meaning of a model's fit to data. We demonstrate the landscaping technique using four models of retention and 77 historical data sets, and show how the method can be used to: (1) evaluate the distinguishability of models, (2) evaluate the informativeness of data in distinguishing between models, and (3) suggest new ways to distinguish between models. The generality of the method is demonstrated in two other research areas (information integration and categorization), and its relationship to the important notion of model complexity is discussed.     

Behavioral treatment of insomnia in older adults: an open clinical trial comparing two interventions

Fifty-five insomniacs, 60 years or above, participated in a behavioral treatment program, comparing two interventions (sleep hygiene+stimulus control vs sleep hygiene+relaxation tape). Half of the subjects were randomized to a waiting-list condition prior to treatment. No significant changes were observed during the waiting-list period. During the treatment period however, the subjects improved on several sleep parameters, and treatment gains were maintained at a 6-month follow-up. The effects of treatment were greater for nocturnal measures (e.g. sleep onset latency and total sleep time) as compared to daytime measures (e.g. life satisfaction, daytime alertness) and not-targeted behavior (medication use). There were no differences in treatment effects for the two interventions.     

Active-imaginal exposure: examination of a new behavioral treatment for cynophobia (dog phobia)

The aims of this study were to investigate exposure-based treatments for cynophobia (dog phobia) and to test a newly developed hybrid imaginal exposure treatment that we have named active imaginal exposure. The treatment introduces an in vivo coping component to imaginal exposure whereby the patient physically performs coping responses to an imagined feared stimulus. Eighty-two participants meeting DSM-IV criteria for specific phobia (animal subtype) were randomly assigned to one of three 30-min. treatments: (a) active-imaginal exposure (AI), (b) imaginal exposure alone (IE), or (c) graduated in vivo exposure (IV). Participants completed a behavioral approach test at pre, post, and four-week follow-up. Significant pre- to posttreatment improvement was observed in all three treatment conditions. Response rates at posttreatment were 51.9, 62.1, and 73.1% for the IE, AI, and IV groups respectively. Likewise, effect sizes at posttreatment were 0.76, 1.41, and 1.55 for the IE, AI, and IV groups respectively. Although in the predicted direction, the between group differences were not significant. A similar pattern of results was observed at follow-up. Further, safety behavior utilization during treatment was associated with less improvement--particularly in the two imaginal treatment conditions. Exposure treatments of dog phobia appear feasible and effective in reducing phobic fear and avoidance associated with dog phobia. Furthermore, preliminary evidence suggests that our active-imaginal exposure treatment may be a viable alternative to in vivo exposure.     

Influence of consonantal context on the pronunciation of vowels: a comparison of human readers and computational models

In two experiments, we found that college students' pronunciations of vowels in nonwords are influenced both by preceding and following consonants. The predominance of rimes in previous studies of reading does not appear to arise because readers are unable to pick up associations that cross the onset-rime boundary, but rather because English has relatively few such associations. Comparisons between people's vowel pronunciations and those produced by various computational models of reading showed that no model provided a good account of human performance on nonwords for which the vowel shows contextual conditioning. Possible directions for improved models are suggested.     

Training parents to implement pediatric feeding protocols

Four different multicomponent training packages were evaluated to increase the treatment integrity of parents implementing pediatric feeding protocols. In Study 1 we exposed 3 parents to a training package that consisted of written protocols (baseline), verbal instructions, therapist modeling, and rehearsal training. Results suggested that the package was successful in increasing treatment integrity of the feeding protocols to high levels. Study 2 investigated three different parent-training packages comprised of components used in Study 1. Two parents were exposed to written protocols, verbal instructions, and modeling; 2 parents were exposed to written protocols, verbal instructions, and rehearsal; and 2 parents were exposed to written protocols and verbal instructions. Results of Study 2 showed that each parent-training package produced very high treatment integrity. Follow-up data in the clinic and home for 5 participants suggested that the results were durable for up to 3 months. These results demonstrate a first step in the transfer and application of research findings into routine clinical practice because we evaluated several methods for training parents to implement behavioral feeding protocols, and we demonstrated that these methods resulted in high levels of treatment integrity in a controlled clinical setting.     

Sequential evaluation of behavioral treatments and methylphenidate dosage for children with attention deficit hyperactivity disorder

We used a sequential approach to evaluate the relative and combined effects of different types of behavioral treatments, as well as dosage of methylphenidate (MPH), on the disruptive behavior of 3 students who had been diagnosed with attention deficit hyperactivity disorder. Results showed that individualized behavioral treatments produced decreases in disruptive behavior equivalent to MPH for all 3 participants and demonstrated the need to evaluate behavioral treatments and medication dosage on an individual basis.     

Remarks on the identifiability of thurstonian paired comparison models under multiple judgment

It is well-known that the representations of the Thurstonian models for difference judgment data are not unique. It has been shown that equivalence classes can be formed to provide a more meaningful partition of the covariance structures of the Thurstonian ranking models. In this paper, we examine the equivalence relations between Thurstonian covariance structure models for paired comparison data obtained under multiple judgment and discuss their implications on the general identification constraints and methods to check for parameter identifiability in restricted models.The author is indebted to Ulf Böckenholt and Albert Maydeu-Olivares for their significant comments and suggestions which led to considerable improvement in this article.     

Global and local covert visual attention: Evidence from a bayesian hidden markov model

Psychological, psychophysical and physiological research indicates that people switch between two covert attention states, local and global attention, while visually exploring complex scenes. The focus in the local attention state is on specific aspects and details of the scene, and on examining its content with greater visual detail. The focus in the global attention state is on exploring the informative and perceptually salient areas of the scene, and possibly on integrating the information contained therein. The existence of these two visual attention states, their relative prevalence and sequence in time has remained empirically untested to date. To fill this gap, we develop a psychometric model of visual covert attention that extends recent work on hidden Markov models, and we test it using eye-movement data. The model aims to describe the observed time series of saccades typically collected in eye-movement research by assuming a latent Markov process, indicative of the brain switching between global and local covert attention. We allow subjects to be in either state while exploring a stimulus visually, and to switch between them an arbitrary number of times. We relax the no-memory-property of the Markov chain. The model that we develop is estimated with MCMC methodology and calibrated on eye-movement data collected in a study of consumers' attention to print advertisements in magazines.We thank Dominique Claessens of Verify International for making the data available to us.     

What do computer models of cognition explain: a reply to Gobet

The differences between Gobet’s views and ours call attention to some points concerning the argumentative status of computational models. If we have two fundamentally different models which reasonably accurately simulate a phenomenon, we must ask, what is the argumentative status of the models in psychological terms. Moreover, if it is possible to present different models of same phenomena, what is the general argumentative power of models in psychology? As the kind of differences between our views and the ones of Gobet’s are common in modelling, we briefly call attention to these foundational issues in our paper.