No description available.

Feedback contextualized to curriculum content and misconceptions is a crucial piece in any learning experience. However, looking through student code and giving feedback requires more time and resources than an instructor typically has available, delaying feedback delivery. Intelligent Tutors for teaching Programming (ITPs) are designed to immediately deliver contextualized feedback of high quality to several students. However, they take significant effort and expertise to develop courses and practice problems, making them difficult to adapt to new situations. Because of this, the most frequently used feedback techniques for immediate feedback systems focus on highlighting incorrect output or pointing out errors in student code. These systems allow for quick development of practice problems and are easily adaptable to new contexts, however, the feedback isn't contextualized to curriculum content and misconceptions. This dissertation explores the implications of the Misconception-Driven Student Model (MDSM) as a model for developing alternatives to the aforementioned methods. I explore the implications and impact of MDSM with relation to feedback through the following thesis: Authoring feedback using a cognitive student model supports student learning of programming. In this dissertation I review relevant cognitive theory and feedback systems and two quasi-experimental studies examining the efficacy of MDSM. / Doctor of Philosophy / Feedback contextualized to curriculum content and misconceptions is a crucial piece in any learning experience. However, looking through student code and giving feedback requires more time and resources than an instructor typically has available, delaying feedback delivery. Intelligent Tutors for teaching Programming (ITPs) are designed to immediately deliver contextualized feedback of high quality to several students. However, they take significant effort and expertise to develop courses and practice problems, making them difficult to adapt to new situations. Because of this, the most frequently used feedback techniques for immediate feedback systems focus on highlighting incorrect output or pointing out errors in student code. These systems allow for quick development of practice problems and are easily adaptable to new contexts, however, the feedback isn't contextualized to curriculum content and misconceptions. This dissertation explores the implications of the Misconception-Driven Student Model (MDSM) as a model for developing alternatives to the aforementioned methods. I explore the implications and impact of MDSM with relation to feedback through the following thesis: Authoring feedback using a cognitive student model supports student learning of programming. In this dissertation I review relevant cognitive theory and feedback systems and two quasi-experimental studies examining the efficacy of MDSM.

The general purpose of this study was to consider the relationship between an individual's perception of his parent-child experiences and the adequacy of a naive cognitive model to describe his judgements of others.

A series of studies were conducted with the aim of showing that the Representational Redescription (RR) model (Karmiloff-Smith, 1992) can be used a general model of cognitive development. In this thesis, 3 aspects of the RR model were explored. The first set of experiments involved analysing the generalisability of RR levels across tasks in a domain. In an initial study, the levels of the RR model were successfully applied to a balance scale task. Then, in a subsequent study, children’s RR levels on the balance scale task were compared with their RR levels on a balance beam task (see Pine et al, 1999). Children were seen to access the same level of verbal knowledge across both tasks. This suggests that it is verbal knowledge which provides the basis for generalisation of knowledge. The second set of experiments involved a consideration of the RR model in relation to the domain of numeracy. The levels of the RR model were applied to children’s developing representations for the one-to-one and cardinality principles. The RR levels were shown to have utility in predicting children’s openness to different types of “procedurally based” and “conceptually based” teaching interventions, with pre-implicit children benefiting from procedural interventions, and children who have implicit and more advanced representational levels benefiting from conceptual interventions. The final study involved a microgenetic analysis of children’s representational levels on the balance beam task. The findings from this study indicated the importance of a period of stability prior to a cognitive advance, and demonstrated that cognitive advances can be driven by changes in the verbal explanations that are offered, rather than changes in successful performance. This provides support for the mechanism of change proposed by Karmiloff-Smith, 1992. Together, the findings indicate that the RR model provides a useful perspective about the cognitive development of children. In particular, the thesis highlights when children can use the same representations for different tasks in a domain and suggests the mechanism that brings about representational change.

No description available.

Cooper and Shallice (2000) implemented a computational version of the Norman and Shallice’s (1986) Contention Scheduling Model (CSM). The CSM is a hierarchically organized network of action schemas and goals. Botvinick and Plaut (2004) instead took a connectionist approach to modeling routine procedural behavior. They argued in favor of holistic, distributed representation of learned step co-occurrence associations. Two experiments found that people can adapt routine procedural behavior to changing circumstances quite readily and that other factors besides statistical co-occurrence can have influence on action selection. A CSM-inspired ACT-R model of the two experiments is the first to postdict differential error rates across multiple between-subjects conditions and trial types. Results from the behavioral and modeling studies favor a CSM-like theory of human routine procedural memory that uses discrete, hierarchically-organized goal and action representations that are adaptable to new but similar procedures. / Office of Naval Research grants #N00014-03-1-0094 and #N00014-06-1-0056

Cognitive control refers to the brain's ability to control attention and other cognitive functions in the service of intention-driven behaviors; moreover, it is an essential aspect in cognition. Cognitive control is commonly evaluated through the so-called conflict adaptation effect, which is revealed through trial-to-trial changes in performance after experiencing cognitive conflict. The conflict monitoring theory is a prominent theory intended to describe conflict adaptation and explain how cognitive control might be engaged in the face of conflict. A passive carryover account, originally aimed at explaining task switching and based on parallel distributed processing models, may represent a superior alternative to the conflict monitoring account. In the carryover account, passive inertia of the cognitive control state explains the trial-to-trial modulation of conflict effects. One problem with conflict adaptation is that the typical paradigms used to create conflict adaptation often include trial-to-trial repetitions that mimic the same performance pattern through priming. Conflict monitoring theory also has difficulty explaining other issues such as whether conflict adaptation is task-specific. A meta-analysis of the so-called conflict adaptation effect suggests trial-to-trial repetitions do not entirely account for conflict adaptation effects, but these effects do appear to be task-specific. The meta-analysis also suggests the withdrawal of control is episodic rather than temporal and conflict adaptation may be sensitive to experimental session length. A novel eye-tracking paradigm addresses the timing of control engagement. The results suggest dynamic regulation of attention coupled with conflict detection. This more tightly coupled detection and regulation process in the context of conflict adaptation is more consistent with a carryover account than what would be expected from the conflict monitoring theory. A subsequent eye-tracking paradigm further supported that trial-to-trial modulations of control are generally more consistent with a carryover account than they are with the conflict monitoring theory. Overall, the empirical evidence points toward a carryover model of cognitive control. / 2016-10-17

Indiana University-Purdue University Indianapolis (IUPUI) / In accord with the cognitive model of poor functioning in schizophrenia, defeatist performance beliefs, or overgeneralized negative beliefs about one’s ability to perform tasks, have been linked to poor functional outcomes, cognitive impairment, and negative symptoms in schizophrenia and are a suggested therapeutic target in Cognitive Therapy for Schizophrenia. However, there is a paucity of research investigating these beliefs in schizotypy, or those exhibiting traits reflecting a putative genetic liability for schizophrenia. This study had three aims: to examine whether defeatist performance beliefs 1) are elevated in schizotypy compared to non-schizotypy, 2) are associated with functioning-related outcomes (i.e., quality of life, working memory, negative schizotypy traits), and 3) mediate the relationships between working memory and both negative schizotypy traits and quality of life. Schizotypy (n = 43) and non-schizotypy (n = 45) groups completed measures of schizotypy traits, defeatist performance beliefs, quality of life, and working memory. Results revealed that the schizotypy group reported significantly more defeatist performance beliefs than the non-schizotypy group. Within the schizotypy group, defeatist performance beliefs were significantly positively associated with negative schizotypy traits and significantly inversely associated with quality of life. No associations were observed between defeatist performance beliefs and positive schizotypy traits and working memory. Further, defeatist performance beliefs did not mediate the relationships between working memory and either quality of life or negative schizotypy traits. Findings are generally consistent with the cognitive model of poor functioning in schizophrenia and suggest that defeatist performance beliefs may be an important therapeutic target in early intervention services.

Understanding the process of memory formation in neural systems is of great interest in the field of neuroscience. Valiant’s Neuroidal Model poses a plausible theory for how memories are created within a computational context. Previously, the algorithm JOIN has been used to show how the brain could perform conjunctive and disjunctive coding to store memories. A limitation of JOIN is that it does not consider the coding of temporal information in a meaningful manner. We propose SeqMem, a similar algorithmic primitive that is designed to encode a series of items within a random graph model. We investigate the feasibility of SeqMem empirically by observing its stability and effects on capacity in our model. We intend to provide value in the use of SeqMem and similar procedures to further develop a neurobiologically plausible theory of mind. Our goal here is to inspire further work in scaling our methods to function at a human-level magnitude of computation.

The current coginitive models of the Obsessive-Compulsive Disorder (OCD) symptoms focuses on the different cognitive factors. Like other nonspecific and noncognitive variables, these factors may also function as vulnerability factors. However, they have been mostly studied separately and majority of the findings in the literature come from the Western samples. Accordingly, the studies examining these factors together and the impact of the culture in these studies are sparse in number. The present study suggested a comprehensive cognitive model for OCD symptoms, including several distal and proximal vulnerability factors. It was aimed to adapt three instruments to examine the interrelationships among the vulnerability factors and OCD symptoms in different cultures. Relevant ten instruments were administered to the university students from Turkey and Canada. The analyses showed that Turkish versions of three instruments had satisfactory psychometric properties for Turkish students. These analyses also revealed some cross-cultural similarities and differences in these factors and OCD symptoms. Neuroticism, age, introversion, OCD beliefs on responsibility/threat estimation, perfectionism/certainty and thought-action fusion in likelihood dimension were found to be associated with the OCD symptoms in both Turkish and Canadian samples. The relational paths between non-specific, appraisal and control factors, and OCD symptoms were also significant in both samples. However, religiousness was only significant factor in OCD symptoms and contributed to several belief and control factors toward these symptoms, only for Turkish subjects. The analyses of the religiousness differences indicated that psychological fusion in general and in morality was more related to the religiosity for Canadian Christians. Besides, Turkish students seemed to utilize worry more for OCD symptoms / whereas, Canadian participants used self-punishment. These common and unique patterns of the relationships were discussed within relevant findings about characteristics of the religion and culture.