Cognitive Science Colloquium - Winter Semester 2019 - 2020
Thursdays 15:30 to 17:00 - Building 57, Room 508
November 07, 2019 - CANCELLED
Speaker: Dr. Joana Costa do Carmo (ISCTE - Lisbon University Institute, invited by Thomas Lachmann)
Topic: Initiation processes in Autism: The role of complementary Long-range Connections of Basal Ganglia and the Cerebellum
Abstract: Long-range connections that both basal ganglia and the cerebellum have with the cerebral cortex indicate their complementary functions dedicated to motor control. Recently, the notion that basal ganglia are solely involved in motor control has been challenged and likewise, a view that stresses the role of cerebellum outside the sphere of motor control is currently acknowledged: the cognitive cerebellum. It has been shown, in a series of experiments, that initiation processes, an executive function, might be impaired in autism spectrum disorder (ASD). We aim to evaluate, in a series of behavioral and imaging studies, the functioning of basal ganglia and the cerebellum and their connections with cortical areas in adults with ASD and matched healthy participants; as well as to estimate the likely contributions of these brain structures to high-level cognition, in particular to the initiation and selection of a response and to the Turn-Taking system. We expect to further extend knowledge on the role of basal ganglia and the cerebellum in cognition and to unveil the neural bases of the Turn Taking system. Moreover, pathological alterations in ASD to the cortical-cerebellar and cortical-basal ganglia recurrent networks could be at the underpinning of the primary symptoms of the disorder.
November 28, 2019
Speaker: Dr. Moritz Schaeffer (University of Johannes-Gutenberg, Mainz - English Linguistics and Translation Studies, invited by Shanley Allen)
Topic: Predicting behavior during translation
Abstract: The current talk will review the theoretical and methodological intersections between two neighbouring disciplines regarding a task which is shared across disciplinary boundaries: translation as a task is employed both in psycholinguistics and translation studies and the theoretical implications of this task for the modeling of bilingual language processing is equally shared. However, despite the fact that there is some overlap between the approaches in these two disciplines, there are important differences in how the study of translation is carried out. The focus of this talk is on both the often stark differences, but also on attempts to bridge these. This talk will contrast theoretical considerations underlying experimental designs and the experimental designs themselves in the hope of finding future avenues which will do justice to the relevant issues raised in both disciplines. In particular, the talk will focus on semantic and syntactic aspects, how these are modeled cross-linguistically and studied experimentally - in both disciplines. It is expected that by incorporating aspects from one disciplinary approach in the other exciting new avenues of research will emerge.
December 12, 2019
Speaker: Prof. Dr. Steve Majerus (PsyNCog: Psychologie & Neuroscience of Cognition, Liège University - France, invited by Shanley Allen)
Topic: Verbal working memory and language: overlap and specificities
Abstract: There is a broad range of working memory models, with one extreme assuming a complete overlap between verbal working memory and language systems while others assume at most indirect interactions between both systems. In this talk, the different theoretical models will be confronted to behavioral, neuropsychological and neuroimaging evidence and the minimal degree of overlap between language and working memory as supported by empirical data will be defined. A special focus will be put on the question of serial order information and the linguistic versus non-linguistic processes that support retention of serial order information in working memory.
January 09, 2020 - CANCELLED
Speaker: Dr. Vivienne Rogers - (Swansea University, invited by Leigh Fernandez)
January 16, 2019
Speaker: Dr. Romy Müller (TU Dresden - invited by Thomas Lachmann)
Topic: Why psychologists should know about chocolate: A domain-centred perspective on human-machine cooperation
Abstract: In the food processing and packaging industry, the daily work of operators and technicians is characterized by a need to handle frequently occurring faults. Fault diagnosis can be supported by assistance systems that store previous experience and make it available in cooperative dialogues. However, to design such human-machine cooperation it is essential to gain a thorough understanding of the domain, because it generates the problem space in which operator actions represent possible solutions. Therefore, the lecture consists of three parts: First, to illustrate why domain specificity is important for psychologists, a comparison between the process and discrete processing industries is presented. The remainder of the lecture focuses on the latter, using the example of a chocolate packaging plant. Functional relations and typical faults are explained, and a look into the tasks and diagnostic strategies of operators and technicians is provided. These considerations generate the requirements and constraints for assistance system design. Most importantly, they generate a bundle of fascinating questions which can be taken to the lab to allow for an in-depth investigation of the processes implied in using different assistance concepts. In the second part of the lecture, one such study is presented, asking how the presentation of sensor data or its automated interpretations will affect performance, information sampling, and knowledge acquisition. Finally, the third part will sketch how we build on our experiences from the field and from the lab to develop an ontology-based, conversational case-based reasoning system for fault diagnosis in a chocolate packaging plant.
January 23, 2020
Speaker: Dr. Simone Malejka (University College London - invited by Thomas Schmidt)
Topic: Continuous Versus Discrete Models of Recognition and Source Memory
Abstract: A vital feature of the human memory system is the ability to identify information that has been encountered before. Performance in a recognition task is influenced by two latent factors: the participant’s ability to discriminate between old and new information (recognition memory) and the participant’s response tendency in case of uncertainty (response bias). To obtain process-pure measures of both, cognitive psychologists apply formal measurement models to recognition data. In the last years, the search for a suitable model culminated in an active controversy: Should recognition memory be measured using models with continuous strength variables or models with discrete cognitive states? While the prevalent opinion favors continuous-strength models (such as signal-detection theory), discrete-state models (such as the two-high-threshold model) are increasingly cast aside. However, many arguments against discrete-state models that were put forward in the literature can be directly refuted. The three remaining arguments constitute the most serious threats to discrete-state models to date. Using model-based evidence as well as simple hypothesis tests in different paradigms, I will show how these arguments can also be refuted. Taken together, discrete-state models are in no way inferior to continuous-strength models and the latter’s supremacy in recognition research seems unjustified.
January 30, 2020
Speaker: Dr. Torsten Schubert - ( Martin-Luther University Halle-Wittenberg, Department of Psychology, invited by Thomas Lachmann)
Topic: The impact of Cognitive Training on Multitasking: Behavioral and neural findings
Abstract: Previous studies showed that complex mental trainings can lead to cognitive changes which may generalize to other non-trained tasks. While several studies showed an impact of Action Video Gaming (AVG) and working memory (WM) training on non-trained single task situations, they left open how training can affect executive functions in multitasking situations. In the talk I will report the results of a series of studies aimed at disentangling the effects of AVG and WM training on executive functions during multitasking. According to the theories in learning and memory (Bjork and Schmidt, 1992), we assumed that the generalization of training effects after complex mental trainings depends on the particular overlap between the processing requirements in the transfer and the training situation. A task analysis showed that AVG involves the fast focusing of attention on selected objects and to be executed actions just as the fast reallocation of attention; on the contrary, WM training improves rather the efficiency of working memory updating processes under increasing memory load. We tested the generalization-by-overlap hypothesis with three experiments in which we investigated the impact of AVG and WM training on performance in different types of multitasking paradigms. The findings of two studies with behavioral measures showed that the transfer effects after both types of training differ and strongly depend on the type of overlap between the processes in the trained and the transfer conditions. This conclusion was further supported by the findings of an additional neuroimaging study showing that extensive WM training leads to a decrease of fMRI activation in general-purpose executive regions of the prefrontal cortex and to a strengthening of the activity in highly-specialized subcortical regions of the striatum, which are related to automatized memory updating processes applied to the same extent in the trained and transferred task. Together, these findings indicate that cognitive training may improve human multitasking and that a generalization of the training-related improvement becomes possible under conditions of overlapping processes in the trained and non-trained Situation.
February 06, 2020
Speaker: Steffen Ronft - (Deputy head of the Center for Empirical and Experimental Business Studies - Mannheim, invited by Tandra Ghose)
Topic: Event Psychology – Integrating interdisciplinary psychology with billion-dollar industry of event management
Abstract: In my talk I will discuss the implications of the psychology of perception for event management. The so-called MICE industry, an umbrella term for Meetings Incentives Conventions Exhibitions, is subject to a global billion-dollar market. Participants are indisputably the most important determinant of each of these events; therefore, human psychology is inherently omnipresent. The science of psychology can be integrated to influence the perception, memory, cognitive performance, and the experience of attending events of the event visitor. Multisensory communication makes messages more effective and integrating insights from business, emotional, and cognitive psychology enables us to improve the efficiency of event content and delivery. In light of sensory overload, increasing competition between marketing events as communication instruments and as entertainment events is an inevitable development for professional event management. The planning and prediction of visitor perceptions, emotions and memories are ultimately the central aspects of modern event planning.
February 13, 2020
Speaker: Dr. Volker Franz - (Tübingen University, invited by Thomas Schmidt)
Topic: Unconscious lie detection as an example of a widespread fallacy in the Neurosciences
Abstract: Neuroscientists frequently use a certain statistical reasoning to establish the existence of distinct neuronal processes in the brain. We show that this reasoning is flawed and that the large corresponding literature needs reconsideration. We illustrate the fallacy with a recent study that received an enormous press coverage because it concluded that humans detect deceit better if they use unconscious processes instead of conscious deliberations. The study was published under a new open-data policy that allows to reanalyze the data with more appropriate methods. We found that unconscious performance was close to chance - just as the conscious performance. This illustrates the flaws of this widely used statistical reasoning, the benefits of open-data practices, and the need for careful reconsideration of studies using the same rationale.