Institutskolloquium Herbstsemester 2018

Donnerstag, 17.15 - 18.00 Uhr
Fabrikstrasse 8, Raum B 306
mit anschliessender Diskussion und Apéro


Dr. Francesca Siclari: " The dreaming brain "
Centre d'investigation et de recherche sur le sommeil, Lausanne


Prof. Sonja Vogt: " The Seduction of Applied Conformity: Delimited Approaches to Ending Female Genital Cutting "
Soziologie, Universität Bern


Prof. Frank Jäkel: " Categorization in Man and Machine "
Technische Universität Darmstadt


Prof. Clara Hill: " Advice-Giving: Is it Advised "
University of Maryland


Francesca Siclari, MD: The dreaming brain

Attending physician, Center for Investigation and Research on Sleep, Lausanne University Hospital, Lausanne, Switzerland

Dreaming is a form of consciousness that occurs during sleep, while we are functionally disconnected from the environment. Traditionally, dreaming has been linked to REM sleep, a behavioral state characterized by fast, desynchronized electroencephalographic activity similar to wakefulness. In recent years however, it has become clear that dreaming can also occur in NREM sleep, in which EEG activity is dominated by slow waves and spindles. This has challenged the understanding of the neural correlates of conscious experiences in sleep. In the present talk I will present a series of studies in which we investigated the EEG correlates of dreaming using a serial awakening paradigm and high-density EEG recordings. More specifically, I will show how local EEG features, including spectral power in different frequency bands, slow waves and spindles relate to the presence and absence of dreaming, and to specific dream contents. These results suggest that local EEG correlates may account for the presence of conscious experiences in behavioral states with radically different global EEG signatures.

Prof. Dr. Sonja Vogt: The Seduction of Applied Conformity: Delimited Approaches to Ending Female Genital Cutting

For a policy maker attempting to reverse a harmful tradition like female genital cutting, childhood smoking, or binge drinking, social influence represents a compelling mechanism.  If an intervention leads one individual to change to a beneficial alternative, this change in behaviour may influence others to follow suit without additional interference from the policy maker.  Accordingly, a key objective is to mobilise social influence to induce these spillovers, and this idea has shaped influential approaches to reversing harmful traditions in various domains, with the abandonment of female genital cutting standing as a canonical example.  We present data from two studies in Sudan that question the basis of relying on social influence to accelerate the abandonment of female genital cutting.  We follow by developing a framework to examine the scope for beneficial spillovers and find that they can be entirely absent even if social influence pervades individual decision making.  Our analysis highlights three critical considerations.  First, if an intervention targets agents amenable to change, social influence typically produces little or no spillovers.  Second, targeting agents resistant to change tends to maximise spillovers, but only if resistant individuals respond to a policy maker's intervention and social networks are not overly homophilous.  Finally, if some groups use the harmful behaviour to distinguish themselves from other groups, social influence takes a xenophobic form.  Xenophobia can severely limit spillovers, and in such cases the policy maker should attempt to break the link between the harmful behaviour and out-group derogation.

Prof. Dr. Frank Jäkel: Categorization in Man and Machine: Bayesian Reverse-Engineering of Perception and Cognition

Bayesian models of cognition and behavior compare human performance to the performance of an ideal decision maker. Such models are particularly promising when they are used in reverse-engineering explanations: explanations that descend from Marr's computational level of analysis to the algorithmic and the implementational level. Unfortunately, it remains unclear exactly how Bayesian models constrain and influence these lower levels of analysis. In several examples ranging from signal detection tasks to inductive reasoning I sketch how Bayesian models are used in psychology.


Weitere Abstracts folgen.