MindTalks

...gemeinschaftlich organisiert von:

Dr. Udo Ernst (Fachbereich 1, Computational Neurophysics Lab)

Dr. Detlef Wegener (Fachbereich 2, Cognitive Neurophysiology)

Prof. Dr. Tanja Schultz (Fachbereich 3, Cognitive Systems Lab)

Im Sommersemester 2025 finden alle Vortr?ge im IW3 (Geb?ude neben Cognium), Raum 0330 statt. 

28.04.2025 | 16:15 – 17:30

IW3 Raum 0330, Am Biologischen Garten 2 

Building a theory of sensory coding for active behavior

Prof. Dr. Wiktor Mlynarski

Sensory systems are the brain's window to the world - they represent the organism's surrounding in order to enable successful action. To instantiate such representations efficiently and accurately, the brain must adapt to the structure of natural environments. However, nothing in the natural world is completely static - environments change, animals' goals and demands fluctuate, and the two are coupled in a closed-loop - the surrounding can be affected by animal's actions. In this talk I will discuss our theoretical attempts to understand how and why should sensory systems adapt to changing environments, varying internal states and dynamic behavior in natural environments.

12.05.2025 | 16:15 – 17:30

IW3 Raum 0330, Am Biologischen Garten 2 

AI for Neuroscience

Prof. Dr. Nergis T?men

Machine learning and neuroscience have a profoundly interconnected history with modern deep learning having its roots in neurophysiology. In the last couple of decades, breakthroughs in artificial intelligence (AI) and neuroscience have advanced our understanding of both domains and unveiled exciting new research opportunities at their intersection. Currently, AI can help support neuroscientific research through various approaches. My talk focuses on recent examples of AI applications in neuroscience, as well as my recent work on using deep neural networks for automating and speeding up optical voltage imaging and superresolution microscopy.

19.05.2025 | 16:15 – 17:30 

IW3 Raum 0330, Am Biologischen Garten 2 

The impact of age-related hearing loss on brain structure and function

Prof. Dr. Christiane Thiel

The presentation will provide an overview of our neuroimaging studies involving elderly volunteers with uncompensated age-related hearing loss. I will demonstrate that hearing loss enhances audiovisual integration but does not trigger cross-modal responses in the auditory cortex. Instead, we found increases in the functional connectivity of the auditory cortex to visual, parietal, and frontal regions during task performance. At rest, this functional connectivity was reduced and correlated with the daily listening effort experienced. Changes in gray matter were also more closely associated with listening effort than with hearing loss itself. Furthermore, I will present findings from several of our neuroimaging studies that did not find evidence of hearing loss affecting brain structure and function, neural activity in working memory tasks, microstructural brain integrity as indicated by diffusion tensor imaging, or signs of increased brain aging in morphometric analyses. Therefore, I propose that the effects of uncompensated hearing loss in healthy, well-educated elderly subjects are subtle, primarily manifesting in changes in functional connectivity and are more related to the daily listening effort than to the hearing loss itself.

02.06.2025 | 16:15 – 17:30

IW3 Raum 0330, Am Biologischen Garten 2 

Adaptivity in deep neural networks: the long road to lifelong learning

Prof. Dr. Martin Mundt

Deep neural networks (DNNs) are often framed as being inspired by the brain. Although their advent has led to remarkable success stories and a plethora of large-scale applications, they however lack a key ability humans posses: the capacity for lifelong learning. That is, DNNs are primarily successful when trained on predefined training and dedicated test datasets. When faced with the breath of novelty and changing experiences in the real world, they become prone to erratic predictions and suffer from catastrophic forgetting. In contrast to the brain’s remarkable adaptation capabilities, this leads to a current practice of frequent and unsustainable re-training. In this talk, I will introduce the elements necessary to shift such prevalent static design towards lifelong machine learning systems; systems that transcend stationary datasets and continually learn in a world full of unknowns. I will outline mechanisms that augment DNNs with the ability to dynamically adapt their structure, robustly deal with novel situations, and efficiently incorporate new knowledge over time. Finally, I will conclude on why these newfound capabilities are but a few of the exciting avenues towards lifelong neural networks, drawing brief parallels to the host of mechanisms posited to contribute to lifelong learning in the brain. 
 

30.06.2025 | 16:15 – 17:30

IW3 Raum 0330, Am Biologischen Garten 2 

Visual distraction – event-related-potential markers and the dimension weighting account

Dr. Heinrich Liesefeld

Despite the vivid subjective impression of a rich sensory world, people actually process only little information from their environment at a time. Which aspects of the world are processed under which conditions is studied in research on selective attention. Ideally, objects would be prioritized based on their relevance to current behavioral goals. But goals are not the only or strongest influence on selective attention, so that attention is often misdirected and irrelevant objects (distractors) receive prioritized processing. Such “failures” of selective attention are subjectively perceived as distraction; they prevent or delay processing of the concurrently presented goal-relevant objects (targets). The dimension weighting account is a long-standing theory of the various influences on priority computations and selective attention from which I derive unique predictions regarding the circumstances under which distraction occurs. As attention allocations cannot be observed directly, I make use of event-related potentials extracted from the electroencephalogram to test these predictions. The N2pc and PD components are particularly useful markers of attentional enhancement and suppression, respectively. With these tools, I show that distractors standing out in the same visual dimension as search targets capture attention, so that the target can be attended only subsequently. By contrast, distractors standing out in a different dimension are suppressed before they can capture attention, so that target processing is only slightly delayed. The aim of this talk is to convince the audience of the theoretical utility of the dimension weighting account and the empirical utility of the employed event-related-potential components.

07.07.2025 –16:15 – 17:30

IW3 Raum 0330, Am Biologischen Garten 2 

From cryptic distortions to distortional cryptography: is there a mosquito in my brain?

Prof. Dr. J?rg Albert

Distortion of signals occurs as a disturbance in most human-made electronics, or human communication at large. At least so goes the general view. Accordingly, human engineers usually take a great effort – and a great pride – in reducing distortions to minimal levels. However, the engineers of evolution may beg to differ. Distortions in the auditory system are essential (that means inevitable) consequences of vital system parameters. In a nutshell, distortions are direct results of the most elementary process of hearing, i.e. of auditory transduction, which comprises the conversion of sound into neural signals. Our ears, one might say, generate distortions as cryptic ‘phantom tones’ of the hearing process. Mathematically speaking, those distortion tones are highly periodic, highly predictable and as I will suggest, highly precise. Using hearing in male malaria mosquitoes, I will try to show the inevitability, and the benefits, of essential distortions not only for mosquito hearing but possibly for neural function at large. Maybe, it turns out that in the end our brain is just a large swarm of mosquitoes, maybe the largest of them all. 

Die Vortr?ge werden gr??tenteils auf Englisch gehalten (siehe Titel).

Für aktuelle Benachrichtigungen k?nnen Sie sich gerne in unsere Mailingliste eintragen lassen. Kontaktieren Sie dazu bitte Agnes Jan?en: ajanssen@neuro.uni-bremen.de