Search Papers | Poster Sessions | All Posters
Poster C94 in Poster Session C - Friday, August 9, 2024, 11:15 am – 1:15 pm, Johnson Ice Rink
'Reusers' and 'Unlearners' display distinct effects of forgetting on reversal learning in mice and artificial neuronal networks
Jonas Elpelt1 (), Jens-Bastian Eppler1, Johannes P.-H. Seiler2, Simon Rumpel2, Matthias Kaschube1; 1Frankfurt Insitute for Advanced Studies, 2Johannes Gutenberg University Mainz
Previous research has indicated that prior learning can be both advantageous or disadvantageous for learning related tasks. Moreover, the speed of learning a related task might be mediated by the extent of forgetting of the original task. Here, we seek to explore the role of forgetting initially learned task representations on reversal learning behavior in both mice and artificial recurrent neural networks. We trained mice to discriminate two auditory stimuli in a go/no-go paradigm. After learning, they had a pause of 2 or 16-days. In general, a shorter pause resulted in better memory retention and faster adaptation during reversal learning with reversed conditions. However, some animals did not benefit from initial learning, suggesting no reuse of initial representations. Similar patterns were observed in artificial neural networks during reversal learning, showing both beneficial reusing and disadvantageous unlearning of previously learned network configurations. Our findings shed light on the use of initial representations during reversal learning and could provide insights into cognitive flexibility in both biological and artificial neural networks.
Keywords: forgetting reversal learning cognitive flexibility recurrent neural networks