A virtual spatial navigation task for multisensory discrimination

Bexter, Alexander; Kampa, Björn M. (Thesis advisor); Spehr, Marc (Thesis advisor)

Aachen : RWTH Aachen University (2021, 2022)
Dissertation / PhD Thesis

Dissertation, RWTH Aachen University, 2021

Abstract

This study’s results help to explain the effects of multisensory visuotactile stimulation on animal behavior. The presented powerful task can capture complex behavioral aspects, while simultaneously accessing brain activity with imaging methods. Mice were head-fixed and placed in a very simple virtual visuotactile corridor in which they were free to run or moved automatically. They encountered different numbers of either visual, tactile or combined visuotactile events on the two walls of the tunnel and had to discriminate between those two and find the side with more events/higher frequency. Visual events were white bars of 1.2 cm width on a black background and tactile events were 40 ms short air puffs to the whiskers of the mouse. After reaching the end of the corridor, the mouse had to give an answer on either the left or the right side, according to the correct side (higher frequency), to obtain a reward. Mice were able to successfully discriminate between target-distractor differences of up to two events in the multisensory condition. Pure visual and tactile task performance was inferior to the visuotactile task performance, which means that there was an increase in performance when congruent visual and tactile information was present at the same time. The psychometric curves are shifted for the unisensory conditions, with higher differences between target and distractor stimulus as discrimination thresholds. The multisensory effect was biggest for medium task difficulty, represented by medium distractors. This means, that mice in a very hard or very easy task did not profit as much from the presence of multiple modalities. Running activity on the wheel had only slight effects on performance. While we found a clear increase in performance with running activity in a pure visual sinusoidal grating discrimination task, this effect is ambiguous in the multisensory experiment. Although there seems to be a trend, the difference in performance is not significant and further testing has to be done to find a possible effect. When we exposed the mice with a modality conflict task, where they were free to choose between equal visual or tactile targets, we found that some animals clearly preferred either visual or tactile cues. In line with these preferences, an analysis of the performance during sessions revealed that animals followed different strategies of relying more on one modality than on the other. This often also changes during a session, whilesome animals were more likely to switch their strategy than others. With the experimental setup shown here, it is possible to get a meaningful read out of animal behavior, which can be easily combined with imaging of neuronal activity, for example with calcium imaging under a 2-photon microscope or widefield macroscope. Those techniques have been successfully tested in the primary visual cortex and higher visual rostrolateral area RL.

Institutions

• Department of Biology [160000]
• Department of Molecular and Systemic Neurophysiology [162320]