SPATIAL REPRESENTATIONS IN THE PIRIFORM CORTEX EMERGE AFTER ODOR-CONTEXT ASSOCIATIVE LEARNING

Olfactory processing depends on prior experience, present context, and animal's internal state. To investigate how learning dynamically reshapes olfactory cortical circuits, we trained mice in a virtual reality-based, context-dependent GO/NO-GO task. Mice explored virtual corridors presenting two odors across two visual contexts, learning to lick for a water reward only for one specific odor-context combination. In previous experiments, we showed that odor-responsive primary piriform cortex (PCx) neurons become mixed-selective after learning, encoding positional, contextual, and associative information. It is still unclear how this shift emerges in the PCx encoding scheme. Here, we performed PCx neuronal recordings throughout the learning process, using behavioral analysis alongside neuronal encoding and decoding models to address this open question. We show that mice learn the task sequentially: they first discriminate odors (intermediate stage) before successfully incorporating contextual information (expert stage). PCx neuronal encoding parallels this behavioral progression. In intermediate sessions, PCx neurons encode odors but lack contextual information. Conversely, expert animals show robust encoding of both parameters. Decoding analyses confirm this shift, revealing that neurons in expert animals exhibit significantly enhanced multiplexing—modulation by a greater number of task parameters—compared to early or intermediate stages. This evidence indicates that the observed contextual encoding in the PCx depends on the associative learning process in which odorant cues are combined with positional information to produce correct behavioral responses in the task. Ongoing experiments focus on the neural mechanisms responsible for making this association in the PCx.