Neural mechanisms of generalization in the ventral hippocampus

Generalization, the tendency to respond in similar ways in different situations, can be adaptive or maladaptive, depending on the context. For example, generalizing fear of a real threat to other similar dangerous contexts can be important for survival; conversely, displaying fear in safe environments is detrimental. The ventral hippocampus (VH) is in a key position to modulate valence and provide contextual information that allows for selective or generalized responses. Yet little is known about how this area modulates these behaviors. The VH has reciprocal connections with the basolateral amygdala (BLA), an area involved in fear learning and extinction (1-4), and is the main hippocampal region sending projections to the prelimbic (PL) and infralimbic (IL) cortices, areas of the medial prefrontal cortex (mPFC) involved in fear expression and extinction, respectively (5-9). Moreover, principal cells in the VH fire in large portion of the environment, displaying overlapping firing fields (10-12) and contain precise spatial information at the population level (10), two characteristics that are necessary for generalization processes. The objective of this proposal is to  determine how the VH modulates selective or generalized responses in different situations. The overarching hypothesis is that the VH modulates these responses though the activity of distinct subpopulations of projection cells with different physiological and computational properties. To determine the validity of this hypothesis, this proposal will address the following questions: 1) How do VH subpopulations respond and synchronize with target regions when animals discriminate or generalize emotional responses? 2) What is the role of different VH projection cells in generalization/discrimination and how do VH representations change when these distinct subpopulations are selectively activated or inhibited? 3) What are the computational and synaptic properties of different VH projection cells? These questions will be addressed by combining in vivo single unit and local field potential recordings along with pathway-specific chemogenetic silencing/activation (Muzzio Lab) and in vitro patch clamping techniques along with optogenetic circuit analysis (Apicella Lab). Understanding  the computations that the VH performs during emotional learning will provide important information regarding how the brain can generate specific vs. generalized responses.



Intellectual merit. The ability to generalize learning is fundamental for survival. However, too narrow or too broad generalization can have extremely negative consequences (13), indicating that adaptive behavioral responses require fine calibration of generalization abilities. Theories have suggested that the hippocampus provides contextual gating of emotional behavior (14, 15) but very few studies have focused on the VH and no studies have assessed how this region modulates generalization. It is becoming clear in the field of neuroscience that brain regions are not homogenous and that understanding the characteristics of different cell types is critical to determine function (16, 17). In this context, using long-range connectivity to classify cell types has been extremely instructive in other brain areas (16), but it has not been studied in depth in the VH. Finally, the importance of understanding the computations underlying how a context serves as an occasion setter for generalization extends beyond emotional learning because the ability to generate similar responses in different scenarios is also critical for artificial intelligence and machine learning (18, 19).



Broader Impacts: This project will train graduate and undergraduate students in a variety of cutting edge techniques. Since the collaboration between the Muzzio and Apicella lab is well established, a primary goal of the broader impacts is to co-mentor students to provide more comprehensive analytical and experimental training. This will be achieved by having common lab meetings to discuss and integrate all findings within the framework of the proposed objectives. UTSA is a minority-serving institution, with 57% enrollment of Hispanics. As such, this is the ideal place to increase research opportunities for diverse groups of students in STEM disciplines. In order to promote science in the community, both PIs will continue participating and leading discussion in the “Neuroscientists Talk Shop”, a podcast that serves to showcase prominent neuroscientists through informal conversations about current scientific issues, organizing yearly public talks by inviting leaders in the field, offering open  public discussions at Texas public radio, and participating in national meetings for underrepresented groups Dissemination will be achieved through presentations at scientific meetings and peer-reviewed publications. Evaluation of the program will be accomplished by keeping a record of the scientific path followed by our trainees.