Ventral hippocampal CA1 and CA3 differentially mediate learned approach-avoidance conflict processing

Abstract

1The ventral hippocampus is thought to play a key role in the resolution of approach-avoidance 2 conflict, a scenario that arises when stimuli with opposing valences are present simultaneously. 3Little is known, however, about the contributions of specific hippocampal sub-regions in this 4 process, a critical issue given the functional and anatomical heterogeneity of this structure. Using 5 a non-spatial cue-based paradigm in rats, we found that transient pharmacological inactivation of 6 ventral CA1 produced an avoidance of a conflict cue imbued with both learned positive and 7 negative outcomes, whereas inactivation of the ventral CA3 resulted in the opposite pattern of 8 behavior, with significant preference for the conflict cue. In contrast, dorsal CA1-and CA3-9 inactivated rats showed no change in conflict behavior. Our findings provide important insight 10 into the functions and circuitry of the ventral hippocampus by demonstrating that the ventral CA1 11 and CA3 subserve distinct and opposing roles in approach-avoidance conflict processing. 12 peer-reviewed)

Double dissociation of learned approach–avoidance conflict processing and spatial pattern separation along the dorsoventral axis of the dentate gyrus

“…Animals were first trained to associate three distinct visuotactile cues with either appetitive, aversive, or neutral outcomes in a 3‐arm Y‐maze and were subsequently presented with a conflict cue, composed of positive and negative cues, in one arm and a neutral cue in another arm. It was found that transient inactivation of the ventral CA3 subfield potentiated approach behavior toward the conflict stimulus, akin to the effects of whole ventral HPC lesions in previous work (Schumacher et al, ), whereas inactivation of the ventral CA1 led to greater avoidance (Schumacher et al, ). This work suggests that the ventral HPC can be further divided into separate motivational circuits when approach–avoidance conflict is evoked, possibly mediated through differential connections to other subcortical structures.…”

“…In the arm with the neutral cue, the animals did not experience any outcomes. Previous studies from our laboratory had demonstrated that these specific schedules and magnitudes of sucrose reward and shock facilitate the development of conditioned approach and avoidance to the appropriate cues without inducing generalized fear of the apparatus and freezing responses to the aversive cue, and results in control animals spending roughly equal times in the conflict and neutral arms during the conflict test (Hamel, Thangarasa, Samadi, & Ito, ; Schumacher et al, , ). The order of arm presentation was varied daily to prevent the animals associating the outcomes with the sequence of arm presentation.…”

“…Much of this evidence comes from tasks that elicit approach–avoidance conflict without prior learning such as the elevated plus maze (EPM), light–dark box, and open field, which are commonly used in rodent studies as tests of anxiety and require the animal to respond based on an innate drive for exploration and an innate aversion to spaces that leave them open to predation (Bannerman et al, ; Kheirbek et al, ; Kjelstrup et al, ; Parfitt et al, ; Trivedi, ; Weeden, Roberts, Kamm, & Kesner, ). More recently, it has also been demonstrated that ventral HPC‐sensitive approach–avoidance conflict can also be elicited by presenting rodents with learned/conditioned stimuli that predict the availability of both positive and negative outcomes (Schumacher et al, ; Schumacher, Vlassov, & Ito, ), which may hold more translational value as real‐life approach–avoidance conflict scenarios usually involve learned stimuli (Ito & Lee, ). Converging with this, human neuroimaging studies have supported the association of the primate anterior HPC with approach–avoidance conflict processing, as tasks that elicit motivational conflict are associated with robust activity in the region (Bach et al, ; Loh et al, ; O'Neil et al, ).…”

“…Notably, the majority of this work has been conducted on the dorsal portion of the HPC, rendering it unclear as to how such processes speak, if at all, to the role of ventral HPC subdivisions in approach–avoidance conflict processing. Recently, Schumacher et al () compared and contrasted the effects of ventral CA3 and ventral CA1 disruption on approach–avoidance behavior by using a mixed valence conditioning paradigm to elicit motivational conflict. Animals were first trained to associate three distinct visuotactile cues with either appetitive, aversive, or neutral outcomes in a 3‐arm Y‐maze and were subsequently presented with a conflict cue, composed of positive and negative cues, in one arm and a neutral cue in another arm.…”

“…The goal of the present study, therefore, was to examine the role of the ventral DG in learned approach–avoidance conflict processing using the same mixed valence Y‐maze paradigm employed in previous work (Schumacher et al, , ). To examine the specificity of any observed effects to the ventral portion of the HPC, we also investigated the impact of dorsal DG inactivation on this task, and moreover, administered a spatial pattern separation task that we expected to be sensitive to dorsal DG manipulation, based on previous data (Clelland et al, ; Knierim & Neunuebel, ; Lee, Hunsaker, & Kesner, ; Leutgeb et al, ; McHugh et al, ).…”