Investigating a role for hippocampus during cost based decision making

Abstract

Hippocampus (HPC) has been implicated in decision-making as context-based features of reward have been shown to bias hippocampal place field properties. Its direct connectivity with brain structures known to be involved in determining the value of response outcomes (e.g. ventral striatum) and impulsivity related to reward-motivated behavior (e.g. orbitofrontal cortex) further suggests a role for the HPC during reward-motivated decision making. Indeed, it has been shown that the expected probability of receiving a reward biases HPC place field remapping: it seems to scale directly with the probability of receiving a reward during a probability discounting task (Tryon et al., 2017). In addition, HPC lesions increase variability in an animal’s preference for large delayed rewards over small immediate rewards in a delay discounting task suggesting that the HPC is involved in cost-benefit decision making (McHugh et al., 2008, Abele and Chudasama, 2013). In this study, we sought to determine whether HPC is necessary for accurately representing and utilizing delay information by temporality inactivating HPC during a delayed discounting T-maze task. In addition, we examined hippocampal CA1 place cell activity during performance of the same task in an effort to determine if they encode delay associated costs information. A delay-based decision making maze task required rats to evaluate the cost of a decision in terms of the amount of time they were willing to wait in order to obtain the large reward (Jo, 2014). The HPC was infused with saline (SAL) or muscimol (MUS), a GABA antagonist, prior to experimental sessions for the inactivation study. For the rats in the recording study, neural activity was correlated with maze region and delay context. Consistent with operant studies, the behavior analysis from 10 rats revealed the strongest preferences for the large reward when it was associated with the 10s delay, but this preference declined as the delay increased. Inactivation results suggest that HPC is necessary for normal delay discounting on this particular task. Current preliminary analysis of hippocampal place fields does not show evidence for encoding the cost of a reward in that population level place field remapping was not observed across trials associated with different delays. This suggests that the intrinsically generated value constructs, such as the expected cost of a reward, may be facilitated by HPC but it is not reflected in broad population level changes in place cell activity.