Print Close

The Role of the Hippocampus in Forecasting Future Rewards during Goal-Directed Behavior

Poster No:

966

Submission Type:

Abstract Submission

Authors:

Jiwoong Park¹^,2^,3, WON MOK SHIM¹^,2^,3

Institutions:

¹Center for Neuroscience Imaging Research, Institute of Basic Science (IBS), Suwon, Korea, Republic of, ²Department of Biomedical Engineering, Sungkyunkwan University (SKKU), Suwon, Korea, Republic of, ³Department of Intelligent Precision Healthcare Convergence, Sungkyunkwan University (SKKU), Suwon, Korea, Republic of

First Author:

Jiwoong Park
Center for Neuroscience Imaging Research, Institute of Basic Science (IBS)|Department of Biomedical Engineering, Sungkyunkwan University (SKKU)|Department of Intelligent Precision Healthcare Convergence, Sungkyunkwan University (SKKU)
Suwon, Korea, Republic of|Suwon, Korea, Republic of|Suwon, Korea, Republic of

Co-Author:

Won Mok Shim
Center for Neuroscience Imaging Research, Institute of Basic Science (IBS)|Department of Biomedical Engineering, Sungkyunkwan University (SKKU)|Department of Intelligent Precision Healthcare Convergence, Sungkyunkwan University (SKKU)
Suwon, Korea, Republic of|Suwon, Korea, Republic of|Suwon, Korea, Republic of

Introduction:

The hippocampus has been reported to exhibit sequential neural activities that correspond to locations before they are actually visited during episodic experiences, involving the anticipation of future events yet to occur (Dragoi and Tonegawa, 2011). This anticipatory process potentially facilitates the simulation of upcoming scenarios. Among the numerous potential future events, predicting a future reward is particularly crucial for learning and adaptive behaviors. A recent animal study demonstrated that alterations in reward structure induce remapping within the hippocampus (Krishnan et al., 2022), indicating the involvement of the hippocampus in reward processing. In this study, we hypothesized that the predictive response of the hippocampus would be engaged in anticipating future rewards during episodic experiences. Specifically, we aimed to investigate the role of the hippocampus in forecasting future events, especially those associated with acquiring rewards during goal-directed behavior in naturalistic settings.

Methods:

As part of the 7T Naturalistic Perception, Action & Cognition (NatPAC) dataset, we introduced a novel Shepherding task that immerses participants in a 3D minecraft world where they engage with various biological entities in real-time by strategically planning and executing hierarchical actions tailored to a complex environment. The task requires herding sheep to a specified goal location while navigating challenges, such as avoiding puddles and evading a randomly spawning fox. To enhance the realism of the shepherding dynamics, we incorporated the behavioral dynamics based on an ethological study of flock behavior (Strömbom et al., 2014). We collected both functional neuroimaging data and diverse behavioral gameplay data as participants actively engaged in this Minecraft-based interactive 3D video game (Fig. 1A). To examine the hippocampal response to potential rewards, we identified three distinct event types during gameplay: "No loss" and "Loss" events, depending on whether the sheep fell into puddles or not, and "Near loss" event if the sheep almost fell but ultimately avoided puddles (Fig. 1B). This approach allowed for a comparison of selective responses to predictable reward loss.

Results:

Our results revealed an increased hippocampal response preceding reward loss compared to events where no loss occurred. Moreover, this heightened hippocampal response before reward loss was also observed in the near loss events, suggesting the potential predictive nature of these responses for losses, when participants anticipated loss based on the risk evaluated before the actual outcome (Fig. 1C, Top). In contrast, we observed that the nucleus accumbens (NAcc), a central hub of the reward system, selectively responded to actual reward loss but not to anticipated but unrealized reward loss (Fig. 1C, Bottom), indicating a dissociation between the hippocampus and NAcc in predictive reward processing.

Conclusions:

Our results suggest that the hippocampus is involved in predictive responses associated with potential reward loss, highlighting its intricate role in anticipating future rewards during episodic experiences. In contrast to the hippocampus, the nucleus accumbens exclusively responds to actual reward loss, suggesting the distinct role of the hippocampus in predicting and encoding potential reward loss in naturalistic settings.

Emotion, Motivation and Social Neuroscience:

Reward and Punishment ²

Higher Cognitive Functions:

Executive Function, Cognitive Control and Decision Making

Reasoning and Problem Solving

Higher Cognitive Functions Other ¹

Keywords:

Cognition

Computational Neuroscience

FUNCTIONAL MRI

^1|2Indicates the priority used for review

Provide references using author date format

Dragoi, G. & Tonegawa, S. (2011) Preplay of future place cell sequences by hippocampal cellular assemblies. Nature 46 9, 397–40.
Krishnan, S., Heer, C., Cherian, C. et al. (2022) Reward expectation extinction restructures and degrades CA1 spatial maps through loss of a dopaminergic reward proximity signal. Nat Commun 13, 6662
Strömbom, D. et al. (2014) Solving the shepherding problem: heuristics for herding autonomous, interacting agents. J R oy Soc Interface 11, 20140719.