Learning, Memory and Decision Making in the Mammalian Brain

  We use rodent models to investigate the physiological mechanisms that underlie information processing and coordinated interactions between multiple brain regions that are necessary for memory and cognition, with a particular focus on hippocampal - prefrontal cortical interactions. The hippocampus is known to be critical for episodic memories, and the prefrontal cortex is involved in executive control, working memory and decision making. Communication between the prefrontal executive system and the hippocampal memory system is key for learning, remembering, planning, prediction and memory-guided decision making. However, the nature of communication between these two regions, the underling neural mechanisms and causal contributions of this pivotal interaction remain largely unknown.

  We address these questions using a combination of techniques, including behavior, large scale multielectrode recordings in awake behaving animals, real time detection and perturbation of neural activity patterns, targeted optogenetic interventions, and computational analysis. We have shown that hippocampal replay during awake sharp-wave ripples (SWRs) is critical for spatial memory, and SWRs are associated with coordinated reactivation of hippocampal-prefrontal neurons during memory-guided decision making. This approach thus allows us to characterize the neurophysiological basis of hippocampal-cortical interactions, and also to provide causal evidence linking specific forms of neural activity to behavior and cognition.

  We posit that neural dynamics at the ensemble level and network coordination still remains a "missing link" that can bridge between molecular/ cellular processes and behavioral phenomena in our understanding of mechanisms that underlie cognitive function and dysfunction. Our findings provide a crucial foundation to investigate if impairments in physiological network patterns lead to deficits in memory and cognition in disorders that involve hippocampus and prefrontal cortex. Our research will thus provide crucial insight into several neurological and neuropsychiatric disorders involving these two key regions, such as dementia, Alzheimer's disease, depression, autism and schizophrenia.

Major Findings and Discoveries

Multiple time-scales of decision making in hippocampus and prefrontal cortex

Slow and fast timescale sequences in the hippocampus and prefrontal cortex

Tang W*, Shin JD*, Jadhav SP (2021), "Multiple time-scales of decision making in the hippocampus and prefrontal cortex", eLife, 10, e66227.
  In this study, we establish neural population sequences at multiple timescales in the hippocampus and prefrontal cortex, and determine their complementary roles in memory-guided decision making. We demonstrate neural population sequences at multiple timescales in the hippocampus and prefrontal cortex, with fast timescales on the order of ~100ms for theta sequences and sharp-wave ripple replay events, and slow timescales on the order of ~seconds for trajectory-dependent firing sequences or “splitter” sequences during behavior. These choice-specific sequences at the behavioral timescale, i.e. trajectory-dependent firing in both CA1 and PFC predicted future choices throughout learning, as expected. In addition, we found compressed theta sequences in both CA1 and PFC nested within behavioral timescale sequences during navigation. We establish PFC theta sequences, PFC theta cycle skipping, and also memory-dependent look ahead of PFC theta sequences, similar to CA1. CA1 theta sequences encoded both actual and alternative choices during decision making, but could not predict upcoming choice. In contrast, PFC theta sequences and coherent CA1-PFC theta sequences predicted actual choice. Thus, theta sequences can support vicarious memory recall. Behavioral timescale or theta sequences could not predict errors. Rather, compressed timescale replay sequences during sharp-wave ripples in inter-trial periods prior to trajectory onset prime decisions, and errors in CA1-PFC replay predict incorrect decisions.
  We thus integrate hippocampal and prefrontal behavioral sequences, theta sequences and replay sequences in a unified framework for memory-guided decision making. These results show that transient dynamics at fast, cognitive timescales can support decisions, and establish cooperative interaction of sequences at multiple timescales for decision making.

We have some neat theta sequence decoding videos for the paper:
Theta Sequence Videos.

Coordinated Hippocampal-Prefrontal Replay Supports Spatial Learning and Memory-guided Decision Making

Activity in hippocampal cells (left) shows a forward replay event representing the path the rat will take next (right)

Shin JD*, Tang W*, Jadhav SP (2019), "Dynamics of awake hippocampal-prefrontal replay for spatial learning and memory-guided decision making", Neuron, 104(6), 1110-1125.
  This study builds on previous findings about sharp-wave ripples (SWRs) and replay to show how coordinated brain activation in hippocampal-prefrontal circuits during awake replay supports spatial learning and memory-guided decision making. In awake replay, place cells show specific patterns of activity during navigation that can be “replayed” later in forward or reverse order, as if the brain were internally fast-forwarding or rewinding through routes the rats have taken. We tracked this replay process by monitoring brain activity continuously throughout the course of learning, and showed that reverse replay mediates the ability to evaluate past actions leading to reward, in order to learn and form memories. In contrast, forward replay supports the ability to think ahead, and imagine and plan choices that will lead to goals in the future. Coordinated activation in the prefrontal cortex uses these retrospective and prospective replay evaluations to recall specific past experiences to make ongoing decisions.
  These findings unify brain mechanisms of past memories and future imagination, which are thought to be linked in the human brain. The involvement of ‘replay’ in memory processes has been observed across many species, including humans, and our study establishes that replay serves as a key neural substrate underlying an internal dialogue across multiple brain regions to support our ability to learn, plan, choose, and imagine.

Featured in a NIMH news article:
"Reading the Brain’s Map: Coordinated Brain Activation Supports Spatial Learning and Decision-Making".
Featured in a Brandeis news article:
"The neuroscience behind remembering the past and plotting the future.".
Also, we have some neat replay decoding videos on twitter @Jadhavlab at this link:
Replay Decoding Videos on Twitter.

Awake and Sleep Replay

SWS and REM sleep detection using hippocampal recordings

Tang W, Shin JD, Frank LM, Jadhav SP (2017), "Hippocampal-prefrontal reactivation during learning is stronger in awake as compared to sleep states", Journal of Neuroscience, 37(49): 11789-11805.
  This study that revealed a surprising difference between awake and sleep replay. Despite the dogma that sleep replay is primarily responsible for reactivating memories of our experiences for permanent long-term storage, we found that replay in the hippocampal-prefrontal cortical circuit is stronger in waking states than in sleep states, and awake replay recapitulates spatial memories much more precisely than sleep replay, which is inherently noisy. These results strongly indicated that awake and sleep replay have different functional roles, and further challenged a major dogma about the role of sleep replay in memory consolidation.

Featured in a Journal of Neuroscience article:
"PFC–Hippocampus Interactions during Sleep and Awake Ripples".
Featured on the Brandeis Science blog:
"Communicating Memory Information Between the Hippocampus and Prefrontal Cortex"
Also, see a related review:
Tang W, Jadhav SP (2018), "Sharp-wave ripples as a signature of hippocampal-prefrontal reactivation for memory during sleep and waking states", Neurobiology of Learning and Memory.

Coherent Spatial Coding Mediated by Theta Oscillations

Spatial decoding during theta oscillations

Zielinski MC*, Shin JD*, Jadhav SP (2019), "Coherent coding of spatial position mediated by theta oscillations in the hippocampus and prefrontal cortex", Journal of Neuroscience, 39(23), 4550-4565.
  This study established that theta oscillations mediate a temporal coordination mechanism for coherent coding of spatial position in hippocampal-prefrontal networks during memory-guided behavior. We found that ensemble activity in the prefrontal cortex encodes animal’s current position coherently with hippocampal populations at a theta-cycle timescale, and theta-phase associated spiking refines this spatial coding in both regions.

Also, see a related review:
Zielinski MC*, Tang W*, Jadhav SP (2020), "The role of replay and theta sequences in mediating hippocampal-prefrontal interactions for memory and cognition", Hippocampus, 30(1):60-72.

Hippocampal-Prefrontal Network for Memory-Guided Decision Making

Multisite, multielectrode recordings in hippocampus and PFC to identify neural coordination mechanisms

Jadhav SP*, Rothschild G*, Roumis DR, Frank LM (2016), "Coordinated excitation and inhibition of prefrontal ensembles during awake hippocampal sharp-wave ripple events", Neuron, 90(1):113-27. doi: 10.1016/j.neuron.2016.02.010
  This study showed that awake SWR replay during quiet waking states (when animals are inactive) corresponds to mental replay of ongoing spatial experiences. The hippocampus, the brain’s memory hub, communicates during awake replay with a decision-making area in the brain’s executive hub, prefrontal cortex (PFC), thus identifying a circuit for experience-informed decision making in rats. We further showed that active online processing of external information in the brain’s executive hub is inhibited during SWR activity bursts, transitioning rapidly to an internal replay process. This study demonstrated that strongly coordinated activity within this hippocampal-prefrontal circuit during awake replay is likely to optimize the brain’s ability to consolidate memories and use them to decide on future actions, and laid the foundation for future investigations of the role of awake replay in memory.

Featured in a NIH news article:
"Circuit for experience-informed decision making".
Also, see a related review:
Shin JD, Jadhav SP (2016), "Multiple modes of hippocampal-prefrontal interactions in memory-guided behavior", Current Opinion in Neurobiology, 40:161-169.

Awake Sharp-Wave Ripple Replay of Mental Experiences Critical for Learning

Schematic showing replay of hippocampal place cell sequences during awake sharp-wave ripples (SWRs)

Jadhav SP, Kemere C, German PW, Frank LM (2012), "Awake hippocampal sharp-wave ripples support spatial memory", Science, 336(6087): 1454-1458.
  The hippocampus is essential for storing and retrieving daily memories, and the neurophysiological basis of memory is a key question. A major focus in the field has been the activity of place cells and its role in spatial navigation, but how this relates to episodic memory is a matter of debate. Although SWRs that occur during sleep after behavior were thought to support memory consolidation, critically, neither place cell activity nor SWRs that occur in the awake state had been linked causally to memory. We hypothesized that awake replay events during SWRs could underlie memory formation and retrieval (Review in Nature Neuroscience, 2011). We addressed this important question by developing innovative techniques to detect specific hippocampal activity patterns in real-time and to eliminate them during behavior, and demonstrated that hippocampal replay during awake SWRs is required for learning and spatial memory.

Featured in a NIH news article:
"Awake mental replay of past experiences critical for learning".

Taste Coding in the Hippocampus

Hippocampal place cell recordings during taste delivery

Herzog LE, Pascual LM, Scott SJ, Mathieson ER, Katz DB, Jadhav SP (2019), “Interaction of taste and place coding in the hippocampus”, Journal of Neuroscience, 39(16), 3057-3069.
  This study reported the discovery of taste-responsive place cells in the hippocampus. A subset of place cells with weak spatial responses discriminated between tastes based on palatability. The hippocampus thus overlays existing mental maps with information about the hedonic values of tastes, providing a mechanism by which animals can use past experience to locate food sources.

Featured in a news release:
"So close, rats can almost taste it".

Refinement and Reactivation of a Taste-Responsive Hippocampal Network

Refinement of taste-responsive cell's place fields with experience

Herzog LE, Katz DB, Jadhav SP (2020), "Refinement and reactivation of a taste-responsive hippocampal network", Current Biology, 30, 1306-1311.
  This study followed up on the discovery of a taste responsive hippocampal sub-network to examine changes in this network during novel taste experience. Our results showed that taste responsiveness is associated with initially larger spatially responsive place fields which are selectively refined with experience. This refinement or reorganization is supported by selective reactivation during sharp-wave ripples (SWRs). This phenomenon can thus underlie a mechanism to signal areas of food availability with experience.