Timescales characterize how fast the observables change in time. In neuroscience, they can be estimated from the measured activity and can be used, for example, as a signature of the memory trace in the network. I will first discuss the inference of the timescales from the neuroscience data comprised of the short trials and introduce a new unbiased method. Then, I will apply the method to the data recorded from a local population of cortical neurons from the visual area V4. I will demonstrate that the ongoing spiking activity unfolds across at least two distinct timescales - fast and slow - and the slow timescale increases when monkeys attend to the location of the receptive field. Which models can give rise to such behavior? Random balanced networks are known for their fast timescales; thus, a change in the neurons or network properties is required to mimic the data. I will propose a set of models that can control effective timescales and demonstrate that only the model with strong recurrent interactions fits the neural data. Finally, I will discuss the timescales' relevance for behavior and cortical computations.

Perception and action are biased by our recent experiences. Even when a sequence of stimuli are randomly presented, responses are sometimes attracted toward the past. The mechanism of such bias, recently termed serial dependence, is still under investigation. Currently, there is mixed evidence indicating that such bias could be either from a sensory and perceptual origin or occurring only at decisional stages. In this talk, I will present recent findings from our group showing that biases are decreased when disrupting the memory trace in a premotor region in a simple visuomotor task. In addition, we have shown that this bias is stable over periods of up to 8 s. At the end, I will show ongoing analysis of a recent experiment and argue that serial dependence may rely on distributed memory representations of stimuli and task relevant features.

The head direction (HD) system is classically modeled as a ring attractor network which ensures a stable representation of the animal’s head direction. This unidimensional description popularized the view of the HD system as the brain’s internal compass. However, unlike a globally consistent magnetic compass, the orientation of the HD system is dynamic, depends on local cues and exhibits remapping across familiar environments5. Such a system requires mechanisms to remember and align to familiar landmarks, which may not be well described within the classic 1-dimensional framework. To search for these mechanisms, we performed large population recordings of mouse thalamic HD cells using calcium imaging, during controlled manipulations of a visual landmark in a familiar environment. First, we find that realignment of the system was associated with a continuous rotation of the HD network representation. The speed and angular distance of this rotation was predicted by a 2nd dimension to the ring attractor which we refer to as network gain, i.e. the instantaneous population firing rate. Moreover, the 360-degree azimuthal profile of network gain, during darkness, maintained a ‘memory trace’ of a previously displayed visual landmark. In a 2nd experiment, brief presentations of a rotated landmark revealed an attraction of the network back to its initial orientation, suggesting a time-dependent mechanism underlying the formation of these network gain memory traces. Finally, in a 3rd experiment, continuous rotation of a visual landmark induced a similar rotation of the HD representation which persisted following removal of the landmark, demonstrating that HD network orientation is subject to experience-dependent recalibration. Together, these results provide new mechanistic insights into how the neural compass flexibly adapts to environmental cues to maintain a reliable representation of the head direction.

Understanding how the brain uses information is a fundamental goal of neuroscience. Several human disorders (ranging from autism spectrum disorder to PTSD to Alzheimer’s disease) may stem from disrupted information processing. Therefore, this basic knowledge is not only critical for understanding normal brain function, but also vital for the development of new treatment strategies for these disorders. Memory may be defined as the retention over time of internal representations gained through experience, and the capacity to reconstruct these representations at later times. Long-lasting physical brain changes (‘engrams’) are thought to encode these internal representations. The concept of a physical memory trace likely originated in ancient Greece, although it wasn’t until 1904 that Richard Semon first coined the term ‘engram’. Despite its long history, finding a specific engram has been challenging, likely because an engram is encoded at multiple levels (epigenetic, synaptic, cell assembly). My lab is interested in understanding how specific neurons are recruited or allocated to an engram, and how neuronal membership in an engram may change over time or with new experience. Here I will describe both older and new unpublished data in our efforts to understand memories in mice.

Neocortical-hippocampal interactions during off-line periods such as slow-wave sleep are implicated in memory processing. In particular, recent memory traces are replayed in hippocampus during some sharp-wave ripple (SWR) events, and these replay events are positively correlated with neocortical memory trace reactivation. A prevalent model is that SWR arise ‘spontaneously’ in CA3 and propagate recent memory ‘indices’ outward to the neocortex to enable memory consolidation there; however, the spatiotemporal distribution of neocortical activation relative to SWR is incompletely understood. We used wide-field optical imaging to study voltage and glutamate release transients in dorsal neocortex in relation to CA1 multiunit activity (MUA) and SWR of sleeping and urethane anesthetized mice. Modulation of voltage and glutamate release signals in relation to SWRs varied across superficial neocortical regions, and it was largest in posteromedial regions surrounding retrosplenial cortex (RSC), which receives strong hippocampal output connections. Activity tended to spread sequentially from more medial towards more lateral regions. Contrary to the unidirectional hypothesis, activation exhibited a continuum of timing relative to SWRs, varying from neocortex leading to neocortex lagging the SWRs (± ~250 msec). The timing continuum was correlated with the skewness of peri-SWR hippocampal MUA and with a tendency for some SWR to occur in clusters. Thus, contrary to the model in which SWRs arise spontaneously in hippocampus, neocortical activation often precedes SWRs and may thus constitute a trigger event in which neocortical information seeds associative reactivation of hippocampal ‘indices’.

Arguably one of the biggest mysteries in neuroscience is how the brain stores long-term memories. The major challenge for investigating the neural circuit underlying memory formation in the neocortex is the distributed nature of the resulting memory trace throughout the cortex. Here, we used a new behavioral paradigm that enabled us to generate memory traces in a specific cortical location and to specifically examine the mechanisms of memory formation in that region. We found that medial-temporal inputs arrive in neocortical layer 1 where the apical dendrites of cortical pyramidal neurons predominate. These dendrites have active properties that make them sensitive to contextual inputs from other areas that also send axons to layer 1 around the cortex. Blocking the influence of these medial-temporal inputs prevented learning and suppressed resulting dendritic activity. We conclude that layer 1 is the locus for hippocampal-dependent memory formation in the neocortex and propose that this process enhances the sensitivity of the tuft dendrites to contextual inputs.