Bernardo Luis Sabatini

Recording and manipulating brain activity during behavior is critical to understanding the underlying mechanisms of decision-making. Linking neural activity to behavior requires behavioral hardware and software tightly integrated with recording and perturbation systems on a shared clock. We built SPOUT (State-machine Platform for Operant Uni/dual-spout Tasks), an open-source, Teensy-driven state-machine platform with a MATLAB interface that runs 10 unique decision-making tasks (with dozens of variations available through user-friendly settings) to study behavior in head-restrained mice. The platform is built on several custom hardware devices: a dual-lick detector, headplate designs for optogenetics and two-photon calcium imaging, a three-axis motorized spout manipulator, and an optogenetics power modulator. The firmware differentiates between one and two lick spout tasks and can be controlled by a user-friendly interface. Task settings can be selected through the interface or by loading predefined settings files. We validated the clock speed and lick detection against an independent, external acquisition system and identified highly precise, sub-millisecond detection of single licks. Using a pseudo-random synchronization pulse generated by SPOUT, we corrected for missing data due to glitches in the acquisition system and clock drift. We showcase the versatility of SPOUT by training mice on an uninstructed lick-left/lick-right task in which the rewarded side switches unexpectedly and found that mice use history-dependent action-outcome associations to guide future behavior. Transiently inhibiting the anterior lateral motor cortex (ALM) during cue presentation induced contralateral deficits, without affecting ipsilateral trials. Finally, two-photon imaging of ALM neurons revealed stronger population responses during contralateral choice licks compared to ipsilateral ones. Together, SPOUT offers an open-source, affordable platform to study decision-making in head-restrained mice while combining neural recordings and manipulations.

Adaptive behavior requires animals to use the outcomes of recent actions (action-outcomes) to guide future decisions. While the striatum is critical for decision-making, it is currently unclear how it is involved in the storage and retrieval of short-term associative memories. Here, we implemented a head-fixed memory-guided decision-making task in which mice use the outcome of a previous choice to determine whether to repeat or switch their next action. We show that dopamine fluctuations in the ventrolateral striatum are modulated by reward receipt or omission and recent outcome history, while the activity of direct and indirect pathway striatal projection neurons encodes recent action-outcome associations and predicts future switch/repeat choices. Closed-loop optogenetic activation and inhibition of direct and indirect pathway neurons during either the action-outcome association period or the delay preceding the next choice bidirectionally biased future actions away from those favored by reward history. Together, these findings suggest that striatal activity maintains a short-term action-outcome associative memory that links completed actions to future motor plans during adaptive decision-making.

Neurons distribute synaptic inputs across their dendritic tree. In layer 2/3 pyramidal cells of primary visual cortex, spines on distal dendrites share somatic orientation preference but have receptive fields displaced in retinotopic space, which supports tuning to visual edges. However, it is not known how synaptic plasticity rules can lead to specialization of tuning properties across dendritic compartments. We demonstrate an experimentally grounded model of compartment-specific spike-timing dependent plasticity (STDP) that accounts for the enrichment of retinotopically-displaced inputs on distal branches. Our previous experimental work revealed compartment-specific calcium signals that predict reduced STDP-mediated depression but preserved potentiation. Based on these findings, we built an STDP model with compartment-specific properties, in which some distal branches are relatively resistant to STDPmediated depression. Synapses on these branches are more likely to stabilize inputs with weaker correlations to postsynaptic spiking. Using a visual input model, we show that compartment-specific reduction in STDP-mediated depression recapitulates in vivo experimental measurements of spine tuning. Furthermore, our experimental results show that reduced STDP-mediated depression is restricted to distal dendritic compartments with complex branching structure and not observed in other distal branches. Therefore, our model makes an untested prediction that complex branches will be hotspots for retinotopically-displaced inputs.

Fast-spiking parvalbumin-positive (PV) neurons provide precisely timed, context-dependent inhibition within cortical circuits. PV neuron firing properties are specialized among cortical neurons, suggesting that they express a unique complement of ion channels. Here, we show that the PV-specific silent voltage-gated potassium (Kv) channel subunit Kv6.4 (encoded by Kcng4 ) modulates both intrinsic and synaptic properties of cortical PV neurons. Kv6.4 does not form functional channels on its own but, as shown in prior work, assembles with Kv2 subunits to create heterotetrameric channel complexes, effectively reducing Kv2-mediated delayed rectifier current. We find that Kcng4 expression is enriched within a distinct Pvalb -expressing subclass in primary somatosensory (S1) and motor (M1) cortex and emerges during postnatal development. In PV neurons, Kv6.4 loss reduces action potential (AP) height and width, hyperpolarizes the threshold and interspike potential, and accelerates AP upstroke particularly during repetitive firing. Kv6.4 loss, potentially due to the changes in AP waveform, also alters GABA release and paired-pulse depression at synapses made by PV onto pyramidal (PYR) neurons. The effects of Kv6.4 loss are amplified during high-frequency firing, within the physiological range of fast-spiking PV neurons, likely due to altered repolarization dynamics that accumulate across successive APs. These findings are thus consistent with the function of Kv6.4 in modifying Kv2-mediated delayed rectifier currents. Hence, Kv6.4 tunes the temporal precision of PV inhibitory output, a feature that may be critical for stable excitation–inhibition ratios and adaptive circuit function underlying learning and behavior.

Credit assignment--how changes in individual neurons and synapses affect a network's output--is central to learning in brains and machines. Noise correlation, which estimates gradients by correlating perturbations of activity with changes in output, provides a biologically plausible solution to credit assignment but scales poorly as accurately estimating the Jacobian requires that the number of perturbations scale with network size. Moreover, isotropic noise conflicts with neurobiological observations that neural activity lies on a low-dimensional manifold. To address these drawbacks, we propose neural manifold noise correlation (NMNC), which performs credit assignment using perturbations restricted to the neural manifold. We show theoretically and empirically that the Jacobian row space aligns with the neural manifold in trained networks, and that manifold dimensionality scales slowly with network size. NMNC substantially improves performance and sample efficiency over vanilla noise correlation in convolutional networks trained on CIFAR-10, ImageNet-scale models, and recurrent networks. NMNC also yields representations more similar to the primate visual system than vanilla noise correlation. These findings offer a mechanistic hypothesis for how biological circuits could support credit assignment, and suggest that biologically inspired constraints may enable, rather than limit, effective learning at scale.

Loss-of-function mutations in AKAP11 (a protein kinase A (PKA)-binding protein) greatly increase the risk of bipolar disorder and schizophrenia. To determine the neurobiological functions of AKAP11, we conduct multi-omic and neurobiological analyses of Akap11 mutant mouse brains. We find that AKAP11 is a key regulator of PKA proteostasis in the brain whose loss leads to dramatically increased levels of PKA subunits and phosphorylated PKA substrates, especially in synapses. Akap11 mutant mice show extensive transcriptomic changes throughout the brain, including prominent decreases in synapse-related genes sets. Gene expression is highly impacted in spiny projection neurons of the striatum, a brain region implicated in motivation, cognition and psychotic disorders. Real-time measurements of PKA activity reveal elevated basal PKA activity in the striatum of Akap11^-/- mice, with exaggerated additional response to dopamine receptor antagonists. Behaviorally, Akap11 mutant mice show abnormally prolonged locomotor response to amphetamine, deficits in associative learning and contextual discrimination, as well as depression-like behaviors. Our study connects molecular changes to circuit dysfunction and behavioral disturbance in a genetically valid animal model of psychotic disorder.

Fast-spiking parvalbumin-positive (PV) neurons provide precisely timed, context-dependent inhibition within cortical circuits. PV neuron firing properties are specialized among cortical neurons, suggesting that they express a unique complement of ion channels. Here, we identify the PV-specific silent voltage-gated potassium (Kv) channel subunit Kv6.4 (encoded by Kcng4 ), whose role in cortical PV neuron physiology was previously unknown, as a modulator of both intrinsic and synaptic properties. Kv6.4 does not form functional channels on its own but, as shown in prior work, assembles with Kv2 subunits to create heterotetrameric channel complexes, effectively reducing Kv2-mediated delayed rectifier current. We find that Kcng4 expression is enriched within a distinct Pvalb -expressing subclass in primary somatosensory (S1) and motor (M1) cortex and emerges during postnatal development. In PV neurons, Kv6.4 loss reduces action potential (AP) height and width, hyperpolarizes the threshold and interspike potential, and accelerates AP upstroke particularly during repetitive firing. Kv6.4 loss, potentially due to the changes in AP waveform, also alters GABA release and paired-pulse depression at synapses made by PV onto pyramidal (PYR) neurons. The effects of Kv6.4 loss are amplified during high-frequency firing, within the physiological range of fast-spiking PV neurons, likely due to altered repolarization dynamics that accumulate across successive APs. These findings are thus consistent with the function of Kv6.4 in modifying Kv2-mediated delayed rectifier currents. Hence, Kv6.4 tunes the temporal precision of PV inhibitory output, a feature that may be critical for stable excitation-inhibition ratios and adaptive circuit function underlying learning and behavior. Significance Statement: Voltage-gated ion channels are broadly expressed yet serve cell-type-specific functions. We demonstrate that the Kv6.4 silent subunit is selectively expressed by fast-spiking parvalbumin-positive (PV) inhibitory neurons among cortical neurons. In PV neurons, Kv6.4 modulates action potential (AP) waveform and GABA release onto excitatory neurons in a frequency-dependent manner, well within the physiological range of PV firing. This mechanism is likely important for preserving the temporal precision of inhibition, which is particularly critical for fast-spiking interneuron output. By shaping short-term synaptic plasticity during high-frequency firing, Kv6.4 provides a molecular and cell-type-specific mechanism for context-dependent tuning of inhibition within cortical circuits. This function may be especially relevant for sensory processing, experience-dependent plasticity, and the maintenance of excitation-inhibition ratios in vivo .

Animals, including humans, use coordinated facial movements to sample the environment, ingest nutrients, and communicate. Rodents, in particular, produce rhythmic facial movements during spontaneous behavior and cognitive tasks. Measuring these movements precisely and linking them to neural activity remains challenging. We introduce face-rhythm, an unsupervised pipeline that combines markerless point-tracking, spectral analysis, and non-negative tensor component analysis to decompose facial video into a small set of interpretable components. Applied to videos of mice during a Pavlovian odor-reward task, a brain-machine interface (BMI) task, and free behavior, face-rhythm recovers human-interpretable behaviors such as whisking, sniffing, licking, and subtler motifs. The resulting components are consistent across animals, are sufficient to decode task variables or internal belief states, and explain cortical activity using a low-rank representation. We also find that the activity of neurons in face-associated primary motor cortex (M1) is predicted well by a phase-invariant spectral transformation of facial movements above ~0.5 Hz, while slower movements retain a phase-variant representation better predicted by the instantaneous position of the face; individual neurons can show either or both forms of tuning. A systematic comparison against deep-learning point-tracking models, contrastive-learning embeddings, and vision-transformer features places face-rhythm competitively across tasks while also achieving the goal of producing a low-dimensional, interpretable description of rodent facial behavior that is closely linked to cortical activity.