The buzziest neuroscience papers of 2023, 2024

It takes time for the true value of any paper to manifest. Before neuroscience assimilates a finding into its canon, other groups must first replicate it, pick at it and build on it. But taking stock of results before they ripen can also provide a sense of the field’s most pressing questions and promising new directions.

The Transmitter set out to compile some of the most exciting neuroscience papers published in 2023 and 2024. A panel of 10 neuroscientists selected the papers from a preliminary list our editorial team assembled by considering invited commentaries in top journals, looking at Altmetric scores and surveying our neuroscientist audience, sources and contributors. We excluded preprints, reviews and perspective pieces from the initial pool.

The resulting collection of 20 papers is a research medley, which we have broken down into several broad categories below, and includes work on the effects of pregnancy on the brain and the first complete fly brain connectome. The studies were conducted in an array of organisms—even penguins! This range exemplifies the multidisciplinary nature of modern neuroscience.

Read more about the 20 buzzy papers, plus what subject-matter experts think they mean for the field—and what they hope comes next.

Barcoding of episodic memories in the hippocampus of a food-caching bird. Chettih S.N., Mackevicius E.L., Hale S., Aronov D. Cell (2024)

Volitional activation of remote place representations with a hippocampal brain-machine interface. Lai C., Tanaka S., Harris T.D., Lee A.K. Science (2023)

Abstract representations emerge in human hippocampal neurons during inference. Courellis H.S., Minxha J., Cardenas A.R., Kimmel D.L., Reed C.M., Valiante T.A., Salzman C.D., Mamelak A.N., Fusi S., Rutishauser U. Nature (2023)

The discovery of place cells in the hippocampus more than 60 years ago led to the theory that the brain area encodes spatial representations. The field has expanded its view since then, as evidenced by these three recent papers examining hippocampal representations of different entities in a variety of models. Dmitriy Aronov and his colleagues showed that a unique and sparse ensemble of neurons—barcodes, similar to engrams—in the avian hippocampus encodes food storage and retrieval in chickadees. This barcode is unique to a particular caching event and does not overlap with other episodes.

In their study, Albert Lee’s lab showed that in rats the hippocampus can, at will, represent not only the present but also remote times and places not currently being experienced. In this way, the hippocampus can simulate different scenarios, similar to Endel Tulving’s idea of our own capacity for mental time travel. And in the third study, Ueli Rutishauser and his colleagues found that making an inference in a task is accompanied by the emergence of abstract, orthogonal representations in the human hippocampus that allow for generalization.

Together, these papers show that the hippocampus is not just a place for storing specific episodes—it can also rapidly build abstract, generalizable codes that help us imagine, reason, infer and flexibly apply knowledge to new situations.

—Sheena Josselyn, senior scientist, Hospital for Sick Children (SickKids); professor of psychology and physiology, University of Toronto

A high-performance speech neuroprosthesis. Willett F.R., Kunz E.M., Fan C., Avansino D.T., Wilson G.H., Choi E.Y., Kamdar F., Glasser M.F., Hochberg L.R. … Henderson J.M. Nature (2023)

A high-performance neuroprosthesis for speech decoding and avatar control. Metzger S.L., Littlejohn K.T., Silva A.B., Moses D.A., Seaton M.P., Wang R., Dougherty M.E., Liu J.R., Wu P. … Chang E.F. Nature (2023)

Single-neuronal elements of speech production in humans. Khanna A.R., Muñoz W., Kim Y.J., Kfir Y., Paulk A.C., Jamali M., Cai J., Mustroph M.L., Caprara I. … Williams Z.M. Nature (2024)

The past few years saw tremendous progress in speech neuroprosthesis research, with major developments in brain-computer interfaces (BCIs) for speech decoding. This is exemplified by the 2023 studies from Jaimie Henderson and Edward Chang’s labs demonstrating high-speed, high-accuracy decoding of attempted speech from neural activity in the sensorimotor cortex. Both studies make advances in several challenging areas: They achieve processing speeds approaching that of natural conversation, apply to large vocabulary sizes and overcome the lack of precise temporal alignment between neural activity and attempted speech elements. Both reveal that the brain preserves cortical codes for speech production after years of paralysis. The two studies leverage distinct spatial scales—spiking activity from small regions within the precentral gyrus in Henderson’s study versus population activity over a much larger region in Chang’s study—revealing both a highly intermixed local and a broader somatotopic organization.

These two neuroprosthesis studies are complemented by work from Ziv Williams and his colleagues, which describes neurons within the middle frontal gyrus that predict phonetic, syllabic and morphological components of words during articulation planning and execution. This region may represent a target for speech neuroprosthesis efforts. Future studies will, hopefully, achieve even lower error rates in BCI performance, use denser or more extensive electrode arrays, develop adaptive decoders for dynamic neural activity and validate the devices in patients with different clinical backgrounds.

—Kirill Nourski, associate professor of neurosurgery, University of Iowa

Psychedelics promote neuroplasticity through the activation of intracellular 5-HT2A receptors. Vargas M.V., Dunlap L.E., Dong C., Carter S.J., Tombari R.J., Jami S.A., Cameron L.P., Patel S.D., Hennessey J.J. … Olson D.E. Science (2023)

Psychedelics reopen the social reward learning critical period.
Nardou R., Sawyer E., Song Y.J., Wilkinson M., Padovan-Hernandez Y., de Deus J.L., Wright N., Lama C., Faltin S. … Dölen G. Nature (2023)

We still do not understand how psychedelics work. These two 2023 studies, however, helped deepen what we do know about psychedelic drug action in the brain by revealing unexpected mechanisms that challenge current views. David Olson and his colleagues evaluated psychedelics with varying lipophilicity and found that those that more readily cross the plasma membrane are more likely to induce neural plasticity—suggesting that intracellular serotonin receptors may play a key role in their activity. This differs sharply from the prevailing view that psychedelics act exclusively on membrane-bound serotonin receptors.

In the other study, Gül Dölen and her colleagues found that although psychedelic compounds such as ketamine, psilocybin, MDMA, LSD and ibogaine bind to different receptors, they all evoke metaplasticity in the nucleus accumbens to influence social reward learning. Plus, the behavioral effects persist longer for drugs that cause a longer acute subjective experience. This suggests that the “trip” may be important for therapeutic effects, which counters the current effort by many labs to develop non-hallucinogenic psychedelic analogs. Overall, both studies are provocative and surprising, opening new avenues for research into the neurobiological basis of psychedelic drug action.

—Alex Kwan, professor of biomedical engineering, Cornell University

Brain-wide representations of behavior spanning multiple timescales and states in C. elegans. Atanas A.A., Kim J., Wang Z., Bueno E., Becker M., Kang D., Park J., Kramer T.S., Wan F.K. … Flavell S.W. Cell (2023)

A cellular basis for mapping behavioural structure. El-Gaby M., Harris A.L., Whittington J.C.R., Dorrell W., Bhomick A., Walton M.E., Akam T., Behrens T.E.J. Nature (2024)

Behaviors do not occur in neat categories: They are guided by continuous sensory inputs integrated with an animal’s internal states. So an established approach of characterizing how individual neurons encode distinct parts of behavior has yielded limited insights. An alternative paradigm suggests that neuronal control emerges entirely from population dynamics, but recent work hints that biology lies somewhere in between.

Steven Flavell and his team imaged comprehensive neuronal activity in behaving Caenorhabditis elegans, which they interpreted in the context of the animals’ behavioral repertoire and connectome. They developed a quantitative model that revealed how cell types represent behavioral modules, and also how neurons flexibly shift their roles according to the animal’s behavioral state. Similarly, Timothy Behrens and his colleagues collected high-density recordings from the medial frontal cortex while mice performed a behavioral task involving multiple flexible goal locations. Some neurons kept track of progress toward individual goals while also guiding future actions according to the abstract task structure.

These and other recent studies suggest that neurons are not simply tuned to stimuli or movements. Instead, they create mixed, hierarchical representations that vary according to timing and internal state. Despite this complexity, interpretable structures exist at the cellular level, especially with our knowledge of connectivity maps. This nuanced view highlights that to link neural computation with behavior—especially natural behavior—you must understand how the nervous system functions in feedback with the behavior.

—Bing Wen Brunton, professor of biology, University of Washington

Neural circuitry for maternal oxytocin release induced by infant cries. Valtcheva S., Issa H.A., Bair-Marshall C.J., Martin K.A., Jung K., Zhang Y., Kwon H.B., Froemke R.C. Nature (2023)

Oxytocin receptor is not required for social attachment in prairie voles. Berendzen K.M., Sharma R., Mandujano M.A., Wei Y., Rogers F.D., Simmons T.C., Seelke A.M.H., Bond J.M. … Shah N.M., Manoli D.S. Neuron (2023)

For decades, researchers have examined the link between the neuropeptide oxytocin and social behavior and dysfunction. Yet we are only beginning to understand the precise neural mechanisms. In one seminal study, Robert Froemke and his colleagues discovered the neural circuit mechanism by which mouse infant cries trigger oxytocin release in mothers—in other words, how the neuropeptide links auditory social cues to maternal neuroendocrine processes essential for pup survival. Pup calls activate oxytocin-releasing neurons in the paraventricular nucleus of maternal mice via the posterior intralaminar thalamus. This circuit forms part of a larger maternal physiological loop connecting infant cries to oxytocin-mediated lactation in mammals.

In another milestone study, Devanand Manoli and his colleagues showed that in the absence of oxytocin receptors, other neurobiological systems can fill in for oxytocin’s essential social functions. Remarkably, monogamous prairie voles that don’t have oxytocin receptors still showed typical partner preference—a gold-standard measure of social bonding in this species—and normal parental behaviors. Though unexpected, this finding underscores the brain’s incredible plasticity in safeguarding the functions of this crucial neuropeptide. Together, these findings emphasize the evolutionary importance of oxytocin: It is tightly integrated to couple external social cues with vital neuroendocrine responses, and its functions are so essential that the brain can work to compensate for them.

—Steve Chang, associate professor of psychology and neuroscience, Yale University

Neural circuit basis of placebo pain relief. Chen C., Niehaus J.K., Dinc F., Huang K.L., Barnette A.L., Tassou A., Shuster S.A., Wang L., Lemire A. … Scherrer G. Nature (2024)

Cardiogenic control of affective behavioural state. Hsueh B., Chen R., Jo Y., Tang D., Raffiee M., Kim Y.S., Inoue M., Randles S., Ramakrishnan C. … Deisseroth K. Nature (2023)

Philosophers have grappled with the mind-body problem, which asks how the conscious mind relates to the physical body, for centuries. Two recent papers tackle that connection head-on. In their 2024 paper, Grégory Scherrer and his colleagues developed a behavioral assay that generates placebo-like anticipatory pain relief in mice. They identified a specific neural circuit projecting from the rostral anterior cingulate cortex to the pontine nucleus that processes these placebo-like feelings. The expectation of pain relief boosted the activity of these neurons and reduced pain. Their findings show that pain perception is not simply a passive readout of a physical sensation but an active, top-down process that can be modulated by cognition.

The 2023 work from Karl Deisseroth and his colleagues marks a crucial step in empirically testing whether emotional states are driven by the brain or the body. The team developed an advanced noninvasive optogenetic pacemaker to manipulate cardiac rhythms and probe whether information flowing from the body drives emotions. An artificially increased heart rate directly influenced anxiety-like behavior in mice, but the effect depended on concurrent brain activity. These findings suggest that emotion is not a localized event tied to one group of cells but rather an emergent property dependent on the coordination of multiple systems.

Together these papers make one thing clear: If we want to move the ancient mind-body debate from the abstract to the verifiable, we must continue to develop the tools.

—M. Maya Kaelberer, assistant professor of physiology at the University of Arizona

Mapping dysfunctional circuits in the frontal cortex using deep brain stimulation. Hollunder B., Ostrem J.L., Sahin I.A., Rajamani N., Oxenford S., Butenko K., Neudorfer C., Reinhardt P., … Li N., Horn A. Nature Neuroscience (2024)

Augmenting hippocampal-prefrontal neuronal synchrony during sleep enhances memory consolidation in humans. Geva-Sagiv M., Mankin E.A., Eliashiv D., Epstein S., Cherry N., Kalender G., Tchemodanov N., Nir Y., Fried I. Nature Neuroscience (2023)

Historically, brain stimulation studies have mostly focused only on the area that is being stimulated. These papers zoom out and think about the brain network being engaged by stimulation.

In their 2024 paper, Andreas Horn and his colleagues conducted a large cohort analysis of deep brain stimulation-implanted patients across four different conditions—Parkinson’s disease, dystonia, obsessive-compulsive disorder and Tourette syndrome. Based on each implant location, the team inferred which white-matter streamlines are likely to be activated and estimated which cortical areas were thus likely impacted by the stimulation. Using this information and patient outcomes, they established which downstream cortical targets were most helpful in different situations (for example, primary motor cortex for dystonia, supplementary motor cortex for Parkinson’s disease). In 2023, Yuval Nir and Itzhak Fried’s team investigated the role of synchronized stimulation between two brain areas in enhancing memory. Specifically, they showed participants a visual stimulus in the evening, performed prefrontal stimulation time-synchronized to temporal lobe activity during sleep, and then tested their recall in the morning. This closed-loop stimulation enhanced memory more than unsynchronized stimulation.

These papers exemplify how the field is moving toward evaluating stimulation through the lens of white-matter connectivity between brain areas and the networks being engaged by the stimulation. This movement in the field is important both for foundational science and for potential clinical applications, including one discussed in our 2023 paper investigating the effects of deep brain stimulation of a target white-matter network for treatment-resistant depression.

—Christopher Rozell, director, Institute for Neuroscience, Neurotechnology and Society at the Georgia Institute of Technology