Abstract
Hunger, thirst, loneliness and ambition determine the reward value of food, water, social interaction and performance outcome1. Dopamine neurons respond to rewards meeting these diverse needs2,3,4,5,6,7,8, but it remains unclear how behaviour and dopamine signals change as priorities change with new opportunities in the environment. One possibility is that dopamine signals for distinct drives are routed to distinct dopamine pathways9,10. Another possibility is that dopamine signals in a given pathway are dynamically tuned to rewards set by the current priority. Here we used electrophysiology and fibre photometry to test how dopamine signals associated with quenching thirst, singing a good song and courting a mate change as male zebra finches (Taeniopygia guttata) were provided with opportunities to retrieve water, evaluate song performance or court a female. When alone, water reward signals were observed in two mesostriatal pathways but singing-related performance error signals were routed to Area X, a striatal nucleus specialized for singing. When courting a female, water seeking was reduced and dopamine responses to both water and song performance outcomes diminished. Instead, dopamine signals in Area X were driven by female calls timed with the courtship song. Thus the dopamine system handled coexisting drives by routing vocal performance and social feedback signals to a striatal area for communication and by flexibly re-tuning to rewards set by the prioritized drive.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The data that support the findings of this study are available at https://osf.io/dcugn/. Source data are provided with this paper.
Code availability
The data collection and analysis code used for this study are available at https://osf.io/dcugn/.
References
Maslow, A. H. A theory of human motivation. Psychol. Rev. 50, 370 (1943).
Schultz, W. Behavioral dopamine signals. Trends Neurosci. 30, 203–210 (2007).
Tsutsui-Kimura, I. et al. Distinct temporal difference error signals in dopamine axons in three regions of the striatum in a decision-making task. eLife 9, e62390 (2020).
Solié, C., Girard, B., Righetti, B., Tapparel, M. & Bellone, C. VTA dopamine neuron activity encodes social interaction and promotes reinforcement learning through social prediction error. Nat. Neurosci. 25, 86–97 (2022).
Hu, R. K. et al. An amygdala-to-hypothalamus circuit for social reward. Nat. Neurosci. 24, 831–842 (2021).
Dai, B. et al. Responses and functions of dopamine in nucleus accumbens core during social behaviors. Cell Rep. 40, 111246 (2022).
Xie, Y., Huang, L., Corona, A., Pagliaro, A. H. & Shea, S. D. A dopaminergic reward prediction error signal shapes maternal behavior in mice. Neuron 111, 557–570 (2023).
Gadagkar, V. et al. Dopamine neurons encode performance error in singing birds. Science 354, 1278–1282 (2016).
Lee, R. S., Engelhard, B., Witten, I. B. & Daw, N. D. A vector reward prediction error model explains dopaminergic heterogeneity. Preprint at bioRxiv https://doi.org/10.1101/2022.02.28.482379 (2022).
Menegas, W., Akiti, K., Amo, R., Uchida, N. & Watabe-Uchida, M. Dopamine neurons projecting to the posterior striatum reinforce avoidance of threatening stimuli. Nat. Neurosci. 21, 1421–1430 (2018).
Skals, N., Anderson, P., Kanneworff, M., Löfstedt, C. & Surlykke, A. Her odours make him deaf: crossmodal modulation of olfaction and hearing in a male moth. J. Exp. Biol. 208, 595–601 (2005).
Allen, W. E. et al. Thirst regulates motivated behavior through modulation of brainwide neural population dynamics. Science 364, 253 (2019).
Lammel, S., Lim, B. K. & Malenka, R. C. Reward and aversion in a heterogeneous midbrain dopamine system. Neuropharmacology 76, 351–359 (2014).
Schultz, W., Dayan, P. & Montague, P. R. A neural substrate of prediction and reward. Science 275, 1593–1599 (1997).
Houk, J., Adams, J. & Barto, A. in Models of Information Processing in the Basal Ganglia (eds Houk, J. C., Davis, J. L. & Beiser, D. G.) 249–270 (MIT Press, 1995).
Sutton, R. S. & Barto, A. G. Reinforcement Learning: An Introduction (MIT Press, 1998).
Iyengar, S., Viswanathan, S. S. & Bottjer, S. W. Development of topography within song control circuitry of zebra finches during the sensitive period for song learning. J. Neurosci. 19, 6037–6057 (1999).
Person, A. L., Gale, S. D., Farries, M. A. & Perkel, D. J. Organization of the songbird basal ganglia, including Area X. J. Comp. Neurol. 508, 840–866 (2008).
Bottjer, S. W., Brady, J. D. & Cribbs, B. Connections of a motor cortical region in zebra finches: relation to pathways for vocal learning. J. Comp. Neurol. 420, 244–260 (2000).
Karten, H. J. et al. Digital atlas of the zebra finch (Taeniopygia guttata) brain: a high‐resolution photo atlas. J. Comp. Neurol. 521, 3702–3715 (2013).
Zann, R. A. The Zebra Finch: A Synthesis of Field and Laboratory Studies, Vol. 5 (Oxford Univ. Press, 1996).
Theunissen, F. E. & Shaevitz, S. S. Auditory processing of vocal sounds in birds. Curr. Opin. Neurobiol. 16, 400–407 (2006).
Woolley, S. C. & Woolley, S. M. in The Neuroethology of Birdsong, Vol. 71 (eds Sakata, J. et al.) 127–155 (2020).
Sakata, J. T. & Brainard, M. S. Social context rapidly modulates the influence of auditory feedback on avian vocal motor control. J. Neurophysiol. 102, 2485–2497 (2009).
Duffy, A., Latimer, K. W., Goldberg, J. H., Fairhall, A. L. & Gadagkar, V. Dopamine neurons evaluate natural fluctuations in performance quality. Cell Rep. 38, 110574 (2022).
Tumer, E. C. & Brainard, M. S. Performance variability enables adaptive plasticity of ‘crystallized’ adult birdsong. Nature 450, 1240–1244 (2007).
Ali, F. et al. The basal ganglia is necessary for learning spectral, but not temporal, features of birdsong. Neuron 80, 494–506 (2013).
Hoffmann, L. A., Saravanan, V., Wood, A. N., He, L. & Sober, S. J. Dopaminergic contributions to vocal learning. J. Neurosci. 36, 2176–2189 (2016).
Chen, R. et al. Songbird ventral pallidum sends diverse performance error signals to dopaminergic midbrain. Neuron 103, 266–276.e264 (2019).
Mohebi, A. et al. Dissociable dopamine dynamics for learning and motivation. Nature 570, 65–70 (2019).
Singh Alvarado, J. et al. Neural dynamics underlying birdsong practice and performance. Nature 599, 635–639 (2021).
Appeltants, D., Absil, P., Balthazart, J. & Ball, G. F. Identification of the origin of catecholaminergic inputs to HVc in canaries by retrograde tract tracing combined with tyrosine hydroxylase immunocytochemistry. J. Chem. Neuroanat. 18, 117–133 (2000).
Ben-Tov, M., Duarte, F. & Mooney, R. A neural hub for holistic courtship displays. Curr. Biol. 33, 1640–1653.e1645 (2023).
Hernandez, A. M., Perez, E. C., Mulard, H., Mathevon, N. & Vignal, C. Mate call as reward: acoustic communication signals can acquire positive reinforcing values during adulthood in female zebra finches (Taeniopygia guttata). J. Comp. Psychol. 130, 36 (2016).
Tokarev, K. et al. Sexual dimorphism in striatal dopaminergic responses promotes monogamy in social songbirds. eLife 6, e25819 (2017).
Schultz, W. Multiple dopamine functions at different time courses. Annu Rev Neurosci 30, 259–288 (2007).
Smith, K. S., Berridge, K. C. & Aldridge, J. W. Disentangling pleasure from incentive salience and learning signals in brain reward circuitry. Proc. Natl Acad. Sci. USA 108, E255–E264 (2011).
Riters, L. V. The role of motivation and reward neural systems in vocal communication in songbirds. Front. Neuroendocrinol. 33, 194–209 (2012).
Acknowledgements
The authors thank A. Mohebi, N. Uchida, B. Ito and members of the Goldberg laboratory for comments; A. Raha for artwork; Z. Zhao for statistical advice; and A. Enzerink and A. Podury for technical assistance. V.G. was supported by a Simons Foundation Postdoctoral Fellowship and a NIH/NINDS Pathway to Independence Award (grant no. K99/R00NS102520), P.A.P. was supported by NIH/NINDS (grant no. F32NS098634), and J.H.G. was supported by NIH/NINDS (grant no. R01NS094667).
Author information
Authors and Affiliations
Contributions
A.R., V.G. and J.H.G. designed the research, analysed data and wrote the paper. A.R., V.G., A.D., P.A.P. and B.K. performed experiments.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature thanks the anonymous reviewer(s) for their contribution to the peer review of this work. Peer review reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Histological verification of fiber placement in Area X and MST.
a, Left: Locations of all Area X fiber implants (n = 19 hemispheres), implanted 1.5 mm lateral to midline. Right: sagittal slices of two hemispheres (Area X, white circle). Scale bars: 250 um. Left: Schematic from the ZEBrA database (http://www.zebrafinchatlas.org)20. b, Sites of MST fibers (n = 9 hemispheres), implanted 0.7 mm lateral to midline and 1 mm posterior to the Area X implant. Left: Schematic from the ZEBrA database (http://www.zebrafinchatlas.org)20. Right: sagittal slices of two MST hemispheres. Scale bars: 250 um. c, Schematic of the songbird auditory system, including forebrain projections to VTA from dorsal arcopallium (AD), ventral intermediate arcopallium (AIV), and ventral pallidum (VP). Schematic adapted from19,22,23,29. Note that Area X and MST fibers were several millimeters anterior this auditory pathway. d, Injection sites of the retrograde tracer cholera toxin beta subunit (CTB) into Area X (left, magenta) and MST (right, green) into the same hemisphere (independently repeated in n = 3 hemispheres; n = 2 birds; Methods). Scale bars: 250 um. e, Zebra finch brain schematic with pink rectangle around VTA (left). MST and Area X projecting neurons co-localize in VTA but are not co-labeled, consistent with previous findings (middle)18. Expanded view at right. Scale bars: 75 um (middle) and 50 um (right). f-g, Core nuclei of the song system project to Area X, but not MST (independently repeated in n = 3 hemispheres; n = 2 birds). f, Schematic of zebra finch brain, with green rectangle surrounding HVC (left). HVC contains Area X projecting neurons (magenta, middle), but no MST projecting neurons in the same hemisphere (right, green), consistent with past work18. Scale bars: 75 um. g, Schematic of zebra finch brain, with blue rectangle surrounding LMANcore (left). LMANcore contains Area X projecting neurons (magenta, middle), but no MST projecting neurons in the same hemisphere (right, green), consistent with past work18,19 (middle and right; approximate boundary of LMANcore denoted by white circle). Scale bars: 75 um.
Extended Data Fig. 2 Reward prediction error signals exist in Area X and MST.
a, Single trial DA responses from one Area X hemisphere in response to a water reward (top) or reward omission (bottom), aligned to spout contacts following ‘reward’ light cues. b, Average z-scored ΔF/F signals across 9 Area X hemispheres, example hemisphere in a indicated by black arrow (top), average across all hemispheres (mean ± SEM; bottom). c-d, Data plotted as in a-b for 9 MST hemispheres. e–f, Reward signals were larger and faster in Area X compared to MST. e, Scatterplots showing average values across all 9 Area X and 9 MST hemispheres (mean ± SD, black) and mean z-scored values for each hemisphere (gray) for reward and reward omission in the 170-500 ms window following spout contact (Area X, reward vs. Area X, no reward: P = 0.0039; Area X, reward vs. MST, reward: P = 1; MST, reward vs. MST, no reward: P = 0.0039; Area X, no reward vs. MST, no reward: P = 0.063; paired, two-sided Wilcoxon signed rank test). f, Latency to peak (rewarded trials) or dip (omission trials) of the z-scored response in the 170–500 ms window after spout contact (mean ± SD, black; single hemispheres, gray; only responses that were greater than a z-score value of 2 within the window were included; Area X, reward vs MST, reward: P = 0.00033; Area X, no reward vs MST, no reward: P = 0.071; unpaired, two-sided Wilcoxon rank sum test). ** P < 0.01, *** P < 0.001, n.s. not significant.
Extended Data Fig. 3 Singing reduces dopaminergic responses to ‘no-reward’ light cues in both Area X and MST.
a, Example ‘no-reward’ light cue trial. b, Water availability under four behavioral conditions. c, Water retrieval probability for ‘no-reward’ (black) and ‘reward’ (gray) light cues across behavioral conditions (Area X implanted bird: triangles; MST implanted bird: circles; n.s.; generalized linear mixed effects model with post-hoc contrast tests; Methods). d, Brain schematics showing recording sites. e-f, DA responses in Area X to the ‘no-reward’ light when the bird was alone. e, DA responses for single trials from a single Area X hemisphere during not singing (top) or singing (bottom) conditions. f, Average z-scored ΔF/F signals across 9 Area X hemispheres, example hemisphere in e indicated by black arrow (top), average across all hemispheres (mean ± SEM; bottom). g-h, Data plotted as in e-f when in the presence of a female. i, Average values across all Area X hemispheres (mean ± SD, black) and mean z-scored values for each hemisphere (gray; n = 9 Area X hemispheres) for the four behavioral conditions in e–h (not singing, alone vs singing, alone: P = 0.00051; not singing, alone vs. not singing, with female: P = 0.98; singing, alone vs. singing with female: P = 0.99; not singing, with female vs. singing, with female: P = 0.00060; 2-way ANOVA and post-hoc Tukey; Methods). j–n, Data plotted as in e-i for DA recordings in 9 MST hemispheres. n, Average values across all MST hemispheres (mean ± SD, black) and mean z-scored values for each hemisphere (n = 9 hemispheres; gray) for the four behavioral conditions in j–m (not singing, alone vs singing, alone: P = 0.0015; not singing, alone vs. not singing, with female: P = 0.27; singing, alone vs. singing, with female: P = 0.91; not singing, with female vs. singing, with female: P = 0.025; 2-way ANOVA and post-hoc Tukey; Methods). * P < 0.05, ** P < 0.01, *** P < 0.001, n.s. not significant.
Extended Data Fig. 4 Modulation of singing-related error signals during courtship is also observed at the level of VTAx spikes.
a, Spectrograms, single trial spiking activity, and raster plots for undistorted (top) and distorted (bottom) renditions recorded in a single antidromically identified VTAx neuron of a bird singing alone, plotted above firing rate histograms (plots aligned to target onset; blue: undistorted; red: distorted). b, Z-scored average from 8 antidromically identified VTAx neurons aligned to undistorted (top) and distorted (bottom) renditions, black arrow indicates example neuron shown in a; bottom: average z-scored response (mean ± SEM). c, Data plotted as in a for the same VTAx neuron recorded during courtship singing. d, Data plotted as in b for the same VTAx neurons recorded during courtship singing. e-f, Scatter plots (gray, n = 8 neurons) and mean ± SD (black) of average z-scored firing rate in the 50–125 ms following undistorted (e; P = 0.026; 2-way ANOVA and post-hoc Tukey; Methods) and distorted (f; P = 0.056; 2-way ANOVA and post-hoc Tukey; Methods) renditions during alone versus female directed singing across all VTAx neurons. g, Z-scored error responses for each neuron (gray; n = 8) and mean ± SD (black) when birds sang alone versus to females (P = 0.008; paired, two-sided Wilcoxon signed rank test). Horizontal bars in histograms (a, c) indicate significant deviations from baseline, P < 0.05 for a one-sided z test. Scale bars in a and c for spiking activity is 1 mV. * P < 0.05, ** P < 0.01.
Extended Data Fig. 5 Singing-related performance error signals are not observed in MST during courtship singing.
a, Spectrograms and single trial DA responses for undistorted (top) and distorted (bottom) renditions recorded in a single MST hemisphere of a bird singing alone, plotted above average ΔF/F signals (different hemisphere from Fig. 2c; plots aligned to target onset; blue: undistorted; red: distorted). b, Z-scored average from 9 MST hemispheres aligned to undistorted (top) and distorted (bottom) renditions, black arrow indicates example hemisphere shown in a; bottom: average z-scored response (mean ± SEM). c, Data plotted as in a for the same hemisphere measured during courtship singing. d, Data plotted as in b for the same hemispheres recorded during courtship singing. e-f, Scatter plots of average across all hemispheres (mean ± SD, black) and mean z-scored ΔF/F value for each hemisphere (gray) in the 150–300 ms following undistorted (e) and distorted (f) renditions in MST during alone versus female directed singing. g, Z-scored error responses when birds sang alone versus to females (mean ± SD, black; single hemispheres, gray). n.s. not significant for a 2-way ANOVA and post-hoc Tukey (e and f; Methods) and for a paired two-sided Wilcoxon signed rank test (g); n = 9 MST hemispheres.
Extended Data Fig. 6 Baseline levels of DA release, but not VTA spiking, increase during both alone and female directed singing.
a, Spectrograms and single trial DA signals for song onsets (top) and offsets (bottom) recorded in a single Area X hemisphere of a bird singing alone, plotted above average ΔF/F signals (purple and orange aligned to song onset or offset, respectively). b, Z-scored average from 18 Area X hemispheres aligned to song onsets (top) and offsets (bottom), black arrow indicates example hemisphere from a; bottom: average z-scored signals (mean ± SEM). c-d, Data plotted as in a-b for the same hemisphere measured during courtship singing. e-f, Mean z-scored ΔF/F value for each Area X hemisphere (gray) in the 0-500 ms window following song onset and offset during alone (e; P = 2.0E-4; paired, two-sided Wilcoxon signed rank test) and female directed singing (f; P = 3.9E-4; paired, two-sided Wilcoxon signed rank test) (black: mean ± SD across all hemispheres). g–l, Data plotted as in a–f for 9 MST hemispheres. k, P = 0.039; paired, two-sided Wilcoxon signed rank test. m, Spectrograms, single trial spiking activity, and raster plots for song onset (top) and offset (bottom) recorded in a single VTAx neuron of a bird singing alone, plotted above firing rate histograms (plots aligned to song onset or offset). n, Z-scored firing rate histograms from 8 VTAx neurons aligned to song onset (top) and offset (bottom), black arrow indicates example neuron shown in m; bottom: average z-scored firing rate (mean ± SEM). o-p, Data plotted as in m-n for the same VTAx neurons recorded during courtship singing. q-r, Scatter plots (gray) and mean ± SD (black) of average z-scored firing rate in the 0-500 ms window following song onset and offset during alone (q) and female directed singing (r) across all VTAx neurons. s–x, Data plotted as in m–r for VTAother neurons. Scale bars in m, s and u for spiking activity is 1 mV, and for o is 0.25 mV. * P < 0.05, *** P < 0.001, n.s. not significant for a paired two-sided Wilcoxon signed rank test; n = 18 Area X and n = 9 MST hemispheres (e,f,k,l) and n = 8 VTAx and n = 14 VTAother neurons (q,r,w,x).
Extended Data Fig. 7 The cued moment of female appearance evokes phasic, but not sustained, activation of DA release.
a, Spectrogram and single trial Area X DA responses to female appearance during the transition between alone and with-female conditions, plotted above average ΔF/F signals (plots aligned to onset of female call playback). Black boxes above spectrogram indicate female call playbacks. b, Z-scored average from 18 Area X hemispheres, black arrow indicates example hemisphere shown in a; bottom: average z-scored response (mean ± SEM). c-d, Data plotted as in a-b for DA signals recorded in 9 MST hemispheres. e, Scatter plots showing phasic response across all hemispheres (mean ± SD, black) and mean z-scored ΔF/F value for each hemisphere (gray; Area X: n = 18 and MST: n = 9 hemispheres) in the 150–300 ms window following female call onset cueing female appearance (triangles: Area X; circles: MST; P = 0.029; unpaired, two-sided Wilcoxon rank sum test). f, Scatter plots of baseline change across all hemispheres (mean ± SD, black) and mean z-scored ΔF/F value for each hemisphere (gray) in the 1–2 s window following cued female appearance (triangles: Area X; circles: MST; Methods; Area X: P = 0.017; MST: P = 0.097; one-sample t test). *P < 0.05, n.s. not significant.
Extended Data Fig. 8 Discharge statistics of VTAx DA neurons do not depend on courtship context during singing and were subtly affected during not-singing periods.
a-b, ISI distribution (a) and normalized spike train autocorrelogram (STA) (b) during singing alone (black) and female directed (green) song for a single VTAx neuron. Insets: mean ± SEM for 8 VTAx neurons. c-f, Mean firing rate (c), median ISI (d), coefficient of variation of the ISI distribution (CV ISI; e), and peak of the STA (f) for 8 VTAx neurons recorded when males sang alone and sang female-directed song. g–l, Data plotted as in a-f for the same 8 VTAx neurons during not singing periods. i, P = 0.008; paired, two-sided Wilcoxon signed rank test. j, P = 0.02; paired, two-sided Wilcoxon signed rank test. m–x, Data plotted as in a–l for 15 VTAother neurons (Methods). q, P = 0.005; paired, two-sided Wilcoxon signed rank test. * P < 0.05, ** P < 0.01, n.s not significant for a paired two-sided Wilcoxon signed rank test.
Extended Data Fig. 9 Socially irrelevant sounds and pre-recorded female calls played during courtship do not drive reliable DA responses.
a–e, Knock responses. a, Spectrograms and single trial Area X DA responses to knock sounds played during not-singing (top) or singing (bottom) periods of the courtship interaction, plotted above average ΔF/F signals (plots aligned to onset of knock) in a single hemisphere. b, Z-scored average from 15 Area X hemispheres, black arrow indicates example hemisphere shown in a; bottom: average z-scored response (mean ± SEM). c-d, Data plotted as in a-b for knock responses recorded in MST. e, Mean z-scored ΔF/F values in the 0.15–0.3 s window following the onset of knock sounds in Area X and MST during singing and not-singing periods (gray, single hemispheres; black, mean ± SD across hemispheres; triangles: Area X; circles: MST). f–k, Random DAF responses. f–j, Data plotted as in a–e for Area X and MST responses to the sound of DAF played randomly throughout the courtship interaction, including during singing and not-singing periods (Area X: n = 10; MST: n = 9 hemispheres). j, Area X, not singing vs. MST, not singing: P = 0.0041; unpaired, two-sided Wilcoxon rank sum test. k, Latencies to peak responses to random DAF (gray, single hemispheres; black, mean ± SD across hemispheres; Methods; Area X, not singing vs. Area X, singing: P = 0.0085; paired, two-sided Wilcoxon signed rank test). l–q, Pre-recorded female calls. Data plotted as in f–k for Area X and MST responses to the sound of pre-recorded female calls played through a speaker at random times throughout a live courtship interaction, including during singing and not-singing periods (Area X: n = 14; MST: n = 9 hemispheres). Note that playback of these calls did not evoke reliable DA responses in Area X or MST during song. p, Area X, female appear vs. MST, female appear: P = 0.029; unpaired two-sided Wilcoxon rank sum test; MST, female appear vs. MST, not singing: P = 0.040; paired two-sided Wilcoxon signed rank test). q, Area X, not singing vs. Area X, singing: P = 3.6E-4; paired two-sided Wilcoxon signed rank test). * P < 0.05, ** P < 0.01, *** P < 0.001, n.s. not significant for a paired two-sided Wilcoxon signed rank test (within Area X and within MST e,j,p) and for an unpaired two-sided Wilcoxon rank sum test (k,q; across Area X and MST in e,j,p).
Supplementary information
Supplementary Information
Supplementary text and references regarding distorted auditory feedback (DAF).
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Roeser, A., Gadagkar, V., Das, A. et al. Dopaminergic error signals retune to social feedback during courtship. Nature 623, 375–380 (2023). https://doi.org/10.1038/s41586-023-06580-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586-023-06580-w
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
