Artificial intelligence
Recent articles
How to collaborate with AI
To make the best use of LLMs in research, turn your scientific question into a set of concrete, checkable proposals, wire up an automatic scoring loop, and let the AI iterate.
How to collaborate with AI
To make the best use of LLMs in research, turn your scientific question into a set of concrete, checkable proposals, wire up an automatic scoring loop, and let the AI iterate.
Tomaso Poggio on his quest for theories to explain the fundamental learning abilities of brains and machines
Thus far, engineering has outpaced theory in the science of intelligence. But Poggio is hopeful that theories can catch up.
Tomaso Poggio on his quest for theories to explain the fundamental learning abilities of brains and machines
Thus far, engineering has outpaced theory in the science of intelligence. But Poggio is hopeful that theories can catch up.
The Transmitter’s favorite essays of 2025
Throughout a tumultuous year in science, researchers opined on policy changes and funding uncertainty, as well as scientific trends and the impact of artificial-intelligence tools on the field.
The Transmitter’s favorite essays of 2025
Throughout a tumultuous year in science, researchers opined on policy changes and funding uncertainty, as well as scientific trends and the impact of artificial-intelligence tools on the field.
AI-assisted coding: 10 simple rules to maintain scientific rigor
These guidelines can help researchers ensure the integrity of their work while accelerating progress on important scientific questions.
AI-assisted coding: 10 simple rules to maintain scientific rigor
These guidelines can help researchers ensure the integrity of their work while accelerating progress on important scientific questions.
Seeing the world as animals do: How to leverage generative AI for ecological neuroscience
Generative artificial intelligence will offer a new way to see, simulate and hypothesize about how animals experience their worlds. In doing so, it could help bridge the long-standing gap between neural function and behavior.
Seeing the world as animals do: How to leverage generative AI for ecological neuroscience
Generative artificial intelligence will offer a new way to see, simulate and hypothesize about how animals experience their worlds. In doing so, it could help bridge the long-standing gap between neural function and behavior.
How will neuroscience training need to change in the future?
Training in computational neuroscience, data science and statistics will need to expand, say many of the scientists we surveyed. But that must be balanced with a more traditional grounding in the scientific method and critical thinking. Researchers noted that funding concerns will also affect training, especially for people from underrepresented groups.
How will neuroscience training need to change in the future?
Training in computational neuroscience, data science and statistics will need to expand, say many of the scientists we surveyed. But that must be balanced with a more traditional grounding in the scientific method and critical thinking. Researchers noted that funding concerns will also affect training, especially for people from underrepresented groups.
How will the field’s relationship to industry change over the next decade? Will a larger neurotechnology sector emerge?
Interactions between academic neuroscience and industry will grow, and the neurotech sector will expand, most survey respondents predict. The current funding upheaval in the United States may accelerate this trend as the field searches for new funding models.
How will the field’s relationship to industry change over the next decade? Will a larger neurotechnology sector emerge?
Interactions between academic neuroscience and industry will grow, and the neurotech sector will expand, most survey respondents predict. The current funding upheaval in the United States may accelerate this trend as the field searches for new funding models.
What are the fastest-growing areas in neuroscience?
Respondents pointed to computational neuroscience, systems neuroscience, neuroimmunology and neuroimaging, among other subfields.
What are the fastest-growing areas in neuroscience?
Respondents pointed to computational neuroscience, systems neuroscience, neuroimmunology and neuroimaging, among other subfields.
What should the field prioritize over the next 10 years?
Respondents pointed to a range of challenges in basic neuroscience—such as understanding naturalistic behaviors, intelligence and embodied cognition—and called for more circuit-level research, more precise brain recordings and more work in alternative models. Just as many pushed for a translational pivot.
What should the field prioritize over the next 10 years?
Respondents pointed to a range of challenges in basic neuroscience—such as understanding naturalistic behaviors, intelligence and embodied cognition—and called for more circuit-level research, more precise brain recordings and more work in alternative models. Just as many pushed for a translational pivot.
The state of neuroscience in 2025: An overview
The Transmitter presents a portrait of the field through four lenses: its focus, its output, its people and its funding.
The state of neuroscience in 2025: An overview
The Transmitter presents a portrait of the field through four lenses: its focus, its output, its people and its funding.
Explore more from The Transmitter
How artificial agents can help us understand social recognition
Neuroscience is chasing the complexity of social behavior, yet we have not answered the simplest question in the chain: How does a brain know “who is who”? Emerging multi-agent artificial intelligence may help accelerate our understanding of this fundamental computation.
How artificial agents can help us understand social recognition
Neuroscience is chasing the complexity of social behavior, yet we have not answered the simplest question in the chain: How does a brain know “who is who”? Emerging multi-agent artificial intelligence may help accelerate our understanding of this fundamental computation.
Methodological flaw may upend network mapping tool
The lesion network mapping method, used to identify disease-specific brain networks for clinical stimulation, produces a nearly identical network map for any given condition, according to a new study.
Methodological flaw may upend network mapping tool
The lesion network mapping method, used to identify disease-specific brain networks for clinical stimulation, produces a nearly identical network map for any given condition, according to a new study.
Common and rare variants shape distinct genetic architecture of autism in African Americans
Certain gene variants may have greater weight in determining autism likelihood for some populations, a new study shows.
Common and rare variants shape distinct genetic architecture of autism in African Americans
Certain gene variants may have greater weight in determining autism likelihood for some populations, a new study shows.