Dr Ben Seymour, Computational and Biological Learning Lab, Trumpington Street, Cambridge CB2 1PZ


Center for Information and Neural Networks, National Institute for Information and Communications Technology (NICT), 1-3 Yamadaoka, Suita, Osaka 565-0871, Japan.

bjs49 AT / seymour AT

F1000 reviews (paywall)



Neurotechnology Science Cafe - Monday 18th November

Join us for an evening discussing the future of neurotechnology. This event is being held in conjunction with the British Consulate General at the Knowledge Salon at the Grand Front Osaka.

Wako Yoshida (ATR) will host a line of speakers including:

"Smart houses", Jamilah Rahim, ATR 
"Mind-reading", Tomoyasu Horikawa, NAIST
"Neurofeedback", Saori Tanaka, Osaka University 
"Cognitive robotics", Eduardo Cantello, CiNet 
"Brain machine interfaces", Ben Seymour, Cambridge University
"Innovation strategies of the future: lessons from neuroscience", Kristen Cardinal, RIKEN Brain Science Institute

If you want to attend (free), please register here:



EFIC in Florence


Research into human pain should less focus on the subjective experience of pain, experts claimed at the Congress of the European Pain Federation EFIC. It was crucial to know how pain acts as a source of information, and understand how that information is processed by the brain’s neural computer. This could lead to a new generation of treatment approaches, involving, in particular, brain-machine interfaces.

Florence, 10 October 2013 – “Research into human pain has focused too much on the subjective experience of pain, and this hampered attempts to understand how the brain processes pain in health and disease”, Dr Ben Seymour (Cambridge, UK and Osaka, Japan) said today at the Congress of the European Pain Federation EFIC in Florence. “Instead, we should focus on what pain ‘does’, not what it ‘feels like’.” Together with Dr Masaki Maruyama, a neuroscientist from Kyoto, Japan, he presented a radically new view of pain.

They suggest that human pain should be conceived as the human brain’s solution to an engineering problem – the problem of identifying and minimizing future harm. “Pain is actually a biological ‘control system’ – a core harm detection and avoidance mechanism that controls and adapts our behaviour in response to harm. It is interesting to explore how the brain actually uses nociceptive information to shape future behaviour away from harm”, Dr Seymour said. Central to this view is treating the brain as a giant biological supercomputer, he said. “The key to understanding pain is to discover the underlying ‘source code’ – the basic computational commands and mathematical equations that are implemented in the brain.”

Some of these basic equations are uncovered by theoretical and human brain scanning work the experts presented at the EFIC Congress. For example they showed how the brain learns about future pain by scanning the brains of healthy volunteers whilst they perform decision-making tasks which involve mild electric shocks and heat pulses. Dr Seymour: “This has allowed us to propose a basic brain ‘circuit diagram’, describing how pain shapes our behaviour.”

“Our view is controversial as it deliberately ignores that fact that the brain is the source of the conscious feeling of pain, and ultimately it is the ‘feeling’ of pain that people in general and pain sufferers care about”, Dr Seymour said. “But we can’t understand the subjective feeling of pain without first understanding why we have evolved the pain system in the first place. This means that we have to know how pain acts as a source of information, and understand how that information is processed by the brain’s neural computer.”

The experts also claim this approach is well suited to an exciting new generation of treatment approaches likely to be available in the next few decades. “In particular, those that involve brain-machine interfaces, rather than drugs, are beginning to transform other areas of medicine such as rehabilitation, where the ability to read the ‘code’ for human movements directly from the brain can be used to control things like a wheelchair or a robotic limb”, according to Dr Seymour. “If we can understand the neural code for pain, it should in theory be possible to interfere or recode pain directly in the brain.”

Sources: EFIC Plenary Session, Seymour, How pain teaches us about harm: The neural basis of pain motivation systems; EFIC Summary Becker, Seymour et al: Brain Mechanism underlying the pain-reward interaction.


New lab members

Maria Joao is joining us as a short-term JSPS fellow, and Harry Barber is joining on a UCL Research Society Summer Studentship. See People page for more details.

We still will likely have an opening for post-doc position based at ATR from November. Please contact me if you're interested.


Post-doc position at CBL, Cambridge from 2015

There's likely to be a position for a post-doc in my lab at CBL Cambridge from approx April 2015, funded by the Wellcome Trust. Informal enquiries welcome. 


Uncertainty increases pain

Check out our new paper: - free online at the Journal of Neuroscience.

Placebo and nocebo effects are well known - and describe how predictions bias experience of pain. In a nutshell, we show that if those predictions are more uncertain, this uncertainty increases pain independently of the bias induced by the basic placebo / nocebo effect. This effect is correlated with fMRI BOLD activity in the PAG.

Uncertainty increases pain: evidence for a novel mechanism of pain modulation involving the periaqueductal gray. Yoshida W, Seymour B, Koltzenburg M, Dolan RJ. J Neurosci. 2013 Mar 27;33(13):5638-46. doi: 10.1523/JNEUROSCI.4984-12.2013. PMID: 23536078