Meet the Expert: Haran Shani-Narkiss

Haran uses Deep Neural Networks to better understand how defensive behaviours are generated in the midbrain. He completed his PhD in the computational neuroscience programme at ELSC, at the Hebrew University of Jerusalem.

Our team caught up with him recently to find out more about his work. 

What is the driving force behind your passion for neuroscience?

Answer: I believe that neuroscience holds the key for understanding humans. I am a musician by background and then I studied psychology. I was always interested in human emotions, behaviour and thoughts. When I was 18 years old, I worked with young people from underprivileged backgrounds and that inspired me to try to understand human behaviour. I figured out that neuroscience would get me the closest to understanding what motivates us and why we act like we do.

What area of your research most excites you and why?

Answer: During my PhD I worked on neurons that are newly generated in the adult brain. I studied the impact of these adult-born neurons on existing circuits in the brain, particularly with regards to olfaction in mice. Although it was fascinating, I saw that with the increasingly sophisticated data collection methods, more effort needed to be invested in understanding all this complex, rich data. Thus I decided to focus on modelling and theory as a postdoc. 

I am very enthusiastic about deep learning and deep neural networks, which are at the forefront of artificial intelligence algorithms. My aim now is to use these algorithms to model the transition from neural activity to behaviour, particularly defensive behaviour, from the data collected by the Branco Group at SWC.

These algorithms are very powerful and are capable of modelling the most complicated relationships. However, real understanding of how they establish this is often lacking. Throughout my time at the Sainsbury Wellcome Centre at UCL, I will try to use cutting-edge methods and develop new ones to better understand how these algorithms do what they do and thus understand how brain activity generates behaviour. 

Can you tell us what interests you about new technologies in neuroscience and the part this plays in your research?

Answer: New recording technologies are acquiring superb quality data at a rapid pace. But there is a paradox in brain science in that the brain needs to understand itself using only itself. It is not even clear if this is possible – it may be that the capacity needed to understand the brain might be too much for the brain to provide. This was stated beautifully in 1911 in the Devil’s Dictionary: “MIND, n. A mysterious form of matter secreted by the brain. Its chief activity consists in the endeavor to ascertain its own nature, the futility of the attempt being due to the fact that it has nothing but itself to know itself with.”  

I think that one of the most exciting things in our era is that we might be able to use artificial intelligence and deep learning algorithms to model processes of neural coding and the translation to behaviour. These algorithms will reflect the computations that are being conducted by the brain and so will provide insights and understanding that was previously not possible.

One thing that really inspired me as a young student when I studied biology is a famous figure¹that shows the amount of energy that every animal needs to invest for every mile that is passes. Humans are bad at this as we have to invest a lot of energy for every mile we travel. However, when you add to this graph a human on a bicycle then we go right to the top of the list. In a similar way, I think deep learning and artificial intelligence might provide us with a tool that can help us to excel on our mission to understand the brain. AI could act as an aid to improve our mental abilities just like the wheel improved our mobility.

What’s your next big challenge in terms of your research?

Answer: My next big challenge will be to have a successful collaboration with my new colleagues here at the Sainsbury Wellcome Centre at UCL and to construct deep learning algorithms that will model the way the brain generates defensive behaviour. 

What do you hope will be the key implications of your research?

Answer: One of the privileges we have as a scientists is studying things for our own curiosity. I am interested in expanding the knowledge of how the brain generates defensive behaviour. This is an innate behaviour that is common to all species, as all animals need to escape or learn to ignore a threat in the environment in an adaptive manner. Satisfying that curiosity is what drives me.

However, I do think it is not too hard to imagine how many implications this research will have for people with conditions like Post Traumatic Stress Disorder (PTSD) for example. This research could be important in better understanding how these conditions arise and how we can help people.

Secondly, I hope that it could be one of several pioneering research studies that are utilising deep neural networks to gain significant insight in neuroscience. Thirdly, I hope to establish a collaborative pipeline of interdisciplinary work between different researchers pursuing a shared goal.

Did you always know that you wanted to follow a career in research?

Answer: Not at all. If you’d told me twenty years ago I would have a career in research, I would have thought you were lying! When I was younger, I saw myself as an artist and I studied music. As an outsider, science looked grey and boring and I didn’t realise how similar arts and science are. Despite a career in research, I still consider myself a musician and I still create music.

What motivates you at the start of each day?

Answer: Nothing motivates me more than the love for my children. If you ask specifically about what motivates me to do science – it is all about the passion to solve important riddles.

What changes do you think are needed for the future of neuroscience? 

Answer: I think we have reached the point where to improve our findings and insight, we have got to be able to collaborate more. Science can often a feel a very lonely place and people are used to working on their own. We all talk about the importance of interdisciplinary research but in reality it is not happening enough. We need to speak to each other as much as possible and keep investing in common pipelines for successful collaborations. We need to acknowledge that we, scientists, are a group with a shared interest, enjoy the things we do and always remember how fortunate we are. 

Further information

References
1.    Wilson, S. S. “BICYCLE TECHNOLOGY.” Scientific American, vol. 228, no. 3, 1973, pp. 81–91. JSTOR, http://www.jstor.org/stable/24923004. Accessed 23 Nov. 2022.

Contact

• Staff profile
• Follow Haran on Twitter - @HShaniNarkiss
• Connect with Haran on LinkedIn
• Branco Lab at SWC
• Sainsbury Wellcome Centre

Interview by April Cashin-Garbutt