Origins of Neuroscience – A history of explorations into Brain Function
It is available online.. Fascinating book on how we have come to know the brain as we see it now(Not really now..)
I don’t know how long will it available for free(Thanks to Michael for the pointer).
Recently there have been many articles about Mind reading devices, based on fMRI. Though there are many ethical issues around such a device. But lets us imagine such a device exists, Is it possible to find out what we really want or hate? Maybe..
An excerpt from the release at Princeton University
A team of scientists from Princeton University has devised a new experimental technique that produces some of the best functional images ever taken of the human brainstem, the most primitive area of the brain.
This region of the brain is sometimes referred to as Reptilian Brain, based on evolutionary view of looking at the brain regions. It is involved in many important human brain functions required for survival including decision making.
Some of the difficulties in looking at this region of the brain are its size, movement effects due to heart pulse, existing experimental and analysis methods.
“For a long time, scientists have tried looking at this area of the brain and have been unsuccessful — it’s just too small,” said Kimberlee D’Ardenne, the lead author on the paper.
The article by Kimberlee D’Ardenne, Samuel M. McClure, Leigh E. Nystrom, and Jonathan D. Cohen appeared in February Issue of Science Magazine. They overcome the above difficulties combining a number of techniques including synchronizing the fMRI with heart rate, using novel pulse sequences, novel algorithms for processing the data.
They look at brain activity of thirsty humans when drops of juice/water/money are given and find that the activity in the brain stem encodes positive reward prediction error, error when an unexpected reward is given, and the probability of getting any reward. But they couldn’t find the negative reward prediction error, error when expected reward is not given.
So they can find what you want but not what you hate. But the take home is the ability to look at brain stem activity.
D’Ardenne K, McClure SM, Nystrom LE, Cohen JD (2008) BOLD responses reflecting dopaminergic signals in the human ventral tegmental area. Science 319: 1264-1267 link
We take decisions all the time from the most simple(food) to the most complex(invest, career, friends).
Ever wondered
One of the decision making theories is based on learning to decide better simply based on rewards and punishments. It has been extensively studied in machine learning, statistics, control communities under the guise of Reinforcement Learning. Temporal Difference learning is one of the simplest models of reinforcement learning used to explain learning from rewards. The learning in the model is based on difference in the expected future reward and actual reward, Temporal Prediction Error.
Decision making behavior of animals was extensively studied and models of their behavior were proposed. Conditioned response to an object in the presence/absence of reward was one of the widely studied behaviors. When animals were repeatedly presented by the object followed by reward, animals started expecting reward on presenting the object.
Models of animal behavior were developed conditioned on just the object and actions taken by the animal. Rescorla-Wagner(RW) model was one of the models used to explain the conditioned reward animal behavior. This model is based on increasing/decreasing the association strength between reward and preceding object presentation. Come to think of it temporal difference learning is very similar to RW model and maybe its offshoot.
Significant amount of work in understanding decision making regions of the brain started after the seminal paper by Wolfram Schultz, Peter Dayan and Read Montague which showed that Dopamine neurons(DA) in Ventral Tegmental Area(VTA) encode the temporal prediction error.
This has led to a new field of research – Neuro-economics – for studying the decision-making aspects of the brain.
Schultz, W., Dayan, P., Montague, P.R. (1997) A Neural substrate of prediction and reward. Science, 275: 5306, 1593-1599. link
R&D 