[TNJC] Neuro Journal Club next week (26/11, 2pm)

[Ben Dongsung Huh]

Dear all,
In next week's TNJC, Dr. Michael Hill from CalTech will present his work. 

Title: Methodological advances and their application in a human single unit recording and microstimulation study

Abstract:
Patients suffering from pharmacologically intractable epilepsy are
regularly implanted with depth electrodes for clinical diagnostics. This
setup allows for the unique opportunity to record single and multi unit
activity as well as to stimulate intracranially in awake human
subjects. A limiting factor in such studies is the small number of
trials that can be run during any given session due to clinical
circumstances. To counterbalance this limitation we developed a new tool
to increase yield without increasing the false alarm rate in the
analysis of peri-stimulus time histograms (PSTH). The method combines an
optimised probability density function with an optimised shuffling
algorithm resulting in a significant increase in power and providing a
potentially valuable tool for any type of PSTH data. In an ongoing study
we used this method to analyse data recorded in the human orbitofrontal
cortex (OFC), the anterior cingulate cortex (ACC) and the amygdala
(AMY). During a reinforcement learning task, subjects had to pick a card
from one of two decks where one deck had a 70% and the other a 30%
chance of winning $10 or $100 randomly. Preliminary analysis from 116
OFC, 201 ACC and 167 AMY units showed value encoding for losing trials
in the OFC and reward prediction error encoding for winning trials in
the ACC. In a second, independent project we attempted to move from such
correlational findings to an investigation of causal relationships. With
this goal in mind we designed and executed the first ever systematic
microstimulation study in humans. So far we have applied
microstimulation in the amygdala of four patients during three different
behavioral tasks; however, we were not able to show any behavioral
effects resulting from stimulation. Nonetheless, the successful proof of
concept for safe application of microstimulation in humans may provide a
useful tool for future investigations. We are also hopeful that in
future sessions we may still find a behavioural effect in response to
microstimulation of the human amygdala.

You can also see the past/future TNJC list here (http://www.gatsby.ucl.ac.uk/tnjc/).
See you then!

Ben Dongsung Huh
Gatsby Computational Neuroscience Unit