Subcellular Computations Within The Brain During Decision-making Revealed

During decision-making, neurons in the brain are capable of much more complex processing than previously thought, new research indicates.

The findings[1] show that during decision-making, there are a multitude of small sections of dendrite throughout each neuron that process information before it is sent to other neurons. This suggests that much more complex processing can occur in the brain through these many, tiny segments of dendrite.

“Now that we have these findings and approaches, we may gain a better understanding of what’s happening in diseases that affect synaptic function and why they affect information processing in the way they do,”

said first author Aaron Kerlin, Ph.D.

Dendrites And Spines

In the study, researchers, including Dr. Kerlin, who is an assistant professor in the Department of Neuroscience and member of the Medical Discovery Team on Optical Imaging and Brain Science at the University of Minnesota Medical School, were the first to develop a microscope that rapidly images large stretches of the dendrite where neurons receive thousands of inputs from other neurons.

Targeted high-speed imaging in behaving mice.

Targeted high-speed imaging in behaving mice. A) Optical layout for high-speed, high-resolution imaging in three dimensions. An x-axis mirror galvanometer, remote focusing arm, and prism-based GDD compensation unit were added to a high resolution (NA = 1.0) resonant two photon microscope. EOM, electro-optic modulator; GDD, group delay dispersion; Res., 8 kHz resonant scanner; PR, pupil relay; Gal., galvanometer; PBS, polarizing beam splitter; QWP, quarter wave plate; RFO, remote focusing objective; VC, voice coil; DM, dichroic mirror; IO, imaging objective; PMT, photomultiplier tube. (B) Maximum intensity projections (MIP) of anatomical stack collected from Syt17-Cre x Ai93 (pia to 306 um depth) mice. Traced dendrite (purple lines) and example targets (red lines) for an example imaging session. (C) Spatial and temporal distribution of the frames that compose the example functional imaging sequences in (B). (D) Average MIP of 30 min of the functional imaging sequence shown in (B, C). (E) Close-up of the dendritic branch outlined in (D) before and after motion correction. (F–I) same as (B–E) for a layer 5 cell (MIP in (F) is pia to 560 um depth). Credit: Aaron Kerlin, et al. CC-BY

Dr. Kerlin believes that future research may include investigating patterns of local processing in neurons within mouse models of autism to determine which dendritic computations are disrupted and over what scale the disruptions occur.

He has also created a web browser that allows the entire dataset to be publicly available to other researchers, which is part of his dedication to advancing the Open Science movement. He hopes that this will encourage theorists and other researchers to make new discoveries with this rich dataset and embark on further research within this field.

[1] Aaron Kerlin, Mohar Boaz, Daniel Flickinger, Bryan J MacLennan, Matthew B Dean, Courtney Davis, Nelson Spruston, Karel Svoboda. Functional clustering of dendritic activity during decision-making. eLife 2019;8:e46966 DOI: 10.7554/eLife.46966