Evaluating methods

What are we measuring?

  • Structure
  • Activity
    • Why not function?

What is question are we asking?

  • Structure X -> Structure Y
  • Structure X -> Function Y

Strengths & Weaknesses

  • Cost
  • Invasiveness
  • Spatial/temporal resolution

Spatial resolution

Types

  • Structural
    • Anatomy
    • Connectivity/connectome
  • Activity/Functional
    • What does it do?
    • Physiology

Structural methods

Cellular

  • Cell/axon stains
  • Cellular types, distribution, concentration, microanatomy
  • Connectivity

Nissl stain – only cell bodies

Franz Nissl

Tracers

  • Retrograde (output -> input)
  • Anterograde (input -> output)

https://openi.nlm.nih.gov/imgs/512/348/3176268/3176268_1471-2105-12-351-2.png

Evaluating cellular techniques

  • Invasive (in humans post-mortem only)
  • High spatial resolution, but poor/coarse temporal
    • Why?

Mapping large-scale structures

Magnetic Resonance Imaging (MRI)

  • Magnetic resonance a property of some isotopes and complex molecules
  • Hydrogen (\(H\)), common in water & fat, is one
  • In magnetic field, \(H\) atoms absorb and release radio frequency (RF) energy
  • \(H\) atoms align with strong magnetic field

https://s.hswstatic.com/gif/mri-steps.jpg

  • Applying RF pulse perturbs alignment
  • Rate/timing of realignment varies by tissue
  • Realignment gives off radio frequency (RF) signals
  • Strength of RF ~ density of \(H\) (or other target)
  • K-space (frequency/phase) -> anatomical space

Structural MRI

  • Tissue density/type differences
  • Gray matter (nerve cells & dendrites) vs. white matter (axon fibers)
  • Region sizes/volumes

Spectroscopy (specific metabolites)

Voxel-based morphometry (VBM)

  • Quantitative analyses of size/volume
  • Example: volume differences in schizophrenic patients vs. controls using statistical maps of size differences

(Pomarol-Clotet et al., 2010)

Diffusion Tensor Imaging (DTI)

  • Structural MRI technique
  • Diffusion tensor: measurement of spatial pattern of \(H_2O\) diffusion in small volume
  • Uniform (“isotropic”) vs. non-uniform (“anisotropic”)
  • Strong anisotropy suggests large # of axons with similar orientations (fiber tracts)

  • Fractional Anisotropy (FA), radial diffusivity (RD), mean diffusivity (MD) measures of “non-uniformity” of diffusion tensor
  • Connecting tensors or fiber tract tracing

Connectome

  • What is the wiring diagram?

Functional (activity) methods

Recording from the brain

Single/multi unit recording

  • Microelectrodes
  • Small numbers of nerve cells

https://www.nature.com/nrn/journal/v5/n11/images/nrn1535-i1.jpg

  • What does neuron X respond to?
  • High temporal (ms) + spatial resolution (um)
  • Invasive
  • Rarely suitable for humans, but…

Electrocorticography (ECoG)**

Single-cell studies ask…

  • How does firing frequency, timing vary with behavior?

Positron Emission Tomography (PET)

  • Radioactive tracers (glucose, oxygen)
  • Positron decay activates paired detectors
  • Tomographic techniques reconstruct 3D geometry
  • Experimental condition - control
  • Average across individuals
  • Temporal (~ s) and spatial (mm-cm) resolution worse than fMRI
  • Radioactive exposures + mildly invasive
  • Dose < airline crew exposure in 1 yr

Functional Magnetic Resonance Imaging (fMRI)

  • Neural activity -> local \(O_2\) consumption increase
  • Blood Oxygen Level Dependent (BOLD) response

  • Oxygenated vs. deoxygenated hemoglobin ≠ magnetic susceptibility

  • How do regional blood \(O_2\) levels (& flow & volume) vary with behavior X?
  • MRI “signals” relate to the speed (1/T) of “relaxation” of the perturbed nuclei to their state of alignment with the main (\(B_0\)) magnetic field.

  • Imaging protocols emphasize different time constants of this relaxation (\(T^1\), \(T^2\), \(T^{2*}\)); \(T^{2*}\) for BOLD imaging

Evaluating fMRI

  • Non-invasive, but expensive
  • Moderate but improving (mm) spatial, temporal (~sec) resolution
  • Spatial limits due to
    • field strength (@ 3T \(~3mm^3\) voxel)
    • Physiology of hemodynamic response
  • Temporal limits due to
    • Hemodynamic Response Function (HRF): ~ 1s delay plus 3-6 s ramp-up
    • Speed of image acquisition (\(TR\) is time of image acquisition, usually 2-3s for whole brain studies)
  • Indirect measure of neural activity

Hemodynamic Response Function (HRF)

Typical analysis…

Generate “predicted” BOLD response to event; compare to actual

Average across individual participants and plot statistical maps (in color space) on top of structural image.

https://en.wikibooks.org/wiki/Cognitive_Psychology_and_Cognitive_Neuroscience/Behavioural_and_Neuroscience_Methods#fMRI

Effects of higher field strengths (3 Tesla vs. 7 Tesla)

(Sladky et al., 2013)

Functional Near-infrared Spectroscopy (fNIRS or NIRS)

Source: https://cibsr.stanford.edu/NIRS_Lab.html

Source: https://cibsr.stanford.edu/NIRS_Lab.html

Source: https://nirx.net

Source: https://nirx.net

Electroencephalography (EEG)

(Cohen, 2017)

https://sfari.org/images/images-2013-folder/images-sfn-2013/20131110sfneeg

Analyze in frequency domains

https://www.peakmind.co.uk/images/frequency.jpg

(Cohen, 2017)

Event-related potentials (ERPs)

Brain Computer Interface (BCI)

https://s.hswstatic.com/gif/brain-computer-interface-3.gif

Magneto-encephalography (MEG)

How do EEG/MEG and fMRI relate?

(Logothetis, Pauls, Augath, Trinath, & Oeltermann, 2001)

Manipulating the brain

Interfering with the brain

Galen

https://en.wikipedia.org/wiki/Galen

Phineas Gage

I highly recommend the work of the late Oliver Sacks.

Stimulating the brain

https://en.wikipedia.org/wiki/Transcranial_direct-current_stimulation

https://en.wikipedia.org/wiki/Transcranial_magnetic_stimulation

Deep brain electrical stimulation as therapy for…

https://www.nimh.nih.gov/images/health-and-outreach/mental-health-topic-brain-stimulation-therapies/dbs_60715_3.jpg

https://youtu.be/KDjWdtDyz5I

Optogenetics

https://www.youtube.com/I64X7vHSHOE

https://youtu.be/FlGbznBmx8M

Simulating the brain

Blue Brain project

Markram, 2006

Or, here’s an example of a brain-like use of artificial intelligence.

Hodgkin-Huxley model of the action potential

Computational neuroscience really began with a set of mathematical models of how neurons generate and send electrical messages down the axon (Hodgkin & Huxley, 1952). The graphs below simulate the time course of voltage \(v\) several difference types of conductance (flow of electrical current) thought to mimic what goes on in an actual neuron. Thanks to https://magesblog.com/post/2012-06-25-hodgkin-huxley-model-in-r/ for the code.

Measuring the body

References

Amo, C., De Santiago, L., Zarza Luciáñez, D., León Alonso-Cortés, J. M., Alonso-Alonso, M., Barea, R., & Boquete, L. (2017). Induced gamma band activity from EEG as a possible index of training-related brain plasticity in motor tasks. PloS One, 12(10), e0186008. https://doi.org/10.1371/journal.pone.0186008

Cohen, M. X. (2017). Where does EEG come from and what does it mean? Trends in Neurosciences, 40(4), 208–218. https://doi.org/10.1016/j.tins.2017.02.004

Hobson, H. M., & Bishop, D. V. M. (2017). The interpretation of mu suppression as an index of mirror neuron activity: Past, present and future. Royal Society Open Science, 4(3), 160662. https://doi.org/10.1098/rsos.160662

Hodgkin, A. L., & Huxley, A. F. (1952). A quantitative description of membrane current and its application to conduction and excitation in nerve. The Journal of Physiology, 117(4), 500–544. https://doi.org/10.1113/jphysiol.1952.sp004764

Lichtman, J. W., Livet, J., & Sanes, J. R. (2008). A technicolour approach to the connectome. Nature Reviews Neuroscience, 9(6), 417–422. https://doi.org/10.1038/nrn2391

Logothetis, N. K., Pauls, J., Augath, M., Trinath, T., & Oeltermann, A. (2001). Neurophysiological investigation of the basis of the fMRI signal. Nature, 412(6843), 150–157. https://doi.org/10.1038/35084005

Logothetis, N. K., & Wandell, B. A. (2004). Interpreting the BOLD signal. Annu. Rev. Physiol., 66(1), 735–769. https://doi.org/10.1146/annurev.physiol.66.082602.092845

Pomarol-Clotet, E., Canales-Rodrı'guez, E. J., Salvador, R., Sarró, S., Gomar, J. J., Vila, F., … McKenna, P. J. (2010). Medial prefrontal cortex pathology in schizophrenia as revealed by convergent findings from multimodal imaging. Mol. Psychiatry, 15(8), 823–830. https://doi.org/10.1038/mp.2009.146

Sejnowski, T. J., Churchland, P. S., & Movshon, J. A. (2014). Putting big data to good use in neuroscience. Nat. Neurosci., 17(11), 1440–1441. https://doi.org/10.1038/nn.3839

Sladky, R., Baldinger, P., Kranz, G. S., Tröstl, J., Höflich, A., Lanzenberger, R., … Windischberger, C. (2013). High-resolution functional MRI of the human amygdala at 7 T. European Journal of Radiology, 82(5), 728–733. https://doi.org/10.1016/j.ejrad.2011.09.025