10 From models to applications

Possible advantages:

1) the ability to test for unique cognitive effects – certain things cannot be tested well in animals, example Numerosity, or speech.

2) training and communication – training is faster because one can communicate with the human subject and describe the desired task, this cannot be done in animals. In humans it avoids overtraining effects.

3) The whole brain – with the techniques that we have, we can measure at a mid-level of spatial and temporal scale brain activity patterns across the entire brain.

Possible disadvantages:

1) The measured signal is not a direct reflection of neural activity, but an indirect reflection of neural activity. This is particularly problematic for comparisons across different populations where a change in the vascular response would be expected (e.g. age).

2) The typical methods for measuring neural activity in animals are unethical in humans,

3) The methods are costly and time intensive.

4) The temporal and spatial resolution is to coarse to see individual neurons or circuits.

1. Strong static magnetic field

2. Radio frequency (RF) or high frequency (HF) pulse

3. Magnetic gradients

• When RF pulse stops, the protons precess back to original state, releasing energy

• signal itself is oscillatory in nature, a free induction decay (FID) signal

• The rate of decay of transverse magnetization is much faster than longitudinal recovery due to dephasing of the protons

• Positron emmission tomography (PET): spatial resolution on the order of 5-10 mm, and temporal resolution of minutes to hours

• Electroencephalography (EEG): spatial resolution on the order of centimeters and temporal resolution of muss less than a ms

• Single cell recordings: high spatial (μm) and temporal resolution (<< ms)

The second question cannot be addressed with human MRI because the spatial scale of the measurements (mm) is too coarse to be able to resolve a single neuron and the fMRI signal cannot resolve inhibitory and excitatory activity (I did not mention this in my lectures, but it is the case).

The Larmor equation is written as follows:

It states that the larmor frequency of a particle, which is its resonant frequency, is equal to the strength of the magnetic field, mutliplied by the particle’s gyromagnetic ratio. Each particle has a unique gyromagentic ratio, which is the ratio between the particle’s magnetic moment to its angular momentum. In MRI, we wish to excite hydrogen protons, so with the hydrogen proton’s gyromagnetic ratio times the strength of the main magnetic field we can determine the main frequency of the radio frequency pulse that is sent into the body or the brain to create and image. It will also be the main frequency of the signal that we receive from the brain.

The signal that is measured with MRI is a damped sinusoidal signal, also called a free induction decay (FID) signal, with a frequency that is centered on the Larmor frequency for hydrogen protons. Differences in brain tissue are measured by tissue specific differences in the speed of decay of the signal (or recovery of the longitudinal magnetization) that are reflected in the brightness of the individual voxels of the image.

A magnetic gradient, that is perpendicular to the 2D slice to be measured, is turned on at the same time as the radio frequency pulse is given. This changes the strength of the main magnetic field along the direction of the 2D slice. Then only those protons in the 2D plane where the Larmor frequency (based on the static magnetic field strength) matches the radio frequency that is given will be excited.

The signal is called the BLOOD OXYGEN LEVEL DEPENDENT (BOLD) signal.

• active neurons require energy

• energy comes from oxygen

• oxygen in the blood surrounds active cells and is consumed

• More oxygenated blood is sent to the active brain region than is needed

• increase in the relative concentration of oxygenated vs deoxygenated hemoglobin

• Oxygenated hemoglobin is LESS magnetic than deoxygenated hemoglobin

• when more oxygenated hemoglobin is present, the hydrogen protons dephase more slowly, leading to a brighter signal

Neurons are active (fire action potentials and signal to other cells through synapses) – this requires energy which requires oxygen. Oxygen is transported, from our lungs through the blood vessels to all tissues of the brain. Through chemical signaling of the neurons and surrounding cells, through neurotransmitters and other molecules in the intercellular space, the local blood vessels surrounding the active tissues dilate, and oxygen is consumed. More oxygenated blood is sent to the active brain region than is needed leading to an increase in the relative concentration of oxygenated vs deoxygenated hemoglobin. Oxygenated hemoglobin is LESS magnetic than deoxygenated hemoglobin. Therefore, when more oxygenated hemoglobin is present, the hydrogen protons dephase more slowly, leading to a brighter signal. At a certain point later, the blood vessels constrict, and the entire system returns to baseline.

• record from individual neurons in the visual cortex of non-human primates and simultaneously measure brain activity with fMRI

• the same physical location neurons that responded with action potentials to a checkerboard visual stimulus in a specific part of the visual field and had a BOLD signal activity in fMRI

• BOLD signal was most closely related to the local field potential (LFP) signal from the electrophysiological recordings

There have been several studies that investigated the relationship between neural activity and the fMRI signal. Two were mentioned in the lecture. First, the group of Nikos Logothetis in Tübingen created a setup in which they could record from individual neurons in the visual cortex of non-human primates and simultaneously measure brain activity with fMRI. They found in the same physical location neurons that responded with action potentials to a checkerboard visual stimulus in a specific part of the visual field and had a BOLD signal activity in fMRI. Additional analyses showed that the BOLD signal was most closely related to the local field potential (LFP) signal from the electrophysiological recordings. LFP is a measure of presynaptic activity more than long-range action potential firing

• a short stimulus was given once, twice and three times (with a pause in between but BOLD signal was not down to the baseline)

• BOLD signal substractions: 2 stimului - 1 stimulus and 3 stimuli - 2 stimuli

• curves were overlayed and aligned to the last stimulus —> highly similar

• —> they can be linearly summed, even if the BOLD signal has not returned to baseline

• BOLD signal were consistent over different brain regions and individulas

• —> we can use a model of the expected BOLD signal to a short stimulus and use it as an impulse response function to create a theoretical or expected BOLD response to ANY stimulus time course

A short stimulus was given once, twice and three times, with a pause in between, but close enough together that the BOLD signal was not back down to baseline. The resulting BOLD signal to one stimulus was subtracted from the resulting BOLD signal to two stimuli, and the resulting BOLD signal of 2 stimuli were then subtracted from the BOLD signal resulting from three stimuli. Then the resulting curves after subtraction were all overlayed on one another aligned to the start of the last stimulus. These curves were highly similar, suggesting that they can be linearly summed even if the BOLD signal has not returned to baseline before the next stimulus is presented. In addition, the size and shape of the BOLD signal is consistent across multiple repetitions of the same stimulus, and across brain regions and individuals. This means we can use a model of the expected BOLD signal to a short stimulus and use it as an impulse response function to create a theoretical or expected BOLD response to ANY stimulus time course. It is the key that makes analyzing fMRI data possible.

The hemodynamic response function is a curve that rises first and peaks at approximately 6 seconds, then lowers and dips below the baseline at approximately 12 seconds after which it returns to baseline. It is modelled as a difference of two gamma curves (gamma distribution function), the first one represents the rise in activity, the second one represents the undershoot. It is used as an impulse response function and convolved with the time course of each of the variables of interest to better estimate the expected BOLD response

Yes, we can measure regional brain activity when we are at rest. This activity is measured by correlating the time course of the brain activity between different voxels of the brain. Those areas of the brain with a high correlation are then thought to be connected to one another and active at rest. The assumption made here is that the correlated brain activity between regions means they are connected, and not for instance, that some external stimulus caused this activity. The most common set of brain regions (or network) found at rest is the Default Mode Network.