Methods for studying the brain Flashcards
What is the localist view?
· The brain is composed of specialised areas.
· Information processing in each of these is local and specific
· Discrete areas are responsible for discrete functions
Areas are discreet
What is the anti-localist view?
· The brain is a collection of networks
· (ii) information processing is ‘distributed’
(iii) All areas are equally responsible for all functions – need all of the brain for it to work altogether
How was the localist view developed?
· Developed without any basis or evidence
· Extreme localist view saying one area is associated with a specific trait
· Gall and Spurzheim (1810):
· Believed function to be localised to specific brain areas.
Reflected in the morphology of the skull (the pseudoscience of ‘Phrenology’).
What did Broca (1861) believe about the localist view?
· More evidence came along to support the localist view
· Reported a case of productive aphasia – unable to produce speech. Lesion in a specific area in the frontal lobe. Nothing else wrong with them other than speech, shows that that specific area was responsible for speech
Brain contained lesion in a very specific area of the frontal lobe, today referred to as ‘Broca’s Area’.
What did Flourens (1822) believe about the anti-localist view?
· Proposed that there were specific functions, but these were mediated by the brain as a whole – need the whole brain to work together
· “All sensations, all perceptions, and all volitions occupy the same seat in these (cerebral) organs. The faculty of sensation, percept and volition is the essentially one faculty.”
· This view was later referred to as the ‘Aggregate field theory’
This alternate view was held by others over time e.g. Lashley (1929) suggested function is shared by all neurons rather than being specific to one.
What is the current view about localist vs anti-localist?
- Have some specialised area
- Still in the middle of the two views
- E.g. there are localised visual areas in the brain (Op de Beeck et al, 2008)
More neurons respond to light looking from straight ahead, other respond better to light from the edges (retopic mapping)
Localized AND distributed information processing
- Anatomical evidence ?
· Localized: Neuronal properties are variable (‘cytoarchitecture’ – means cells), so the way they process information must also be variable. Different types of cells do different jobs
Distributed: The brain consists of massively interconnected networks of neurons, so there must be distributed processing in the brain. Across the brain we see fibres in different areas
Localized AND distributed information processing
- Evidence from brain activity studies ?
· Localized: Electrophysiology and functional brain imaging shows that activity in individual brain areas is task-specific.
Distributed: The activity of one brain area influences the activity of its connected areas. Activity of one brain area influences activity in another area
Localized AND distributed information processing - Evidence from lesion studies ?
· Localized: The effects of carefully applied lesions on tasks are functionally specific.
Distributed: But some lesions affect the ability to perform tasks only when applied to more than one area. Some lesions are not very specific
Structure-function relationships in the brain: - Current thinking?
· Evidence for both localised and distributed information processing
· An important aim of neuroscience is to explain brain-behaviour
· relationships in terms of localised and distributed processing
· New terminology for similar concepts: do not look at local vs non-local
· Functional Segregation: How does function emerge from specialised information processing in individual brain areas?
· Functional Integration: How does function emerge from distributed information processing across the networks of these areas?
Modern neuroscience seeks to integrate these levels of explanation
The insufficiency of lesion studies?
· Permanent lesions in patients and animal models are the traditional method of studying structure-function relationships. Are they sufficient for understanding how the intact brain works?
· There are many interpretations to lesion studies so, on their own, they cannot tell us about why lesions do/don’t cause deficits.
· Lesions that impair functions…
· Essential information processing for the task takes place in that area. Essential information processing in connected areas is disrupted. May not mean that area does a task
· Lesions that don’t impair functions…
· Information processing in that area is not needed for the task.
Information processing is needed for the task in the normal brain, but the lesioned brain compensates to restore function.
What is Brain Architecture (Anatomy) made up of?
· Cytoarchitecture
· Connectivity: Tracers (non-human primates) - Diffusion-tensor imaging (humans)
Gross morphology - Structural Magnetic Resonance Imaging (MRI). Overall shape of the brain
How can Brain Activity be measured?
· Neurophysiology (single cell) · Electroencephalography (EEG) · Magnetoencephalography (MEG) · Positron Emission Tomography (PET) · Functional Magnetic Resonance Imaging (fMRI) All look at the activity of the brain
How do you measure brain stimulation?
· Transcranial magnetic stimulation (TMS)
· Deep Brain Stimulation (DBS)
Trying to interfere with what the brain is doing to understand behaviour
What is Brain architecture: cytoarchitecture?
· Each area of the brain has different packings of neurons
· Brodmann areas – different amount of neurons
· The cellular composition of the cortex varies
· They way that information is processed across the cortex therefore cannot be uniform
· Mapping the variation can distinguish cortical areas involved in different forms of information processing
What is Brain architecture: connectivity?
· The flow of information across brain networks is an integral part of information processing in the brain
Understanding the architecture of these networks is important for understanding the relationships between neural and cognitive processes
How is brain connectivity measured?
· Measured using tracers – inject a chemical into a neuron and is transported to wherever that neuron connects to
· The flow of information across brain networks is an integral part of information processing in the brain.
· Understanding the architecture of these networks is important for understanding the relationships between neural and cognitive processes.
· In animal studies, tracers injected into one end of a neuron in one area are transported to the other end.
· Histological analysis reveals the terminal sites.
New Magnetic Resonance Imaging (MRI) technology makes it possible to study the trajectories of fibre pathways in the brain
What is Diffusion Tensor Imaging (DTI)
· Diffusion of a water molecule within a fiber bundle. Diffusion is less restricted along the axis of the fiber bundle than across it.
· Diffusion can be described by an ellipsoid, or tensor. The tensor is fully characterized by its three orthogonal eigenvectors and their associated lengths (or eigenvalues). This method can say any given area what the water flow is going, e.g. direction, how constrained it is
· The shape of the ellipsoid (called fractional anisotropy, FA) contains information about the directional dependency of the diffusion signal. High FA, can only flow in one direction, low FA: can travel freely
· Simple, streamlining tractography proceeds by tracing a line through the tensor field, following the principal diffusion direction. The schematic shows a grid of voxels; the grayscale reflects FA [ranging from zero (black) to one (white)].
Johansen-Berg and Rushworth (2009)
How can you measure brain activity?
· Behavioural Neurophysiology · Electroencephalography (EEG) · Magnetoencephalography (MEG) · Positron-Emission-Tomography (PET) Functional Magnetic Resonance Imaging (fMRI)
What is Behavioural neurophysiology?
· Sometimes it is important to record the activity of neurons directly using neurophysiological methods
· Monkey’s given different types of handles, been trained to use each one. Is there a part of the brain which encodes seeing the handle and carrying out the action?
· When a light shows, monkey has to grasp the handle
Filippini et al (2017)
What is single and multi-unit recordings?
· Electrodes are inserted into or directly onto a small area of the brain.
· Single-unit recordings: Recording the activity of individual neurons
· Multi-unit recordings: Recording summed activity from small populations of neurons – not clear where each neuron is coming from
· Advantages: Important in animal studies for determining exactly where the signal is coming from, at very high temporal spatial resolution.
High spatial: know exactly where the neuron is coming from
Can single cells be recorded in humans?
· Some people suffering from pharmacologically intractable epilepsy are implanted with depth electrodes to localize focus of seizure onset – can’t be treated with drugs
· Recordings from single neurons in the human medial temporal lobe
Showed different visual pictures – was able to get different recordings from different areas of the brain. Spikes when images were shown of specific actresses, suggesting there is a single neuron which controls this - Quian Quiroga, Reddy, Kreiman, Koch, & Fried (2005)
What is Electroencephalography (EEG)?
· Non-invasive, direct measure of changes in electrical activity
· Low spatial resolution on the scalp, electrodes places quite far apart
· Electrical potentials produced by single neurons is small, but when several neurons are active together, the resulting signal can be measured by an electrode on the scalp.
· EEG records with a very high temporal resolution, but the spatial resolution is relatively poor because of the way that the signal is distorted by the tissues between the brain and the electrode (e.g. bone).
· By averaging the traces to cancel out noise, we can detect EEG responses that are specific to particular signals (event- related potentials, ERP’s).
· ERP’s indicate how neural activity changes over time
· Also can measure oscillations at different frequencies.
E.g. Alpha oscillation occur in visual cortex when eyes are closed (8-13 Hz)
What is Magnetoencephalography (MEG) ?
· MEG measures the magnetic signal changes that result from changes in electrical potential.
The detectors, called ‘SQUIDs’ (superconducting quantum interference device), are very sensitive to these signals.
How does Magnetoencephalography (MEG) measure?
· Relies on measuring changes from the scalp, measuring changes in magnetism. Magnetic signal changes that results in changes of electrical potential
· Not picking up on one single neuron, need enough action potentials for the MEG to pick up the change
· Magnetic changes are very small
· MEG has a very high temporal resolution (in the order of ms), and in addition has better spatial resolution (than EEG) (around 1 cm and less) because magnetic signals are not distorted as in EEG.
· Evoked MEG responses can be detected by averaging as in EEG, and can indicate how neural activity changes over time.
· As with EEG can measure the same types of oscillations.
What are the pros and cons of MEG?
· Sources generating the measured signals can be estimated. The sensitivity and therefore the spatial resolution of MEG source imaging is uneven across the brain: for instance, superficial cortical sources produce MEG signals up to 100 times stronger than deeper, subcortical structures at equivalent current strengths.
· MEG more expensive than EEG and more physically restrictive
What is Positron-Emission-Tomography (PET)?
· PET scanners measures changes in blood flow that are stimulated by changes in neural activity.
· If an isotope containing radioactive oxygen (e.g. H2O15) is injected into the bloodstream, it is absorbed into the brain. Drawback of PET, need to inject participants, radioactive isotope which labels the oxygen
· There is a series of sub-atomic events resulting in the emission of photons emitted in opposite directions.
· Increases in blood flow are measured by increases in coincidences at opposite detectors in the PET scanner.
· Reasonable spatial resolution and can image deep structures as well as the cortex.
Very poor temporal resolution. Indirect measure.
What is MRI?
- Each hydrogen atom’s proton rotates about its axis, acting as a small magnet with its own north-south dipole. Normally the protons are randomly diffused, so the tissue has no net charge
- When placed in a magnetic field, the protons align in parallel
A radiofrequency pulse applied to the tissue pushes the protons to their sides, causing them to wobble about their axes and about their north-south orientation. This is motion called precession, produces measurable vertical and horizontal magnetic fields
What is Functional magnetic resonance imaging (fMRI)?
· To do with blood flow
· Hemoglobin with oxygen (oxy-Hgb), and without (deoxy-Hgb)
Hemoglobin (Hgb) can bind up to four oxygen (O2) atoms
· Higher spatial resolution than EEG and MEG (around 3x3x3 mm)
Lower temporal resolution (in the order of seconds)
fMRI data analysis
· In an fMRI experiment with two conditions (A and B), the stimulus function is convolved with a canonical HRF to obtain two sets of predicted BOLD responses. The responses are placed into the columns of a design matrix X and used to compute whether there is significant signal corresponding to the two conditions in a particular time course.
Statistical parametric maps (SPM) are used to present the results of the statistical analysis. Voxels whose p-values are below a certain threshold are color-coded to signify that they contain significant task-related signal. The results are superimposed onto a high-resolution anatomical image for presentation purposes.
There are different measures of neuronal activity:
· Single-unit activity: The activity of single neurons
· Multi-unit activity (MUA): Reflects the output of a population of neurons (within ~200microns of electrode tip)
· Local field potentials (LFP): Averaged input signals of a neuronal population (within a few mm of electrode tip)
BOLD activation may reflect more the neural activity related to the input and the local processing in any given area, rather than the spiking activity commonly thought of as the output of the area
What is Transcranial Magnetic Stimulation (TMS) ?
· Temporary lesion
· Brief and rapidly changing high intensity electrical current is passed through loop of conducting wire onto an area of the scalp.
· Generates a powerful magnetic field that is capable of inducing an electric current in excitable tissue.
· Applied onto cortical areas, the induced current depolarizes nearby located neuron assemblies located beneath the coil and generates neurophysiological and/or behavioral effects depending on their contributing functions.
· Can deliver disruption for a very small amount of time at a very localised area of cortex.
· Can therefore target very specific processes that occur in particular time points in experiments.
Like a temporary lesion
How does TMS work in V1 and M1?
· TMS of the primary visual cortex (V1): induces the percept of phosphene reported as appearing in the central visual field.
· TMS of the primary motor region (M1): triggers involuntary twitches of muscles in segments of the upper limb contralateral to the stimulated hemisphere, and following their somatotopic organization in the motor cortex
Response can be characterized quantitatively by recording the latency and amplitude of the Motor Evoked Potentials (MEP), by means of electromyography (EMG) on specific hand or forearm muscles.
