Methods and the Brain Flashcards
Dependent variable
The variable that is measured in the experiment
Independent variable
The variable that is manipulated in the experiment
Confounding Factors
Uncontrolled variables that could also influence the
dependent variable. Should be kept constant across manipulations
Statistical significance
If it is sufficiently unlikely that a result occurred by chance (< 5%), the IV is said to have a “statistically significant” effect on the DB; describes how reliable an effect is, not how large it is
Statistical test to determine if something is statistically significant
Mean/variance
Factorial design
An efficient way to combine variables — you perform all
combinations of manipulations; possible results are main effect and interaction
Main effect
The effect of one IV on the DV, ignoring all other IVs
Interaction
When the effect of one IV on the DV depends on the level of the other IV
Marr’s 3 levels of analysis
- Computational (broadest)
- Algorithmic (intermediate)
- Implementation (specific)
Computational
What is the purpose?
Algorithmic
How does it work?
Implementation
Physical processes behind it
How does understanding how the brain supports cognition at an implementation level give insights into how cognition works at computational and algorithmic levels?
- If two cognitive abilities are neurally independent, we can assume that they are cognitively independent
- Long-standing cognitive debates often can’t be resolved with behaviour alone — sometime neural findings show that both theories are right, just
implemented in different parts of the brain! - Discoveries in the brain can give us new clues about how cognition works
Materalist/physicalist philosophy
Argues that “the mind is what the brain does” - mental states can be generated in so many ways. We can only understand how human mental states work by knowing how they are generated by the brain
Dualist philosophy
Mind and the brain are two separate entities and biology can’t explain how the mind works
What do some materialists argue that opposes the view that we need to fully understand the brain to understand the mind?
That the mind is best explained at computational
and algorithmic levels; the implementation level is too reductionist
Arguments against reductionism
Higher levels of analysis may emerge but not be fully explained by lower levels - we might not be able to build up from those very basic concepts and they might not even be relevant to the problems we want to solve
Santiago Cajal
Discovered neurons
Neural Doctrine
- Neurons are the basic units (cells) of the brain
- Neurons contain cell bodies, dendrites and
axons - Neurons communicate via synapses, not
physical connections - Neurons are connected in circuits, not
randomly
Resting potential
Difference in the electrical charge between the inside and outside of an axon: -70mV
Action potential
- APs travelling along an axon maintain the same
amplitude; you can’t have a big or small AP - APs trigger the release of neurotransmitters
from the axon terminals -> can excite dendrites
of other neurons
What triggers an action potential?
When neuron is excited, the membrane
potential increases; if it reaches -55mV an
Action Potential (AP) is triggered
Strength of a sensation
Proportional to the rate of neural firing
Qualities of an experience or thought are related to _____
which neurons fire
Feature Detectors
Neurons identified in primary visual cortex (V1) that respond to specific visual features, such as orientation, size, or the more complex features that make up environmental stimuli.
Blakesmore and Cooper’s experiment on how experience shapes perception and the brain
Kittens raised in vertical striped environments only
perceive vertical lines + only have feature detectors for
vertical lines
Hierarchical processing
Progression from feature detectors in lower brain regions to more complex preferences in higher
regions (visual cortex which senses simple stimuli -> temporal lobe which senses more complex stimuli; neurons that respond to simple stimuli -> neurons that respond to complex geometrical shapes -> neurons that respond to faces)
Neurons in the temporal lobe of monkeys
Have preferred stimuli — they prefer specific complex objects (e.g. hands or faces)
Sensory coding
How neurons represent various characteristics of our environment - includes specificity, population and sparse coding
Specificity coding
Individual neurons represent (only fire in response to) individual stimuli/concepts — “grandmother cells”
Population coding
Stimuli/concepts are represented in the pattern of firing across a large number of neurons
Sparse coding
Like population coding but the pattern is coded within a small proportion of neurons — most neurons are relatively silent
What does the specificity of cognitive impairments suggest?
That functions are localized in different areas of the brain
Occipital lobe damage
Damage to soldiers’ brains demonstrated that this
lobe is necessary for vision
Prosopagnosia
Damage to a region of the temporal lobe selectively impairs face perception
Double dissociation
- Lesioning one brain area impairs cognition A but
not B; lesioning a different area impairs B but not A. - Required to demonstrate independence of cognitive functions
fMRI
A brain imaging technique that measures how blood flow changes in response to cognitive activity.
How fMRI is used in perception experiments
In perception experiments, participants look at different types of pictures while undergoing fMRI; many regions have preferred stimulus types
Fusiform face area (FFA)
An area in the temporal lobe that contains many neurons that respond selectively to faces.
Parahippocampal place area (PPA)
An area in the temporal lobe that contains neurons that are selectively activated by pictures of indoor and outdoor scenes.
Lateral occipital complex
Responds most strongly to pictures of objects
Extrastriate body area
An area in the temporal cortex that is activated by pictures of bodies and parts of bodies, but not by faces or other objects.
Distributed representations
Even if some functions are localized, the
representation of an experience is distributed
across areas of the brain
Brain networks
How areas of the brain are connected to each other - some areas are connected more strongly than others and these strongly connected regions communicate with each other to support cognitions (e.g. regions of frontal and parietal lobe –> attention)
How can we study brain networks?
Through neuroimaging tools that measure physical connections (DTI) and functional connections (functional connectivity)
Converging methods
Researchers use multiple techniques with complementary strengths to answer the same question
Lesion techniques
Assess cognitive impairments which occur following damage to a brain region (can be used to infer causality)
Lesion mapping
Determining the probable source of the cognitive deficit by identifying a set of patients who all have the
same cognitive deficits and identify where
the damage overlaps
Transcranial magnetic stimulation
- Can be used to temporarily inhibit a brain region (type of lesioning)
- More precise localization but can only be performed
on certain brain regions & only reduces function
EEG
Electrodes placed on the scalp measure coordinated neuronal responses over large areas of the brain
Event Related Potentials
The measured brain response that is the direct result of a specific sensory, cognitive, or motor event, the result of many thousands of neurons near the electrode that fire together.
Strengths/weaknesses of EEG
Has a high temporal resolution (milliseconds) but poor spatial resolution — you learn more about when neural processes occur but little about where they occur
MEG
Provides similar temporal information as EEG but has a higher spatial resolution
How can you enhance spatial resolution in EEG?
EEG can also be recorded with implanted intra-cranial electrodes to further increase spatial resolution
How does MRI work?
Allows us to take pictures of the brain by producing very strong magnetic fields, and radio waves disrupt the alignment between atoms and the magnetic field. Different tissue (fat, water) re-align at different rates and MRI machines measure the time taken for this to happen.
Strengths of MRI
Provides high spatial resolution (~.3mm) images of
brain anatomy
How does fMRI work?
Works like anatomical MRI, but:
- Takes a picture of the whole brain every ~2 seconds
- Targets the concentration of oxygen in the blood
- Blood oxygen increases when neurons in a brain region are active
Strengths/weaknesses of fMRI
- Provides high spatial resolution (~2.5mm) images of brain function
- Temporal resolution is relatively low (1-2 seconds) and the measurement is indirect
Diffusion tensor imaging (DTI)
- Conducted using MRI scanner
- Measures the strength of anatomical connections
(structural connectivity) between brain regions by
detecting the movement of water molecules - Water tends to move along the main axis of axons
What does DTI allow?
Allows researchers construct the brain’s “connectome” - a wiring diagram of brain networks
How is functional connectivity measured?
Measured using fMRI (sometimes MEG)
How is functional connectivity determined?
- Involves analyzing fMRI data in a different way — asking whether regions activate and deactivate
together, called a correlation - Often measured when people are told to “rest” to identify resting state networks
- Regions within resting state networks tend to have
strong anatomical connections, co-activate during
tasks, and be reliable across people
Principle of neural representation
Everything a person experiences is based on representations in the person’s nervous system.
Broca’s aphasia
A condition associated with damage to Broca’s area, in the frontal lobe, characterized by labored ungrammatical speech and difficulty in understanding some types of sentences.
Wernicke’s aphasia
A condition, caused by damage to Wernicke’s area (in the temporal lobe), that is characterized by difficulty in understanding language, and fluent, grammatically correct, but incoherent speech.
