PSY 392 brain waves and human cog exam 1

Cognitive Neuroscience

an interdisciplinary field studying the biological basis of mental processes (cognition) in the brain and investigating how neural circuits support cognitive functions using neuroscience and behavioral methods

cognitive psychology

the scientific study of internal mental processes, exploring how people cquire, process, store, and use information, covering areas like perception, memory, attention, language, problem-solving, and decision-making

neuroscientific methods

fMRI, EEG, electrocorticography (ECoG, or iEEG), TMS, etc.

What is an electrode

a conductor (like metal or graphite) that allows electricity to enter or leave a nonmetallic part of a circuit, serving as a gateway for currentflow in devices like batteries, medical sensors, and electronic components.

ERP components

voltage deflections in the ERP waveform

Naming of ERP components

• NAming of ERP components

  • Ordinal position vs. latency of the peak

  • p1, p2, N1, N2, P3 vs. P300, N400

  • paradigm-based, e.g., no-go N2

  • Function based name, e.g., ERN(error-related negativity), FRN (feedback-related negativity), RewP (Reward Positivity)

• Which way is up?

  • Historically, negative is up, so pay attention to the Y axis

    • more recent papers switch this

what leads to PSPs

constant brain activity leads to constant variations in the pattern of PSPs across the billions of neurons in your brain, this lead to a constantly varying EEG on the scalp

many different brain waves are what at the individual scalp electrodes

combined, creating a complicated mixture

one portion of the mixture of brain signals during an EEG that are brief, transient responses to internal and external events

ERPs

What activity of EEG is not driven by discrete events

much of the event driven activity is oscillatory in nature, reflecting feedback loops in the brain

EEG oscillations are mainly classified according what?

frequency bands

• alpha band (8-13 Hzz)

• Delta band (<4Hz)

• Theta band (4-8 Hz)

• beta band (13-30 Hz)

• gamma band (>30Hz)

• It is not generally true however that a specific frequency band reflects a specific process

What is EEG

• Electorencephalogram

  • a non-invasive measure of changes in electrical potentials at the scalp produced by neural activity. It is a continuous and direct measure of brain activity in real time

EEG indexes broad mental states and large signals

• mentalarousal level

• epileptic seizures

• large muscle movements

  • some patterns topographically recognized

ways EEG is used in the medical field

• sleep disorders

• epilepsy and seizure disorder

• brain tumor

• brain damage from head injury

• stroke

• drug intoxication

• monitor anesthesia

• confirm brain death

• other brain disorders

EEGs can be used for brain-computer interfaces (BCIs) using motor imagery

• large muscle movements

• EEG during imagined movements allows direct control of devices without using limbs

What is EEG good for

• EEG indexes broad mental states and large signals

• EEG is used extensively in the medical field

• EEGs can be used for brain-computer interfaces (BCI’s) using motor imagery

• EEGs can be used to ask questions about the mind (mind-behavior link) and factors that influence mental states

  • How to go beyound slide 12’s early “ERPology”

What are ERPs good for

• Assessing the time course of processing

• Identifying potential neurocognitive processes

  • is a process absent or present

• Covert measurement of processing

• A link to the brain? (Maybe, but more about the mibd and the relationship between the mind and behavior, not so much about the brain circuitry)

• Biomarker? (maybe…)

What are ERPs bad for

• Waveforms recorded on the scalp represent the sum of many underlying components and it is difficult to decompose this mixture into the individual underlying neural porcesses

• ERPs are small signals, sensitive to head, mouth, eye movements

  • signal to noise ratio - not great

• “ERP” require the use of measurable events

• can’t measure cognitive function that extends beyond a few seconds (e.g., long-term memory consolidation)

• fNIRS

• fNIRS (functional Near-Infrared Spectroscopy) is a non invasive brain imaging technique that uses near infrared light to measure chnage in blood oxygenation in the brain

fMRI

(functional Magnetic Resonance imaging)

non invasive brain imaging technique that maps brain activity by detecting changes in blood flow, revealing which areas are working hardest during specifiic tasks like thinking, speaking, or moving, by measureing blood oxygen levels

PET

Positron emission tomography

is a nuclear medicine imaging technique that creates #D pictures of metabolic activity inside the body, revealing how tissues and organs are functioning at a cellular level, often beforestructural changes appear on other scans like CT or MRI

MEG

Magnetoencephalography

non-invasive brain imaging technique that measures the tiny megnetic fields produced by electrical currents from brain activity, offering high temporal and spatial resolution to map brain functions like langauge and movement, and pinpoint seizure origins for epilepsy surgery

which design best allows a researcher to make causal claims

experimental manipulation

in within subject design each participant

serve as their own control

which of the following is a key disadvantage of a within subject design

order effects

counterbalancing is used to

control for order effects

which statement best describes reliability

the consistency of a measurement

a confounding variable is

a variable that varies with the independent variable and affects the dependent variable

in psych research a result is considered significant when

• the probability of observing the data if the null hypothesis is true

random assignment is used to

equate groups on all extraneous variable

Current

actual flow of lectricity (charged particles) through a conductor. It is a measure of the number of charge units (electrons, or protons) that flow past a given point in a specific amount of time.

• unit: amperes (A)

Voltage

(or electrical potential) the pressure that pushes the electrical current through the conductor.

• unit: volt V

• Microvolt (UV)

Resistance

the ability ofa substance to keep charged particles from passing. the inverse of “conduction”. the length, diameter, composition of the substance determines the resistance

• unit: ohms (

Electricity and magnetism

the flow of current through aconductor is always accompanied by a magnetic field that flows around the conductor

• this is what MEG measures

Action potentials

discrete voltage spikes that travel from the beginning of the axon at the cell body to the axon terminals, where neurotransmitters are released

Postsynaptic potentials (PSP)

the voltages that arise when the neurotransmitters bind to receptors on the membran of the postsynaptic cell, causing ion channels to open or close and leading to a graded change in the voltage accross the cell membrane

• drives action potentials

ERPs almost alway reflect postsynaptic potentials (PSP) rather tha cation potentials. why?

• action potentials last about 1- ms, whereas PSP lasts 10- 100+ ms, allowing PSP from many neurons to summate and record at a great distance

  • scalp ERPs are thought to arise from cortical pyramidal cells (main input-output cells of the cerebral cortex

• Cortical pyramidal cells are all oriented perpendicular to the cortical surface with the apical dendrite heading in the direction of the cortical surface and the cell body and basal dendrites located closer to the white matter

Dipole

• a pair of positive and negative electrical charges seperated by a small distance (e.g., a pyramidal cell)

Equivalent current dipole

• summed dipoles from individual neurons within a folded sheet of cortex

• Arrowhead indicates positive voltage

Volume conduction

when a dipole is present in a conductive medium (e.g. the brain) current is conducted through that medium until it reaches the surface

• note that we’re measuring coltage, not current. thus we’re measuring a PSP in a set of neurons that produce an instantaneous voltage field throught the entirety of the head, with no delay

• however the skill (high resistance) causes the voltage to be even widely distributed. Thus the scalp distribution of an “ERP component” is usually very broad

The position and orientation of the equivalent current dipole determines the

distribution of positive and negative voltages recorded at the surface of the head

Electrodes that are perpendicular to the equivalent current dipole will of a voltage of

0

Magneticencephalogram (MEG) measures

the magnetic field of the quivalent current dipoles

• magnetic fields are perpendicular to the current

Dipoles that are tengential (parallel) to the skull can be recorded with MEG, but not those that are ______ to the skull

radial (perpendicular)

• this is opposite for EEG

Why is EEG better for measuring post-synaptic potentials than action potantials?

APs are rapid changes in electrical potential whereas PSPs are slower and graded

the slower more sustained PSP allows the signals to summate creating a stronger measurablesignal

Why isn’t EEG sensitive to all activity within the brain?

• the signal is may be weak if neurons are not synchronously active (timing of neurons)

• the signal is very weak or non-existent if synchronsly active neurons are in opposite alignment or randomly aligned (orientations of neurons)

• The signal is strongest when synchronously active neurons are spatially aligned

Forward problem: from dipoles to scalp ERPs

C= underlying component or source waveform generated by a dipole

E = electrodes on the scalp

• cannot directly dipole activty from electrical activity

  • combo of different sources and different weights

• Voltage measured at a given electrode sight is a weighted sum of all the underlying components

• the weights will be negative on one side of the head and positive on the other, with a “narrow” band where the weights are zero transitioning between positive and negative sides of the dipoles

• a given electrode will pick up at least some voltage from almost every component in the brain. How many components in an experiment?

• Picton et al.,: at least 10 different sources was found in a brief period from 50-200 ms after the onset of an auditory stimulus in a simple target detection task!!

Superposition probelm

• we’re interested in the “underlying components” not the mixture recorded at a given scalp electrode

• superposition problem - hwo to ‘recover’ the underlying components from this (scalp) mixture?

  • Dipole localization method, principle component analysis independent component analysis, fourier analysis, time-frequency analysis etc

• but the reality is that there no way to solve the superposition problem

  • all of the analysis methods above are based on assumptions that are either known to be false or are not known to be true…

Challenges of ERP localization

• A given voltage distribution can be produced by an infinite number of different dipole orientations and locations

Localizing ERP

• when the data are noisy, the problem becomes even worse

• one way to help ERP localization is to add external constraints (e.g., structural MRI to constrain the dipoles to be in the gray matter) to improve the non-uniqueness of the solutions

• another way is to do hypothesis testing and deductive reasoning

  • e.g., P3 was hypothesized to be generated in the hippocampus, but P3 was found to be intact in patients with medial temporal lobe lesions

Is an equivalent current dipole the same thing as an ERP component?

no

• a single ERP component can be explianed by multiple dipoles or distributed sources

  • a single dipole can contribute to multiple apparent ERP components across time or conditions

• An equivalent current dipole is a modeling abstraction of neural source activity, whereas an ERP component is a deceptive feature of the scalp-recorded waveform; the two should not be equated

Waveforme ERP peaks vs. underlying components

• because of the superimposition problem, it is oftern very misleading to look at ERP waveforms and interpret them as undeerlying components, why?

  • 6 rules of ERP interpretation

Waveform ERP peaks vs. underlying components cont..

• peaks dont equal components

  • observed ERP peaks do not usually have any particular physiological or psychological meaning

  • observed ERP peaks are usually unrelated to the time course of any individual underlying component

    • peak 1 in A does not equal c1’s peak in B

  • theories of cognition or brain processes do not usually say much about when a process peaks

• It is impossible to estimate the time course or peak latency of an underlying component by looking at the local part of the observed ERP waveform

  • e.g. C2’ vs. C2 are both possible underlying components producing the peak 2 in the observed ERP waveform

• An experimental effect (e.g. condition X vs baseline) during the time period of a particular peak may not rflect a modulation of the underlying component that is usally associated with the peak

  • looked at peaks → falsely thinking that at least 2 underlying components are modulated by condition x

  • or, falsely drawing inference that increased amplitude in peak 3 reflects an incres in the amplitudes of long-latency positive component

  • or it could be reflecting a 50% decrease of C2’ (i.e., an inytermediate-latency negative component)

• Difference as waves as a solution

• However, differences in peak amplitude do not necessarily correspond with differences in component size and differences in peak latency do not necessarily correspond with changes in component timing

• averaged ERP waveforms fo not always represent the individual waveforms that were averaged together

  • i.e., averaged ERP waveform (B) does not equal the single trial ERP wavefomr in A

• An ERP effect observed in one experiment may not reflect the same underlying brain activity as an effect of the same polarity and timing in previous experiments

  • e.g., if you conducted an experiment comparing two conditions (e.g., bilinguals reading nouns in two different languages) that supposedly elicit a larger effect around 400 ms, one cannot be sure of the underlying source

  • specificially, it would be unclear whether the eefct reflects an increased N400 or a decreased p3

the 6 rules of ERP interpretation

• peaks and components are not the same thing

• you can’t infer the time course or a peak latency of an underlying ERP component from an observed ERP waveform

• An effect during the time period of a particular peak may not reflect a modulation the underlying component that is usually associated with that peak

• you can’t infer differences in peak amplitudes as differences in component size, and you can’t infer differences in peak latency as changes in component timing

• onset and offset times in the average waveform represent the earliest onsets and latest offsets from the individual trials or individual subjects that contribute to the average

• An ERP effect observed in one experiment may not reflect the same underlying brain activity as an effect of the same polarity and timing in previous experiments

Difference waves

• tool for isolating components

  • help revealing the time course of an underlying component

  • a well-constructed difference waves always contain fewer components than the parent waveforms, thus less opportunity for confusion due to the mixing of components

  • logic of differences waves: see “forward” problem solving

• Importance of measuring scalp distributions from difference waves

  • gives an idea where things are coming from

• Difference waves can help avoid vuisual illusion when lookign at ARP effects on raw waveforms

Difference waves are great but…

• effects observed in difference waves cannot be uniquely attributed yto asingle somponent, as they may arise from multiple underlying component modulations

• Always noisier than the parent waveforms

  • however, a t-test against zero on a difference wave statistically equivilent to a one-way ANOVA on the parent waveforms

What is an “ERP component”?

• Historically, it was defined by its polarity, latency and general scalp distribution, but these are superficial features

  • but they are still helpful in determinging what an experimental effect reflects

• Conceptual definition: an ERP component is a scalp-recorded neural signal that is generated in a specific neuroanatomical module when a specific operation is performed (useless definition because too many unknowns)

• operational def: an ERP component is a set of potential changes that can be shown to be functionally relatedto an (well-controlled) experimental variableor a combination of them

  • ignoring that ERP components are generated by specific neuroanatomical modules

  • ignoring the presence of spontaneous or correlational variability beyound the well-controlled experimental variables

• Luck (ch2, ERP techniques

  • an ERP component can be operationalluy defined as a set of voltage chnages that are consistent with a single neural generator site and that systematicall vary in amplitude across conditions, time, individuals and so forth

  • an ERP component is a source of systematic and reliable variability in an ERP data set

Given we have a conceptual and an operational definition for ERP components, how can we determine which underlying components are responsible for observed differences?

• its difficult to compare between experiments (both waveformsand scalp distributions)

  • a slight change in the experimental paradigm can change ERP components and/or scalp distributions

  • fancy techniques such as source localization, ICA, etc can help

  • but its easier to find evidence against the hypothesis that two effects reflect the same component than in favor of a hypothesis

• start with well-designed experiments, clearly articulated hypotheses, and the accumulation of converging evidence

are source waveforms the same thing as observed scalp ERP waveforms?

no

source waveforms are generated by modeled dipoles, whereas ERP waveforms reflect the observed mixture of multiple source waveforms at the scalp

Amplifier

amplifies (adds voltage from power source) and filters voltage data

scalp electrodes

measure participants neural responses

EEG data acquisition computer

periodically samples and records voltage from the amplifier

Data monitor

displays EEG waves (channels) as software connects sampled time points

what do you need if you want to record neural responses to a specific event presented in an experiment (ERP data)

• Stimulus monitor

  • presents experiment to participant

• response device

  • collects participants behavioral response

• stimulus computer

  • controls experiment and presentation of stimuli

• stimulus monitor 2

  • mirrors experiment to experimenter

• cables

  • send event codes from stimuli and responses to EEG data file

Electrical potential (voltage)

The potential for current to flow due to the difference in charge between two points is measured in volts or microvolts

measuring voltage in EEG is similar to how we measure voltage in a battery

• electrical potential is a relative measure reflecting a difference in charge between two locations

• it takes two electrodes to measure voltage

  • active electrode (data) relative to the

  • reference electrode (baseline)

strength and polarity of voltage measurment

• the voltage strength is indexed by the absolute value of the difference in chrage between the two electrodes

  • active - reference

• the polarity (+ or -) depends on the relative positions of the two electrodes

what happens if we switch the positions of the active and reference electrodes?

the strength remains the same, but the polarity becomes negative

you need ____ electrodes to measure scalp voltage

2

• active electrode

  • the signal of interest

• ground electrode

  • primary reference electrode

• active - ground = brain signal + system noise

Adding asecond reference cancels what?

system noise

• active electrode

  • the signal of interest

• reference electrode

  • secondary reference (baseline)

• ground electrode

  • primary reference electrode

(active - ground) - (reference - ground) = active - reference

most studies use

multiple active scalp electrodes

electrode capes and configurations

• caps are used to save time and maintain standard scalp locations

• standard 10-10 configuration

  • all distance is 10% from NZ

• Standard 10-20 configuration

EEG electrode positions

refere to locations on the scalp NOT to locations in the brain

Electro-oculogram (EOG) electrodes

Electro-oculogram (EOG) electrodes index eye movements that create large voltage fluctuations in the EEG

horizontal electro-oculogram (HEOG) electrodes

electrodes are placed to the right and left to capture horizontal eye movements

vertical electro-oculogram (VEOG)

electrodes are placed below one eye to capture vertical eye movements and blinks

averaging ERPs to reduce

noise

Data contains signal +

signal + noise

Intro to simulation: the odball paradigm

• presents rare targets (red circles) among common standards (black circles)

• participants press one button for targets, and another for standards

Intro to simulation: noise can mask the effect

• P3 component: compared to standards, targets produce a greater positive deflection in the stimulus-locked waveform around 300 ms after stimulus presentation

• the difference between targets and standards in masked by single trial noise

Average multiple trials in each condition boosts

the signal to noise retio by decreasing the noise

Why does averaging reduce the noisebut maintain the signal?

• on each trial the event-related signal is always aligned to time 0, the onset of the stimulus, so averaging these data will provide a consistent measure of the event-related signal

• The noise is not time-locked to the stimulus, and so averaging over many trials reduces the noise

what are ERP components

• ERP components are deflections (peaks and troughs) in the ERP waveform that reflect the summation of neural activity in response to an event

LAbeling ERP components

• ERP components are typically labeled by their:

  • polarity (negative or positive)

  • order or peak latency

• Example: the P3 component

  • third positive deflection

  • sometime called P300 because it peaks around 300 ms after the stimulus

Waveform plot

shows how electric potentials change over time at a particular scalp location (e.g., a single electrode or set of averaged electrodes)

• positive polarity can be plotted up or down

Topographical map

• shows how distribution of electric potential change over scalp locations at a particular time

• these are scalp locations, not brain locations

We can described the physical properties of ERP components

• amplitude: how large is the deflection (uV)?

• latency: when does the deflection begin or peak (ms)?

Scalp distributions

• the scalp distribution for each component represents a different time-frome during which the component amplitude is maximal

• scalp distributions show which alectrodes have the largest amplitude for a particular component

• Amplitude and latency measurements for a components are reported from electrodes which typically have the highest amplitudes for that component

How would you describe the difference between targets and standards

the amplitude of the p3 component is higher for targets than standards

raw data contains information about

• the event (signal) plus information from other sources

3 main categories

• exogenous sensory components: triggered by the presence of a stimulus (butmay be modulated to some degree by top-down processes (e.g., goals, expectations, attentions, etc)

• endogenous components: reflecting neural processes that are task-dependent

• motor components: accompanying the preparation and execution of a motor response

naming conventions: P = positive going; N = negative going; P”300”: peaks at 300ms when it was first discovered; P”2”: 2nd major positive peak; functional names, e.g., “error-related negativity (ERN)”

Common ERP components

• CNV (contingent negative variation)

• C1, P1, N1

• N2 family (N2a, N2b = anterior N2, N2c = posterior N2

  • N2a, e.g., the Mismatch Negativity (MMN): rather automatic/pre-attentive (deviant vs. standard)

  • anterior N2 e.g., response inhibition, conflicts between response alternatives

  • posterior N2 (almost like the P3) e.g., N2pc, PD, CDA

• P3 family

  • Frontal P3a

  • parietal P3b (P3, P300)

• N400, ERN (error-related negativity), LRP (lateralized readiness potential)

• Steady-state ERPs

ERP components evoked by visual tasks

Link ERPs with neural and cognitive processes

• its not possible to get an one-to-one mapping between one specific ERP component and a specific functional process

• list of antecedent vs. a functional theory of what an ERP component reflect (e.g. the computational process served by the circuit that generate that ERP component

  • e.g., P3 is a manifestation of a process invoked in the service of the uupdating process, not necessarily the updating per se

• “reverse inference” error e.g. if P entails X, X does not necessarily entail P

• forward inference problem

• getting around the inference problem

  • accepting claims or less precise conclusion (1) if compontnt Y occurs, process P is probably active (2) settle for conclusions that do not involve the specification of highly precise cognitive functions

  • conducting a very involved program of resaerch designed to asses the relationship (boot strapping approach)

Reverse inference error

If P entails X, X does not necessarily entail P

Forward inference problem

testing the hypothesis that comonent Y occurs if and only if process P is active. But… if we don’t know when process P is active, we cannot test that hypothesis in the first place

• getting around the forward inference problem

  • accepting claims or less precise conclusion (1) if compontnt Y occurs, process P is probably active (2) settle for conclusions that do not involve the specification of highly precise cognitive functions

  • conducting a very involved program of resaerch designed to asses the relationship (boot strapping approach)

Example cognitive association: N1 and visual discrimination

how do we know?

• simple response

  • press button for any colorful stimulus

• discrimination response

  • press on button if red is present another button when red is absent

• The N1 component has a larger amplitude when visual discrimination is required

N170

• Jeffreys (1989): comparing faces vs. non-face stimuli, observed a difference from 150 to 200-ms at Cz → VPP (vertex positive potential)

• Bentin et al: N170 at lateral occipital sites

  • N170 and VPP are likely the opposite sides of thesame dipole

  • one of subcomponent of the N1 wave

• Does the N170 effect (face vs. non-face difference waves) truly reflect face-specificprocessing?

  • hypothesis:the N170 effect reflects face specific processing

Bentin et al

Q will simple stimuli (two dots) be perceived as part of faces elicit N170 when subjects are primed to perceive faces?

ERP components might be affected by

multiple cognitive processes