PSYC358 Midterm

brain development - structural vs functional changes

structural changes - shape, size, components
functional changes - activation patterns, connectivity, efficiency

how many nuerons are in the brain?

128 billion

action potential

electrical signal for neurons to communicate through neurotransmitters

organization of neurons in the brain

- neurons are organized as clusters
- wire brain to balance btwn speed and cost
- airport metaphor - not every neuron is connected to everything to be more efficient
- con: if one hub goes down it's really bad

functional selectivity

neurons that respond to similar information are grouped together (what fires together wires together)

white matter

- tissue of the NS primarily made of axons and support cells
- allows for connections btwn diff regions
- more is better

white matter tracts

bundles of axons tied together
- white matter tracts project between diff cortical regions in the same hemisphere (left PFC to right PFC), between diff cortical regions in between hemispheres (frontal lobe to parietal lobe), between cortical and subcortical structures (parietal lobe to amygdala), and subcortical regions to subcortical regions

grey matter

- primarily the cell bodies
- less is better, indicates how much synapses have proliferated
- peaks at age 2

frontal lobe

integration of other functions, attention, decision making, and planning

parietal lobe

motor function

temporal lobe

memory, hearing, language, social functions

occipital lobe

sight, sensory systems/awareness of world, movement planning

gyri vs sulci

gyri - top of folds
sulci - valleys of folds

corpus callosum

white matter tract connecting the 2 hemispheres

ways to grow brain system

- thicker myelination
- reorganize for efficiency
- add more neurons (more dendrites)

growing the brain from the neuronal level

- new neurons, growth in dendrites, thicker myelin coating on axons (faster flights)

well-tuned = more efficient = more likely to be used

pruning

- because you can't just keep adding stuff
- less used connections become weaker and die off --> efficiency
- cell either dies or dendrites decrease
(use it or lose it)

GM vs WM development

grey matter:
tuning+pruning - big boom then decrease

white matter:
- big boom in first few years then gradual increase
- not as much pruning bc we need myelin>cell bodies

how do growth patterns vary across different brain regions?

basic functions (back of brain) mature first while complex functions mature later (front of brain)

how to grow the brain system

- at network level (connectivity) - subcortical connectivity is faster bc closer together + fewer ppl
- short-long range connections

childhood-->adolescence-->young adult:
1. subcortical-subcortical
2. subcortical-cortical
3. cortico-subcortical
4. cortico-cortical

methods to measure brain structure (mainly WM/GM)

- post-mortem brain tissue analysis
- structural MRI
- diffusion tensor imaging

post-mortem brain tissue analysis

- shows brain structure changes w age through staining and tracing

staining: identifying diff types of cells (ex: measures myelin thickness in diff ages to show changes over time)
tracing: injecting dyes to see connections btwn neurons; NTs/receptor growth

structural MRI

- reads signal from hydrogen molecules in the brain
- H molecules are responsive to behaviour
- behaviour of H after scanner pulse is diff in various tissues
- can measure size of diff regions and compare GM and WM in diff parts
- H have a resting state then is perturbed by some behaviour so we measure time to return to OG state
- can determine if it's GM vs WM based on that time it takes

diffusion tensor imaging

- measures water molecules in WM tracts
- can map out efficiency of water in the tissue
- water goes thru the tube and myelin axon is fatty which repulses water so water going straight thru is most efficient (high fractional anisotropy)

the brain (mostly) develops until age 25

- long development period compared to other species
- brain dvpt is guided by environment (epigenetics)
- brain is plastic (experience-expectant vs experience-dependent)

overview of measuring brain activity

- can measure other stuff than brain structure
- localization of brain response, amplitude of brain response, communication btwn regions

temporal vs spatial resolution

temporal resolution: how specific can you get about timing?
spatial resolution: how precisely can you see where a given activity is happening?

electrophysiological measures

- look at electric current, specifically changes in the electrical responsiveness of a neuron
- EEG (more common) and MEG

hemodynamic measures

- look at blood flow or blood oxygenation to infer activity
- fMRI (more common) and PET (rare bc radioactive lesions so only rly used in clin when necessary)

electroencephalography (EEG)

- comes from post-AP current in synapse (btwn neurons)
- EEG principles based on idea that if a group of neurons are active synchronously, electrical field will be detectable at the scalp
- must look at cluster in a functional hub, not just 1 neuron
- look at electrical activity in certain brain region and compare to another reference point (often nasal site)
- EEG waves: oscillations from neurons firing at same rate (can have diff frequency of waves: gamma, beta, alpha, theta, delta)
- oscillations arise bc large groups of neurons are often in temporal synchrony
- decrease in power when observing someone else do smth (MN)

how do we use EEG

- time-lock to event: put marker down when P does smth and look at waves
- average across many trials: to signal out noise from random activity (hard w kids bc they get bored)
- compute ERPs: see timing and amplitude of peaks (tells u when smth is happening before any response but poor spatial resolution)

N170

- ERP w negative peak at 170 ms
- selective for face processing (includes human, animal, and emoji faces)
- at right posterior temporal electrode sites

EEG pros

- good temporal resolution at ms level
- cheap per person
- accessible
- better for kids bc they can move

EEG cons

- bad spatial resolution (not precise)
- can't get to subcortical

alternatives to EEG

- single-cell recording/electrocorticography (take off skull)
- magnetoencephalography (MEG)
- fMRI

functional magnetic resonance imaging (fMRI)

- helps determine if it's GM vs WM based on tissue properties (same as MRI)
- infers activity of neurons based on blood oxygen level in the brain (brain/neurons need oxygen to function)
- more active neurons need more oxygen so when a brain region uses more blood we can infer that area's more active
- fMRI measures concentration of oxygenated blood relative to deoxygenated blood in a given area, relative to stimulus timing

deoxygenated vs oxygenated blood in fMRI

oxy: clear signal
deoxy: full of magnetic properties, noisy signal so bad quality in scanner

blood oxygen level dependent (BOLD) signal

- higher when there's more oxy>deoxy hemoglobin
- deoxy molecules disrupt magnetic field and reduce MR signal
- neuron calls for oxygen (showing they're active) so neurons eat the oxygen so those molcules become deoxy and now mess w signal
- from this we generate the hemodynamic response function (HRF)

hemodynamic response function (HRF)

- demonstration of BOLD signal when that process happens
- fMRI measures shape of HRF in diff regions at diff times
- looking to fit that curve to ur blood oxy signal
- higher BOLD signal=active neurons of that area
- there's always blood in the brain so must time-lock signal to event
- fMRI - compare signal for a given event to a comparison event - contrast

fMRI analysis: activation patterns

- brain is separated into voxels (not real, grid we impose onto 3d brain)
- cluster of voxels (larger than that of change) is required to be activated

fMRI functional connectivity

- networks of regions that work together on a task
- can be studied using task-based or at rest
- doesn't need structural connectivity

fMRI pros

- high spatial resolution
- reaches subcortical structures

fMRI cons

- poor temporal resolution (blood flow is slow)
- may be hard for kids to stay still
- expensive, inaccessible (tight space and bad w medal)

components of emotions

1. subjective feeling
2. physiological response
3. cognition
4. expression
5. action tendencies
6. neural response

interoception

brain's ability to sense the current state of it's internal organs

features of emotions

- transient (compared to mood which extends over time)
- have hedonic value (liked/disliked)

functions of emotions

SURVIVAL VALUE
- survival, communication, social bonds
- central to learning
- centralizes attn to important stuff
- makes events more salient --> memory

historical theories of emotion

darwin: expressions/interpretations are innate
james-lange: stimulus-->physio-->feeling
schachter-singer: bodily experiences can enhance conscious emotional experience but don't create emotions
cannon-bard: hypothalamus is central to emotions

basic emotion theory (panksepp and ekman)

- emotion kinds can be binned into categories
- universal and innate
- emphasis on facial expressions as indicators
- look angry=feel angry
- facial affect coding system (action units)
- happiness, fear, surprise (mb), disgust, anger, sadness

constructivist theory of emotion (LF barrett)

- emotions are arousal+valence dimensions
- differentiation of emotions is based on context, language/label of experience
- emotion requires appraisal
- same emotion can be dimensional (ie hot vs cold anger)

representation of emotions in the amygdala: monkey study

- monkeys had amygdala removed and were suddenly v submissive, change in social hierarchy so does amygdala=aggression

representation of emotions in the amygdala: rat study

- classical conditioning rat shock studies
- remove amygdala=no more fear so does amygdala=fear
- mb bc memory and learning

representation of emotions in the amygdala: epilepsy study

- 20 adults w/o epilepsy
- 19 w left temporal lobe epilepsy
- 18 w right temporal lobe epilepsy
- usually damage to amygdala and HC
- shown contrast - fear faces>landscapes to isolate neural response
- Ps showed reduced neural response in amygdala

amygdala role in emotion

- responds more to neg>pos stimuli
- projects to sensory areas (fusiform gyrus) which may influence early stages of sensory processing and modulate cortical response
- projects to brainstem, HT, anterior cingulate cortex which may influence autonomic and endocrine processes
- projects to HC which may influence stored info/memory
SALIENCE DETECTION (signals to other systems)

orbitofrontal cortex role in emotion

- coding motivational value of rewards and changes the value of rewards based on context
- tracks changes to motivation (drive towards smth)
- decision making in context of emotion

PFC role in emotion

- connected to many sensory areas, amygdala, and temporal poles
- integration and regulation: integrates external and internal cues and modulated emotional behaviour

PFC and amygdala connectivity across age

- interaction btwn top-down processes and affective processes
- PFC - cortical, integration
- amygdala - subcortical, salience detection
- less regulatory capacity of PFC in adolescence --> more risk-seeking behaviour

anterior cingulate cortex role in emotion

ventral ACC
- prediction
- connected w amygdala (salience), OFC (motivational value of stimuli), etc
- prediction and DM = if i do x, what will happen

dorsal ACC
- connected w PFC, motor areas, etc
- cog processing
- attn and exec functioning
- monitoring of responses/errors (errors are v salient)
- computing costs/benefits of an outcome

error-related negativity (ERN)

negative deflection in ERP 50ms after error, in dorsal ACC electrodes
- direction of attn; monitoring errors
- linked to anxiety in adults
- marker of where attn is being directed

ERN and anxiety: flanker task

- 8-13 y/o
- parent report of anxiety
- tap side the middle fish is looking
- brain launches little ERN when u make mistake
- adults w anxiety=bigger dip (ERN)
- 13 y/o: more anxiety-->bigger ERN mb bc more social expectations
- 8 y/o: more anxiety-->less ERN

insula role in emotion

- interoception (sensing internal organs), perceiving one's own body state (taste, pain perception)

central nervous system

brain and spinal cord

peripheral nervous system (PNS)

somatic NS and auto NS (SNS and PsNS)

measuring PNS: electromyography

- measures muscle contractions
- somatic NS measure
- places electrodes on ppls muscles and measures APs in the muscle fibers
- measures microvolts as a measure of how active they are
- compare to reference electrode
- can measure small muscles in face
ex: study looked at whether 3 y/o mirro these facial muscles when seeing diff expressions (EMG shows frown muscle used more when kids see anger expression)

measuring PNS: skin conductance response

- sweat response, measures ANS
- place electrodes on 2 fingers and apply a weak electrical current
- electrical signal flows better when u sweat more
- heightened arousal=sweat more=better electrical signal flow

measuring PNS: pupillometry

- dilation of pupils, measures ANS
- often look at response to stimuli
ex: chatroom task: greater pupil dilation when rejected (even more-so in 17>9 y/o bc social feedback is more salient)

measuring behaviour

- performance: accuracy
- performance: response time (ex: kahoot - faster face perception for happy>neutral faces)
- observational measures: frequency of behaviour, specific action, looking patterns (preferential looking and habituation paradigms)
- survey measures
- neuropsych tests (often used in lesion studies)

lesion methods: naturally occurring

- reverse engineering: based on logic of functional specialization
- double associations: inform involvement of X in a given function (broca's and wernicke's aphasia)

lesion methods: with transcranial magnetic stimulation

- apply temporary stimulation with magnets or electrical current
- simulate lesions, better able to choose region to look at
- gets w topographic representation in the brain

fMRI - challenges with kids

- takes a while
- braces disrupt image
- expensive
- change in brain structure/BOLD w age

EEG - challenges with kids

- cap discomfort
- requires many trials

performance measures - challenges with kids

- motor requirements (adults>kids)
- attn to task
- change in competence or sensitivity of test

observational measures - challenges with kids

- reliability of coding

survey measures - challenges with kids

- validity/reliability w sample age
- lack of self-report capacity for young kids

bruce and young model

- structural encoding of facial features
- matching to memory representations determines identity, links to conceptual knowledge, and matches to 3d structural representation so u can recognize face from diff pics/angles
- gaze processing (where they're looking, attending to, feeling)
- expression analysis (controversial bc interactive w facial recognition)

streams the brain process faces through?

ventral visual stream: what is it? face perception
dorsal visual stream: where is it?

how does the brain process faces? haxby et al. model

specialized regions:
- occipital face area (OFC)
- fusiform face area (FFA)
- superior temporal sulcus (STS)

extended system
- auditory cortex
- limbic system (amygdala, insula...)
- temporal poles

occipital face area (OFA)

selectivity (greater BOLD response to faces>other things)
- early coding of physical facial features
- BOLD signal habituates (smaller BOLD response over time bc eventually those neurons need less oxygen to activate) to 'same' stimuli
- no habituation to diff facial features meaning it's coding for those
- OFA codes face composition
- OFA is sensitive to physical changes in the stimulus
- OFA projects to FFA and STS

fusiform face area (FFA)

selectivity (greater BOLD response to faces>other things)
- codes invariant aspects of the face (identity)
- BOLD signal habituates to diff angle of same person bc it codes invariant aspects of the face
- no habituation for different people bc FFA is sensitive to changes in identity
- if u don't realize the person is diff then ur FFA will habituate
- is FFA specialized for faces or expert discrimination (u can make people experts at anything like greebles and FFA will have same response patterns)

prosopagnosia

- inability to recognize familiar faces
- some experts in other things have this
- shows you can be a visual category expert w/o having good face perception

superior temporal sulcus

- codes changeable aspects of the face (eye gaze, expression, etc)
- when u see 2 diff ppl, ur FFA activates bc ur making decisions about identity
- when seeing same person of 2 diff orientations, ur STS activates bc u make decisions about things like eye gaze
- responds to dynamic info (dots walking vs random - STS responds to walking)
- multi-sensory inputs - integrate diff modalities

development of face processing

- activity improves but we also get different neural representations and strategies at different ages
- we become face experts with experience

development of facial perception

by 1 month - preference for face-like>non-face-like stimuli (tested w habituation or preferential looking paradigm)
- system is shaped by input it receives (ingroup faces>other stimuli)
by 4 days - infants look longer at moms face>strangers
- bad visual acutity so test for olfactory cues, external facial features, and vocal cues (effect goes away w/o talking for 4 days) so probs a mix of features
by 1 month - discriminate static mom pictures vs strangers
by 3 months - discriminate different angles of mom pics vs strangers

face recognition: feature vs configural processing

featural processing: using different features to differentiate
configural processing: using configuration of the features to differentiate

proof of configural vs feature processing

- inversion effects (slower RT when saying faces are same vs different bc need to switch from configural to featural)
- adults recognize faces from relational info > features
- measure configural processing thru impairment from inversion
- see impairment from inversion at 10 yrs

proof of configural vs feature processing in infants

- just bc impairment is small doesn't mean it doesn't happen
- study habituated 7 mos to 2 faces
- then showed them familiar, novel, or switched face
- if kids use configural processing, they'll dishabituate (switched>familair)
- when inverted, familiar and switched were equal
INFANTS USE CONFIGURAL PROCESSING but less often/robustly than adults

neural responses to faces throughout development

adult - N170 upon face presentation (disrupted by inversion
6mos - P400 upon face presentation (preceded by N290)
12mos - P400 upon face presentation (preceded by N290) + start of N170, disrupted by inversion

infants (mostly featural)-->adults (mostly configural)
- developmental change in N170 latency reflects increased processing speed w age

neural responses to faces - 2 month old study

- control diodes < faces
- FFA - higher activation for faces
- location of activation consistent w adults but also more diffuse
- infants have a multi-modal response to faces
- facial activation for infants is more diffused across the brain compared to adults whose facial activation is more localized

neural responses to faces - fMRI study

- 5-8, 11-14, and adults
- younger kids don't show specialization of FFA, OFA, and STS response to faces>other visual stimuli

neural representation of emotion in faces - pathways

2 pathways - subcortical and cortical

fast evaluation of stimulus via subcortical pathway (inside brain, midbrain)
- superior colliculus, pulvinar, amygdala
- may be more relevant in dangerous systems
- subcortical --> impression from amygdala is fast
- amygdala is connected to FFA so FFA responds to emotional faces

slow evaluation of emotion via cortical pathway
- OFA, STS, FFA

proof: damage to 1 pathway doesn't interfere w the other
- ex: ppl w visual cortex damage (occipital areas) can still perceive emotions in faces w/o having the conscious face

neural representation of emotion in faces - integration with distributed network of brain regions

- emotion representation (limbic regions) - subjective feeling of emotions - embodiment of them)
- conscious labelling (frontal regions) + interpretation of emotion
- sensorimotor simulation (motor/sensory regions) - mirror neurons?

fusar-poli et al.

meta-analysis of fMRI studies on processing of emotional faces (vs neural)
- fusiform gyrus (FFA), amygdala (many studies looking at fear), ACC (error monitoring, DM in context of reward), insula

does the neural representation of emotion in faces change with age?

- emotion influences the N170 in adults but not 7mo infants
kids/adolescents show
- greater amygdala activation and less PFC activation to emotional faces
- positive connectivity btwn PFC and amygdala
- amygdala leads the PFC in adolescence and in adulthood it's a more equal or opposite response

moore et al.

- 9-11 and 12-14
- emotional faces vs nothing
- puberty status starts predicting neural activation (amygdala, VS, PFC, temporal pole) around 12-14 yrs
- positive correlation btwn pubertal status and activation in extrastriate (visual) cortex, thalamus, amygdala plus at 12-14 yrs temporal pole, vlPFC, vmPFC

neural representation of emotion in bodies

fusiform body area (FBA) and extrastriate body area (EBA)
- early visual input about body representations
- equivalent to FFA - respond more to bodies>other things
- FBA - whole bodies>parts
- EBA - body
- FBA responds in graded way

superior temporal sulcus
- involved in bio motion perception

neural representation of emotion in bodies - fast evaluation of stimulus via subcortical pathway

- amygdala, basal ganglia motor circuits, pulvinar, superior colliculus
- similar to fast/slow face eval

neural representation of emotion in bodies - slow evaluation of emotion via cortical pathway

- visual analysis in EBA, FBA, STS
- action observation network (premotor cortex, inferior frontal gyrus, intraparietal sulcus)
- mentalizing network (TPJ, temporal pole, mPFC)

proof of parallel between body and face processing

- when does the smile drop study
- people w pen in mouth say it drops later

development of response to emotional bodies

3.5 months - discriminate emotional bodies
5-6 months - recognize most emotions in bodies above change
- development of capacity to read emotional bodies thought to be supported by simulation with experience (MN) and mentalizing

ross et al.

6-11, 12-17
- passively viewed happy, angry, and neutral body movements
- showed increase in response to emotions