Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Advanced Methods > Brain Connectivity > Flashcards

Brain Connectivity Flashcards

1
Q

How many types of brain connectivity there are?

A

There are 3 types:

  • functional
  • structural/anatomical
  • effective
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is anatomical connectivity?

A

Anatomical connectivity refers to a network of physical or structural (synaptic) connections linking sets of neurons or neuronal elements, as well as their associated structural biophysical attributes encapsulated in parameters such as synaptic strength or effectiveness

Set of physical structures linking neuronal units at a given time
- local circuits to large state networks and inter-regional pathways
Relatively static and shorter time scales (Seconds-minutes), but can be dynamic and longer time scales e.g. hours to days (learning and development)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is functional connectivity?

A

It captures patterns of deviations from statistical independence between distributed and often spatially remote locations, measuring their correlation, spectral coherence or phase-locking

Functional connectivity refers to the functionally integrated relationship between spatially separated brain regions.

  • Time dependant (hundreds of Ms) and measures statistical interdependence without explicit reference to causal effects
  • Different methodologies of measuring the brain activity will result different statistical estimates of functional connectivity
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is effective connectivity?

A

effective connectivity is defined as the influence one neural system exerts over another
It requires the specification of a causal model
inferred though perturbations of observations of the temporal order of neural effects

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Explain the interaction between different types of connectivity

A
  • Mutually interrelated. E.g. functional and effective connectivity are constrained by structural connectivity
  • Structural inputs and outputs of a given cortical region, its connectional fingerprint are major determinants of its functional properties.
  • Functional interaction can contribute to the shaping of the undelaying anatomical substrate either though activity dependant synaptic modification or Through affecting an organism’s perceptual, cognitive or behavioural capabilities, and thus its adaptation and survival
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Define neural synchrony

A

Concerted interactions among neuronal populations or Direct reciprocal exchange of signals between two Populations, whereby the activity in one population Influences the second, such that the dynamics become Entrained and mutually reinforcing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are the types of measuring neural synchronisation?

A

Linear and nonlinear

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What are some examples of linear methods?

A
  • Linear correlation
  • Coherence (magnitude squared coherence and partial coherence)
  • Granger causality
  • Multivariate modelling (directed transfer function and partial directed coherence)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are some examples of nonlinear methods?

A
  • Nonlinear correlation
  • Information theory (mutual information and transfer entropy)
  • Phase synchrony (wavelet, mean phase coherence, Hilbert+Shannon)
  • Generalised synchrony (similarity index and families; mixed predictability; cross prediction)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What are the advantages of linear methods?

A
  • Simple to calculate
  • Easier to interpret
  • Availability of confidence limit
  • Backing of strong mathematical support
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are the disadvantages of linear methods?

A
  • Only linear coupling can be revealed
  • No directional information
  • Requirement of stationarity condition
  • Requires strong visual similarity between two signals
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are the advantages of non-linear methods?

A
  • Simple to calculate
  • Easy to interpret
  • Availability of confidence limit
  • Frequency related coupling information
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What are the disadvantages of non-linear methods?

A
  • Stationarity, linearity
  • Poor estimation for limited data
  • No directional information
  • Mixing amplitude and phase information
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are the advantages and disadvantaged of Granger causality?

A 
Advantages 
•	Availability of directional information 
•	Stochastic formulation 
Disadvantages 
•	Only linear coupling 
•	Causality is not always well-defined 
•	Biased estimation for limited data 
•	Parametric formulation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is Partial Directed Coherence (PDC)?

A
  • Detection of causality
  • Sensitivity on directed coupling
  • Separation of common driver
  • Multivariate modelling Complete co-variance structure
  • Frequency specific measure
  • Applicable to linear as well as nonlinear systems
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are the advantages and disadvantages of partial directed coherence (PDC)?

A 
Advantages 
•	Considers the structure of the whole data set 
•	Directional influence 
•	Faster computation 
•	Frequency related information 
Disadvantages 
•	Only linear interaction 
•	Limited statistical confidence 
•	Sensitive to pre-processing 
•	Parametric modelling
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What are the advantages and disadvantages of mutual information?

A

Advantages
• Reveals both linear and nonlinear coupling
• Reveals statistical interdependency
• Good theoretical backing

Disadvantages
• Difficult to estimate for limited data
• Sensitivity on noise
• Problem in estimation for high dimensional system
• No freq. related information

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is Phase synchrony?

A
  • Classically, phase synchrony is equivalent to adjustment of frequencies
  • Phase & Amplitude – Not necessarily convey identical information
  • Role of nonlinearity and internal noise

• !!! Detection of synchronisation is only possible in statistical sense!!

19
Q

What are the advantages and disadvantages of phase synchrony?

A

Advantages
• Neurophysiologic ally meaningful analysis
• Suitable for nonlinearity, nonstationary
• Sensitive to weak coupling
• Generalized formulation with fewer assumptions

Disadvantages
• Phase is meaningful only for narrowband signal
• Directional information is not immediately clear

20
Q

What is generalised synchronisation (GS)?

A
  • There exists a smooth mathematical map between the dynamics of two processes.
  • Dynamical Interdependency
21
Q

What are the advantages and disadvantages of GS?

A

Advantages
• Reveals both linear and nonlinear coupling
• Reveals dynamical interdependency
• Sensitive to weak interdependency

Disadvantages 
•	Coarse time resolution 
•	Computationally expensive 
•	Influenced by dimension mismatch 
•	Influenced by spatially heterogeneous attractors
22
Q

Definition of Phase slope index (PSI)

A
  • Based on the concept nonvanishingimaginary part of the coherence cannot be explained by volume conduction effect
  • So PSI detects noninstantaneous functional relationships
  • In short, PSI measures connectivity between two signals based on the slope of the phase of their cross spectrum
23
Q

What is PSI?

A
  • Very sensitive to the flow of information
  • Robust against volume conduction effect (i.e. mixture of neural sources)
  • Provides a relatively easy way of estimating significance
24
Q

What PSI isn’t?

A
  • Not a measure of instantaneous connectivity For bidirectional connectivity, PSI refers to the net direction
  • Poor representation for mutually identical information flow
  • Does not cover all forms of causal relationships
25
Q

Summary of Neuronal synchrony measures

A
  • Nonlinear measures (GS, PS) works better than linear measures (coherence, correlation), mutual information
  • For nonstationary data, PS is a better choice than GS
  • GS is more sensitive for weak interaction
  • For data with low SNR and few time samples, rely on linear methods
  • Apply more than one method
26
Q

What is network analysis?

A
  • Graphs as Models for Complex Systems
  • Complex Systems have properties that are neither completely random nor regular Showing non-trivial characteristics
  • Interconnected between sub-parts
  • Connections across scales
  • Interactions between bottom-up and top-down processes
  • Big systems (comprising millions of agents/units)
27
Q

What is the definition of a graph?

A
  • It is an abstract representation of a network
  • It consists of Vertices (or nodes)
  • Edges (or links, connections between the nodes)
  • The presence of an edge between two vertices indicates the presence of some kind of interaction or connection
  • In EEG/MEG sense, vertices correspond to electrodes/sensors, and edges correspond to some measure of functional/effective connectivity between pairs of electrodes/sensors.
28
Q

What is a matrix?

A
  • Matrices are useful for visualization of networks but contain no meaningful spatial or topological information
  • It offers a compact description of the pairwise connectivity between all nodes of
29
Q

What is a diagonal element?

A

connectivity of each node with itself, but usually encodes some intrinsic property of the node

30
Q

what is off-diagonal element?

A

Misrepresents the connectivity between pairs of distinct neural elements
• The value and the range depend on the type of methods used for connectivity

31
Q

What is directionality?

A
  • If we have different values in the upper and lower triangle, then Cis asymmetric, and the matrix represents a directed graph (digraphs) and the asymmetry encode the directions of connections
  • Different conventions of directionality but do pay attention on how the matrix is constructed
32
Q

What are binary graphs?

A
  • Brain networks can also be represented as unweighted or binary graphs
  • Binary networks tell us where connections are in the network, but provide no information about variations in connectivity weights between nodes
33
Q

What is an adjacency matrix?

A
  • In graph theory, two nodes that are directly connected by an edge are said to be adjacent or neighbours
  • The matrix defines the pattern of pairwise adjacency between nodes
34
Q

Why do we need thresholding?

A
  • Understanding the topological patterns of connections between nodes, irrespective of weights variations.
  • Focusing only on connections and their spatial configurations can provide insights into the basic architectural plan of a brain network
  • Many measures are much simpler for binary networks and easier to interpret
35
Q

How to choose the global threshold?

A
  • Choose in a way that is independent of the condition of comparisons Pool connectivity values across all conditions
  • Select the threshold
  • Apply to individual adjacency matrix
  • Select separate thresholds for individual frequency bands
  • A threshold of one s.t.dabove the median is often used
36
Q

What are two important things to remember?

A
  • Whatever threshold you have chosen, you do need to repeat your subsequent analysis for a range of threshold values and inspect the impact of your choice of threshold
  • The adjacency matrix forms the basis for all subsequent network analysis
37
Q

What types of degrees there are and what do they mean?

A
  • Network degree: the total number of edges (2E)

* Connectivity degree: for each node, count the number of edges (finding a hub)

38
Q

What is a degree distribution?

A
  • The likelihood P(k) that a randomly chosen node will have degree kis given by the degree distribution
  • It is a plot of P(k) vs k
  • It can have different forms: Gaussian, binomial, Poisson, exponential or power law
  • The degree distribution is an important property of network
39
Q

What is clustering coefficient?

A
  • Can be measured both at the local node and at the global network level
  • It is an index of local structure, and has been interpreted as a measure of resilience to random omission (i.e. if a node iis lost)
40
Q

What is Path Length?

A
  • It is the average shortest path length between all possible pairs of nodes in a network
  • Where liis the average shortest path length from node ito all other nodes and lijis the shortest path length from node jto node i
  • It is a measure of integration in brain networks
  • A network with a short characteristic path length are supposed to integrate information more efficiently between nodes
  • A short path length means that information, on average, be routed between pairs of nodes using only a few edges Minimizes the metabolic cost, requires less energy, and hence provides faster, more direct and less noisy information transfer
41
Q

What are the three core measures?

A
  • The degree distribution, clustering coefficient, and characteristic path length can distinguish major types of networks Ordered
  • Random
  • Small-world
  • Scale-free
42
Q

what are the network features of a healthy brain

A
  • High clustering
  • Short characteristic path length
  • Hierarchical modularity
  • Scale-free degree distribution
  • Presence of hubs (that are interconnected in a rich club)
43
Q

Summary of network analysis

A
  • This new approach provides new insight into the topological and dynamical properties of brain networks
  • It allows us to understand the correlations between network structure and the processes takes place within the network
  • Different network measures can be used as biomarkers
44
Q

What are some issues of network analysis?

A
  • Interpretation problems Methods are not developed to study brain in mind
  • Depends on the types of methods (imaging and data analysis)
  • Lack of proper control or reference networks
  • Thresholding problems
  • Multiple comparison problems
  • Mindless applications

Advanced Methods (9 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions