Cognitive Neuroscience Methods Independent Variables

Post on 13-Jun-2015

561 views 2 download

Tags:

Transcript of Cognitive Neuroscience Methods Independent Variables

Cognitive Neuroscience Methods

• Independent Variables– Cognitive psychology– Computer modeling– Animal experiments– Neurology and neuropsychology

Cognitive Neuroscience Methods

• Dependent variables– EEG and ERP– MEG and ERF– CT– MRI– PET/SPECT– fMRI– Performance: RT % correct– Subjective– Other physiological

Cognitive Psychology

• Information processing– Mental representations– Transformations (stages)– Chronometric methods

Cognitive Psychology

• Information processing– Mental representations

• Many be generated simultaneously (in parallel)• Posner letter matching task

Cognitive Psychology

• Information processing– Mental representations– Transformations (stages)

• Representation 1 as input• Mental operation performed• Representation 2 as output• S. Sternberg serial comparison task

Cognitive Psychology

• Information processing– Mental representations– Transformations (stages)– Chronometric methods

• Method of subtraction (Donders)• Additive factors (S. Sternberg)

Cognitive Psychology

• Chronometric methods– Method of subtraction (Donders)

• Latency of stages can be determined by comparing RT from different tasks

• A task - simple RT• B task - choice RT• C task - disjunctive (go-no go) RT• C RT minus A RT = stimulus identification time• B RT minus C RT = response selection time

Cognitive Psychology

• Chronometric methods– Additive factors (S. Sternberg)

• RT is sum of independent (nonoverlapping) stages that perform a transformation

• Transformation is independent of its own duration and the duration of preceding stages

• Two experimental manipulations that effect a common stage will produce an interactive effect, whereas two experimental manipulations that effect different stages will produce an additive effect

Cognitive Psychology

• Information processing• Constraints on information processing

– Subtractive method• Assumption of pure insertion

– Additive factors• Words and word categories as stimuli• Automaticity/experience• Unequal stimulus probabilities

– Dual-tasks

Cognitive Neuroscience Methods

• Independent Variables– Cognitive psychology– Computer modeling– Animal experiments– Neurology and neuropsychology

Computer Modeling

• Simulation

• Artificial intelligence (AI)

• Explicit

• Testable

Computer Modeling

• Production systems (condition-action or if-then)– ACT-R – 4CAPS – EPIC– SOAR

• Computational (neural nets)

Computer Modeling

• Production systems (condition-action or if-then)

• Computational (neural nets)– PDP

Cognitive Neuroscience Methods

• Independent Variables– Cognitive psychology– Computer modeling– Animal experiments– Neurology and neuropsychology

Animal Experiments

• Single-cell recording

• Lesion studies

• Genetic manipulations– Selective breeding– Knock-out

Cognitive Neuroscience Methods

• Independent Variables– Cognitive psychology– Computer modeling– Animal experiments– Neurology and neuropsychology

Neurology and Neuropsychology

• Neurology = pathology of CNS

• Neuropsychology = assessment of functional psychological consequences of pathology

Neurology and Neuropsychology

• Neurological disorders – Vascular (Strokes)

• Ischemic – Embolism– Hypotension

• Hemorrhagic– Aneurysm

Neurology and Neuropsychology

• Neurological disorders – Vascular (Strokes)– Tumors

Neurology and Neuropsychology

• Neurological disorders – Vascular (Strokes)– Tumors– Degenerative and infectious diseases

Neurology and Neuropsychology

• Neurological disorders – Vascular (Strokes)– Tumors– Degenerative and infectious diseases– Trauma– Epilepsy

• "Split-brain" patients

Converging Methodologies

• Integration of multiple methodologies– Cognitive neuropsychology = cognitive

psychology + neuropsychology – ERP and fMRI

• Limitations– Brain damage may effect function of undamaged

regions– General vs specific loss of function– Appropriate control conditions

Converging Methodologies

• Integration of multiple methodologies• Single and double dissociations

– Single• patient group differs from controls on one task but not on a second

task• assumes tasks are equally difficult

– Double• one patient group differs from controls on first task but not on

second task• second patient group differs from controls on second task but not

on first task

Converging Methodologies

• Integration of multiple methodologies

• Single and double dissociations

• Groups vs individuals

Converging Methodologies

• Integration of multiple methodologies

• Single and double dissociations

• Groups vs individuals

• Transcranial Magnetic Stimulation (TMS)– "Virtual" lesion

Cognitive Neuroscience Methods

• Dependent variables– EEG and ERP– MEG and ERF– CT– MRI– PET/SPECT– fMRI– Performance: RT % correct– Subjective– Other physiological

EEG and ERP

• Electrical activity of populations of neurons

• Inexpensive

• Good temporal resolution

• Poor spatial resolution

• EEG "smeared" by meniges, skull, & scalp

EEG and ERP

• Dipole Modeling

MEG and ERF

• Magnetic activity of populations of neurons

• Good temporal resolution

• Better spatial resolution than EEG

• SQUID - Super-cooled Quantum Interference Device

CT or CAT

• Computed Tomography

• X-ray absorption

• Multiple 2-D "slices"

• Poor spatial resolution

• Poor structural resolution

MRI

• Magnetic Resonance Imaging

• Multiple 2-D "slices" (pixel)

• 3-D reconstruction by computer (voxel)

• Good spatial resolution

• Good structural resolution

MRI PrimerSam Patz, Ph.D.

http://www.med.harvard.edu/AANLIB/sigsors.html

• When protons are placed in a magnetic field, they oscillate. (protons + neutrons = odd number)

• The frequency at which they oscillate depends on the strength of the magnetic field.

• Protons are capable of absorbing energy if exposed to electromagnetic energy at the frequency of oscillation. After they absorb energy, the nuclei release or reradiate this energy so that they return to their initial state of equilibrium. This reradiation or transmission of energy by the nuclei as they return to their initial state is what is observed as the MRI signal.

MRI Primer (cont.)Sam Patz, Ph.D.

http://www.med.harvard.edu/AANLIB/sigsors.html

• The return of the nuclei to their equilibrium state does not take place instantaneously, but rather takes place over some time.

• The return of the nuclei to their initial state is governed by two physical processes: – the relaxation back to equilibrium of the component of the nuclear

magnetization which is parallel to the magnetic field, and – the relaxation back to equilibrium of the component of the nuclear

magnetization which is perpendicular to the magnetic field. • The time that it takes for these two relaxation processes to take place

is roughly equal to: – time T1 for the first process (longitudinal magnetization)– time T2 for the second process (transverse magnetization).

MRI Primer (cont.)Sam Patz, Ph.D.

http://www.med.harvard.edu/AANLIB/sigsors.html

• The strength of the MRI signal depends primarily on three parameters. – Density of protons in a tissue: The greater the density of protons, the

larger the signal will be. – T1 – T2

• The contrast between brain tissues is dependent upon how these 3 parameters differ between tissues.

• For most "soft" tissues in the body, the proton density is very homogeneous and therefore does not contribute in a major way to signal differences seen in a image.

• However, T1 and T2 can be dramatically different for different soft tissues, and these parameters are responsible for the major contrast between soft tissues.

MRI Primer (cont.)Sam Patz, Ph.D.

http://www.med.harvard.edu/AANLIB/sigsors.html

• T1 and T2 are strongly influenced by the viscosity or rigidity of a tissue. Generally speaking, the greater the viscosity and rigidity, the smaller the value for T1 and T2.

• It is possible to manipulate the MR signal by changing the way in which the nuclei are initially subjected to electromagnetic energy. This manipulation can change the dependence of the observed signal on the three parameters: proton density, T1 and T2. Hence, one has a number of different MR imaging techniques ("weightings") to choose from, which accentuate some properties and not others.

• Dark on T1-weighted image: – increased water, as in edema, tumor, infarction, inflammation, infection,

hemorrhage (hyperacute or chronic) – low proton density, calcification – flow void

• Bright on T1-weighted image: – fat – subacute hemorrhage – melanin – protein-rich fluid – slowly flowing blood – paramagnetic substances: gadolinium, manganese, copper – calcification (rarely) – laminar necrosis of cerebral infarction

MRI Primer (cont.)Sam Patz, Ph.D.

http://www.med.harvard.edu/AANLIB/sigsors.html

• Bright on T2-weighted image: – increased water, as in edema, tumor, infarction,

inflammation, infection, subdural collection – methemoglobin (extracellular) in subacute hemorrhage

• Dark on T2-weighted image: – low proton density, calcification, fibrous tissue – paramagnetic substances: deoxyhemoglobin,

methemoglobin (intracellular), iron, ferritin, hemosiderin, melanin

– protein-rich fluid – flow void

MRI Primer (cont.)Sam Patz, Ph.D.

http://www.med.harvard.edu/AANLIB/sigsors.html

Normal tissue

MR-T11 MR-T21 xray-CT2

dense bone dark dark bright

air dark dark dark

fat bright bright dark

water dark bright dark

brain "anatomic"3 interm. interm.

1. Bright means high signal intensity, dark means low, and interm. means intermediate.

2. Bright means high density/high attenuation of x-rays, dark means low.

3. Grey matter appears grey, white matter white.

MRI Primer (cont.). http://www.med.harvard.edu/AANLIB/

Abnormal tissue

MR-T1 MR-T2 xray-CT

infarct dark bright dark

bleed bright bright bright

tumor dark bright dark

MS plaque dark bright

MRI Primer (cont.). http://www.med.harvard.edu/AANLIB/

PET/SPECT

• Positron Emission Tomography• Single Photon Emission Computed Tomography• Measures rCBF• Invasive

– Injection of radioactive isotope that metabolizes with blood glucose

• Poor spatial resolution• Poor temporal resolution

fMRI

• Functional MRI• Measures rCBF• Noninvasive - can be obtained with MRI

– T2* = T2 relaxation time– BOLD response

• Change in venous oxygenated hemoglobin

• Good spatial resolution• Better temporal resolution