Cognitive Neuroscience: Introduction and Neuroanatomy

Introduction to Cognitive Neuroscience

• Cognitive Neuroscience definition: An interdisciplinary field examining how cognitive functions like perception, memory, decision-making, and language are implemented in the brain
• Combines the methodology of psychology, neuroscience, and computational modeling to study the neural mechanisms of human thought and behavior

Basic Neuroanatomy

• Knowledge of brain lobes and their functions is important to interpret research in cognitive neuroscience
• Frontal Lobe: Located at the front, it is responsible for higher-order cognitive functions like decision-making, problem-solving, voluntary motor control, impulse regulation and personality
• Parietal Lobe: Situated behind the frontal lobe, it processes sensory information, integrates spatial awareness, and coordinates movement and perception
• Temporal Lobe: Found on the sides of the brain, it is critical for auditory, language, memory storage and emotional processing
• Occipital Lobe: Located at the back of the brain, it primarily processes visual information like shape, color, and motion perception

Subcortical Structures

• Hippocampus: Critical in forming new memories and spatial navigation
• Amygdala: Implicated in processing emotions, particularly fear and reward
• Basal Ganglia: Important for movement coordination, habit formation, and reinforcement learning
• Thalamus: The brain's relay center and directs sensory information to appropriate cortical areas

Brain Orientation & MRI Imaging Directions

• Knowledge of brain orientations is needed to understand neuroimaging data
• Axial View: a horizontal slice of the brain showing top-down perspectives
• Sagittal View: A vertical slice dividing the brain into left and right halves
• Coronal View: A vertical slice dividing the brain into front and back sections
• Directional Terms:
• Dorsal (Superior): Toward the top
• Ventral (Inferior): Toward the bottom
• Rostral (Anterior): Toward the front
• Caudal (Posterior): Toward the back

MRI and fMRI

• MRI Physics: works by placing a subject in a strong magnetic field to align hydrogen atoms
• Scanner hardware:
• Superconducting wire: Used to generate a strong, stables magnetic field
• Liquid helium: Cools the superconducting magnets to maintain efficient functioning
• RF (Radiofrequency) coils: Transmit radio waves into the subject and receive the returning signals to detect changes in hydrogen atom alignment
• MRI Safety:
• Magnetic objects should not be near the machine to avoid serious injury
• Quench: Emergency shutdown where helium is rapidly released, eliminating the magnetic field, with risks like oxygen displacement
• fMRI Principles: Measures brain activity by detecting BOLD signals
• Oxyhemoglobin vs. Deoxyhemoglobin:
• Oxyhemoglobin is non-magnetic (diamagnetic) and does not distort the magnetic field
• Deoxyhemoglobin is weakly magnetic (paramagnetic), reducing MR signal strength
• BOLD Signal: Detects neural activity based on oxygen consumption
• Magnetic Susceptibility: Variations in magnetization caused by different tissue properties can influence signal detection

The Subtraction Method

• "Lights up" definition: Increased activation in one condition compared to another, indicating relative neural activity differences
• fMRI and multiple conditions:
• BOLD signals do not provide measure activity absolutely, only relative changes
• The brain is constantly active, so control conditions are needed to isolate specific cognitive processes
• Good Contrasts in fMRI: Isolates a variable of interest while minimizing other influences
• Example: Studying biological motion vs. random motion to examine motion-specific brain regions
• Poor contrast choices can lead to confounds and misinterpretation of results

Confounds in fMRI Research

• Confounds: Extraneous variable that influences dependent and independent variables and distorts results
• Common Confounds in fMRI:
• Eye Movements: Differences in fixation can alter neural activation patterns
• Task Difficulty: Increased cognitive load in one condition might increase activity regardless of the specific cognitive processing being tested
• Motion Artifacts: Head movement can introduce noise and reduce data reliability
• Alternative Methods to Subtraction:
• Conjunction Analysis: Identifies brain regions consistently activated across multiple tasks.
• Factorial Design: Examines interactions between different cognitive processes.
• Parametric Studies: Investigates how brain responses vary with task difficulty.
• Adaptation Studies: Measures changes in neural response with repeated stimulus exposure.

fMRI Statistics

• General Linear Model (GLM): Converts voxel time series into activation values via regressors for each experimental condition
• Steps:
• Design regressors for each condition
• Convolve with the Hemodynamic Response Function (HRF) to predict expected BOLD responses
• Solve for beta (β) values in the equation: Y = Xβ + e, where:
• Y = observed signal
• X = predicted response
• B = activation magnitude
• e = error term
• Statistical Concerns:
• Noise Sources:
• Motion artifacts distort signal
• Scanner drift leads to gradual signal changes
• Physiological noise from breathing and heartbeat
• Type 1 Error (False Positive): Incorrectly detecting activity where none exits
• Type II Error (False Negative): Failing to detect actual activity
• Multiple Comparisons Problem: Thousands of voxels analyzed → risk of false positives
• Correction Methods:
• Bonferroni Correction: Adjusts p-values to reduce Type 1 errors
• Cluster-based Correction: Groups neighboring active voxels to reduce noise effects
• Permutation Testing: Uses randomized datasets to determine statistical significance

Group Analysis, ROI, and Circularity

• Group Analysis:
• Individual differences can obscure results so group analysis helps identify generalizable patterns
• Multiple Comparisons Problem:
• Thousands of voxels analyzed → increased risk of false positives
• Correction Methods: Family-wise error correction, cluster-based correction, permutation testing
• Region of Interest (ROI) Analysis:
• Focuses on pre-defined brain area instead of analyzing the entire brain
• Steps: Functional localizer → ROI selection → Experiment
• Advantages:
• Enables targeted hypothesis testing
• Reduces the multiple comparisons problem
• Allows cross-region analysis of cognitive functions
• Disadvantages:
• Risk of Circularity: Selecting voxels based on the same dataset used for analysis which can bias results
• Potential for missing unexpected brain activations outside the ROI

Face Processing

• Are category-selective areas innate or experience-based?
• Protocortex vs. Protomap Hypothesis:
• Protocortex Hypothesis: The entire cortex starts as a homogenous structure, with functional specialization emerging based on experience
• Protomap Hypothesis: Brain regions are genetically pre-specified for certain functions, with plasticity
• Retinotopic Mapping: Early visual processing is spatially organized; face-selective areas may emerge based on retinal input distribution
• Evidence from Face Deprivation Studies:
• Monkeys raised without exposure to faces fail to develop normal face-selective regions
• Suggests that face-processing areas require visual input to mature
• Human Studies:
• Face perception abilities emerge early in infancy
• Infants prefer face like stimuli and support biological predisposition for face processing
• fMRI studies show activation in the Fusiform Face Area (FFA) even in young infants

Neuroeconomics

• Key Concepts:
• Reward Prediction Error (RPE): The difference between expected and actual crucial for learning and decision-making
• Brain Regions Involved:
• Ventral Medial Prefrontal Cortex (vmPFC): Computes subjective value and preferences
• Ventral Striatum: Processes reward anticipation and reinforcement
• Insula: Assesses cost, risk, and effort in decision-making
• Study Example:
• fMRI studies show ventral striatum activation when viewing preferred products like luxury cars
• Brand perception influences neural response to consumer goods

Love and Memories

• Neural Correlates of Love:
• Ventral Tegmental Area (VTA): Reward system involvement
• Caudate Nucleus; Correlates with love intensity
• Posterior Cingulate Cortex (PCC): Memory-related love effects
• Love & Pain Studies:
• Viewing a partner reduces pain perception
• Activation in reward areas (Nucleus Accumbens, Caudate) linked to pain relief
• Potential confounds: Familiarity effects in love related-brain activation

Explore cognitive neuroscience, an interdisciplinary field studying the neural basis of cognition. Learn about basic brain regions and their broad functions. Key areas covered include the frontal, parietal, temporal, and occipital lobes, and their roles in perception, memory, and decision-making.