Cognitive neuroscience Study Guide

📖 Core Concepts Cognitive neuroscience: study of how brain biology (neurons, circuits, lobes) produces mental functions. Neurons: discrete cells that transmit signals; the basic unit linking brain activity to cognition. Localizationist view: specific mental abilities map to distinct brain regions (e.g., Broca’s area = speech production). Aggregate field view: all brain regions contribute to every mental function; emphasizes network dynamics. Bottom‑up attention: stimulus‑driven capture of salient items; relies mainly on the ventral visual stream. Top‑down attention: goal‑directed focus on task‑relevant objects; relies mainly on the dorsal visual stream. Marr’s three levels: (1) computational (what problem is solved), (2) algorithmic‑representational (how it’s solved), (3) physical (neural implementation). Functional neuroimaging: fMRI (BOLD signal), PET (metabolic activity), EEG/MEG (electrical/magnetic fields) – tools to link brain activity with cognition. --- 📌 Must Remember Broca’s area – posterior inferior frontal gyrus → speech production. Wernicke’s area – left posterior superior temporal gyrus → language comprehension. Brodmann areas – 52 cytoarchitectonic cortical regions defined by cellular composition. Alpha rhythm – 8–12 Hz EEG wave first identified by Hans Berger; indicates relaxed wakefulness. Bottom‑up → ventral stream (object identity). Top‑down → dorsal stream (spatial location, goal‑directed selection). fMRI measures blood‑oxygen‑level‑dependent (BOLD) contrast; temporal resolution ≈ seconds, spatial ≈ mm. PET tracks radiotracer metabolism; useful for neurotransmitter studies. EEG/MEG provide millisecond temporal resolution; limited spatial precision. --- 🔄 Key Processes Designing a cognitive‑neuroscience experiment Define the cognitive process → select a behavioral task (psychophysics, memory test, etc.). Choose measurement modality (fMRI for spatial maps, EEG/MEG for timing, TMS for causal interference). Collect baseline data → run task → compare activation patterns or performance between conditions. Lesion inference Identify patients with focal brain damage. Test cognitive abilities → map deficits to damaged region → infer function of that region. Attention network activation Present salient stimulus → ventral stream activation (bottom‑up). Provide goal cue → dorsal stream activation (top‑down). Marr’s analysis workflow Computational: state the problem (e.g., “identify object”). Algorithmic: propose representation (e.g., edge detection) and steps. Physical: map steps onto neural circuitry (e.g., V1 → V2 → IT). --- 🔍 Key Comparisons Bottom‑up vs. Top‑down attention Stimulus‑driven vs. goal‑driven. Ventral visual stream vs. Dorsal visual stream. fMRI vs. PET fMRI → BOLD signal, non‑radioactive, higher spatial resolution. PET → radiotracer metabolism, can image neurotransmitter systems, lower spatial resolution. EEG vs. MEG EEG → measures electric potentials; affected by skull conductivity. MEG → measures magnetic fields; less distortion, but requires magnetically shielded rooms. Localizationist vs. Aggregate field Specific region ↔ specific function vs. distributed network for all functions. --- ⚠️ Common Misunderstandings “fMRI shows neurons firing” – fMRI reflects hemodynamic changes, not direct neuronal spikes. “Lesion studies prove causality” – they suggest necessity but may involve compensatory network changes. “Bottom‑up attention ignores the brain” – it still depends on cortical processing (ventral stream). “All cognition is localized” – many higher‑order tasks emerge from large‑scale networks (aggregate view). --- 🧠 Mental Models / Intuition “Neural city map”: think of each cortical lobe as a district; specific streets (Brodmann areas) host specialized services (speech, vision). “Spotlight vs. floodlight”: top‑down attention is a spotlight (focused, intentional), bottom‑up is a floodlight (brightens anything salient). “Three‑layer cake” (Marr): computational question is the top layer (what), algorithmic layer is the middle (how), physical layer is the base (where in the brain). --- 🚩 Exceptions & Edge Cases Dual‑function areas: e.g., dorsolateral prefrontal cortex participates in both working memory and decision‑making. Plasticity after lesions: nearby or contralateral regions can take over lost functions, blurring strict localization. Cross‑modal attention: auditory salience can still trigger ventral visual stream activation under certain tasks. --- 📍 When to Use Which Choose fMRI when you need high spatial detail (e.g., mapping language areas pre‑surgery). Choose EEG/MEG for millisecond timing (e.g., event‑related potentials in attention). Choose PET to study neurotransmitter systems (e.g., dopamine in reward). Use TMS to test causality: temporarily disrupt a region and observe behavioral change. Apply lesion studies for natural “knock‑out” evidence, especially in clinical neuropsychology. --- 👀 Patterns to Recognize Activation of left inferior frontal gyrus + temporal regions → language production/comprehension tasks. Increased BOLD in posterior parietal cortex → spatial attention or top‑down control. Alpha‑band suppression in EEG → heightened visual processing or attentional engagement. Consistent ventral stream activation when subjects report object identity, regardless of task. --- 🗂️ Exam Traps Distractor: “fMRI measures electrical activity.” – Remember it measures blood flow (BOLD), not direct electricity. Distractor: “Bottom‑up attention is always faster than top‑down.” – Speed depends on stimulus salience vs. task demands; not a strict rule. Distractor: “Lesion of Broca’s area eliminates all speech.” – Patients often retain comprehension and can use non‑verbal communication; lesion effects are selective. Distractor: “EEG has better spatial resolution than fMRI.” – EEG’s spatial resolution is poorer; it excels in temporal resolution. ---