Deep Dive: Cognition and Perception

Welcome to Deep Dive: Cognition & Perception — Exploring how the brain predicts and perceives the world. Each post blends scientific explanation with interactive illusions to reveal how our mind shapes what we see, hear, and remember. (Friston 2005; Rao & Ballard 1999; Fenn & Hambrick 2011)

[25deepdivecognitionclassproject]

Deep Dive: Cognition & Perception — Project Overview

Deep Dive: Cognition & Perception — Project Overview

This Tumblr serves as my Deep Dive term project for PSYCH 351. The central theme is predictive coding — the idea that the brain is a prediction machine that constantly anticipates what it will see, hear, and experience, and then updates its beliefs when those predictions are wrong.

The posts on this blog are organized into several connected sections:

Predictive Coding — Core Concepts Introduction to the predictive brain framework and why perception is more than just bottom-up input.

Memory & Expectation — Predicting What Comes Next How prior experience and stored knowledge guide perception and interpretation.

Visual Illusions — When the Brain’s Predictions Beat Reality Classic illusions (checker-shadow, hollow mask, Ponzo) that reveal the gap between the retinal image and what we perceive.

Auditory Illusions — Hearing What We Expect Examples like the McGurk Effect and phoneme restoration that show prediction in hearing.

Prediction Error & the Bayesian Brain How mismatches between expectation and input drive learning and model updating.

Attention, Expectation & Perceptual Errors Inattentional blindness, change blindness, and how selective attention shapes what we notice.

Motor Prediction — The Brain’s Internal Simulator How the brain predicts the consequences of our movements and why actions feel smooth and automatic.

Reflection — Seeing the Mind Seeing Final reflections on what this project taught me about perception, prediction, and my own thinking.

Together, these posts aim to show that perception is not a simple recording of reality, but an active process where the brain constantly predicts, tests, and revises its model of the world.

[25deepdivecognitionclassproject]

Week 12: Functional Divisions of Mental Simulation

🎯 Learning Goal

Explain the functional divisions of “David’s Integrated Neural System for Mental Simulation.”

📚 Core Concepts

Professor Havas’ Integrated Neural System for Mental Simulation breaks down how the brain simulates actions and outcomes using distinct but interconnected systems:

• Perception–Action Cycle:

Every perception leads to an action, and every action leads to a new perception. This cycle allows the brain to continuously update its internal model of the world.

Motor Cortex Hierarchy:

SMA (Supplementary Motor Area): Forms intentions and abstract action plans.

PM (Premotor Cortex): Specifies and selects action sequences.

M1 (Primary Motor Cortex): Executes fine motor movements.

“Actions are planned with respect to a goal… and unfold through a hierarchy of motor areas.” — Week 12 Motor Cognition Lecture

• Mirror Neuron System:

Observing others’ actions activates the same motor areas in the observer’s brain, enabling simulation, empathy, and social understanding.

Internal vs. External Information Streams:

External: Affordances — objects automatically suggest possible actions (e.g., a chair affords sitting).

Internal: Goals, memories, and emotional states influence whether or how we act.

🔬 Evidence from Research

• Lecture Insight:

The brain is a prediction machine. It learns the structure of the world from the bottom up and uses top-down projections to simulate what will happen next.

“The brain’s predictions are corrected by the world… and the brain learns from its mistakes.” — Week 12 Motor Cognition Lecture

• Saygin et al. (2004):

Even simplified point-light animations of human motion activate the premotor cortex, showing that the brain simulates actions even when visual input is minimal.

“These stimuli can also recruit action observation networks… characterized by motion cues alone.” — Saygin et al., 2004

• TED Talk (Biological Motion):

Point-light walkers reveal how we extract rich social and emotional information from motion alone — gender, emotion, even personality traits.

“There’s lots of information we can retrieve from the way someone moves.” — Biomotion Lab

🧩 Integration & Synthesis

This system supports predictive coding: the brain constantly simulates what will happen next based on past experience. The SMA, PM, and M1 form a functional hierarchy that allows us to mentally rehearse actions before we perform them.

The mirror neuron system allows us to simulate others’ actions, which supports empathy, language comprehension, and social cognition. This is why we flinch when we see someone else get hurt — our motor system simulates their experience.

Example: Seeing a cup of coffee activates motor plans for grasping — even if we don’t act. Example: A mug with a handle affords grasping — its shape and orientation suggest how to interact with it. But whether we actually reach for it depends on internal factors like thirst, memory of its contents, or social context (e.g., “Is this my mug?”). Example: Watching someone kick a ball activates our own motor cortex as if we were kicking.

🎨 Visuals

“The motor system is organized from intention (SMA) to execution (M1).”

“Perception and action form a continuous loop of simulation and feedback.”

“Even minimal motion cues activate motor simulation areas.”

[25deepdivecognitionclassproject]

Week 13: Aphasia and the Breakdown of Language

🎯 Learning Goal

List the defining characteristics and causes of three forms of aphasia.

📚 Core Concepts

Aphasia is a language disorder caused by damage to specific areas of the brain, often due to stroke or traumatic brain injury. The three major types are:

Broca’s Aphasia (Nonfluent/Expressive Aphasia):

Defining Features: Halting, effortful speech; short phrases; relatively preserved comprehension.

Cause: Damage to Broca’s area in the left frontal lobe, typically from a stroke.

Example: A person might say “walk dog” instead of “I will walk the dog.”

“Broca’s aphasia is the most common nonfluent type… speech is limited, but comprehension is often intact.” Verywell Health

Wernicke’s Aphasia (Fluent/Receptive Aphasia):

Defining Features: Fluent but nonsensical speech; poor comprehension; frequent neologisms.

Cause: Damage to Wernicke’s area in the left temporal lobe.

Example: A person might say “The dog smoothed the table with happiness.”

“They do not realize that their language is impaired, and they don’t understand others—or even themselves—when they speak.” Verywell Health

Global Aphasia:

Defining Features: Severe impairment in both speech production and comprehension.

Cause: Extensive damage to multiple language areas in the left hemisphere, including both Broca’s and Wernicke’s areas.

Example: A person may only be able to produce a few sounds or words and struggle to understand spoken language.

“Global aphasia… affects all aspects of language, making both speaking and understanding very difficult.” Stroke Assoc...

🔬 Evidence from Research

• Lecture Insight:

Aphasia reveals how language is not a single function but a network of specialized systems. Damage to different areas results in distinct deficits, showing how modular and localized language processing is in the brain.

• Neuroscience Support:

fMRI and lesion studies confirm that Broca’s area is critical for speech production, while Wernicke’s area is essential for comprehension. Global aphasia typically results from large strokes affecting the middle cerebral artery, which supplies both regions.

🧩 Integration & Synthesis

These aphasia types illustrate the functional specialization of the brain’s language system. They also connect to earlier course themes:

• Motor Cognition: Broca’s aphasia disrupts the motor planning of speech, linking it to the premotor and motor cortex functions discussed in Week 11.

• Mental Simulation: Wernicke’s aphasia impairs the ability to simulate and interpret meaning, showing how prediction and comprehension are intertwined.

• Predictive Coding: All forms of aphasia disrupt the brain’s ability to generate and test linguistic predictions — a core idea from Weeks 2 and 12.

🎨 Visuals

“Broca’s area (frontal lobe) and Wernicke’s area (temporal lobe) are key to speech production and comprehension.”

“Large strokes affecting the middle cerebral artery can lead to global aphasia by damaging multiple language areas.”

[25deepdivecognitionclassproject]

Week 11: Motor Cognition and Mental Simulation

🎯 Learning Goal

Describe three distinct cognitive roles for different motor areas of the brain.

📚 Core Concepts

Motor cognition isn’t just about movement — it’s the foundation of how we think, plan, and simulate actions. The three key motor areas and their cognitive roles are:

Supplementary Motor Area (SMA):

Responsible for intention formation — setting up abstract action plans.

Example: Deciding to go to the store or study for an exam.

Premotor Cortex (PM):

Handles specification and selection of action sequences.

Example: Choosing how to reach for a cup — with your left or right hand, from which angle, etc.

Primary Motor Cortex (M1):

Executes fine motor movements by directly controlling muscles.

Executes fine motor movements by sending signals to muscles.

These divisions show how motor cognition is not “primitive” but central to planning, simulating, and executing complex actions.

🔬 Evidence from Research

“The brain really seems to be some kind of a prediction-making machine.”

— Motor Cognition Lecture

• Lecture Insight:

The perception–action cycle shows that every perception leads to an action, and every action leads to a new perception. This loop is central to how we simulate actions before doing them.

• Saygin et al. (2004):

Using fMRI, researchers found that even point-light biological motion (just dots moving like a person) activates the premotor cortex, showing that our motor system helps us interpret and simulate others’ actions — even when visual information is minimal.

“Saygin et al. (2004): Using fMRI, researchers found that even point-light biological motion (just dots moving like a person) activates the premotor cortex, showing that our motor system helps us interpret and simulate others’ actions — even when visual information is minimal.

“Point-light biological motion activates human premotor cortex.” — Saygin et al., 2004

“These stimuli can also recruit action observation networks, although they are very simplified and characterize actions by motion cues alone.” — Saygin et al., 2004

• Daniel Wolpert’s TED Talk: Wolpert argues that “we have a brain for one reason and one reason only — to produce adaptable and complex movement.” This supports the idea that cognition evolved primarily to serve motor control.

"Brains evolved not to think or feel, but to move." - Daniel Wolpert, TED Talk (2011)

🧩 Integration & Synthesis

Motor cognition supports predictive coding — the brain constantly generates hypotheses about what will happen next based on past experiences. SMA, PM, and M1 form a hierarchy that allows simulation at multiple levels: planning, sequencing, and execution.

This also connects to mirror neurons: when we observe someone else performing an action, our own motor areas activate as if we were doing it ourselves. This underlies empathy, social learning, and language comprehension.

Example: Seeing a coffee cup activates motor plans for grasping — even before we touch it.

Example: Watching someone reach for a cup activates our own motor cortex, helping us understand their intention.

This supports the predictive coding model introduced in Week 2 — the brain uses prior experience to simulate and anticipate action outcomes.

“The brain forms internal models based on past experiences and uses them to predict incoming sensory input. These predictions guide both perception and action.” — Biology Insights, 2025 biologyinsig...

🎨 Visuals

The motor cortex is organized hierarchically, from abstract planning (SMA) to execution (M1).

Every perception leads to an action, and every action leads to a new perception — forming a continuous simulation loop.

Even simplified motion cues like point-light walkers activate premotor cortex — showing how the brain simulates action

[25deepdivecognitionclassproject]

Motor Prediction — The Brain’s Internal Simulator

Motor Prediction — The Brain’s Internal Simulator

The brain predicts the consequences of movement before the movement happens. This allows actions to feel smooth, controlled, and automatic.

The cerebellum builds “forward models” that simulate the expected outcome of a movement. When reality matches prediction, movement feels effortless. When prediction fails, movements feel clumsy or inaccurate.

⚡ Examples

Why tickling yourself doesn’t work (the brain predicts the sensation).

Why athletes anticipate ball direction before it moves.

How motor learning uses prediction error to refine skills.

[25deepdivecognitionclassproject]

Attention, Expectation & Perceptual Errors

Attention, Expectation & Perceptual Errors

Attention determines which predictions the brain prioritizes. When attention is focused elsewhere, prediction takes over and perception can fail in surprising ways.

🙈 1. Inattentional Blindness

We miss obvious events when attention is absorbed elsewhere (like the “gorilla experiment”).

🔄 2. Change Blindness

Large visual changes go unnoticed between flickers or camera cuts.

🎯 Why This Happens

Attention filters what the brain predicts will be important.

Unexpected details are ignored or overwritten.

The brain prioritizes prediction over accuracy.

[25deepdivecognitionclassproject]

Prediction Error — How the Brain Learns

Prediction Error — How the Brain Learns

Prediction error is the difference between what the brain expects and what actually happens. Predictive coding theory proposes that the brain constantly updates its model of the world by minimizing these errors (Friston 2005).

When predictions match reality → processing becomes efficient. When predictions fail → the brain updates its internal model.

🔢 Bayesian Brain Model

The brain combines:

Priors (what it expects)

Likelihood (the current sensory input)

Posterior (updated understanding)

Perception = Prior × Evidence → Updated Belief

🌫 Why Illusions Cause Large Prediction Errors

Illusions exploit situations where sensory input is ambiguous or contradicts expectation. The brain “chooses” the most likely explanation, even when incorrect.

[25deepdivecognitionclassproject]

Reflection — Seeing the Mind Seeing

Working through the Deep Dive project changed the way I think about perception. I used to imagine the brain as a processor that responded to the world moment-by-moment. Now I understand perception as a prediction cycle, shaped by experience, memory, and expectation. The brain constantly tests its assumptions, corrects its errors, and updates its understanding of reality.

This project helped me appreciate how much of what we “see” is actually generated internally. Our experiences influence our expectations, and those expectations guide attention, interpretation, and behavior. Predictive processing explains why illusions trick us, why we sometimes miss important details, and why we react automatically in familiar situations.

Overall, this Deep Dive made cognitive psychology more personal: it showed me how my own mind interprets the world, and how awareness of these predictive mechanisms can help reduce bias, improve learning, and make perception more intentional.

[25deepdivecognitionclassproject]

Real-World Applications — Cognitive Prediction in Daily Life

Predictive processing influences everyday behavior in subtle but powerful ways. For example, expectations shape what we hear in conversations, making speech comprehension faster but also prone to misunderstanding. In visual perception, the brain fills in details we never consciously notice, allowing us to move efficiently through complex environments.

Prediction also plays a role in emotional and social situations. The brain anticipates others’ reactions, making social interactions smoother — but also leading to anxiety when predictions are negative or uncertain. In learning, prediction errors help refine skills as the brain continually adjusts its internal model based on feedback (Fenn & Hambrick 2011).

Even habits operate through predictive loops: the brain expects certain outcomes and triggers behaviors automatically. By understanding how these predictions work, we can become more aware of how perception, memory, and action interact to shape real-world decisions.

[25deepdivecognitionclassproject]

Memory & Expectation — Predicting What Comes Next

Memory plays a central role in predictive processing. The brain uses stored experiences to anticipate patterns, fill in missing information, and guide perception. Instead of reacting to the present moment from scratch, we interpret the world by referencing prior knowledge — what we have seen, heard, and learned before (Rao & Ballard 1999).

This helps explain why two people can view the same event and perceive it differently: different memories create different expectations. It also clarifies why familiar environments feel easier to navigate — the brain can predict what will happen with greater accuracy.

Prediction is most visible in ambiguous situations. When sensory information is weak or noisy, the brain leans heavily on memory to “guess” what’s happening. These guesses feel automatic, but they shape everything from reading text quickly to recognizing faces to responding to potential threats. Understanding these memory-based predictions helps explain errors such as false memories, biases, and perceptual distortions.

[25deepdivecognitionclassproject]

Auditory Illusions — When Hearing Deceives the Brain

Auditory Illusions — Hearing What We Expect

Auditory Illusions — Hearing What We Expect

Just like vision, hearing is shaped heavily by prediction. Auditory illusions show that the brain fills in missing sound, corrects noise, and interprets speech using expectation.

🗣 1. McGurk Effect

When the sound “ba” is paired with a video of lips saying “fa,” people hear “fa.” Vision rewrites hearing.

Why it matters: the brain predicts speech using multi-sensory integration, not sound alone.

🔊 2. Phoneme Restoration Effect

A missing sound in a word is “filled in” by the brain even though the sound never occurred.

Example: “It was found that the *cough* eel was on the axle.” You hear “wheel,” but the “w” never existed.

🎵 3. Shepard Tone (the “infinite” pitch)

A tone seems to rise forever, but never actually gets higher.

This illusion shows how the brain predicts musical structure and continuity.

[25deepdivecognitionclassproject]

Title: Visual Illusions — When Seeing Is Misleading

Visual Illusions — When the Brain’s Predictions Beat Reality

Visual illusions reveal how strongly the brain relies on prediction rather than raw sensory input. Instead of processing the world pixel-by-pixel, the brain uses expectations, context, lighting, and memory to interpret ambiguous information. Illusions expose the difference between what is on the retina and what the brain thinks should be there.

🟩 1. The Checker-Shadow Illusion

A and B look like different shades of gray, but they are identical. The brain assumes the cylinder is casting a shadow and “corrects” for lighting automatically.

Image placeholder: Insert the Checker-Shadow Illusion image here.

Why it matters: This illusion demonstrates how the brain predicts lighting conditions and adjusts perception before we become aware of the image.

🎭 2. The Hollow Mask Illusion

A concave (inward) mask of a face appears convex (outward). The brain has a strong prior: faces are normally convex. It overrides the actual sensory signal.

🚆 3. The Ponzo Illusion

Two identical lines appear different lengths because the brain interprets background “rails” as depth cues.

Image placeholder: Insert Ponzo illusion graphic.

✨ What Illusions Reveal

The brain does not show us the world as it is.

It shows us the world as it predicts it should be.

Context + expectation override raw vision.

[25deepdivecognitionclassproject]

Predictive Coding — How the Brain Builds Reality

Predictive coding proposes that the brain is not a passive receiver of sensory input but a prediction machine. Instead of waiting for the world to tell it what is happening, the brain continuously generates expectations about incoming information and compares those predictions to the actual sensory signal (Friston 2005). When predictions match reality, processing becomes efficient. When there is a mismatch, the brain updates its model of the world to reduce future errors.

This theory reframes perception as top-down inference rather than bottom-up detection. We do not simply “see what is there”; we see what the brain expects to be there, corrected by small amounts of sensory evidence. This helps explain why illusions work, why we overlook unexpected objects, and why attention guides what we notice.

Predictive coding is foundational to the rest of my Deep Dive. It helps explain how memory guides perception, how biases form, and how we adapt to uncertainty. In future posts, I connect these ideas to everyday situations — including misperceptions, learning, and decision-making.