Perception and Attention

Learning Objectives

By the end of this page, you should be able to:

• Distinguish sensation from perception
• Explain the main theories of selective attention (Broadbent, Treisman, late-selection models) and the classic experiments behind them
• Describe divided attention and the cost of multitasking
• Explain inattentional blindness and change blindness with real experimental evidence
• Apply the concept of perceptual set to explain everyday misperceptions
• Identify practical applications of attention research (driving, ADHD, design)

Quick Answer

Perception is the brain's process of organizing and interpreting raw sensory signals into a meaningful experience of the world; attention is the mechanism that selects which of those signals get prioritized for further processing. They work as a team: attention acts like a spotlight or filter, and perception is what happens to whatever that spotlight lands on. Both are limited-capacity systems — you cannot perceive everything reaching your senses at once, and you cannot attend to everything you perceive. This matters because these limits explain real phenomena: why eyewitnesses miss obvious details, why texting while driving is dangerous even hands-free, and why magicians can fool a room full of adults using nothing but misdirected attention.

Core Concepts

Sensation vs. Perception

Definition: Sensation is the raw detection of physical stimuli by sensory receptors (eyes, ears, skin); perception is the brain's interpretation and organization of that raw data into a coherent experience.

Explanation: Your eyes register light wavelengths as raw sensory data — that's sensation. Your brain then organizes those wavelengths into edges, objects, depth, and eventually recognizes "a red apple on a table" — that's perception. The same sensory input can produce different perceptions depending on context, expectation, and past experience.

Example: The retina detects two-dimensional patterns of light and dark; perception constructs a three-dimensional scene with depth, using cues like relative size and overlap.

Real-World Example: Radiologists sense the same X-ray image as anyone else, but their trained perception lets them "see" a tumor that an untrained viewer's brain organizes as unremarkable noise — the sensory input is identical, the perceptual interpretation isn't.

Why It Matters: Understanding this split explains why two people can look at the same event (say, a car accident) and sincerely report seeing different things — their eyes registered the same light, but their brains organized and interpreted it differently based on where their attention was and what they expected to see.

Common Misunderstanding: Students often use "sensation" and "perception" interchangeably. Distinguishing them is critical for understanding perceptual errors — an error isn't necessarily a sensory malfunction (like poor eyesight) but a normal, predictable interpretive process going a different way than reality.

Selective Attention

Definition: The cognitive process of focusing mental resources on one source of information while filtering out others.

Explanation: Because working memory and conscious processing are limited-capacity, the brain must select what to prioritize. Classic research asked exactly where in the processing stream this filtering happens.

Donald Broadbent's filter theory (1958) proposed that unattended information is blocked very early, based purely on physical characteristics (like which ear it came in, in dichotic listening tasks), before its meaning is even processed. But this couldn't explain the cocktail party effect — you can be deeply engaged in one conversation at a party, tuning out all the surrounding chatter, yet instantly notice when someone across the room says your name. If unattended information were fully blocked, you'd never register your name in it.

Anne Treisman's attenuation theory (1964) revised this: unattended information isn't blocked completely, just turned down in volume (attenuated), so highly meaningful stimuli (like your own name) can still break through if they exceed your personal threshold.

Later "late-selection" theorists (Deutsch & Deutsch) argued all incoming information is processed for meaning automatically, and selection only happens at the response stage, determining what you consciously act on or remember.

Example: In a classic dichotic listening experiment, participants wear headphones playing different messages in each ear and are told to "shadow" (repeat aloud) only one ear's message. They can report almost nothing about the unattended ear's content — except very salient features, like a change from a male to a female voice, or their own name.

Real-World Example: A parent asleep can sleep through traffic noise, a television, and conversation, but wakes instantly at a baby's faint cry — the meaningful stimulus breaks through the attentional filter even though it's physically much quieter than the other sounds.

Why It Matters: These competing theories explain a huge amount of practical human error: why you miss instructions given while you're focused elsewhere, and why safety warnings need to use salient, meaning-triggering cues (like your name or an alarm pattern) rather than just being louder.

Common Misunderstanding: Students often think unattended information is "not perceived at all." Treisman's research shows it's processed at a shallow level — enough to detect its physical properties and sometimes salient meaning — it just doesn't reach conscious awareness or memory under normal conditions.

Divided Attention and the Cost of Multitasking

Definition: The attempt to attend to and process two or more sources of information or tasks simultaneously.

Explanation: True simultaneous processing of two demanding tasks is very limited; what feels like multitasking is often rapid switching between tasks, which carries a measurable "switch cost" in time and accuracy. Divided attention succeeds only when at least one task is highly automatic (requiring little conscious control).

Example: You can walk and talk at once because walking is automatized, requiring minimal attention — but you cannot solve a difficult math problem while also composing a coherent argument in a debate, because both require controlled attention.

Real-World Example: Strayer and Drews (2007) used a driving simulator to show that drivers talking on cell phones — even hands-free — had reaction times and following distances similar to legally drunk drivers, because the conversation itself (not just holding the phone) consumed attentional resources needed for hazard detection. This directly contradicts the popular belief that hands-free calling is safe.

Why It Matters: This research reshaped traffic safety laws; many jurisdictions ban both handheld and hands-free phone use while driving, because the danger comes from divided attention, not just occupied hands.

Common Misunderstanding: People assume they're good at multitasking if they can technically perform two tasks "at the same time." Research consistently shows that performance on at least one task degrades, often invisibly to the person doing it — self-rated multitasking ability correlates poorly, or even negatively, with actual multitasking performance.

Inattentional Blindness and Change Blindness

Definition: Inattentional blindness is failing to notice a fully visible but unexpected object because attention is engaged elsewhere; change blindness is failing to notice a change between two scenes when attention isn't directed at the changing element.

Explanation: Both phenomena show that perception is not a passive recording of everything in the visual field — you only consciously perceive what attention selects, however "obvious" the missed information seems in hindsight.

Example / Real-World Example: In Simons and Chabris's famous 1999 "invisible gorilla" experiment, participants were asked to count basketball passes made by players in white shirts in a video. About half failed to notice a person in a full gorilla suit walk through the middle of the scene, thump their chest, and walk off — because their attention was locked onto counting passes. This has since been used to train radiologists and pilots on the danger of tunnel-visioned attention: radiologists focused on searching for nodules have been shown to miss a gorilla image inserted into a lung scan.

Why It Matters: This has serious real-world consequences: eyewitnesses can genuinely fail to see a clearly visible event because their attention was elsewhere, and this isn't lying or bad eyesight — it's a normal limit of attention. It's also why safety-critical professions (air traffic control, surgery, driving) train people to actively scan rather than passively watch.

Common Misunderstanding: People assume "if it was right in front of me, I would have seen it." Inattentional blindness demonstrates this is false — visibility and awareness are not the same thing; awareness requires attention.

Top-Down Influences on Perception: Perceptual Set

Definition: A perceptual set is a mental predisposition to perceive certain stimuli in a particular way, based on expectations, context, culture, or recent experience.

Explanation: Because perception is partly constructed by the brain rather than a direct readout of reality, prior expectations bias what we perceive — especially with ambiguous stimuli.

Example: Shown an ambiguous image that could be either a duck or a rabbit, people primed with "farm animals" tend to see the duck, while people primed with "pets" more often see the rabbit.

Real-World Example: In police lineups, if an officer inadvertently signals which person is the suspect, witnesses' perceptual set shifts and they become more likely to "recognize" that person — a key reason for double-blind lineup procedures in modern police practice.

Why It Matters: Perceptual set explains cultural and contextual differences in perception, and is central to understanding bias in eyewitness identification, medical diagnosis, and everyday misunderstandings.

Common Misunderstanding: Students think perception is objective and expectation-free unless something goes wrong. In fact, expectation-driven perception is the default, normal mode of operation, not an occasional glitch.

Visual Learning

This diagram captures the core exam idea: attention sits as a gatekeeper between raw sensation and conscious perception, and prior knowledge shapes both what passes the gate and how it's interpreted once it does.

Real-World Applications

• Traffic safety: Divided-attention research (Strayer & Drews) directly informs distracted-driving legislation, including hands-free phone bans.
• Aviation and radiology: Training programs use inattentional-blindness demonstrations (like the gorilla-in-the-lung-scan study) to teach professionals to actively scan rather than passively look.
• Eyewitness testimony: Understanding perceptual set and attentional limits has driven reforms like double-blind police lineups to reduce suggestive influence.
• User interface design: Designers place critical alerts in high-salience positions (color, motion, center of visual field) because peripheral or unexpected information is easily missed under inattentional blindness.
• Clinical diagnosis of ADHD: Attention research provides the theoretical basis for tasks (like continuous performance tests) used to assess sustained and selective attention deficits.

Key Terms

Term	Definition
Sensation	Raw detection of a physical stimulus by sensory receptors
Perception	The brain's process of organizing and interpreting sensory input into meaningful experience
Selective attention	Focusing on one source of information while filtering out others
Cocktail party effect	The ability to notice highly meaningful stimuli (like your own name) in an unattended channel
Filter theory (Broadbent)	Theory that unattended information is blocked early, based on physical features, before meaning is processed
Attenuation theory (Treisman)	Theory that unattended information is turned down, not blocked, allowing meaningful content to break through
Divided attention	Attempting to process two or more sources of information simultaneously
Inattentional blindness	Failing to notice a fully visible, unexpected object because attention is engaged elsewhere
Change blindness	Failing to notice a change between two scenes when attention isn't on the changing element
Perceptual set	A mental predisposition to perceive stimuli in a particular way based on expectation or context

Common Mistakes

Misconception 1: Hands-free phone calls are safe while driving. Why it's wrong: The danger of phone use while driving comes primarily from divided cognitive attention, not occupied hands. Strayer and Drews's simulator studies found hands-free conversation impaired driving as much as handheld use. Correct understanding: The conversation itself consumes attentional resources needed for hazard perception and reaction time, regardless of whether hands are on the wheel.

Misconception 2: If something is clearly visible, you will notice it. Why it's wrong: The invisible gorilla experiment (Simons & Chabris, 1999) showed roughly half of attentive observers miss a large, unmistakable, unexpected object because their attention was engaged in a different task. Correct understanding: Conscious awareness requires attention, not just visibility; inattentional blindness is a normal cognitive limitation, not a sign of carelessness or poor vision.

Misconception 3: Unattended information is completely blocked out and never processed. Why it's wrong: The cocktail party effect shows meaningful unattended stimuli (like hearing your name) can break through, which Broadbent's strict early-filter model could not explain. Correct understanding: Treisman's attenuation theory better fits the evidence — unattended information is weakened, not eliminated, so especially salient or meaningful content can still be noticed.

Comparison and Connections

Feature	Broadbent's Filter Theory	Treisman's Attenuation Theory	Late-Selection Theory
When is selection made?	Very early, based on physical features	Early, but selectively weakened rather than blocked	Late, after meaning is processed
Explains cocktail party effect?	No — predicts unattended info is fully blocked	Yes — meaningful content can exceed threshold	Yes — all info is processed for meaning
What happens to unattended info?	Completely filtered out	Attenuated (turned down), not eliminated	Fully processed but not acted on/remembered
Main limitation	Cannot explain hearing your name in the unattended ear	Harder to specify exact attenuation mechanism	Implies wasteful full processing of everything, even irrelevant channels

Practice Questions

Recall

1. Define selective attention and divided attention. Answer guidance: Selective attention is focusing on one information source while filtering others; divided attention is attempting to process multiple sources or tasks at once.
2. What did Simons and Chabris's 1999 gorilla experiment demonstrate? Answer guidance: That people engaged in a demanding attentional task (counting basketball passes) often fail to notice a fully visible, unexpected event (a person in a gorilla suit) — inattentional blindness.

Understanding

3. Explain why the cocktail party effect was a problem for Broadbent's filter theory, and how Treisman's theory resolved it. Answer guidance: Broadbent's theory said unattended information is blocked before meaning is analyzed, so hearing your name in an unattended conversation shouldn't be possible. Treisman proposed attenuation (turned down, not blocked) so highly meaningful content, like your own name, could exceed a personal recognition threshold and break through.
4. Why does perceptual set sometimes lead people to "see" things that aren't there, or miss things that are? Answer guidance: Because perception is partly a top-down, constructive process; expectations and context bias interpretation of ambiguous or incomplete sensory data, especially when the input itself is unclear.

Application

5. A factory manager wants workers to notice a rare safety malfunction light among dozens of normal indicator lights. What attention research should inform the panel's design? Answer guidance: Use salience (color, size, blinking) to increase the chance the malfunction light "breaks through" selective attention, similar to how meaningful stimuli break through in the cocktail party effect; avoid designs that rely on passive scanning, given inattentional blindness risk.
6. Why might an experienced radiologist still miss an obvious abnormality inserted into a scan, even though they're an expert? Answer guidance: Their attention is tuned to search for a specific pattern (e.g., nodules), and inattentional blindness can cause them to miss an unexpected, unrelated but visible object (like an inserted gorilla image), since expertise narrows the attentional search rather than broadening general vigilance.

Analysis

7. Compare early-selection (Broadbent) and late-selection (Deutsch & Deutsch) models of attention in terms of how efficient they assume the brain to be. Answer guidance: Early-selection models assume the brain conserves resources by filtering out irrelevant information before deep processing (efficient but can't explain cocktail-party effect); late-selection models assume the brain processes everything for meaning first, only filtering at the response/memory stage (explains cocktail-party effect but implies a lot of "wasted" processing of irrelevant input).
8. A student says, "I can text and listen to a lecture at the same time, so multitasking isn't really a myth." Analyze this claim using divided-attention research. Answer guidance: The student's texting and lecture-listening both require controlled, non-automatic attention, so true simultaneous processing isn't happening — they are rapidly switching, incurring switch costs, and likely missing pieces of the lecture without realizing it, similar to how drivers on phone calls underestimate their own performance decline.

FAQ

What's the real difference between sensation and perception? Sensation is the raw physical detection of a stimulus (light hitting the retina); perception is the brain's interpretation of that raw input into a meaningful experience (recognizing it as a face). The same sensation can lead to different perceptions depending on attention and expectation.

Why can't humans truly multitask on two demanding activities? Because controlled attention is a limited-capacity resource; two tasks that both require conscious control compete for the same limited pool, causing rapid switching (with a time and accuracy cost) rather than true parallel processing. Only when one task is highly automatic can two things run smoothly together.

Is inattentional blindness the same as not paying attention at all? No — the person is often paying close attention, just to a different aspect of the scene. Inattentional blindness happens precisely because attention is fully engaged elsewhere, not because the person is distracted or careless in a general sense.

Does perceptual set mean we can't trust our own perception? Not entirely — perceptual set biases interpretation especially with ambiguous or degraded stimuli, but it doesn't mean perception is generally unreliable. It does mean eyewitness accounts and quick judgments should be treated with appropriate caution, especially when expectations were primed beforehand.

How is attention research used outside of psychology labs? It shapes real policy and design: distracted-driving laws, double-blind eyewitness lineup procedures, safety-critical interface design (aviation, medicine), and training programs that teach professionals to actively scan environments instead of relying on passive vigilance.

Quick Revision

• Sensation = raw detection by receptors; perception = brain's organized interpretation of that data.
• Selective attention filters what gets deep processing; divided attention splits resources across tasks.
• Broadbent's filter theory: early filtering by physical features — can't explain cocktail party effect.
• Treisman's attenuation theory: unattended info is turned down, not blocked — explains hearing your name.
• Late-selection theory: all info processed for meaning; filtering happens at the response stage.
• Strayer & Drews (2007): hands-free phone conversations impair driving as much as handheld — attention, not hands, is the danger.
• Simons & Chabris (1999) "invisible gorilla": about half of observers doing a counting task miss an obvious unexpected object — inattentional blindness.
• Change blindness: failing to notice a change between scenes when attention isn't on the changing part.
• Perceptual set: expectations and context bias how ambiguous stimuli are interpreted (duck/rabbit illusion, biased lineup identification).
• Awareness requires attention — visibility alone does not guarantee perception.

Related Topics

Prerequisites: Introduction to Cognitive Psychology, basic sensory physiology (vision, hearing)

Related Topics: Sensation and the senses, consciousness and awareness, eyewitness testimony research

Next Topics: Memory and Learning, Problem Solving and Decision Making