Attention and Perception

Attention and Perception: Your EPPP Guide

Your senses pick up far more than you could ever use. Attention decides what gets through. Perception turns it into a world that makes sense. The EPPP tests this area through definitions, research classics, and "which theory explains this?" questions.

Why This Matters for Psychologists

A child with ADHD who drifts off during reading. A stroke patient who ignores the left side of space. A driver on a hands-free call who misses a red light. The models help you tell a sensing problem from an attention problem from a decision problem.

Sensation: Turning Energy Into Signals

Sensation happens when a sensory receptor detects a stimulus. Perception is how the brain organizes and interprets that input into conscious experience.

Transduction

Receptors convert stimulus energy (light, sound waves, molecules) into action potentials. That conversion is transduction. It's like a translator at a border crossing. Light and sound arrive speaking different languages, and receptors translate both into the one language the nervous system reads: nerve impulses.

Thresholds

Absolute threshold: the smallest amount of stimulus energy you can detect 50% of the time. That 50% detail is a favorite exam target.

Difference threshold, also called the just noticeable difference (JND): the smallest change between two stimuli that you can detect. It grows as the starting stimulus gets stronger. Light one candle in a dark room and everyone notices. Add that same candle to a table already holding fifty, and nobody does. Weber's law states the rule: the JND is a fixed proportion of the starting stimulus.

Sensory Adaptation

Sensory adaptation means you stop perceiving a stimulus that stays the same for a long time. You notice the hum of the office air conditioner when you first sit down. An hour later it is still humming, but you no longer hear it. Adaptation shows sensation and perception coming apart: the input keeps arriving while your awareness of it fades.

Subliminal Perception

A subliminal stimulus reaches the senses but stays too weak to reach awareness. Lab studies show people can still register and react to such input. In daily life, though, hidden messages in ads or music barely change what people do.

The Psychophysics Laws in One Breath

The Biopsychology lesson on sensation and perception covers these in full. Fechner used Weber's law to argue that felt intensity grows with the logarithm of physical intensity. Stevens later showed that direct ratings of intensity follow a power law (Johnson et al., 2002).

Signal Detection Theory: Can You Tell, and Will You Say?

Signal detection theory (SDT) says that detecting a faint stimulus depends on two separate things: how well you can tell the signal from background noise, and how willing you are to say "yes, it's there." It grew out of 1950s radar research on picking real blips out of noise (Anderson, 2015).

In SDT math, d′ does not depend on where you set β, so the theory pulls ability apart from willingness. Payoffs and base rates move the criterion. Rare targets or costly false alarms push people toward a conservative criterion. Common targets or costly misses push them toward a liberal one (Lynn & Barrett, 2014).

A smoke detector shows both measures. A cheap sensor that can barely tell smoke from shower steam has a low d′. Flip it to its most "alarm-happy" setting and you have lowered β: it now catches more real fires and also shrieks every time someone showers.

Exam-style example: believers in the paranormal report more patterns in random noise than skeptics do. Studies show this reflects a looser criterion, not sharper perception (Anderson, 2015).

Perception: Building Meaning From Input

Bottom-Up and Top-Down Processing

Bottom-up processing is driven by the stimulus itself. Top-down processing is driven by your knowledge, goals, and expectations. Top-down attention, such as choosing where to look, is goal directed and voluntary.

A fire alarm going off mid-session grabs your attention whether you like it or not. That's bottom-up. Scanning a packed waiting room for a client you know wears a red coat is top-down: your goal tells your eyes where to look.

Perceptual Set

Perception runs on perceptual hypotheses, educated guesses shaped by your personality, experiences, expectations, and schemas (taught in the Thinking, Problem Solving, and Language lesson). Together they create a perceptual set, a readiness to see things a certain way. Verbal priming, for example, biases how people interpret ambiguous figures. A new parent in the shower keeps "hearing" the baby cry through the running water. Culture matters too. People raised among straight-edged, "carpentered" buildings were more prone to the Müller-Lyer illusion than people from settings built around round huts (Segall et al., 1963).

Organizing the Scene: Gestalt Principles

Gestalt psychology (Wertheimer, Köhler, Koffka) holds that the whole is different from the sum of its parts.

Perceptual Constancy: A Stable World From a Shifting Image

The image on your retina changes every time you move, yet objects seem stable. That stability is perceptual constancy.

Constancy is strong but not perfect. For size, the brain appears to combine the retinal image with distance cues.

Depth Perception: Binocular vs. Monocular Cues

Closeness to the horizon is another monocular cue. Monocular cues let you see depth in paintings and judge depth with one eye, and you have more of them than binocular cues.

Hold a pen at arm's length and wink one eye, then the other. The pen jumps sideways. That jump is binocular disparity, and your brain uses it to judge depth.

Color Vision: Two Theories, Two Stages

Both theories are correct. They describe different stages. Negative afterimages are the classic evidence for opponent-process theory, and single-cell recordings in animals later found neurons excited by some wavelengths and inhibited by others.

Stare at a bright green square for 30 seconds, then look at a white wall. A reddish square appears. Green and red are partners in one channel, which is exactly what opponent-process theory predicts.

Pitch Perception: Place, Frequency, and Volley

Pitch is your perception of a sound's frequency.

One nerve cell cannot fire on every cycle of a very high-pitched sound, so frequency theory alone cannot cover the 20 to 20,000 Hz range. Up to about 4,000 Hz, firing rate and place both contribute. Above that, only place works.

Think of a team clapping a fast beat. No one person can clap that fast, but if they take turns, the group keeps the tempo.

Selective Attention: The Bouncer at the Door

The Cocktail Party Effect

The cocktail party effect (Cherry, 1953): you follow one voice in a noisy room, yet highly relevant input, like your own name, can still grab you from a conversation you were ignoring. About one-third of listeners notice their name in a message they were told to ignore (Moray, 1959; Wood & Cowan, 1995). Twist: people with lower working memory capacity are more likely to catch it, likely because they block out distraction less well (Conway et al., 2001).

Dichotic Listening and Shadowing

In dichotic listening, headphones play a different message to each ear. In shadowing, you repeat one ear's message aloud while ignoring the other. People later remember almost nothing of the ignored ear's words. These studies supplied most of the evidence behind the early attention theories.

Early, Attenuated, or Late Selection?

Broadbent's filter is a bouncer who checks only outfits at the door. Treisman's attenuator is a volume knob turned way down, so a shout of your name still gets through. Deutsch and Deutsch let everyone into the club and pick who gets the microphone later.

The own-name finding and studies where the shadowed message switched ears were used against a strict early filter. The debate is not closed: with attention tightly controlled, some studies find no sign that truly unattended input gets identified (Lachter et al., 2004). For the exam, know the classic mapping.

Feature Integration Theory: Pop-Out vs. Conjunction Search

Anne Treisman's feature integration theory (Treisman & Gelade, 1980) has two stages:

1. Preattentive stage: basic features such as color and orientation are registered in parallel across the whole visual field.
2. Focused attention stage: attention "binds" features into whole objects, one or a few items at a time.

Key support: illusory conjunctions. When attention is spread thin, people sometimes pair the color of one letter with the shape of another, seeing a blue X when a red X and a blue O were shown. That is what loose, unbound features would produce (Treisman & Paterson, 1984; Cohen & Ivry, 1989).

The theory also predicts two kinds of search:

• Feature search (pop-out): a target with one unique feature pops out. Search time barely changes as you add distractors.
• Conjunction search: a target defined by a combination of features (a red vertical line among red horizontal and green vertical lines) needs attention item by item, so reaction time rises with the number of items.

Finding the one red sock in a drawer of white socks is pop-out. Finding the one red ankle sock among red knee socks and white ankle socks is conjunction search: you check them one by one.

Later work found conjunction searches are often faster than the theory predicted, because feature information can guide attention toward likely targets (Wolfe's Guided Search).

Divided Attention: How Much Can You Juggle?

Kahneman's Capacity Model

Filter theories picture a structural bottleneck. Daniel Kahneman (1973) proposed a capacity model instead: there is a general limit on how much mental work you can do at once, and you can divide that capacity across tasks fairly flexibly. He used pupil dilation as an index of mental effort.

Filter models picture a narrow doorway. Kahneman pictures a fixed monthly budget: you can split it many ways, but when one bill grows, something else gets less.

Dual-Task Costs and "Multitasking"

People often struggle to do even two simple tasks at the same moment. One key bottleneck sits at choosing a response (Pashler, 1994). So "multitasking" often means waiting at that bottleneck or switching back and forth. Each switch has a switch cost: you are slower and more error-prone right after changing tasks. Preparation shrinks the cost but does not erase it (Monsell, 2003).

In a simulated driving task, cell phone conversations, handheld or hands-free, doubled missed traffic signals and slowed reactions. Listening to the radio or an audiobook did not (Strayer & Johnston, 2001). The problem is the attention a conversation takes, not the hands.

Automatic vs. Controlled Processing

In Shiffrin and Schneider's terms, automatic processing is fast, low-effort, and hard to stop, while controlled processing is slow, effortful, and voluntary. The full comparison lives in the Thinking, Problem Solving, and Language lesson (System 1 vs. System 2). The attention point: practice moves a task from controlled to automatic, which frees capacity. Your first week of driving took every ounce of focus. Years later you pull into the driveway with no memory of the turns.

The Stroop Effect

In the Stroop task, you name the ink color of a color word. Naming is slower when word and ink conflict (RED printed in blue) than when they match. Reading is the more automatic process, so you must hold it back to name the ink. The Stroop Color and Word Test measures this ability to inhibit interference, and patients with frontal lesions make more errors on conflict items, such as reading the word instead of naming the ink.

It's like trying not to read a billboard as you drive past. Your eyes read it before you can stop them.

Looking Without Seeing

Whether people notice the unexpected object depends on how similar it is to the other objects on screen and on how hard the main task is. Attention is a flashlight in a dark room. Anything outside the beam can be right in front of you and still go unseen.

Visual-Spatial Processing

Two Visual Streams

From the occipital lobe, vision splits into two parallel pathways.

When you reach for your coffee, the ventral stream knows it's your mug. The dorsal stream shapes your hand to grab the handle.

Mental Rotation

Shepard and Metzler (1971) showed people two drawings of 3D objects and asked whether they were the same shape. The time needed rose linearly with the angle between the two orientations, and rotations in depth took no longer than flat, picture-plane rotations. It's like turning a puzzle piece in your hand: the farther you have to turn it, the longer it takes.

Mental Imagery

Kosslyn and colleagues (1978) found that scanning a longer distance across a mental image takes more time, and larger mental images take longer to scan. Images seem to preserve real spatial layout. Picture your neighborhood. Traveling from your front door to the corner store in your head takes longer than traveling to your mailbox. Imagery shares brain machinery with seeing, plays a role in many mental disorders, and can be used in treatment (Pearson et al., 2015).

Attention in the Clinic and the Cockpit

ADHD: a persistent pattern of inattention and/or hyperactivity-impulsivity. Signs include trouble with tasks that need sustained attention, easy distraction, and forgetfulness.

Hemispatial (unilateral) neglect: failure to attend to one side of space, usually the side opposite the lesion. Left neglect after right-hemisphere damage is more frequent, more severe, or both, compared with the reverse. Typical lesion sites include the posterior parietal lobe and the temporo-parietal junction.

Human factors psychology (ergonomics) designs machines, displays, and workspaces to fit human limits, including vigilance, spotting signals in noise, and divided attention. In one bank security center, operators who made more decisions per day produced more false alarms but no more misses. For a bank, that is the cheaper error, since a missed breach costs far more.

EPPP Traps: Common Mistakes

Trap 1: "Absolute threshold is the faintest stimulus you can ever detect."

• Reality: It is the level detected 50% of the time.

Trap 2: "A clinician with more hits is more sensitive."

• Reality: Not if false alarms rose too. More hits plus more false alarms can simply mean a looser criterion (β), not a higher d′.

Trap 3: "Treisman's attenuation model is late selection."

• Reality: Attenuation is the middle ground. Late selection belongs to Deutsch and Deutsch.

Trap 4: "Feature integration theory says all visual search is slow and serial."

• Reality: Single features pop out in parallel. Only conjunctions need item-by-item attention.

Trap 5: "Hands-free phones make driving safe."

• Reality: The conversation itself drains attention. Hands-free and handheld calls both increased missed signals.

Trap 6: "Inattentional blindness and change blindness are the same thing."

• Reality: Inattentional blindness misses an unexpected object. Change blindness misses a difference across views.

Trap 7: "Mental rotation time is constant."

• Reality: It rises linearly with the angle of rotation.

Trap 8: "Negative afterimages support trichromatic theory."

• Reality: They are the classic evidence for opponent-process theory.

Memory Aids for the EPPP

Filter theories: Broadbent Blocks. Treisman Turns it down. Deutsch and Deutsch Decide late.

SDT: d′ = discrimination (can you tell?). β = bias (will you say?).

Visual streams: Dorsal = Doing (where/how). Ventral = Vocabulary (what it is called).

Search: one feature = flash (pop-out). Two features = trudge (conjunction).

Key Takeaways

• Absolute threshold = detected 50% of the time; the JND is a constant fraction of the starting stimulus (Weber's law).
• SDT separates sensitivity (d′) from criterion (β); payoffs and base rates shift β.
• Bottom-up is stimulus driven; top-down is knowledge driven; perceptual set shapes what you see.
• Gestalt grouping and constancy turn a shifting retinal image into stable objects.
• Depth: binocular (disparity, convergence) vs. monocular (linear perspective, interposition, relative size, texture gradient, shading, motion parallax).
• Color: trichromatic at the retina, opponent-process beyond it; afterimages support opponent-process. Pitch: place for high, timing for low, volley to extend timing.
• Selective attention: Broadbent (early), Treisman (attenuation), Deutsch and Deutsch (late).
• Feature integration: features pop out in parallel; conjunctions need focused attention; illusory conjunctions show unbound features.
• Kahneman: attention is a limited, flexible capacity; dual tasks and switches carry costs.
• Stroop effect: automatic reading collides with controlled color naming.
• Inattentional blindness misses the unexpected; change blindness misses changes.
• Ventral = what, dorsal = where/how; mental rotation time rises linearly with angle.
• Clinical links: ADHD (sustained attention), left neglect after right-hemisphere damage, human factors design.

Now cover the tables and sketch the SDT grid and the three filter theories from memory. Retrieval beats rereading.