Chapter 1 (In depth)

Sensation

Detecting physical stimulation. It begins with sensory receptors and provides information about the environment.

Perception

Interpreting sensory information to create a meaningful representation that helps us understand and respond to the world. “Sensation gives us information; perception gives that information meaning.” We can’t access the physical world directly, only our own sensory experiences.

Transduction

Converting physical energy into neural signals.

Path from stimulus to experience

Stimulus → sensory receptors → transduction → neural processing → sensation → perception → action.

Vestibular system

Senses balance, movement, acceleration, and physical orientation. Its sensory organs are located in the inner ear.

Interoception

Sensing information from inside the body. Whether it counts as a separate “sense” is “not so simple.”

Number of senses

The five-senses model is too simple (we also sense balance, internal states, temperature, itch, and tickle). One organ can detect multiple types of information, so there is no single, useful number of senses.

Ambiguous sensory information

The same sensory information can have multiple possible sources, so perception requires interpretation. Physical stimulus ≠ perceptual experience.

Six ways to study perception

Thresholds (how much is needed to detect), scaling (perception across intensities), signal detection (difficult decisions), time course (when perception occurs), sensory neuroscience (what the nervous system does), and computational models (how the brain might process information).

Gustav Fechner

German scientist and philosopher (medicine, physics, math) who asked whether subjective experience can be measured scientifically. Founder of psychophysics.

Psychophysics

The quantitative study of the relationship between physical events and mental experiences.

Absolute threshold

The minimum intensity of a stimulus that can be detected.

50% detection

The absolute threshold is defined as the intensity detected on ~50% of trials, because there is no sharp boundary between “detected” and “not detected.”

Method of constant stimuli

Several intensities, each presented many times in random order; the participant reports detected or not. 50% detection = threshold.

Method of limits

Intensity changes in order. Ascending (faint → intense: “when do you first detect it?”) and descending (intense → faint: “when do you stop?”). Threshold = the average of the crossover points.

Adaptive methods

Use smaller steps near the threshold and adjust the stimulus based on previous responses. They improve precision and reduce unnecessary trials.

Method of adjustment

The participant controls the stimulus intensity, adjusting it until it is just detectable.

Difference threshold (JND)

The smallest change in a stimulus that can be detected (“Can I detect a change?”). Also called the just noticeable difference.

Ernst Weber

Did the weight-lifting experiment and found that the JND depends on the starting intensity.

Weber’s law

The JND is a constant proportion (the Weber fraction, K) of the original stimulus (e.g., +1 g noticed at 40 g, +10 g noticed at 400 g). We notice relative changes, not absolute ones.

Fechner’s law

As physical intensity increases, sensation increases more slowly. The relationship is logarithmic, and Fechner built it on Weber’s law.

Fechner’s law

As physical intensity increases, sensation increases more slowly. The relationship is logarithmic, and Fechner built it on Weber’s law.

Scaling

Measures perceived intensity, differences between people, and changes across a stimulus range.

Physical vs. perceptual measures

Physical → perceptual pairs: sound pressure → loudness; frequency → pitch; light intensity → brightness; chemical concentration → sweetness; temperature → warmth. Perception transforms physical information.

Magnitude estimation

Observers assign numbers to their experiences, using either their own scale or a standard reference value.

Stevens’s power law

Perceived sensation depends on intensity raised to an exponent. Exponent < 1: sensation grows more slowly. = 1: grows proportionally. > 1: grows more quickly.

Cross-modality matching

Matching the perceived intensity of one sense to another (e.g., matching a taste’s strength to a sound’s loudness).

PROP tasters

People differ dramatically in tasting the bitter chemical PROP: nontasters (almost tasteless), medium tasters (moderately bitter), and supertasters (extremely bitter). Same chemical, different experience.

Signal detection theory

Treats detection as a decision under uncertainty: is there a signal within the noise?

Signal

The stimulus we’re trying to detect (e.g., your phone ringing).

Noise

Everything that makes detection uncertain (e.g., the sound of the shower).

Internal noise

Activity generated by the nervous system itself. Neurons fire even without stimulation, and perception varies from moment to moment.

External noise

Variation in the environment: background sounds, visual clutter, similar-looking objects.

Criterion

The decision boundary: above it you say “signal,” below it you say “no signal.” It reflects how much evidence you require.

Hit

A signal is present and you say yes.

Miss

A signal is present and you say no.

False alarm

No signal is present but you say yes.

Correct rejection

No signal is present and you correctly say no.

Sensitivity (d′)

How well you can distinguish signal from noise, based on how much the “noise” and “signal + noise” distributions overlap. Low d′ = heavy overlap and poor discrimination; high d′ = little overlap and excellent discrimination.

Liberal criterion

“Better safe than sorry.” More hits, more false alarms, fewer misses.

Conservative criterion

“Don’t call it unless you’re sure.” Fewer false alarms, more misses, fewer hits.

ROC curve

Plots hits (y-axis) vs. false alarms (x-axis) across criteria. The diagonal = chance; the upper-left corner = perfect; higher d′ = a curve closer to the upper-left.

Time course of perception

Asks when we detect something. Measured with accuracy, reaction time (~200–250 ms for sound, light, and vibration; >400 ms for vestibular), and stimulus duration. Stimulus duration ≠ processing time.

Bloch’s law

For very brief flashes, visibility depends on luminance × duration, so a very bright, very brief flash can still be detected.

Masking

A second stimulus (the mask) follows the target to interrupt processing. It is used to measure processing time.

SOA (stimulus onset asynchrony)

The time between the onset of the target and the onset of the mask. Longer SOA gives more time to process the target; shorter SOA gives less.

Doctrine of specific nerve energies

Johannes Müller’s idea that the brain interprets signals according to which nerve pathway is activated, not what activated it. Example: pressing on your closed eye makes you “see” light (phosphenes).

Capsaicin / menthol

Capsaicin activates warmth fibers (you feel heat without a temperature rise); menthol activates cold fibers (you feel cooling). Perception reflects neural activity, not always the stimulus.

Cranial nerves

12 pairs connecting the brain with the body. They can carry sensory information, motor information, or both.

Sensory cranial nerves

I olfactory (smell), II optic (vision), VIII vestibulocochlear (hearing + balance).

Motor cranial nerves

III oculomotor, IV trochlear, VI abducens (they control movement).

Mixed cranial nerves

Carry both sensory and motor information: V trigeminal, VII facial, IX glossopharyngeal, X vagus.

Multisensory integration

Combining information from multiple senses.

Neuron parts

Dendrites receive; the cell body maintains the neuron; the axon carries the action potential; axon terminals communicate with other cells.

Action potential

A neuron “firing.” A stimulus causes depolarization: Na⁺ (sodium) rushes in, then K⁺ (potassium) flows out to restore the resting state. The signal travels down the axon like a stadium wave.

Synapse

The gap/connection point between neurons. Neurons do not physically touch.

Neurotransmitter

A chemical released from the axon terminal that crosses the synapse and binds to receptors, making the next neuron more or less likely to fire.

Recording neuronal activity

Electrodes placed near neurons measure action potentials and electrical/chemical activity to find out what information neurons encode.

Rate coding

How often a neuron fires.

Spike timing

When a neuron fires. Phase locking = firing at the same point in a tone’s cycle.

Population coding

How many neurons respond together. Perception is usually a product of neural populations, not a single neuron.

Tuning curve

Shows a neuron’s response across stimuli. It peaks at the stimulus the neuron responds to best (e.g., a tone near 2000 Hz), where its threshold is lowest.

Neuroimaging

Studies activity across thousands or millions of neurons in the living brain. Key question: where and when is the brain active?

EEG

Scalp electrodes measure electrical activity. Excellent temporal (“when”), limited spatial resolution.

ERP

Repeated EEG responses to the same stimulus, averaged together.

MEG

Detects tiny magnetic fields from neural activity. Excellent temporal resolution and better spatial resolution than EEG (“when + where”), but very expensive and less portable.

MRI

Detailed images of brain structure/anatomy (“what does it look like?”).

fMRI

Measures changes in blood oxygenation to show which areas become more active. Better spatial (“where”), slow temporal resolution.

Computational models

Model how sensory information becomes perception, using probability/statistics, math equations, or neural networks. They make quantitative predictions and can be tested against human behavior.

Artificial neural networks

Inspired by biological neural networks: many interconnected units whose connections change with experience, so they can learn patterns and categories.

Deep learning

Networks with many layers and millions of connections, used for object recognition, speech recognition, and medical images. Performing like humans ≠ working like humans.

Prior knowledge in perception

Sensory input + past experience + context + expectations → perception. The brain uses predictions, prior knowledge, heuristics, and context, so different people can experience the same stimulus differently.