Visual Search + Display Controls

Lectures 5F-6B

Visual search

Process of scanning a visual field to find a specific target among distracting items.

Target (signal)

The object or event you are trying to detect in a visual search or SDT task.

Distractor (visual noise)

Objects in the visual field that are not the target and make detection harder.

Serial search model

Model where items are inspected one at a time until the target is found or the field is exhausted.

Average inspection time (I)

The average time required to inspect one item during a serial visual search.

Search time formula (T = N×I/2)

Equation estimating average time to find a target, assuming on average half the items must be inspected.

Law of diminishing returns (search)

In random, unstructured search, each extra unit of time yields a smaller increase in detection probability.

Conspicuity

How much a target stands out from its background and distractors, drawing attention.

Bottom-up processing

Attention driven by properties of the stimulus itself (e.g., bright, moving, odd-colored items).

Top-down processing

Attention guided by expectations, knowledge, and goals about where and what the target is.

Feature search / Disjunctive search

Efficient search where the target differs from distractors on a single, salient feature and can be found in parallel.

Efficient search

Visual search where detection time does not grow much with the number of items, often due to strong conspicuity.

Discriminability from background

How visually different a target is from other items in color, size, shape, brightness, or motion.

Simplicity of target definition

Describing targets using a single feature rather than multiple complex combinations to make them easier to find.

Automaticity

When a familiar or highly meaningful target (e.g., your name) pops out with little effort.

Signal Detection Theory (SDT)

Framework that separates sensitivity from decision bias when detecting signals under uncertainty.

Noise

Background sensory activity or irrelevant stimuli present whether or not the signal is there.

Signal + Noise

Condition where the target signal is present on top of the existing noise.

Hit

Saying 'yes, signal present' when the signal is actually present.

Miss

Saying 'no, signal absent' when the signal is actually present.

False alarm

Saying 'yes, signal present' when the signal is actually absent (noise only).

Correct rejection

Saying 'no, signal absent' when the signal is truly absent.

Sensitivity (discriminability)

How well an observer can distinguish signal+noise from noise alone; higher means better detection.

d′ (d-prime)

Quantitative measure of sensitivity; distance between signal+noise and noise distributions in standard deviation units.

Response criterion

Internal decision threshold for saying 'yes, signal' versus 'no, noise' in an SDT task.

Beta (β)

Numerical measure of response criterion; ratio of signal+noise to noise probability densities at the criterion.

Risky criterion (low β)

Decision strategy that favors saying 'signal present', leading to more hits and more false alarms.

Conservative criterion (high β)

Decision strategy that favors saying 'signal absent', leading to fewer false alarms and more misses.

Optimal β (probability-based)

Value of β that minimizes errors when based only on relative probabilities of signal and noise.

Payoffs in SDT

Values or costs assigned to hits, misses, false alarms, and correct rejections that influence optimal decision strategy.

Optimal β (payoff-based)

Criterion that maximizes expected value by combining event probabilities with payoffs and costs.

Sluggish beta

Tendency for people to adjust their criterion less than optimally when event probabilities or payoffs change.

Receiver Operating Characteristic (ROC) curve

Plot of hit rate vs. false-alarm rate showing trade-offs as the decision criterion shifts.

Iso-sensitivity curve

ROC curve for a fixed value of d′; moving along it changes criterion without changing sensitivity.

Useful Field of View (UFOV)

Region of the visual field within which a target can be detected quickly if present.

Eye-mind hypothesis

Assumption that where the eyes are looking is closely related to what the mind is processing.

Eye tracking

Technique for measuring where and for how long a person looks on a visual display.

Pursuit eye movement

Smooth eye movement used to track a moving object at roughly constant velocity.

Saccadic eye movement

Rapid, discrete eye movement that shifts gaze from one fixation point to another.

Fixation

Brief period when the eyes remain relatively still and visual information is taken in.

Saccade

The quick jump of the eyes between fixations during visual scanning.

Scanpath

Sequence of fixations and saccades that shows how the eyes move over a display.

Initiation latency

Time between stimulus change and the start of a saccadic eye movement.

Movement time (eye movement)

Time taken by the eyes to execute a saccade from one point to another.

Destination time / fixation duration

Time spent fixating on the target after a saccade, within the useful field of view.

Display

Any device or interface element that presents information about system state to the user.

Static display

Display that presents fixed information that does not change over time (e.g., road sign).

Dynamic display

Display whose information changes over time (e.g., gauges, dashboards, HUDs).

Quantitative reading

Use of a display to extract precise numerical information.

Qualitative reading

Use of a display to judge approximate level or trend (e.g., low/medium/high).

Checking reading

Use of a display to confirm whether a value is within normal or abnormal limits.

Situation awareness display

Display used to inform upcoming events.

Heads-up display (HUD)

Transparent display that overlays key information in the user's forward field of view.

Auditory display

Use of non-speech or speech sounds to convey information to the user.

Vibrotactile display

Use of vibration on the body to convey information through the sense of touch.

Salience compatibility (Principle 1: Attention)

Important and urgent information should stand out and capture attention.

Minimize information access cost (Principle 2: Attention: Display)

Frequently needed information should be easy and quick to access with minimal visual movement.

Proximity compatibility principle (Principle 3: Attention)

Information that must be mentally integrated should be placed close together in the display.

Avoid resource competition (Principle 4: Attention)

Distribute information across different sensory and cognitive channels to reduce interference.

Make displays legible/audible (Principle 5: Perceptual)

Ensure text, symbols, and sounds are clear and easy to perceive under expected conditions.

Avoid absolute judgment limits (Principle 6: Perceptual)

Do not rely on users distinguishing many levels of a single sensory dimension (e.g., many similar colors).

Support top-down processing (Principle 7: Perceptual: Display)

Design displays that conform to users' expectations and prior knowledge when possible.

Exploit redundancy gain (Principle 8: Perceptual: Display)

Use multiple cues (e.g., color AND shape or position) to convey important information.

Make discriminable (similarity causes confusion) (Principle 9: Perceptual)

Reduce unnecessary similarity and highlight distinctive features so items are not confused.

Knowledge in the world (Principle 10: Memory)

Place needed information and cues in the environment so users do not rely solely on memory.

Support visual momentum (Principle 11: Memory)

Organize displays to help users maintain orientation and follow changes smoothly across views.

Provide predictive aiding (Principle 12: Memory)

Give users information about future states to support proactive decisions.

Principle of Consistency (Principle 13: Memory)

Use consistent formats, codes, and locations so skills transfer and errors are reduced.

Principle of pictorial realism (Principle 14: Mental Model)

Display should look like the variable it represents (e.g., high on display means high in the system).

Principle of moving parts (Principle 15: Mental Model)

Movement in a display should be congruent with the system’s physical movement or user's mental model.

Orders of control

Classification of control systems based on how control input affects system state (e.g., position, rate).

Control–display gain

Ratio of the movement of the display or cursor to the movement of the control device.

Hick-Hyman Law (Principle 12:Response Selection: Control)

Model showing that the time it takes to make a decision increases logarithmically as the number of choices increases (RT = a + b·log2 N).

Fitts's Law (Principle 13:Response Selection: Control)

Model that predicts movement time to a target based on distance and target width (MT = a + b·log2(2A/W)).

Movement time (MT)

Time required to move a control or pointer from start position to target.

Amplitude of movement (A)

Distance that the control or pointer must move to reach the target.

Target width (W)

Size of the target region that must be acquired in a pointing or control task.

Index of difficulty (ID)

Quantifies how hard it is to reach a target; log2(2A/W) in Fitts's Law.

Throughput

Combined measure of speed and accuracy in pointing tasks, typically ID divided by MT (bits/second).

Make controls accessible (Principle 3: Perceptual: Control)

Controls should be within easy reach and convenient for the operator to use.

Make controls discriminable (Principle 4: Perceptual: Control)

Controls should look and feel distinct so they are easily identified and not confused.

Exploit redundancy gain in controls (Principle 5: Perceptual: Control)

Use multiple codes (shape, size, color, labeling) to help users identify controls.

Avoid absolute judgment limits (Principle 6: Perceptual: Control)

Do not depend on users distinguishing too many levels of one control code (e.g., many similar colors).

Location compatibility (Principle 9: Mental Model: Control)

Controls should be located near the displays or system elements they affect.

Movement compatibility (Principle 10: Mental Model: Control)

Direction of control movement should match the corresponding system or display movement.

Avoid accidental activation (Principle 11:Response Selection: Control)

Design controls and forcing functions to prevent unintended activation that could cause errors.

Decision complexity advantage  (Principle 14:Response Selection: Control)

For some tasks, one complex decision can be faster than several simple sequential choices.

Provide feedback (Principle 15:Response Selection: Control)

Controls should give immediate, perceivable feedback that an action has been carried out.

avoid resource competition (Principle 2: Attention: Control)

When human are performing more than one

task at the same time, multiple resource theory predicts a benefit of

dividing the tasks across different mental resources.

knowledge in the world  (Principle 7: Memory: Control)

The actuation of the control should be reflected in the

control itself

be consistent   (Principle 8: Memory: Control)

Consistency makes it possible for people to apply skills from one

situation to another, reducing errors and response time (standardization)