Donald Broadbent's early filter (bottleneck) theory of attention proposed that:
A: All incoming stimuli are fully processed for meaning before selection occurs
B: A selective filter operates early, on physical features, allowing only the attended channel through to full processing
C: Attention is an unlimited resource that can be distributed across any number of tasks simultaneously
D: Selection is based primarily on the emotional significance of stimuli
Correct: A selective filter operates early, on physical features, allowing only the attended channel through to full processing
Broadbent (1958) proposed that the human information-processing system has limited capacity, and a selective filter operates before full semantic analysis. This filter uses physical properties (e.g., voice pitch, location) to select one channel for further processing, while others are held briefly in a short-term store. Evidence came from Cherry's (1953) dichotic listening studies, where listeners following one ear's content could extract little from the unattended ear. Later work by Treisman (1964) showed the filter does not block out all unattended content but rather attenuates it — the "cocktail party effect," where your own name in the unattended channel still breaks through.
The cocktail party effect describes the phenomenon where:
A: Background noise disrupts focused concentration in proportion to the number of speakers present
B: Despite attending to one conversation, people can detect personally significant stimuli (e.g. their own name) in an unattended channel
C: People cannot follow more than two conversations simultaneously, regardless of training
D: Alcohol impairs the ability to filter out irrelevant background speech
Correct: Despite attending to one conversation, people can detect personally significant stimuli (e.g. their own name) in an unattended channel
Cherry (1953) demonstrated that in a cocktail party environment, people can focus on one conversation but still detect their own name or other personally significant content from a simultaneously occurring unattended conversation. This challenged Broadbent's strict early-filter model, because it implies that at least some unattended content is analysed for meaning. Treisman's (1964) attenuation model resolved this by proposing that the unattended channel is attenuated — processed at reduced gain — rather than completely blocked, allowing high-salience stimuli to exceed the attentional threshold.
Treisman's Feature Integration Theory (FIT) proposes that visual search differs depending on what kind of target is sought. According to FIT, which of the following would be found in parallel across a display (pop-out search)?
A: A red X among red Os and blue Xs — a conjunction of colour and shape
B: A red circle among green circles — a single-feature difference
C: A tilted line among vertical lines of the same colour
D: Both B and C, because single-feature targets always pop out regardless of display size
Correct: Both B and C, because single-feature targets always pop out regardless of display size
Treisman and Gelade (1980) proposed that individual features (colour, orientation, size) are processed in parallel across the visual field by separate feature maps in a pre-attentive stage — producing "pop-out" when a target differs from distractors on a single feature. Finding a red item among green items or a tilted line among vertical lines requires no attention; reaction time is flat as display size increases. By contrast, conjunction targets (defined by a combination of features, like a red X among red Os and blue Xs) require serial, attention-demanding binding — reaction times increase with display size. The binding of features into coherent objects is accomplished by the spotlight of focal attention.
The attentional blink (AB) demonstrates that after detecting a target in a rapid visual stream, there is a brief period (~200–500 ms) during which a second target is frequently missed.
Answer: True
Raymond, Shapiro, and Arnell (1992) showed that when observers monitor a rapid serial visual presentation (RSVP) stream for two targets (T1 and T2), T2 is frequently missed if it appears within ~200–500 ms of T1 — the "attentional blink." This is thought to reflect a temporary depletion of attentional resources engaged in consolidating T1 into working memory, leaving T2 unprotected. The AB does not occur when T2 immediately follows T1 (the "lag-1 sparing" effect), suggesting that T2 can sometimes be processed while the T1 consolidation window is still open.
Inattentional blindness (Simons & Chabris, 1999) refers to:
A: Failure to perceive a stimulus because of peripheral vision limitations
B: Failure to perceive a clearly visible, unexpected object or event because attention is occupied elsewhere
C: The gradual loss of sensitivity to a stimulus under prolonged focused attention
D: Difficulty switching attention between tasks in patients with frontal lobe damage
Correct: Failure to perceive a clearly visible, unexpected object or event because attention is occupied elsewhere
Simons and Chabris (1999) demonstrated inattentional blindness dramatically: observers counting basketball passes failed to notice a person in a gorilla suit walking through the scene. The stimulus was not outside the field of view; it was simply unexpected and task-irrelevant, and attention was fully committed to the counting task. The phenomenon shows that phenomenal consciousness of a stimulus does not guarantee noticing it — what is attended is noticed; much of what is visually available goes unnoticed. Closely related is change blindness: the failure to notice large changes in a scene during brief interruptions.
The Stroop effect — slower response when naming the ink colour of colour words printed in an incongruent colour — is best explained by:
A: Poor working memory capacity causing interference between two simultaneous tasks
B: Automatic processing of the word's meaning competing with the controlled process of naming the ink colour
C: Hemispheric differences in colour and language processing
D: A general slowing of response times caused by the difficulty of reading coloured text
Correct: Automatic processing of the word's meaning competing with the controlled process of naming the ink colour
Stroop (1935) found that naming the ink colour of a word like "RED" printed in blue takes longer than naming the colour of a neutral stimulus. Reading is more automatised than colour naming (for literate adults), so word meaning is processed rapidly and without intention — this automatic process produces a conflicting response (saying "red") that must be inhibited while executing the controlled response of naming the ink colour ("blue"). The Stroop task is widely used in neuropsychology as a measure of executive attentional control and inhibition, and is sensitive to frontal lobe damage and ADHD.
Posner's spatial cueing paradigm (1980) distinguished between exogenous and endogenous attention. Which pairing is correct?
A: Exogenous = voluntary, goal-directed; Endogenous = reflexive, stimulus-driven
B: Exogenous = reflexive, stimulus-driven (e.g. a sudden flash); Endogenous = voluntary, goal-directed (e.g. a directional arrow at fixation)
C: Exogenous attention is controlled by the frontal lobes; endogenous by the cerebellum
D: Exogenous and endogenous attention are now understood to rely on identical neural mechanisms
Correct: Exogenous = reflexive, stimulus-driven (e.g. a sudden flash); Endogenous = voluntary, goal-directed (e.g. a directional arrow at fixation)
Posner (1980) distinguished two modes of spatial attention. Exogenous (bottom-up/reflexive) orienting is driven by external stimuli — a sudden onset, a flash, or movement at a peripheral location captures attention automatically and rapidly. Endogenous (top-down/voluntary) orienting is driven by internal goals — a central arrow or instruction directs attention to a predicted location. Exogenous cues produce faster RT benefits at shorter SOAs; endogenous cues show benefits at longer SOAs, reflecting the time required for voluntary allocation. Both benefit detection at the cued location and impair detection elsewhere (inhibition of return for exogenous cues at long SOAs).
Corbetta and Shulman's (2002) two-network model of attention identifies a dorsal and a ventral fronto-parietal network. Which is the correct description?
A: The dorsal network is right-lateralised and responds to unexpected salient stimuli; the ventral network is bilateral and drives voluntary, top-down allocation
B: The dorsal network (bilateral; IPS/FEF) supports voluntary top-down attention allocation; the ventral network (right-lateralised; TPJ/IFG) acts as a circuit-breaker detecting unexpected, behaviourally relevant stimuli
C: The dorsal network processes spatial attention; the ventral network processes feature-based attention, and they are anatomically separate
D: Both networks are strongly left-lateralised, consistent with the left hemisphere's role in language and executive control
Correct: The dorsal network (bilateral; IPS/FEF) supports voluntary top-down attention allocation; the ventral network (right-lateralised; TPJ/IFG) acts as a circuit-breaker detecting unexpected, behaviourally relevant stimuli
Corbetta and Shulman (2002) proposed two anatomically distinct fronto-parietal networks. The dorsal network — bilateral, involving the intraparietal sulcus (IPS) and frontal eye fields (FEF) — implements voluntary, goal-directed (endogenous) attention allocation. The ventral network — predominantly right-lateralised, involving the temporoparietal junction (TPJ) and inferior frontal gyrus (IFG) — acts as a "circuit-breaker" or alerting system, detecting unexpected, behaviourally relevant stimuli and reorienting attention to them. This model explains hemispatial neglect: right hemisphere damage disrupting the ventral network produces profound left-sided inattention, while left hemisphere damage causes milder neglect because the right hemisphere's ventral network compensates.
Attention
Donald Broadbent's early filter (bottleneck) theory of attention proposed that:
About this quiz
Attention is the cognitive mechanism that selects, prioritises, and sustains focus on particular information while filtering out competing input. Far from being a single faculty, attention is a family of distinct but interacting processes — and understanding its limits and architecture is central to cognitive neuroscience.
This quiz covers the major theoretical models of attention (filter theory, spotlight model, feature integration theory), landmark empirical phenomena (the cocktail party effect, attentional blink, inattentional blindness, the Stroop effect), and the neural systems that support them — including the fronto-parietal attention networks and the distinction between bottom-up (salience-driven) and top-down (goal-directed) control.