Lecture 24: Attention and Awareness

What this lecture covers

This lecture examines why humans cannot process everything reaching their senses at once and how attention selects what gets through. It moves from everyday demonstrations of limited capacity to formal distinctions between kinds of attention, the brain regions that generate and are modulated by attentional signals, and finally to experiments that isolate perceptual awareness from the raw stimulus itself. It is one of the final lectures in the course.

After watching, you should be able to distinguish covert from overt attention and stimulus-driven from voluntary attention, describe how attention modulates activity in sensory cortex even before a stimulus appears, identify the frontoparietal attention network and how it differs from domain-specific regions like the fusiform face area, and explain how binocular rivalry and the attentional blink are used to study awareness independent of the stimulus on the retina.

Key ideas

• Limited processing capacity: people cannot fully process everything hitting their retina or cochlea at once, illustrated by tasks such as reporting colored letters in a briefly flashed array, where accuracy drops sharply once more than one target item is present.
• Change blindness: a phenomenon where substantial changes to a scene, or even a person, go unnoticed between viewings, showing that a felt sense of rich perception does not match what is actually retained.
• Covert versus overt attention: covert attention shifts processing to a location without moving the eyes, while overt attention involves eye movements toward the fovea, the high-resolution center of gaze that is disproportionately represented in early visual cortex.
• Stimulus-driven (exogenous) versus voluntary (endogenous) attention: attention can be automatically pulled by a salient stimulus, such as a pop-out color or a web pop-up, or deliberately directed by the observer, as tested with spatial cueing experiments.
• Spatial versus feature-based attention: attention can be directed at a location in space or at a feature such as a color or shape, independent of where that feature is.
• Frontoparietal attention network: a set of parietal and frontal regions active during attention shifts and difficult tasks generally, contrasted with domain-specific regions such as the fusiform face area; damage to this network bilaterally can produce Balint syndrome, in which a person can only perceive one object at a time.
• Binocular rivalry: a technique using glasses that send different images to each eye, producing an alternating percept that lets researchers study brain responses to a change in awareness while the retinal input stays constant.
• Attentional blink: a brief window after detecting one rapidly presented target during which a second target can go completely unreported, used to test how far an unseen stimulus is processed in the brain.

Walkthrough

Limited capacity and the toaster model of cognition (0:10)

The lecture opens with the everyday example of driving while talking on the phone to motivate the idea that mental processing draws on some shared, limited resource, described informally as the "toaster model" where dimming the lights in one place reflects less capacity elsewhere. A demonstration with a flashed array of colored letters shows that people reliably report a single target letter but miss others when several compete for attention, while irrelevant, unattended letters have no effect on performance.

Change blindness and unnoticed information (12:17)

Building on the idea of limited capacity, the lecture shows a street scene that flashes repeatedly with small changes, such as an awning color or a car model, which most viewers fail to notice despite feeling they perceived the scene richly. A related video demonstration and a study of commercial pilots landing in a flight simulator, some of whom failed to see another plane sitting on the runway, are used to show that attention acts as a filter that can let major, even safety-relevant, information go unnoticed.

Covert and overt attention (20:53)

The lecture introduces Helmholtz's observation that attention can shift independently of eye position, distinguishing covert attention, which redirects processing without eye movement, from overt attention, which uses eye movements to place high-acuity foveal vision on a target. It explains that the fovea is served by a disproportionately large share of both photoreceptors and visual cortex, making eye movements a highly effective way to sample the world, and shows eye-tracking footage of a person's gaze shifting rapidly between speakers' faces in conversation.

Stimulus-driven versus voluntary attention, and spatial versus feature-based attention (27:58)

The lecture distinguishes stimulus-driven (exogenous) attention, automatically captured by salient properties like a pop-out color, from voluntary (endogenous) attention, directed deliberately as in a spatial cueing task where subjects fixate a cross and respond faster to targets at a validly cued location. It then introduces feature-based attention as a separate dimension, searching for an object by a known property such as color rather than a known location.

Neural sources and targets of attentional modulation (33:02)

Using early fMRI studies, the lecture shows that simply cueing attention to a location increases baseline activity in early visual areas such as V1 and V2 before any stimulus appears, and that attending to faces versus houses versus a crosshair modulates responses in the fusiform face area even when the same stimuli are physically present throughout. This modulation by attention extends broadly across sensory cortex, including primary auditory cortex and even the lateral geniculate nucleus. The source of these attentional signals is identified as the frontoparietal attention network, active during shifts of attention and difficult tasks generally, in contrast to domain-specific regions like face or scene areas; bilateral damage to this network can cause Balint syndrome, in which patients can perceive only one object at a time.

Studying awareness: binocular rivalry and the attentional blink (42:12)

The lecture uses a binocular rivalry demonstration, viewed with colored glasses showing a face to one eye and a house to the other, to show that perception alternates between the two images even though the retinal input never changes. An fMRI experiment tracking these switches shows the fusiform face area and parahippocampal place area rising and falling in step with what the subject reports seeing, not with the unchanging stimulus. The lecture then turns to the attentional blink, in which a second rapid target is often missed if it follows too soon after a first target that must be reported, and shows an fMRI study in which the parahippocampal place area still responded differently to a missed scene than to a correctly rejected absence of a scene, indicating some processing of the stimulus even without conscious awareness.

Before you watch

• Review the earlier lectures on the fusiform face area and parahippocampal place area, since their responses to faces and houses are used throughout this lecture's attention experiments.
• Familiarity with basic fMRI logic (comparing conditions with identical stimuli but different tasks) from earlier lectures is helpful for the attention-modulation experiments described here.

Check your understanding

1. What do the flashed-letter and change blindness demonstrations suggest about the difference between what hits the retina and what is actually perceived?
2. How does covert attention differ from overt attention, and why is the fovea so central to overt attention's power?
3. Describe the spatial cueing experiment used to demonstrate voluntary attention, and what the reaction time results showed.
4. Why is the frontoparietal attention network described as domain general, in contrast to regions like the fusiform face area?
5. How did the binocular rivalry experiment separate the effects of the stimulus from the effects of awareness, and what did the fMRI results in the FFA and PPA show?