Questions: Feature-Based Attention in Visual Cortex

Question 1

A person is told to search for a red dot while keeping their eyes fixed at center. How does attending to 'red' affect neural processing of red items at peripheral, spatially unattended locations?

• A. Processing at peripheral locations is unchanged — feature-based attention only affects the currently fixated location
• B. Neural responses to red items increase at all locations, including spatially unattended peripheral ones
• C. Attention narrows the receptive fields of color-tuned neurons so they respond only to centrally located red items
• D. Only locations where red items appear are enhanced, creating a spotlight that hops sequentially between targets

Answer: (B)

Feature-based attention operates globally — it enhances responses to the attended feature across the entire visual field regardless of where spatial attention is directed. This 'global gain' effect is the defining signature of feature-based attention: turning up the signal from all neurons tuned to the attended feature value everywhere at once. Options A, C, and D all assume a spatially constrained mechanism, which describes spotlight attention, not feature-based attention.

Question 2

Under feature-based attention to 'red', what changes in the neural response of a color-tuned neuron in V4?

• A. Its tuning curve shifts so it now responds more strongly to red than before, suppressing other color preferences
• B. Its overall responsiveness to red stimuli increases, but its tuning curve shape remains unchanged
• C. Its receptive field expands to capture more red stimuli across the visual field
• D. It fires spontaneously in anticipation of red stimuli even when no red is present

Answer: (B)

Feature-based attention modulates gain — the overall response magnitude — without changing the tuning curve. A neuron that preferred red still responds maximally to red, and its selectivity profile is unchanged; only its signal strength is scaled up for the attended value. Tuning curve reshaping (option A) would imply the neuron changed its preference, which is not what attention does. Gain modulation increases signal-to-noise ratio while preserving the representational code.

Question 3

Feature-based attention can enhance neural processing of stimuli at locations where spatial attention is simultaneously directed elsewhere.

• T. True
• F. False

Answer: True

This is the core empirical finding distinguishing feature-based from spatial attention. Experiments show that when spatial attention is explicitly directed to one location, the neural responses to the attended feature are elevated at *other* locations as well — ones where spatial attention is not directed. The two systems are functionally distinct and can operate independently; when both converge on the same stimulus, the effects are roughly multiplicative.

Question 4

Feature-based attention sharpens a neuron's selectivity by making it more discriminating — more selective for the attended feature value and less responsive to similar but non-identical features.

• T. True
• F. False

Answer: False

Feature-based attention does not change tuning selectivity — it modulates overall gain. The neuron's tuning curve shape (its selectivity profile) is unchanged; only the amplitude of its response is scaled up for the attended feature value. It is like turning up the volume on a radio channel already selected, not retuning the dial. Sharpened tuning would change the bandwidth of selectivity; gain modulation changes the amplitude.

Question 5

Why is the global nature of feature-based attention's gain mechanism functionally useful for visual search, and how does it differ from what a purely spatial spotlight would accomplish?

Model answer: A global gain mechanism pre-filters the entire visual field simultaneously: attending to 'red' boosts the salience of all red items everywhere at once, guiding subsequent spatial attention to likely target locations without serial scanning. A spatial spotlight would have to move location by location. This is why visual search for a distinctive feature produces near-instant 'pop-out' — the feature-based mechanism has already highlighted targets before deliberate search begins.

The functional connection to pop-out search is the payoff of understanding the global mechanism. Feature-based attention scales with feature prevalence, not spatial arrangement, making it ideal for pre-attentive filtering. The combination of feature-based (global) and spatial (local) attention is more powerful than either alone: feature-based attention directs spatial attention toward probable target locations, allowing the two systems to work in tandem.