Questions: Motion Perception and Middle Temporal (MT) Area

Question 1

A V1 neuron with a small receptive field detects a moving diagonal edge. Why is this neuron's output alone insufficient to determine the true direction of motion?

• A. V1 neurons are tuned for color and contrast, not motion direction, so they provide no directional information
• B. The neuron can only detect the component of motion perpendicular to the edge's orientation — the true direction of the object remains ambiguous
• C. V1 neurons have refractory periods that prevent them from responding to rapid motion
• D. V1 neurons fire equally to all motion directions and must be averaged across the cortex

Answer: (B)

This is the aperture problem. A V1 neuron with a restricted receptive field sees only a small patch of an edge. For a diagonal edge, the neuron can detect motion perpendicular to the edge (normal motion) but cannot distinguish whether the object is moving horizontally, vertically, or obliquely — any of these would produce the same signal at the local level. Area MT solves this by pooling signals from V1 neurons tuned to different edge orientations: the global motion direction consistent with all local signals is unambiguous. This pooling operation is why MT is essential for coherent motion perception.

Question 2

A patient with MT damage can accurately describe the shape, color, and texture of a cup of water being poured, but reports that the pouring liquid appears as a series of static snapshots that teleport between positions. This deficit is called...

• A. Prosopagnosia — an inability to recognize objects by their form
• B. Hemianopia — loss of the visual field on one side
• C. Akinetopsia — a selective inability to perceive motion while static form recognition remains intact
• D. Visual agnosia — a general inability to recognize objects from visual information

Answer: (C)

Akinetopsia (literally 'motion blindness') is the selective loss of motion perception following MT damage. The patient's preserved ability to describe shape, color, and texture shows that static form processing pathways are intact — the deficit is specific to the motion computation that MT performs. This double dissociation (motion impaired, form spared) is the key evidence for MT's role as a dedicated motion-processing area rather than a general visual area. Prosopagnosia is face-specific; hemianopia is a field loss; agnosia is a general recognition failure — none of these match the selective motion deficit.

Question 3

MT neurons inherit their direction selectivity directly from V1 neurons, since V1 already encodes motion direction for each local region of the visual field.

• T. True
• F. False

Answer: False

V1 neurons encode oriented edges, not motion direction. A V1 neuron tuned to a vertical edge will respond to both leftward and rightward motion of that edge (as long as the edge stays within the receptive field). MT achieves direction selectivity by integrating signals from V1 neurons with different orientation preferences — the combination of multiple locally ambiguous signals yields an unambiguous global direction. MT's direction selectivity is thus computed at the MT level, not simply passed up from V1. This integration step is precisely what solves the aperture problem.

Question 4

The motion aftereffect — seeing a stationary cliff appear to drift upward after staring at a waterfall — provides evidence for MT's role in motion perception because it reflects adaptation of direction-selective neurons in MT.

• T. True
• F. False

Answer: True

During prolonged viewing of downward motion (the waterfall), MT neurons tuned for downward motion reduce their firing through adaptation. When you shift gaze to the stationary cliff, the adapted downward-preferring neurons are suppressed relative to baseline, while upward-preferring neurons are relatively more active (they haven't adapted). This imbalance produces the percept of upward motion even though nothing is moving. The fact that the aftereffect is direction-specific and arises in a motion-selective area like MT (demonstrated by fMRI and single-unit recording) links it directly to MT's direction-tuned population.

Question 5

How does the aperture problem arise in motion perception, and what does area MT do to solve it?

Model answer: The aperture problem arises because any single neuron with a small receptive field — like a V1 neuron — sees only a local patch of a moving edge. For a moving edge, only the motion component perpendicular to the edge's orientation is detectable; the component parallel to the edge is invisible locally. This means a single neuron's response is consistent with many different true directions of motion — the direction is locally ambiguous. Area MT solves this by pooling responses from many V1 neurons with different orientation preferences across a larger region of space. Because each orientation preference is ambiguous about different directions, the combination of all their signals has a unique globally consistent interpretation: the one true direction of motion that best explains all the local signals simultaneously. This integration over orientation and space converts local ambiguity into global certainty.

The aperture problem is one of the clearest examples of why hierarchical cortical processing is necessary: no single local detector can solve a problem that requires integrating information across multiple locations and feature dimensions. MT's pooling operation is the neural implementation of a global motion computation that V1 cannot perform alone.