Stereoscopic vision Flashcards
• Barlow et al (1967):
Recording V1 neurons in anaesthetised cat and mapped their RF of left and right eye. Found that stimuli located correctly in the visual fields of both eyes are more effective in driving units than a monocular stimulus, and much more than is only correctly positioned in one eye. Different units will be optimally excited by objects lying at different distances. This suggest that neurons require different disparities
Prince et al (2002)
Recording responses of single V1 neurons in response to a dynamic random dot stereogram and found no evidence of clustering indicating that cannot be divided into discrete tuning types and can conclude that a continuum of tuning types is present. Aso showed that both phase and position based mechanisms can encode disparity
• Ohzawa et al (1990):
The essence energy model Recording neurons of cat V1 with grating stimuli and identify them as simple, complex and whether they are disparity selective and thenroposed a model as a disparity detector that involves a hierarchical arrangement of cortical cells Found that in simple cells both the position and contrast polarity effects the response profile whereas for complex cells their response is independent of the contrast polarity (they reject mismatch contrast polarities) and so complex cells appear especially suited for detection of binocular disparities and therefore propose that behaviour of complex cells can be achieved by combing inputs from small number of simple cells
• Cummings and Parker (1997):
Recording V1 neurons of monkeys to see if they correspond directly to the conscious perception of binocular stereoscopic depth by with RDS and acRDS. Found that neurons in V1 still respond to acRDS and so cannot unambiguously signal stereoscopic depth , even though the depth tuning was shifted
• Cummings and Parker (1999):
recorded disparity selective neurons in the V1 of awake monkeys and found that these neurons maintained consistent firing for respect to absolute disparity, none of them showed a consistent relationship with relative disparity.
• Thomas et al (2002):
recording neurons in the V2 of macaque to random dot patterns (centre and annular) and altered the relative and absolute disparity. Found that some neurons were selective for relative disparities and some show partial
• Tanabe et al (2004)
Recorded responses from macaque V4 neurons showed that most responses showed attenuated responses to acRDS, suggesting the response is reduced from the V1
• Umeda et al (2007):
Recorded responses in the V4 from macaques in with RDS (centre and annulus) and found that these V4 neurons showed significant shifts with changing the relative disparity. These shifts were greater than for V2 but less than for an ideal relative disparity-selective cell so suggests that processing from V2 to V4
• Janssen et al (2003)
recording neurons responses to IT neurons to RDS and found that these neurons selectivity for RDS portraying curved convex surfaces but were not selective for the anticorrelated RDS and so may underlie depth percept
• Masson et al (1997):
): Found that anticorrelated patterns initiate vergence responses that are similar but in the opposite direction, despite humans failing to perceive depth in the stimuli suggesting that their visual input is derived from earlier than later where depth percepts are elaborated.
• Takemura et al (2001):
Recordings of single units in MST of monkeys in response to random dot patterns and found that when summed disparity tuning curves together it fitted with animals vergence responses and the latency of activity suggest the activity of the MST occurs early enough to play a part in the generation of vergence
• Krug et al (2004):
Recording responses of neurons in V5 of macaques to rotation of binocular cylinder stimuli and ne with anticorrelated properties and found that these neurons respond systematically to the sign of anticorrelated binocular disparities and that neuron singular or a fixed pool of them could underlie stereo depth percept.
• Krug and Parker (2011):
Recorded neurons in the MT/V5 of macaque to 2 superimposed, transparent planes composed of dots mobbing in the opposite direction to relative and absolute depth and found that many neurons were tuned to relative disparity of individual planes. They found that for 2 plane stimuli, neuronal responses were often linearly related to the responses to absolute disparity but some aspects of relative disparity tuning was not explained by linear combination. This relative disparity tuning may be important for segmentation and depth order of moving visual features
• Uka and DeAngelis (2004):
Look at responses of MT to performance of 2 depth discrimination tasks a course task (absolute disparities in the presence of noise) and fine task (very small differences In relative disparity) and found that electrical stimulation of MT during the fine task does not affect perceptual decisions although many individual neurons have sufficient sensitivity to account for behavioural performance however microstimulation during the coarse task does bias perception. So this would suggest that MT encodes absolute depth not relative depth
• Taira et al (2000):
recorded surface-orientation selective cells in the CIP of macaque to RDS or solid figure stereogram without perspective cue using a delated matching to sample tast and found that all neurons showed orientation selectivity for both. Found that for the RDS, neurons that show orientation sensitivity compute surface orientation from the gradient across the surface whereas for the SFS its from the gradient along the contours. So likely this will be area of higher order processing b integrating the input signals from many disparity sensitive neurons with different disparity tunings
