multisensory integration 2 Flashcards
McGurk effect
Conflict between visual and auditory stimuli. Saying one thing but lips move like they are saying another thing. Hear a different thing when you shut your eyes
Why do we have sensory uncertainty
Perceptual limits- e.g. needing glasses
neural noise- we don’t hear things perfectly
cognitive resource limits- attention
Visual dominance theory
People look at an object through a lens and feel it with their hands, the lens makes the object appear smaller than it is.
People think that the object is smaller, because thats what they can see, regardless of what they can feel.
This theory states that vision dominates- visual cortex is much more developed than our other senses
Modality precision hypothesis
Which modality dominates will depend on the task
Spatial- vision
Temporal task- audition- hearing is more sensitive to time than vision
Proof that audition can dominate vision
Participants were asked to count how many flashes they saw on a screen, when beeps were played simultaneously they incorrectly identified the number of flashes. e.g. when there were 4 beeps and 1 flash, people recorded seeing 2 flashes
Bar task- uncertainty of one sense improves the other
Present participants with a bar that is 50mm tall but looks 60mm tall.
Compare it to a bar without sensory conflict- 55mm tall
Determine point of subjective equality (PSE)- point where people cannot distinguish which is taller- 50% say its taller 50% say its shorter
They then manipulated the sensory uncertainty of visual feedback by introducing visual noise
Without visual noise visual input biased perception of the bar height (compared to touch)
When they introduced 100% visual noise, PSE is accurat, people are using both visual and haptic senses
Normative model
How a problem should be solved (optimal solution)
Based on theory
Process model
How a problem is actually solved
Based on data
Integrating probabilities
A normative model of multisensory integration would combine the uncertainties of the haptic and visual conflict an create an output that is less uncertain than the two others
if once signal is less uncertain than the other (has less variance) the combined estimate is biased towards that one
Correspondence problem
How does the brain decide that two signals are coming from the same source?
