6 - multi sensory integration II Flashcards
why do we have sensory conflicts?
perceptual limits: senses aren’t perfect e.g. visual resolution determined by spacing of photoreceptors in the fovea
neural noise: none of our senses are completely fool proof, we don’t always get a clean signal e.g. synaptic noise
cognitive resource limits: e.g. attention which strengthens signal
evidence for visual dominance?
create sensory conflict using reducing lens, so object looks small but feels big
participants thought object was small, therefore vision dominates perceived object size
how does attention interact with multi sensory integration?
attention can affect how multi sensory signals are integrated (McGurk effect)
multi sensory signals can guide attention (reading papers with read aloud on allows us to read faster)
multi sensory integration is probably both pre and post attentive
what are theories of sensory conflicts?
- visual dominance
- modality precision
- weighting based on uncertainty
what is the modality precision hypothesis?
the dominance of modality depends on task:
- spatial task: vision
- temporal task: audition
evidence of the modality precision hypothesis
participants presented with flashes and auditory beeps during flashes
if there was 1 flash and 2 beeps, participants perceived this as 2 flashes
audition dominated vision in temporal task
evidence of weighting based on uncertainty?
present virtual bar with sensory conflict between haptic and visual information
compare against bar without sensory conflict
determine point of subjective equality (PSE) - point where people cannot distinguish which is taller
then we manipulate sensory uncertainty of visual feedback
more visual noise moves PSE towards haptic size
when uncertainty of visual information increases, dominance of vision decreases
what is a normative model?
how a problem should be solved (optimal solution)
- based on theory
- establishes bounds
what is a process model?
how a problem is actually solved
- based on data
what is a normative model of multi sensory integration?
Q: how should we solve problem of sensory integration
A: pick integration method that minimises uncertainty
how is variability affected by sensory uncertainty
high variance = high sensory uncertainty
low variance = low sensory uncertainty
what is the ideal solution to sensory conflict? (normative model)
haptic and visual stimuli create a combined signal that is less variable and uncertain than either estimate alone
when both signals have same amount of uncertainty, combined signals is in the middle
if signals have different amount of uncertainty, combined signal is biased towards less uncertain stimuli, no longer in the middle
how to test how people actually solve sensory conflict?
compare PSE to what computational model predicts is PSE (combined signal of both stimuli)
found that humans use uncertainty during multi sensory integration, they also follow optimal sensory integration rules (normative model)
