Presentation Flashcards
Slide 1:
Hi, my name is Gitanjali Sharma and today I will be doing a presentation on Epstein and Kanwisher paper called a cortical representation of the local visual environment
Slide 2 - (2)
There is a region in the parahippocamapl gyrus called the parahippocampal place area , that has a higher response to images to places as compared to non-place images like objects or faces (PPA is in the yellow arrow).
Episten and Kaniwhser’s study wanted to investigate whether this specific region demonstrates whether it shows selectivity to scenes/places using region of interest (ROI) analysis and standard functional magnetic resonance (fMRI) design.
Slide 3 (3)
In the first experiment, participants viewed videotapes of black and white photographs of faces, common objects, houses and scenes that were scrambled or intact (point to the images)
The procedure for all the experiments the researchers conducted was that during the scan the participants either viewed the photographs passively or performed a 1 back repetition detection task which involved participants to press a button when they saw two identical pictures in a row
They found that PPA responds selectively to visually presented scenes even low-level features were subtracted out (point to the bottom column by subtracting the MRI percent signal for both intact and scrambled) and when subjects merely view the stimuli passively with no task required for the image.
Slide 4 - (6)
In their second experiment, participants view photographs of faces, singular or groups of objects, landmarks on white background, landscapes, unfamiliar rooms that were empty or furnished and familiar outdoor scenes (i.e., partiicpants’ MIT university campus).
They found that:
PPA responded much more strongly to places so photographs of landmarks, landscapes, empty rooms, furnished rooms, outdoor scenes than it did to faces, objects or multiple objects.
PPA’s response to empty rooms was as strong as its response to the same rooms furnished
PPA’s response to the same rooms furnished was twice as strong as compared to its response of many objects with no spatial context (i.e., in no room).
PPA’s response to multiple objects was not significantly greater than to single objects
Slide 5 - (4)
In their third and final experiment, they showed participants images of faces, objects, intact rooms, fractured rooms (relative positions of the surfaces of room were preserved) and fractured +rearranged rooms (positions of surfaces of room were arranged so they no longer defined a space)
The researchers found that:
- PPA responded more strongly to both intact and fractured rooms than fractured + re arranged
- More specifically, PPA was selective of the spatial layout as they found PPA responded higher to fractured than fractured + rearranged
Slide (6) - (6)
The experiments showed that:
1. There is a region of parahippocampal cortex that responds selectively to visual scenes depicting places
2. PPA activation occurs automatically even when no explicit cognitive test is required
3. This response is found even if there are no discrete objects in the scene
4. Critical factor in this response is the presence in the stimulus of info about the layout of local space
- Together results suggest that PPA performs an analysis of the shape of the local environment that is critical to determine where we are.
- ROI (region of interest) is where researchers involves extraction of
fMRI signals from specified regions in this case Para hippocampal cortex
- The one-back task ensured the participants were
attended at least as much to uninteresting stimulus sets (e.g., scrambled images) as to the more interesting sets (faces and scenes)
- Explain the low-level differences -
They wanted to distinguish the changes in activation in PPA resulting from high-level differences from stimulus types to changes in activation resulting from low-level features between stimulus types by subtracting per cent signal for scrambled photographs (low-level) to those of intact photographs (high)
- Why did they do a second experiment? - (2)
o The preference for PPA for photographs of scenes over everything is suggestive but does not show PPA encodes for spatial layout
o If PPA is involved in representing the shape of the local environment, it would be strongly activated by scenes even when they show bare spatial layout without any discrete objects
Why did they do a third experiment?
- In a final test of the spatial layout hypothesis
