Prof Spence - Psychology
@11 on Wednesday, 03 May, 2000
complex 3D shapes paired, either identical or mirror image � subjects were asked which
linear relation between angle and reaction time
but� why should our ability to rotate mental images be limited to 60� rotation rate?
should always say �imagining the rotation of an object�, NOT �rotating the mental image of an object�
rate of mental rotation independent of figural complexity (2D) (60�/�)
though there are some factors which do alter the speed
presented subjects with wire 2D shapes (no part is ever obscured) � v difficult to rotate
people were unable to
why?
not ecologically valid � no surface area
they argue that it�s a consequence of the difficulty of performing the comparison
in Consciousness Explained
complex block, with one surface being painted red: can it be seen through a peephole, requiring the subject to rotate the shape � most people can�t do this
apparently, lose information, e.g. symmetry in axes
presented random polygons
could rotate 60�/sec
seems strongest evidence that people can perform mental rotation
Kosslyn (1980, 1994) � argues for analogue format for mental imagery. caricatured meatphor for image generation = cinema projector
Pylyshyn (1981) - disagrees, arguing that everything is stored in a propositional format. caricatured metaphors for image generation = compture graphics (generation of images from propositional form)
overhead map of an island, with landmarks � ask subject to remember their position, point to it, then imagine going from one place to another, measuring time
as the distance the spot will have to travel increases, so will reaction time (as in mental rotation)
but: modulated by experimental expectancy effects
didn�t stand up to Kosslyn�s criticisms
dots in random positions � then shown an arrow and asked if it points to one of the dots, as remembered
no mention of imagining the dot�s movement here � showed linear correlation, takes longer to scan further
shows that we can scan mental images, probably analogue fashion
set in England, map, showing towns + names
then asked, is Hull north or south of Sheffield
for towns close together, much harder � opposite of the 2 previous scanning studies � implying propositional account
3x3 grid, write GAS OIL DRY in the three lines � ask them to scan for the 3 letters downwards � v difficult
could be because words are stored in some more abstracted format which makes them less open to the mental image scanning
shown complicated image for 5 secs, looks like rabbit/duck � designers/architects were asked to draw the image � couldn�t then re-interpret their mental image to see the ambiguity
recently: controversy over whether C&R trained their subjects right
imagine a capital letter, �D�, rotate the figure 90�, then add �J� underneath
people were able to re-interpet their image as an umbrella
C&R argue that though you can see the umbrella, you never lose sight of the �D� and the �J�
whereas the duck/rabbit replace each other � also require you to reverse back/front
cognitive psychologist � why can we mentally imagine things? because it enables us to predict future outcomes, by carrying an action out on my mental imagery
if so, then should see it if you take more real images � takes pictures of real world scenes which could only have one direction of motion, e.g. someone jumping off a wall, glasses smashing etc.
we should be able to see the first image, then imagine what will happen in a few seconds
present still video clips, then black screen, then second image � same or different? if shown a few frames later, then they will take longer because are comparing it with their projected mental image. if shown a few frames earlier, faster at telling that the image is definitely different.
extensive ability to transform + manipulate mental images: rotate, scan, zoom in, maybe re-interpret
Rock�s pipe-cleaner figures, Dennett�s cube-peephole
evidence seems to support mainly analogical, not propositional, view of mental imagery
mental imagery must have some functional role
1. Shepard & Meltzer - max rate = 60�/what?
2. Shepard & Meltzer - could it be because there is no way to predict what the shapes might look like?
3. Wilton - does familiarity with the map make a difference?
4. Chambers & Reisberg � simply cannot remember the image in sufficient detail, or store it with a bias