Pattern recognition: bottom up approach, we identify complex arrangements of sensory stimuli

Perceptual organization: the organizational proccesses that influece the shapes we see (top-down approach)

• Laws of grouping
• proximity or nearness
• similarity
• good continuation
• closure
• common fate (moving in same direction)
• Law of pragnanz: general princple encompassing all gestalt laws of grouping
• figure ground relationship
• ambiguous figures and ground relationships

General principle encompassing all gestalt laws

Of all the possible perceptual organizations, we pick the one that is the best,  simplest, & most stable

-we like some shapes better than others (circle better than squares, squares better than rectangles)

-we like symmetry

when 2 areas share a common boundary, the figure is the distinct shape with clearly defined edges

-the rest of the stimulus is the background

Law of simplicity or good figure: we assume shapes are complete and just partially hidden

1. methodology is phenomenological observation: nonrigorous research methods, participants just describe immediate perception, not sure of more detail bc you aren't suppose to probe
2. laws of grouping are vageue and not mutually exclusive: don't know which law explains for grouping so you can't test them
3. most shapes are already perceived as whole so saying "perceive the whole" isnt real explanatory, dont know how we perceive it that way
4. no neurological, anatomic, or physiological explanation 

1. gestalt approach

2. prototype-matching approach: top down

3. distinctive features approach: bottom up

Top-down processing

Idealized, abstract prototype in memory

e.g. prototype of 'M' in memory and it allows u to match all of the 'M's we see

Different from template-matching: must have each form of the figure stored in memory to perceive it (relearn each time you saw diff handwriting)

-template doesn't hold up for complex patterns

bottom-up

-we have definitive features of items that help us perceive them

-compatible w/ visual anatomy and physiology (specific cells respond to certain line orientations)

e.g. for letters we would detect straight v. curved lines, then detect intersections of lines.

top down effects

parsing paradox: how can we know the context unless we recognize individual parts? BUT how can we recognize ind parts until we know the context?

word superiority effect/word apprehension effect:

l3 R O N Z E vs. l360428 (perceive B or 13)

we can perceive facial parts w/out detail when on a face or with detail when not on face, but cdon't know what they are without context and detail

-remember specific angles better if they are arranged into a right triangle

demonstrates illusory or subjective contours: edge that we see, even thought it isn't physically present, see an edge or contour based on context alone

inducing areas or contours (lines): area that cause the illusion, changing the inducing areas/contours changes what we see (can make it brighter or change the shape)

illusion: incorrect perception

Backwards masking occurs in the visual system

-present 1 shape briefly (150msec) and present another that spatially overlaps the first

-we can't report what the first shape was

-second shape masks the first

No forward masking in visual system

-maladaptive (wouldn't know where a car is now bc where it previously was would mask current location)

-if material from every fixation pause were retained during later fixation pause you would see letters that overlapped when reading (maladaptive)

-backwards masking lets each spot be read clearly