17.+Visual+Knowledge

April 21, 2015
 * Visual Knowledge**

Concepts Applications
 * Outline**
 * Mental images are like pictures
 * Mental images aren't like pictures
 * Dual coding and interactive imagery
 * Long-term visual memory
 * Eidetic memory


 * Concepts**
 * Mental Image** - the mental representation of objects or behaviors that are not actually present
 * Two main theories of how mental images are stored:
 * **Depiction**: the representation resembles the object
 * It's like a picture
 * Evidence:
 * **Description**: we have an abstract, linguistic representation
 * It's words
 * Kosslyn, Ball, & Reiser (1978) - Subjects asked to memorize a map.[[image:visualmap.png width="119" height="132" align="right"]]
 * Asked to:
 * Imagine it
 * Scan between 2 landmarks
 * Subjects press a button when they "arrive"
 * Greater distance = longer scanning times
 * Kosslyn (1975)
 * "Imagine a rabbit with a __fly__ next to it."
 * "Does a rabbit have a pink nose?"
 * "Imagine a rabbit with an __elephant__ next to it."
 * "Does a rabbit have a pink nose?"
 * You have to "zoom in," so participants took a little longer in the elephant + rabbit condition
 * Mental Rotation[[image:rotation.png width="129" height="171" align="right"]][[image:enantiomers.png width="144" height="81" align="right"]]
 * Shepard and Metzler (1971): Participants are shown two images and asked if they are the same. Participants have to rotate the image to decide.
 * Decision time is a function of the amount of rotation (participants take longer to answer if they have to rotate it further)
 * The image actually turns in your mind's eye
 * About 1 degree every 60 ms
 * It seems that mental rotation involves depiction (not description)
 * Depiction vs. Description
 * Evidence seems to support depiction (reaction time was measured in all three cases)
 * Scanning
 * Zooming
 * Rotation
 * Visual vs. Spatial Imagery
 * Evidence that visual imagery is different from spatial imagery:
 * People blind from birth
 * Same rotation effects
 * Same scanning effects
 * "Vivid imagers" vs. "non-imagers"
 * Vivid imagers are better on tasks that specifically require visual imagery
 * Equal on tasks that depend on spatial imagery
 * Imaging in the brain - brain evidence
 * What parts of the brain active during visual imagery?
 * Predictions:
 * Description: verbal areas should be active
 * Depiction: visual areas should be active
 * Le Bihan et al. (1993): Participants see patterns or imagine them. Primary visual cortex activity recorded using fMRI
 * Visual imagery happens in brain areas responsible for vision
 * We can engage in visual imagery without seeing the stimuli (i.e., no input)
 * Similarly, we can engage in motor imagery without doing the movements (i.e., no output)
 * When you imagine going bowling:
 * Your visual cortex can "see" a bowling alley
 * Your motor cortex can "bowl"
 * Message: brain areas that underlie mental simulation also underlie actually sensory/motor activity
 * Disrupting visual-processing areas with transcranial magnetic stimulation (TMS) also disrupts mental imagery
 * People who have trouble recognizing objects may also have trouble imaging those objects
 * living vs. nonliving things
 * faces
 * colors
 * Evidence that visual images are NOT just like pictures:
 * Chambers & Reisberg (1985): Show Duck/Rabbit. Remove it.[[image:duckrabbit.png width="164" height="108" align="right"]]
 * Ask participants if they can reinterpret it
 * Success rate: 0%
 * Ask them to draw it from memory and then try to reinterpret it
 * Success rate: 100%
 * Reference frames - the mental image is different from the actual image
 * Understood as a duck, associated with the reference frame "facing to the left"
 * Imagery goes beyond the perceptual information
 * Summary:
 * Depiction theory explains the findings best
 * Visual imagery is a lot like visual perception
 * RT (reaction time) data
 * Brain data
 * Images aren't perfect pictures
 * They're far less accurate and detailed
 * They have reference frames and schemas attached.

Dual Coding
 * Applications**
 * Dual coding : encoding both the word and the picture
 * Benefits from the words
 * I need to grab my backpack
 * Benefit from the image
 * Form mental image of backpack
 * Some hotels do this by putting a picture next to your room number to help you remember which room you're in
 * Interactive imagery: images cue each other when they interact[[image:toast.png width="169" height="108" align="right"]]
 * Example: If you're going shopping and you want to remember eggs and bread
 * You could form images of eggs and bread
 * But what's even better is if you form an image of them together as egg in a hole

How good is visual memory?
 * Mitchell (2006)
 * Longitudinal group[[image:camel.png width="169" height="79" align="right"]]
 * Participants studied line drawings 3x each
 * 17 years later they got a package in the mail
 * Control group
 * Never studied the line drawings
 * Got the same package
 * Package said: "identify these things"
 * [[image:outline.png width="195" height="90"]]
 * Participants who'd seen the drawings before did better
 * ...even though some participants didn't even remember doing the experiment
 * Nonconscious priming after 17 years
 * Visual memory after very brief exposure many years ago
 * Dissociation between explicit and implicit memory

Eidetic memory - extremely detailed form of visual memory