Transcript Cognition – 2/e Dr. . Daniel B. Willingham

Cognition – 2/e
Dr. Daniel B. Willingham
Chapter 5:
Memory Encoding
PowerPoint by Glenn E. Meyer, Trinity University
©2004 Prentice Hall
What Determines What We
Encode in Memory?
•
•
•
•
Factors that Help Memory: Emotion &
Depth
Factors that Don’t Help Memory: Intention
to Learn and Repetition
Problems with the Levels of Processing
Theory
Match Between Encoding and Retrieval:
Transfer Appropriate Processing
©2004 Prentice Hall
2
Factors that Help Memory: Emotion and
Depth
•
Laboratory Studies of Emotion and Memory - continued:
•
Cahill, et al. (1996) – PET analysis during recall of emotional or
nonemotional film clips found
•
Emotional clips better recalled
• Activity in amygdala correlated with recall. Amygdala suggested to
modulate memory based on its evaluation of fear and disgust
•
•
•
Canli, et al. (2002) – women remember more emotional pictures
than men and show more left amygdala activation
Buchanan, et al. (2001) left amygdala more important than right in
enhancing emotion effects on memory.
Burke, et al. (1992) – slide show with surgical scenes better
remember than one with scenes of auto repair. But result might be
due to distinctiveness rather than emotion
©2004 Prentice Hall
3
Factors that Help - Continued
•
Flashbulb memories
•
•
•
Definition: A very rich, very detailed memory that is encoded when
something that is emotionally intense happens. For example:
Where were you when you heard of 9/11? The Kennedy
assassination? The Challenger explosion?
Brown and Kulik (1977) – first study of such, asked about Kennedy
assassination.
•
Suggested flashbulb memories have three properties:
•
•
•
•
•
•
Very complete
Immune from forgetting
Are accurate
Due to a “NOW PRINT” process that takes a memorial snapshot of
what is happening under great emotional duress.
Modern thought is that there is NO special flashbulb memory
system
Shmolck, Buffalo and Squire (2000) studied OJ verdict memory
and found that over time, memories do contain major distortions.
Many details were reported, seemingly like a snapshot, but were
incorrect.
©2004 Prentice Hall
4
Factors that Help - Continued
•
Depth of Processing
•
•
Definition: A description of how one thinks about material at
encoding. Depth refers to the degree of semantic involvement (that
is, the word’s meaning).
Craik and Lockhart (1972) – Levels of Processing framework
•
What is remembered based on depth of processing. The deeper the
level the better the memory. Processing level can be:
•
•
•
•
•
Deep – greater degree of semantic involvement and thinking about such
Shallow – thinking about the surface characteristics of the item
Examples are seen in Table 5.1 from Craik and Tulving (1975)
Craik and Tulving (1975) demonstrated that words are better
remembered if processed more deeply according to their definitions
of depth as seen in Table 5.1 and Figure 5.2
Notions similar to depth of processing:
•
•
Elaborative rehearsal: A type of encoding in which new material is
related to material one already knows - would aid memory
Maintenance rehearsal: A type of encoding in which one repeats new
material over and over to oneself . Not much of an aid to getting info
into secondary memory according to research
©2004 Prentice Hall
5
Factors that Help - Continued
•
Depth and Elaboration
•
•
•
Elaborative processing does not always add to deep
processing and better memory
Elaboration has to be relevant to what you are trying to
remember
Bradshaw and Anderson (1982):
•
•
•
•
•
•
Subjects heard sentence about a famous person
Some then heard info relevant to original sentence
Some heard info that was unrelated to original sentence.
Folks who heard just original sentence had 38% recall, those
who heard the irrelevant extra fact only had 32% recall.
Folks who heard the extra relevant fact had 61% recall
Thus, info must be relevant to aid in deep processing
and memory performance.
©2004 Prentice Hall
6
Factors that Don’t Help Memory:
Intention to Learn and Repetition
•
Intention to learn
•
•
•
Depth of processing studies indicate memory not affected by effort
to learn as tested with incidental memory tests (subject not told
they will be test on materials they are dealing with). Intentional
memory test inform subjects that they will be tested.
Hyde and Jenkins (1973) compared incidental and intentional
memory by informing some subjects they would be tested on
memory of items they were processing deeply or shallowly. As seen
in Fig. 5.3, intention did not aid in performance. Only the shallow
vs. deep variable was important. The latter aided performance.
Repetition
•
•
Nickerson and Adams (1979) as seen in Fig. 5.4 demonstrated that
memory for the common Lincoln cent was quite bad even though
it had been seen by most many, many times. Is Lincoln facing left
or right? On which side is the date?
Watkins (1973) demonstrated time an item in primary memory had
little effect on LTM.
©2004 Prentice Hall
7
Problems with the Levels of Processing
Theory
1) Concept of Deep Processing is Circular


Circular theory
A theory that uses term A to define term B but then
also uses term B to define term A, leaving unclear
what terms A and B mean.
Critics claimed the only evidence for deep processing
was better memory and better memory was due to
deep processing – that is a circular argument
2) Level of Processing Theory said little about the
importance of memory retrieval
©2004 Prentice Hall
8
Match Between Encoding and Retrieval:
Transfer Appropriate Processing
•
Key Points:
•
•
•
Morris, et al. (1977):
•
•
•
•
Must consider retrieval when discussing encoding
This was a flaw of level of processing theories
Varied encoding and retrieval by having subjects do either
rhyming or semantic analysis on a sentence. The latter should
induce more depth of processing.
When tested with rhyming cues, the subjects in that group
demonstrated a reversal of the depth effect as seen in Fig. 5.6.
Indicates match between encoding and retrieval processes
important as compared to standard “depth” explanation
Hypothesis based on results:
•
•
Transfer appropriate processing -The concept that when similar
processes are used at encoding and retrieval times, retrieval will be
more successful than if different processes were used
Criticized as being circular – similar to depth of processing
problems
©2004 Prentice Hall
9
What is Remembered and What is
Forgotten – Brain Activity
•
Brewer, et al. (1998), Wagner et al. (1998)
as seen in Box 5-2
• Subjects made judgments about pictures or words during
an FMRI.
• Then took a recognition test during FMRI
• Findings:
•
Right dorsolateral prefrontal and bilateral parahippocampal
cortex – more activity for items that would be remembered
• Word targets – activity restricted to left parahippocampal
cortex
• Picture targets – bilateral activity in parahippocampal cortex
©2004 Prentice Hall
10
Why Do We Encode Information as
We Do?
•
Prior Knowledge Reduces What
We Must Remember
• Prior Knowledge Guides the
Interpretations of Details
• Prior Knowledge Makes Unusual
Things Stand Out
©2004 Prentice Hall
11
Prior Knowledge Reduces What We
Must Remember
•
Use of Chunking Reduces Memory Load
•
•
Chunk A unit of knowledge that can be decomposed into smaller
units of knowledge. Similarly, smaller units of knowledge can be
combined (“chunked”) into a single unit of knowledge (e.g.,
chunking the numbers 1, 9, 0, and 0 into a unit to represent the
year 1900)
Chase and Simon (1973)
•
•
•
Master level chess players and nonexperts viewed middle game of a
chess board
Master player memory of position far superior due to chunking of
pieces into larger meaningful units using their expert knowledge
Reingold, et al. (2001) – replicated results with a sophisticated eye
movement tracking study
©2004 Prentice Hall
12
Prior Knowledge Guides the
Interpretations of Details
•
•
Prior knowledge can be thought of as a set of related facts
Facts can come in a packet known as a schema (term
introduced by Bartlett (1932)) which can guide memory
processes:
•
Schema: A memory representation containing general
information about an object or an event. It contains
information representative of a type of event rather than of a
single event
• Default value: A characteristic that is a part of a schema that
is assumed to be true in the absence of other information.
For example, unless one is told otherwise, one assumes that a
dog is furry; furriness is a default characteristic for dogs
©2004 Prentice Hall
13
Prior Knowledge Guides the Interpretations
of Details - Continued
•
•
Schemas aid in interpreting ambiguities .
(Bransford and Johnson, 1972) had subjects try to recall the
following:
The procedure is actually quite simple. First you arrange things into different
groups. Of course one pile may be sufficient depending on how much there is
to do. If you have to go somewhere else due to lack of facilities that is the next
step, otherwise you are pretty well set. It is important not to overdo things.
That is, it is better to do too few things at one time than too many. In the short
run this may not seem important but complications can easily arise. A mistake
can be expensive as well. At first the whole procedure will seem complicated.
Soon, however, it will become just another facet of life. It is difficult to foresee
any end to the necessity for this task in the immediate future, but then one can
never tell. After the procedure is completed one arranges the materials into
different groups again. Then they can be put into their appropriate places.
Eventually they will be used once more and the whole cycle will then have to be
repeated. However, that is a part of life.
•
Memory was much improved if the subjects knew that the paragraph
referred to “washing clothes” as that knowledge reduced the ambiguity
of the paragraph. Similar results found by Anderson and Pritchard
(1978)
©2004 Prentice Hall
14
Prior Knowledge Makes Unusual Things
Stand out
•
Prior knowledge leads to expect what usually happens
• Knowledge about common situations postulated to be organized
into scripts (Schank and Abelson, 1977).
•
•
Definition: Script - A type of schema that describes a series of
events
Example:
•
•
•
What to do in a restaurant
Visiting the doctor
Bower, et al. (1979) found good agreement among subjexts on basic
scripts of American Culture as seen in Table 5.2
•
Studies show that memory is best for details that are not part of
the script and relevant to the goals of the script – Bower, et al.
(1979)
• Zachs, et al. (2001) – scripts used at retrieval and to understand
ongoing behavior.
©2004 Prentice Hall
15