Memory

1
Chapter 8

• Memory

2
What is Memory?
Often when we use the word memory we are
referring to the conscious recollection of some
past experience. e.g., What did you wear
yesterday? However, in Psychology, we define
memory much more generally. My definitions would
go something like Memory influences are any
influence by which past experiences affect
current performance. Given this more broad
definition, there may be a large number of ways
in which memory can influence us for example
3
Alphabetic Arithmetic Example

• A3D
• C2F
• T2W
• A3D
• S3V
• C2F
• S3V
• T2W

• C2F
• S3V
• A3D
• T2W
• A3D
• C2F
• S3V
• T2W

4
Rough Processing Model of Memory
Stimulus
aka, iconic or echoic memory
Sensory Memory
aka, short-term memory
Working Memory
Long-Term Memory
aka, memory
5
Sensory Memory
Often a sensory trace or the stimulus remains
after the stimulus is gone. These traces are
termed sensory memory, and they tend to be very
short-lived. Sensory memory was most extensively
studies by a cognitive psychologist named
Sperling. Sperlings studies focused on visual
sensory memory which he termed iconic memory
heres how they worked.
6
Iconic Memory - Full Report Condition
Nine items will briefly be presented in the box
below, then they will disappear. How many can
you remember?
7
Iconic Memory - Full Report Condition
Nine items will briefly be presented in the box
below, then they will disappear. How many can
you remember?
K L W D S P H J A
8
Iconic Memory - Full Report Condition
What was your subjective impression? Did you
think you saw them all for a short while then
they faded away?
9
Iconic Memory - Partial Report Condition
This time, report only the row that is indicated
by the arrow that comes up after the letters are
gone.
10
Iconic Memory - Partial Report Condition
This time, report only the row that is indicated
by the arrow that comes up after the letters are
gone.
S J U B M Q A R P
11
Iconic Memory - Partial Report Condition
If we multiply your number recalled here by 3, we
likely get a larger number than your full report
number, right?
12
Echoic Memory
There is also an auditory version of sensory
memory that is called echoic memory. You likely
have noticed this form of memory in action. For
example, the what? effect. As a further
example, Steve will now do an auditory demonstrati
on of echoic memory his so called
5-3-5-7-2-stop game. While iconic memory
disappears in approximately 1 second, echoic
memory seems to last about 4 seconds.
13
Rough Processing Model of Memory
Stimulus
aka, iconic or echoic memory
Sensory Memory
aka, short-term memory
Working Memory
Long-Term Memory
aka, memory
14
Short-Term or Working Memory
Steve will now read out a set of numbers, try
your best to remember them. That process you
feel is something Cognitive Psychologists call
working memory. As Steve will now demonstrate,
this form of memory is fairly fragile and
capacity limited. It seems to require a great
deal of mental effort to keep things in working
memory and, once the leave, they are
gone. Sometimes we use this memory for
short-term storage, though it also seems
necessary for transferring info to long-term mem
15
The Relation Between Working Memoryand Long-Term
Memory
The purpose of Working Memory is not to simply
transfer information into long-term memory. In
fact, some would argue that working memory is
what we sometimes call thinking and long-term
memory clearly enters into it. Try some of the
following gt D6K, true or false? gt 5 X 13
? gt Imagine yourself sitting on a camel, how
high could you reach?
16
Primacy and Recency Effects
Primacy Effect
If I gave you a long list of words to remember,
then asked you to just recall all the words you
remember, you would likely remember words at the
beginning and end of the list best.
Recency Effect
The primacy effect is typically attributed to
additional rehearsal of items earlier in the list
- Long-term Memory. In contrast, the recency
effect is typically attributed to
either short-term memory readout or even echoic
memory.
17
How long do things stay in Working Memory?
If a person is allowed to rehearse, information
will stay in working memory for as long as it is
rehearsed. However, if not allowed to rehearse,
Peterson Peterson (1959) showed that
information decays from working memory fairly
quickly. In their experiment, rehearsal was
prevented by making subjects count back from
some number by threes while remembering letter
trios (e.g., JDK, LPD)
18
How are things lost from Working Memory
So, this disappear quite quickly from working
memory if they are not rehearsed what makes
them disappear? One possibility is that the
items just decay over time. A second possibility
is that new items coming into working memory
actually push out things currently in it. This
second possibility seems most reasonable given
the data to the right.
19
The Capacity of Working Memory
As we already discussed, working memory has a
limited capacity. Specifically, the limit seems
to be 7 plus or minus 2 chunks. What is a chunk?
Time for another memory experiment! Hopefully
the demo showed that we can greatly increase our
ability to keep things in working memory by
chunking the information.
20
Rough Processing Model of Memory
Stimulus
aka, iconic or echoic memory
Sensory Memory
aka, short-term memory
Working Memory
Long-Term Memory
aka, memory
21
Long-term Memory
So, a simplistic view on what weve said so far
is that things that are rehearsed enough end up
being stored in long-term memory. Things that
are not rehearsed are not. This transfer of
items from working memory to long- term memory is
called consolidation, and the theory of
consolidation is primarily supported by
concussion studies (e.g, the football
example). However, this simplistic view is not
sufficient.
22
Depth of Processing - Shallow

• flame
• patch
• sonic
• bless
• fleet
• pears
• spade
• bliss
• forth

• peels
• speed
• block
• freak
• pints
• spice
• blush
• frost
• pluck

23
DOP - Shallow - Recall
24
Depth of Processing - Deep

• spoon
• bonds
• glass
• ports
• spray
• boots
• goose
• prize
• steam

• brand
• grass
• quart
• stink
• bride
• green
• queen
• story
• brown

25
DOP - Deep - Recall
26
Depth of Processing - Overview
Clearly then, the way that you rehearse
information effects the likelihood that the
information will enter long-term
memory. Elaborative rehearsal (i.e. deep) tends
to produce superior memory on conceptual tasks
like most memory tasks. Maintenance rehearsal
(what we typically do when trying to remember a
phone number for a little while) is not nearly
as good at transfering info to long-term
memory. Thus, let us re-visit our attractive
person gives you phone number when you do not
have paper example.
27
Learning Without Rehearsal
As the depth of processing section demonstrates,
we can learn things via elaborative rehearsal
this type of process is sometimes referred to as
an effortful process. However, we often remember
things that we did not rehearse in an effortful
manner. The formation of memories for things we
did not perform effortful processing on is called
automatic processing. The exact processes
underlying automatic transfer to long- term
memory are still largely unknown but its very
existence challenges a simple consolidation view.
28
Improving Your Memory via Mnemonics
Given all these studies of memory, what have
we learned about improving memory
skills? Techniques used to improve memory are
called mnemonic strategies and, as the name
implies, typical involve some form of effortful
processing. Well consider three
techniques (1) the method of loci, (2) the
peg-word method, (3) creating a narrative,
and (4) creating acronyms
29
The Method of Loci
In a country far far away (Greece), at a time
long long ago (sixth century BC), oratory skills
were prized, and paper was rare (OK, it didnt
exist). Orators had to come up with ways to
memorize long speeches. They devised the method
of loci. This method simply involves forming an
image of some route you are familiar with (say
the drive to Scarbra), and then placing images
of the concepts you want to remember along this
route. Then, by retracing the route in your head
and examining the images, you can reconstruct the
concepts in order.
30
The Peg-Word Method
This method is similar to the method of loci
except, instead of putting images along a route,
you associate them (via imagery) with nouns that
rhyme with the numbers. The typical
ones one bun image representing concept
1 two shoe image representing concept
2 three tree image representing concept 3
. 32 dirty shoe image representing concept 32
31
Other Techniques
Another way of remembering a list of items is
to create a story (or song) that links the
concepts together in some ordered manner - this
is called forming a narrative. Yet one more
method is to form an acronym that represents the
concepts you want to remember Roy G. Biv The
A.B.C.s of first aid. Begin With Review And
Friend or Big Women Really Are Fun!
32
Memory, One Structure or More?
Currently, one of the debates in memory concerns
whether we have a single, or multiple long-term
memory systems. Those who believe in multiple
memory systems typically talk about things like
the following Episodic Memory - Our memory of
very specific events in our lives tends to
contain rich detailed info. E.g. - What did you
do last night? Semantic Memory - Our general
world knowledge. E.g. - What city is the
capitol of Manitoba? Procedural Memory - Our
memory of how to do things. E.g. - How to ride a
bike, or kill without thinking.
33
Implicit vs. Explicit Memory/Tasks
Another distinction that is often made is the
distinction between implicit and explicit
memory. Implicit memory tasks are ones that test
memory without specifically directing subjects to
think about the study items. In contrast,
explicit memory tasks do direct the subject to
try and use study items when completing the
memory test. Implicit and explicit memory are
the memory structures these tasks are thought to
tap. An example if you please ...
34
Implicit/Explicit Example
Graf Mandler (1984) showed subjects lists of
words and ask them either to perform deep
processing (how much do you like the word) or
shallow processing (how many letters does the
word have) on them
The subjects were then shown stems corresponding
to the items (e.g., spice --gt spi__) and
were given either implicit (complete with the
first word that comes to mind) or explicit
(complete with a study item) instructions.
35
The Biological Basis of Memory
So how does this all relate to the brain? Most
of what we know about the biological basis of
memory comes from research in two areas (1)
Neuropsychological studies of human brain-
damaged patients. (2) Psychophysiological
studies with animals
36
Neuropsychological Studies
There are generally two types of memory
impairments that can occur as a result of brain
damage Retrograde amnesia refers to the
condition where patients cannot remember events
that occurred prior to the head trauma. Most
recent events are the most likely to be lost, and
the amount of loss can be from minutes to
years. Anterograde amnesia is a condition
wherein patients can remember past events just
fine, but they have an inability to form new
long-term memories some types at least.
37
H.M. Alcoholics
One of the most influential cases of anterograde
amnesia was the case of H.M., a patient who
underwent bi-lateral removal of his hippocampus
and amygdala to treat severe epilepsy. The good
news is, his epilepsy was cured. The bad news
is, he ended up with a very profound case of
anterograde amnesia. This condition also is a
common result of a form of alcoholism termed
Korsokoffs syndrome. Some alcoholics get all
their nourishment from the liquor, causing
vitamin deficiency. A lack of one vitamin in
specific leads to profound anterograde amnesia.
38
What is lost and what is not?
Patients with anterograde amnesia do have memory
for events that occurred prior to the trauma, so
clearly they have an intact memory retrieval
system. Thus, it seems their real problem is in
storing new memories but not all kinds of new
memories, just episodic memories it seems. Thus,
the hippocampus appears critical for the
formation of episodic memories
39
Psychophysiological Studies
Critters also appear to form episodic
memories which help them to do things like
remember locations where they have already
searched for food. If their hippocampus is
destroyed, they also appear to suffer from the
loss of episodic memory.
40
Summary of Long-Term Memory So Far
So we know that there are a number of ways things
can get into long-term memory, and various
strategies can be used to facilitate this
process. We also know that there at least seems
to be different types of long-term memory and
episodic memory seems to be the most fragile of
these. Finally, we also know that the
hippocampus appears critical for the formation
of new long-term episodic memories, with
destruction of the hippocampus leading to
anterograde amnesia.
41
Remembering
Generally, psychologists believe that the actual
process of retrieving information from memory is
an automatic process. An automatic process is
one that (a) is carried out very quickly (b)
is not under the control of consciousness (c)
does not interfere with other ongoing
processes Things are assumed to become automatic
via a process of overlearning the assumption
being that if some process is performed over and
over again, it will become automatic
42
Reading as an Automatic Process
As an example of an automatic process, remember
the Stroop experiment we did early in the year?
RED GREEN BLUE RED GREEN RED BLUE GREEN BLUE RED G
REEN BLUE GREEN BLUE RED BLUE
RED GREEN BLUE RED GREEN RED BLUE GREEN BLUE RED G
REEN BLUE GREEN BLUE RED BLUE
43
The Importance of Retrieval Cues
If retrieval from memory is automatic, why does
it sometimes seem so effortful to retrieve
something from memory? For example, the
tip of the tongue phenomenon Retrieval is
automatic IF useful retrieval cues are present in
the environment. The effortful part of retrieval
is trying to come up with effective retrieval
cues that will make retrieval happen.
44
Tranfer-Appropriate Processing
The most effective retrieval cues are those that,
in some manner, re-create part of the original
learning environment. The importance of
recreating the learning environment was first
shown in what may be the only (so far)
psychology experiment with SCUBA divers. That is
why when you lose something people will
often suggest that you retrace your steps from
some point in time when you had that thing. It
is also related, but not identical, to the
concept of state- dependent learning.
45
Decay from Long-Term Memory
Often students feel like they study hard for an
exam and, as soon as the exam is complete, the
information they studied is gone! Is it? How
long does information stay in memory, and how can
we scientifically study memory decay? This issue
was first addressed by Ebbinghaus (1895) and his
results and techniques are still
interesting. Ebbinghaus taught himself thirteen
nonsense syllables (e.g., dax, wuj) and then
tested his memory after various delays.
46
Ebbinghaus Results and Real World
Ebbinghaus work suggests we retain some of the
info for at least 30 days, even when it has no
meaning. Similar real world studies suggest
that we can retain information we learned over 40
years ago or more.
47
Relearning Gone but not Forgotten
Ebbinghaus also showed that even when information
feels like it has been lost from memory, it is
still there. He demonstrated this using a
relearning task in which he had to relearn a list
he had already studied a long time ago. Even
when he felt that the information had been
completely lost from memory, it took him less
time to relearn that information than it had the
first time. So, even though you think you forget
stuff after you write your exam, it is still
there, and you will be able to get it back
quickly if you should ever need to.
48
Remembering Part Fact, Part Fiction
When we remember something, some of what we
remember is fact, and some is a reconstruction
that fits with our ideas of the world and our
current context. How about an experiment?
49
Remembering Part Fact, Part Fiction
When we remember something, some of what we
remember is fact, and some is a reconstruction
that fits with our ideas of the world and our
current context. How warm was it yesterday?
50
Remembering Part Fact, Part Fiction
When we remember something, some of what we
remember is fact, and some is a reconstruction
that fits with our ideas of the world and our
current context. How cold was it yesterday?
51
Remembering Part Fact, Part Fiction
What we seem to do is remember only certain
details that we experienced, then, while
remembering, we create a story that includes
these details. The story we create is biased
by a number of things including (a) the current
context including the question that lead to
recounting, (b) a desire to tell a
coherent, sensical story, (c) our current mental
state and beliefs. Moreover, the confidence we
have in our memories seems largely unrelated to
the accuracy of those memories, often we are most
confident in the memories we created.
52
A Final Experiment (for now)
SLEEP YARN PINK THREAD HAMMER PILLOW MATRESS TI
RED CHAIR THIMBLE SHOT BLANKET TETINUS CACTUS SHE
ET CANDLE STICHES DREAM NIGHT PIERCE
53
A Final Experiment (for now)
PINK? NEEDLE? ROSE? WATCH? CANDLE? TICKET? BED?
BOOK?
54
A Final Experiment (for now)
SLEEP YARN PINK THREAD HAMMER PILLOW MATRESS TI
RED CHAIR THIMBLE SHOT BLANKET TETINUS CACTUS SHE
ET CANDLE STICHES DREAM NIGHT PIERCE
PINK? NEEDLE? ROSE? WATCH? CANDLE? TICKET? BED?
BOOK?