Psy 342 Learning & Memory
Storage &
(Terry Ch. 10) 


A. A basic question connected to the storage of information is “How long is information retained?”

1. Ebbinghaus’s research and classic forgetting curve suggested that not much information is stored in LTM.

2. Other research, in more naturalistic settings, shows that memories can be stored for decades.

a. A critical point to also consider here, however, is that the inability to recall information might indicate bad retrieval, as opposed to memory loss.

B. Bahrick’s (1975) work studied the memories that people had of their high school experiences.

1. Depending on how much time had passed since one’s high school graduation, subjects freely recalled between 20-40% of people’s names; recognition accuracy rates were between 70-80%.

2. Studying memory in such a naturalistic manner poses several problems, from a control perspective.

C. Others have extended Bahrick’s work to study retention of course material

1. Research by Conway, et al. (1991) shows that memory for course material from a cognitive psychology course shows the largest amount of forgetting during the first 3-4 years post-course, and after that knowledge remains at a stable, above-chance level.

2. A critical determinant of retention is initial acquisition level. Those who complete difficult courses and earn high grades seem to remember more than those with easier courses and lower grades.

D. The overall nature of how memories are stored has also been given attention by memory researchers.

1. One of the more dominant storage models is that of semantic memory, which proposes a very logical organization of general information.

a. Such semantic systems differ from the storage of more personal, episodic memories, in the sense that episodic memories are stored in a more idiosyncratic manner across all individuals.

2. Semantic network models argue that items are always connected by associative pathways that vary in strength and illustrate the relationship between items.

3. Collins & Quillian’s model is one of the more common semantic networks, organizing information in a hierarchical manner, with general categories at the top of the hierarchy and more subordinate elements at the bottom of the hierarchy.

4. The retrieval of information in semantic networks is said to proceed by a Spreading of Activation, where the activation of one item in memory leads to the activation of other items connected to the initial item, and so on.

a. Spreading of activation can be limited, however, by the fact that activation tends to weaken as it expands throughout the network, and the fan effect can inhibit information processing.
b. The fan effect occurs when a particular item has so many connections that the activation of any particular connection is slowed in comparison to the activation of concepts that have fewer connections.

5. At a more microscopic biological level, the connections between individual neurons are also pointed to for evidence of how memories are organized.
a. The process of Long-Term-Potentiation (LTP) suggests that repeated activation of synaptic connections leads to a biological type of spreading of activation, and this process underlies how memories are formed in a physical sense.
b. Kandel’s research on the LTP process, within Aplysia, has revealed possible explanations for how short-term (STM) and long-term memory (LTM) occur at the neuronal level.

6. Consolidation Theory offers an explanation as to how memories are transformed from temporary to permanent traces.
a. When information is first attended to, neural or environmental events can disrupt memory processing causing memory loss.
b. In the absence of such forgetting, acquired information is eventually moved from short-term memory and into long-term memory.



A. In order for memories to be useful, they must be retrieved. However, retrieving information from episodic memory is different than semantic retrieval.

B. Several factors play a role in episodic retrieval.

1. Memories that are more distinct than others tend to stand out in our memory, as retrieval cues for such information tends to be very specific.

2. Retrieval is also enhanced by practicing the retrieval of information; such testing effects show that memory is enhanced when multiple retrieval attempts are made concerning a particular event.

a. This enhanced memory over successive retrieval attempts is referred to as hypermnesia, as it goes against the tendency to lose information over time (e.g., Ebbinghaus’s classic forgetting curve).

b. Why does hypermnesia occur? Additional tests provide additional retrieval time, and successive retrieval attempts force people to process information from a variety of perspectives, increasing the likelihood of successful retrievals.

C. What types of stimuli serve as good retrieval cues?

1. Mantyla (1986) required experiment subjects to generate up to 3 properties of a presented word, and followed this with a recall test that used participants’ own properties, or those generated by others, as retrieval cues.

a. One week following initial property generation, participants recalled 60-65% of the target words when using their self-generated cues, but only 20% when using properties generated by other individuals.

2. The principle of encoding specificity argues that good retrieval cues are those cues that were also present at encoding; if those cues can be matched, the chances of retrieval are high.  Try this

a. Related to encoding specificity is the notion of State-Dependent Learning, retrieval is enhanced when one’s physiological state at the time of retrieval matches one’s physiological state at the time of initial learning.... Mood-Dependent Recall involves matching mood

Goodwin et al. (1969) demonstrated state-dependent memory using alcohol. 


Why does a shift from intoxication to sobriety impair memory more than a shift from sobriety to intoxication? 

What is encoded? 

D. One’s level of emotional arousal can also influence the manner in which retrieval occurs.

1. When people experience highly arousing emotions such as fear and anxiety, research shows that retrieval of information can be inhibited.

2. This type of memory problem is not an all-or-none effect, however, as some people are able to retrieve information that can be considered unpleasant.


A. Several variables influence the manner in which we manipulate our memory abilities to store and retrieve information.

B. Prospective Memory (PM) is one form of retrieval manipulation, and involves remembering to perform some future action.

1. This ability can be enhanced by external memory aids, such as lists, planners, and electronic reminders.

2. Internal cues can also enhance PM, such as when one engages in spontaneous recall of information in the absence of any external reminders.

3. PM is also considered to be a particularly challenging type of memory to retrieve, as one must remember both what the information is and when it is to occur.


A. A misconception about memory is that once something is stored in LTM, its form never changes (reproductive retrieval).

B. A more accurate conceptualization of memory is that it is reconstructive, in the sense that when retrieval occurs, bits and pieces of an original memory, plus general information connected to the memory, is retrieved.

1. This collection of general and specific information is merged to form a retrieved memory, but it is usually less accurate than the initial memory that was encoded.

2. One of the earliest studies that highlighted this reconstructive nature of LTM was Barlett’s “War of the Ghosts” experiment from 1932.

a. Bartlett introduced the term schema

People’s expectations are based on experience--personal or vicarious    

Script--A schema describing a typical sequence of events, such as attending a lecture or going to a restaurant.  

 Scripts and schemas are generic; they describe typical events and allow for some variability (e.g., restaurants differ, from coffee shops to five-star establishments).  

 Based on script, listeners/readers infer actions not explicitly stated in a text or conversation.

 For instance, we have no difficulty understanding:

John went to a restaurant. He ordered chicken. He left a large tip.


Try this activity

A. Retrieval failure can occur when an individual mistakenly recalls something that they believe to be a true memory, when in effect it is not.

1. This notion of retrieval failure is directly connected to the controversial false memory/recovered memory debate that has arisen in therapeutic circles.

B. Why does false retrieval occur?

1. One explanation is that the spreading of activation that accompanies normal retrieval can sometimes activate information that is related to a target, but not directly connected to a target.

a. This process is highlighted within the Deese-Roediger-McDermott (DRM) procedure.

 Researchers have demonstrated that people can create a false memory for a word they have not actually seen.   

Ss are asked to remember a list of words they had seen earlier. Accuracy is then measured.

 Roediger and McDermott (1995) found a false recall rate of 55%.

 People may “recall” events that are related to their actual experiences, but these events never really occurred.

 Ss can construct false memories for events in their own lives that never actually happened.

  Hyman and colleagues (1995) asked parents of Ss to supply information about several events.

 Interviewed each student individually about several events that really did occur.  Also, during the interview, researchers also planted several false memories.

 Some students gradually began to create a false memory (25% of participants overall).


Loftus & Palmer (1974)—Preliminary study showed that the way a question is asked influences subsequent recall. 

“How fast were the cars going when they _____ each other?”

 Different verb intensity—smashed, collided, bumped, hit, contacted 

Intensity of verb influenced Ss estimate of speed

2. In addition to spreading activation, another explanation of false retrieval involves errors made while trying to determine a source memory.

a. Source Memory involves making a distinction between events that one actually experienced and those that did not occur but are believed to have occurred (e.g., through dreams, imagination, etc.).

b. In this light, a false memory will occur when one cannot recall the actual source of a memory.

3. False retrieval can also occur as a result of repeated attempts to imagine an event happening, a process known as Imagination Inflation.

C. Retrieval failure can also be connected to the presentation of information about an event after an event has passed (post-event information).

Misinformation effect
In general--
People first view an event and afterward are given misleading information about the event. Later, they incorrectly recall the misleading information, rather than the event they actually saw (Zaragoza et al, 1997). 

 Loftus, Miller, and Burns (1978)-- the classic experiment on the misinformation effect.

 Ss saw a series of slides. A sports car stopped at an intersection, and then it turned and hit a pedestrian.  Half saw yield sign at the intersection; the other half saw stop sign. 

Critical question contained information that was either consistent with a detail in the original slide series, inconsistent with that detail, or neutral (i.e., did not mention the detail).  

Ss were then shown two slides (stop sign or yield sign) and asked to select which they had previously seen. 

People who saw the inconsistent information were much less accurate than people in the other two conditions. 

Warning: A confident eyewitness may not necessarily be an accurate eyewitness.

1. People’s memories can be altered by misleading information, making people recall events that never occurred.

2. This post-event information can also be argued to serve as retroactive interference for the critical information.

a. In this light, it is possible that the postevent information replaces the initial memory, or the initial memory has been forgotten and the postevent information is all that is available for recall.

b. The Reconsolidation Hypothesis argues that when a memory is retrieved, its placement into short-term memory makes it vulnerable to disruptive information, which can in turn alter the initial memory in such a way that the reconsolidated data differs from the initially-stored information.

B. Factors affecting accuracy of eyewitness testimony    

Errors are More Likely If: 

(1) The witness’s attention has been distracted at the time of the event. 

(2) The misinformation is plausible. 

(3) There is social pressure to provide a specific answer.

(4) Eyewitness has been given positive feedback.

 (5) The response is based on process of comparison or elimination, rather than just "knowing"

 (6) Simultaneous line-up procedure (encourages relative judgments)

    What's the alternative?



A. A current controversy within memory research focuses on whether or not memories of childhood abuse can be repressed.

1. Some psychotherapists claim that recovery of repressed memories is a real process.

2. Laboratory research, however, has shown that suggestions can lead to false memories, and that emotional events tend to be well remembered.

B. One of the critical problems within this controversy centers around the fact that in many cases of repressed/recovered memories, it is difficult to verify the claims, based on the time lapse between the event and the recall, as well as the lack of hard evidence supporting the abuse claim.