Recall, Recognition, Encoding Specificity

1
Recall, Recognition, Encoding Specificity

• Rationale
• Since Ebbinghaus (1885), two measures have been
  used to study human memory
• Recall
• Recognition

2

• Recall
• Free recall or Cued recall
• Subjects have to retrieve information
• Recognition
• Yes/No or Multiple Choice
• Subjects identify correct items among distractors.

3
Big Question

• What is the difference between these measures?
• A number of theories.

4
Strength Model

• The oldest.
• Claim
• The difference is the strength of memory trace
  they require to be successful.
• Recall requires stronger memory trace than
  recognition.

5

• Support
• Recognition is often easier than recall.
• Problem
• Word frequency effect
• High frequency words are easier to recall than
  low frequency words.
• Low frequency words are easier to recognize than
  high frequency words.

6
New Model

• Kintch (1971)
• Difference is the number of processes they
  require.
• Recall - Generate Recognition
• Recognition - Recognition
• Generate-Recognize model
• Support - Word Frequency Effect

7
But, is it?

• Tulving Thompson (1973)
• Maybe not.
• Whats wrong with it?

8
Which one is easier?

• Generate-Recognize model
• Recognition is easier than recall
• Recall generate recognition
• Recognition recogition

9
Can recall be easier than recognition?

• Generate-Recognize model
• No
• Recall generate recognition
• Recognition recognition

10
Tulving Thompson (1973)

• Phase I - Memorized the capitalized words.
• pretty - BLUE
• Phase II - Free association and recognition
• lake - water, ocean, blue, deep
• Phase III - Recall
• pretty - _________

11
Found

• Recall was easier than recognition.
• How can it be?
• The difference is not the type of processing.
• The difference is the amount of overlap in cues
  between encoding and retrieval.
• Often - overlap is greater for recognition than
  recall
• But - you can manipulate it to make recall easier.

12
Encoding Specificity

• Performance depends on the amount of overlap in
  cues between encoding and retrieval.
• Explain many phenomena
• State dependent memory
• Environmental reinstatement effect

13
Eich et al. (1975)

• Encoding
• THC no THC THC
• Retrieval
• no THC

14
Eich et al. (1975)

• Encoding
• THC no THC THC
• Retrieval
• no THC

15
This weeks experiment

• Manipulate the amount of overlap in cues between
  encoding and retrieval.
• Test - whether recall can be made easier than
  recognition
• If so - Encoding Specificity Hypothesis is
  supported.
• If not - Generate-Recognize model is supported.