Retrieval Induced Forgetting in Coherent Narrative Text.

Diane Poulos 3

Retrieval Induced Forgetting in Coherent Narrative Text

        The definition of forgetting is to lose the memory of something from the mind. Two possible reasons one forgets are either the memory is not in the brain or it is there but just can not be found. Forgetting can be useful for allowing new memories of similar items to be remembered. This would be useful in a situation like trying to remember a new boy/girl friends name and forgetting the old ones. Yet it also works the other way around. Not being able to remember the new boy/girl friends name because the old on is stuck in your head. This would not be useful.
        Retrieval induced forgetting is a phenomenon which says that in addition to retrieving desired memories, the act of remembering information inhibits related memories. Many studies have been done looking at this phenomenon. Anderson, Bjork, and Bjork (1994) did a study to see if retrieval induced forgetting occurred in groups of categories. In the experiment the researchers gave people lists of categorized words and then asked them to retrieve some of the words from some of the categories. Then during another memory test, hit rates were lower for items that had not previously been retrieved but that came from a category from which other category members had been retrieved than for items that were in category groups from which no items had been retrieved. The researcher’s explanation was that activation of related words causes retrieval inhibition of words from the same semantic category.
        Macrae and Macleod (1999) ran three experiments, experiment 1 and 3 being

Diane Poulos 4

impression formation and experiment 2 as a mock examination. Experiment 1 was to test whether or not retrieval induced forgetting applies to issues and processing operations in social cognition. The experimenters informed the participants that their task was to form impressions of two men, John and Bill. Participants were to study index cards that had 10 traits about John and ten traits about Bill. Each trait set of cards had a subgroup that had five traits to form a practiced and unpracticed set of items. After the studying, a surprise test was given. The participant was shown a card with the targets name and a hint about his trait (the first two letters of the trait.) The results showed that recall of Rp+ items was significantly greater than Nrp items. However, recall of Rp- items was significantly lower than Nrp items. This demonstrated the consequences of retrieval practice such as retrieval induced forgetting.
Macrae and Macleod’s (1999) second experiment was to determine the possible memorial consequences of retrieval inhibition for perceivers. Participants were informed that they would be taking a mock geography test. They were to perform as well as possible on this test. Each participant was presented with 20 facts about two fictitious islands called Tok and Bilo. Like in experiment 1, each fact group contained a subgroup with 5 facts to create a practiced and unpracticed set of facts. After studying, the participants were to complete a retrieval practice phase in which they were presented with cards that probed their recollections of half the facts for one island. Each card had the name of the island and a hint about the fact (i.e. the first letter of an export or native language.) There was a filler task in which participants had to count backwards by 3 from

Diane Poulos 5

2,000 for 5 minutes before allowed to recall as much information about the islands as possible. Unlike the first experiment, there was a condition in which no retrieval practiced information was presented. Participants were to give facts about general geographical knowledge like the capitals of actual states. The recall of Rp+ items was significantly greater than the recall of Nrp items. The recall of Rp- items was significantly lower than that of Nrp items. The results also showed retrieval practice enhanced recall performance on practiced items from the previously practiced island. However, the retrieval practice also impaired recall on unpracticed items relative from one island to another. Macrae and Macleod (1999) concluded that there were in fact consequences of retrieval inhibition for perceivers.
In Macrae and Macleod’s (1999) experiment three, the objective was to determine whether retrieval induced forgetting is moderated by the frequency of retrieval practice. This experiment was a replication of experiment 1 discussed earlier only here participants were randomly assigned to one of three conditions where they were presented with the trait items either once, three times or six times, and Macrae and Macleod found that frequency of repeated retrieval practice did not increase the magnitude of temporary forgetting.
Other research has been done to determine how retrieval induced forgetting occurs. Researchers had used category cues in the recognition phase to study retrieval induced forgetting and the studies show that the phenomenon occurs. Veling and Knippenberg (2004) wanted to see whether or not retrieval induced forgetting occurs

Diane Poulos 6

without using categories as cues. Veling and Knippenberg (2004) ran 2 experiments. The first experiment was to show that retrieval induced forgetting could be obtained using recognition latencies without using categories as cues in the recognition phase. The expected results were that participants would react slower to the rp- items than to either the rp+ or nrp items. Three categories (fruits, sports and animals) were constructed as stimuli for this experiment. Participants were to study word pairs that belonged to one of the three categories. For example, the categories were presented as SPORTS-football. Then after the practice phase, the participants were to recall words by looking at cues. Then after the retrieval practice phase, participants were shown words, either new or used in the practice phase, and asked to indicate as quickly as possibly if the word was on the study page. Though the average recall rate was 99.1%, participants were slower to recognize rp- items relative to nrp items. Also, rp- items had longer recognition latencies than nrp items. These results occurred without presenting the categories as cues in the recognition phase and supported their idea of the cue independent explanation of retrieval induced forgetting. After this the experimenters still wanted to test their theory by eliminating the possibility of spontaneous category activation. Experiment 2 was designed to test whether retrieval induced forgetting could be found in an implicit test of memory. Experiment 2 used 12 categories. The procedure was exactly the same as experiment one except this time the category names were not on the instruction screen of the study phase and the category example pairs were shown in two blocks of nine presentations. Showing 2 category example pairs as fillers before, between and after the 2

Diane Poulos 7

This is a preview of the whole essay

Diane Poulos 7

experimental blocks prevented primacy and recency effects. After the study phase, participants were asked to identify as quickly as possible whether or not a string of letters was a word. There were 18 words that were used in the examples of the study phase and 18 words that were new. Then there were 36 non words. The results showed that people were slower to identify rp- items as words than both rp+ and nrp items. The experimenters concluded that retrieval induced forgetting does in fact occur in implicit memory tests.
Lists of facts and concrete information were used in previous experiments. We chose to use coherent connected text such as short stories in narrative form for our experiment to see if this kind of text works the same as strings of facts with retrieval induced forgetting. In this experiment, we set out to determine whether retrieval induced forgetting occurred in reading short stories. Our purpose is to see if text works the same as strings of facts with retrieval induced forgetting.
We set up the experiment to have participants read short stories and then report on information from the text. The participant was allowed to read each story then asked to recall information that was in the text of both stories to see if retrieval induced forgetting had occurred. Like Macrae and Macleod’s (1999) second experiment, the control group had a section where a short general knowledge form was to be filled out.
Previous studies used concrete facts and our study uses coherent connected text. If text works the same way as strings of facts, we will get the same effect as those tests that use episodic questions. If text does not work the same as strings of facts than we wont get the effects.

Diane Poulos 8

Method
Design
The first part is the experimental group. This study has one within-subjects experimental variable with three levels and a control group. The experimental variable was retrieval practice. The three levels were Retrieval practice+ or Rp+, Retrieval practice- or Rp-, and no retrieval practice or NRP. The control group was never any retrieval practice, or NARP.
Participants
There were 104 people who participated in this study. One test was discarded due to the fact that the person had already read the stories at an earlier point making the number of participants 103. All participants were friends or acquaintances of psychology students at a small north eastern college. The participants agreed to participate by request of a psychology major that needed participants for the experiment. Sixty-nine individuals made up of 34 females, 31 males and 4 people choosing to not give there genders were in the experimental group. Their mean age was 22.1. The control condition had thirty-four participants made up of 26 females and 8 males. Their mean age was also 22.1.

Materials

In this study we used 2 Russian Folk Tales as our coherent narrative text. See Appendix A. One was called The Just Reward, and the other was called Masha and the

Diane Poulos 9

Bear. Both have been used in other experiments. There were questions developed about each story that were constructed by the experimenter. See Appendix B.

The experimental packet started with a page to fill out age, year in college, sex, and if English was the participants first language. The next page had the instructions, stating to read the instructions for each task carefully, to follow the instructions concerning turning pages, and a statement explaining that this is so all participants spend the same amount of time on each task. Then the instructions for the first task were listed. This stated that the participant would be reading two passages, and would later be taking a test on the contents of each. Participants were informed to read carefully because of this. It was also explained to read each passage only once thoroughly and then to turn the page. Page 2 had either The Just Reward or Masha and the Bear on it. Page 3 had whichever passage was not on the page two. The fourth page had 5 questions on 1 of the passages; this was the retrieval practice page. Page 5 was the wait page. This page was when the experimenter did the filler task with the participant. Page 6 had 20 questions, 10 about each passage.
There were eight counterbalanced experimental versions of this packet. Four versions of these packets had The Just Reward first, and the other 4 had Masha and the Bear first. With the 4 versions of the packets having The Just Reward first, 2 of them had retrieval practice questions about The Just Reward, while the other 2 had retrieval practice questions about Masha and the Bear. With the 4 versions of the packets with

Diane Poulos 10

Masha and the Bear first, 2 of them had retrieval practice questions about Masha and the Bear, with the other 2 retrieval practice questions about The Just Reward. The control packet also had the first instruction page, then both Russian Folk Tales. Then there were 5 general knowledge questions instead of the retrieval practice page. This is the difference between the experimental and control packet. After this was the wait page. Then, on the next page there were 20 questions, 10 about each passage. The control packets were counterbalanced. Half of them had1 passage first, and the other half had the other passage first.

Procedure
Participants were alone in a room sitting at a desk. Each test took about 15 minutes to complete. First, participants were handed an informed consent form and asked to read it and sign it. Then packets were randomly selected for each participant and handed out. The participant then filled out the information on the first page, were asked if they had any questions, and preceded with the instruction of the experimenter. The instructions were that there was going to be a test. After the participant read both tales, the experimenter noted the time of their arrival at the retrieval practice page, giving them about a minute to complete the 5 questions. When subjects got to the “wait” page they were instructed to count backwards by threes from 100 to 0 with the instructor. This took about 3-4 minutes. After the waiting exercise, participants were instructed to finish the last page. This took about 5 minutes. The experimenter collected the test and thanked the participant for their participation.

Diane Poulos 11

Results
We first counted how many of the 5 items on the retrieval practice page were right for each person in the experimental group. We categorized each item as rppluse (5), rpminus (5), or nrp (10) for the 20 items on the test page. Then we counted how many of each category they got right. There was no retrieval practice page for the control group and all 20 questions were of the same status. We called them narp (never any retrieval practice). We counted how many of the 20 they got right by using the criterion of what was meant by their answer not necessarily exact wording. Then we wrote on the top of the question page how many of each group was correctly answered for each packet.

We got the means of the scores on all the tests. We then converted the means into percentages. The mean percentage of items correct in each category was rppage with 59.7 and a standard deviation of 24.4. The prpplus with 60.6 and a standard deviation of 25. The prpminus was 49.8 with a standard deviation of 26.9.The pnrp was 52.9 with a standard deviation of 25, and pnarp was 50.9 with a standard deviation of 18.7. See table 1.

Discussion
Our experiment’s main goal was to see if narrative coherent connected text worked the same way as strings of facts. We ran the experiment to see if the no retrieval practice questions were answered with more accuracy than the retrieval practice- group. The results for the experimental condition showed no retrieval practice items were more easily recalled than the retrieval practice – items, with the results for the control condition as our baseline. We found that no retrieval practice was in fact greater than retrieval

Diane Poulos 12

practice – concerning narrative coherent connected text; however, the difference was not significant. We therefore did not find retrieval induced forgetting in coherent narrative text.

If someone is studying a story for a test these results mean that retrieval induced forgetting will not occur so they can study all or some of the information without having to worry about the consequences of retrieval induced forgetting. Because we used narrative text and textbooks use expository text, we do not know if students in school apply to this study. We would have to conduct a study where the participants read expository text and are given study questions concerning half of one of the topics. This study would also have to have a control group that did not get any retrieval practice.

In our study we tested to see if narrative text worked the same way as discrete facts, in that it produces retrieval induced forgetting, we found that they do not result in the same way and retrieval induced forgetting does not occur in narrative text.

Diane Poulos 13

References
Anderson, M.C., Bjork, R.A., & Bjork, E.L. (1994). Mechanisms of inhibition in long-term memory: A new taxonomy. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20, 1063-1087.
Macrae, N.C., & MacLeod, M.D. (1999). On recollections lost: When practice makes imperfect. Journal of Personality and Social Psychology, 77, 463-473.
Veling, H. & Knippenberg, A.V. (2004). Remembering can cause inhibition: Retrieval -induced inhibition as cue independent process. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30, 315-318.

Diane Poulos 14 Appendix A

The Just Reward
The king of a certain country lost his ring while riding through his hunting grounds. He had his sheriff ride throughout the kingdom with an announcement promising that whoever might find and return his ring would receive a large reward of money. A simple private was lucky enough to find it. “What shall I do?” thought the soldier. “If I report my find at regimental headquarters, the whole affair will be referred to my superiors, each in his turn, from the sergeant to the lieutenant, from the lieutenant to the captain, from the captain to the colonel, and from the colonel to the general; there will never be an end to it. I would rather not go through all of that.”
He went straight to the palace. The officer on guard, who was doing paperwork, looked up and asked him: “What do you want here?” “I have found the king’s ring,” said the soldier. “Very well, brother! I will announce you, but only on condition that I get half the reward that the king gives you.” The soldier thought to himself: “For once in my life I have had a piece of luck, and now I have to share it.” However, he answered the officer on guard: “Very well, I agree. But, you have to give me a note stating that half the reward is for you and half for me.” The officer did so.
The officer brought the soldier in and announced him to the king. The king praised the soldier for having found the ring. “Thank you, brave soldier!” he said. “I shall give you two thousand rubles.” But the soldier said : “No, Your Royal Majesty, that is not a soldier’s reward. A soldier’s reward is two hundred lashes.” “What a fool you are,“ said

Diane Poulos 15

the king, and ordered the whips to be brought in.
While the soldier was undressing and unbuttoning his tunic, the note dropped on the floor. “What paper is that?” asked the king. “Your majesty, this is a note stating that only half the reward is for me, and that the other half must go to the officer on guard.” The king laughed, called the officer on guard, and ordered that he be given a hundred lashes. The order was carried out, and when it was time to count the last ten lashes, the soldier drew near to the king and said: “Your majesty, since he is so greedy, I will give the other half of the reward to him too.” “How kind you are!” said the king, and ordered that the officer on guard be given the second hundred lashes. After this reward the officer could hardly crawl home. As for the soldier, the king gave him an honorable discharge from the service and presented him with three thousand rubles.

Masha and the Bear
Once upon a time there lived a Grandfather and Grandmother, and with them was their little granddaughter Masha. Masha wanted to go into the forest to gather mushrooms. “Go little granddaughter, only don’t get lost!” they said. Masha walked and walked about the forest and was lost. Suddenly she saw it, a little wooden izba! The little girl went into the izba, but no one was there. “Who lives here? Where is the owner?” she called. But in that izba lived a big bear! In the evening he returned home, saw Masha and was so happy! “You will live with me now. You will stoke my stove and you will make my porridge. You will feel me porridge. I will never let you go!”

Diane Poulos 16

Masha began living in the bear’s izba. In the morning the bear went to gather food in the forest. “You don’t go anywhere!” he said. “And if you run away, I will catch you, and eat you up!” Masha began to think, how could she get away from the bear? The forest was all around, and she didn’t know the way home. Masha thought and thought and thought some more.
She baked some peach pies and put them in a great big basket. “Bear, please let me have a day off so I can go into the village,” she said. “I want to take these pies to Grandmother and Grandfather.” “No, I will take the pies myself,” said the bear. “On the way, don’t open the basket,” warned Masha. “Don’t eat the pies. I will be in a tall tree in the forest and I will see everything you do!” “Alright,” answered the bear. Masha laid the pies in the basket. “Go outside to look to see if it’s raining.” she told him. The bear went outside, and Masha climbed into the basket. She put the big plate with the pies on her head. The bear came back in, picked up the basket and went to the village.
The bear walked and walked through the forest and got tired. “I will sit on this stump and eat the pies” he said. “I SEE YOU I SEE YOU! Don’t sit on the stump, don’t eat the pies. Take them to Grandmother and Grandfather!” shouted Masha. The bear looked up at the trees. Masha is sitting high in the tree, she sees everything, he thought.
The bear walked into the village. “Open up quickly, I have brought pies to you from Masha!” he called. The dog saw the bear and ran straight at him! The bear was frightened, dropped the basket, and there was Masha! Grandfather and Grandmother rejoiced. They began hugging and kissing Masha, calling her a clever girl!

Diane Poulos 17
Appendix B
Test questions used.
The Just Reward
1. Where was the king riding when he lost his ring?
____________________________________________________________
2. How did the king announce that he was giving a reward?
____________________________________________________________
3.What rank was the soldier who found the ring?
____________________________________________________________
4.Why didn’t the soldier want to report his find at regimental headquarters?
____________________________________________________________
5. Instead of going to regimental headquarters, where did the soldier go?
____________________________________________________________
6. What did the soldier say to the king that a soldier’s reward is?
____________________________________________________________
7. What did the king order to be brought in to give the “reward?”
____________________________________________________________
8. How did the note drop on the floor?
____________________________________________________________
9. When did the soldier suggest that the officer get the second half, too?
____________________________________________________________
10. How much money did the king finally give the soldier?

Diane Poulos 18
Masha and the Bear
1. What did Masha want to gather in the forest?
____________________________________________________________
2. What was the little izba made from?
____________________________________________________________
3. What did the bear tell Masha she would do for him?
____________________________________________________________
4. When did the bear go to gather food in the forest?
____________________________________________________________
5. What kind of pies did Masha bake?
____________________________________________________________
6. After Masha placed the pies in the basket, what did she tell the bear to go outside and do?
____________________________________________________________
7. What did Masha put on her head?
____________________________________________________________
8. Where did the bear want to sit to eat the pies?
____________________________________________________________
9. What did the dog do when he saw the bear?
____________________________________________________________
10. What did Grandfather and Grandmother call Masha?
____________________________________________________________

Diane Poulos 19

Table 1
Means and standard deviation percentages of items correct for all 3 conditions.

_______________________________________________________________
Actual Retrieval Retrieval Practice Status
__________________________

Condition Practice Task Rp+ Rp- NRP
_________________________________________________________________

Experimental 59.7(24.4) 60.6(25.0) 49.9(27.0) 52.9(25.0)
Control ------ ----- ---- 50.9(18.7)
__________________________________________________________________

Note: Standard deviations in parentheses.