Schwartz, Bennett L.; Fisher, Ronald P.; Hebert, Kellye S. (1998). The Relation of Output Order and Commission Errors in Free Recall and Eyewitness Accounts. Memory, 6, 257-275.

-These studies explore the relation between relation output order and the likelihood of a commission error in free recall under both laboratory and eyewitness conditions.

-Output Order: refers to the order that information is reported.

-Commission Error: the reporting of events that did no occur.

Participants

-55 undergrads

Procedure

-all participants were told that they would be presented a list of 20 words on a computer program. After each word they were told to make judgements of learning (likelihood that they would be able to remember the later)

-each word was presented for 4 seconds after this the participants were prompted to make their learning judgements.

-after participants had completed the list; they were given a 5 minute arithmetic task.

-at the end of the arithmetic task, participants recalled as many words as they could, then typed a "?" to indicate they had completed the task.

-response were assigned into a quartile

Results

-participants reported a mean of 8.7 words out of the list of 20 with and 89.9 accuracy rate.

-serial positioning effect

-largest number of commission errors found in the fourth quartile. (Fig. 1)

Participants

-119 undergrads (however 21 failed to give descriptions of one or more of the targets, or gave descriptions that could not be coded), leaving 98 participants

Procedure

-participants watched a video of two men being robbed by a man and a woman.

-after viewing the video, the participants were given retrieval packs which contained instructions and 4 questions.

-questions were counterbalanced so that half the participants described the sequence of events first and the other half describe the criminal or truck first.

-response were assigned to a quartile.

Results

-participants made more responses to sequences of events questions then to criminal or truck questions.

-for the description questions, fewer errors were made in describing the truck; fewer errors were reported in the first quartile. (Fig. 2)

-for sequential questions, more errors were made in the second and third quartile. (Fig. 3)

Participants

-68 undergrads (one was excluded for not following directions)

Procedure

-basically the same as exp. 2 but with different content in the video.

-the video depicted a female psychologist interviewing a male patient.

-all participants described the man first, then the lady

Results

-no difference in amount of correct or incorrect information for the man and woman.

-more commission errors were made in the fourth quartile. (Fig 4)

-the authors offer two possible explanations for their findings. 1) response inhibition 2) relaxing the accuracy criterion.

-response inhibition: the act of searching through memory reduces the effectiveness of subsequent memory searches.

-relaxing the accuracy criterion: as one progresses further into the output, the accuracy criterion for producing a response is relaxed.

-Uses for this study include helping police determine the accuracy of items in an eyewitness's report, therapist could find this information useful in determining accurate and inaccurate information in long detailed descriptions of autobiographical experiences, and journalist and historians could us it when piecing together mixed recollections to formulate historical accurate events.