The sample was made up of 39 introductory educational psychology students. The story (see below) had 72 discreet ideas - 15 that were related to burglary and 13 that were related to purchasing a home. The participants were randomly allocated to one of two conditions - the "robber condition" or the "home buyer condition."

They were given two minutes to read the story as a prospective burglar or a prospective home buyer. Then they were given 12 minutes to take an 84 item vocabulary test.  This was both a distractor task and control for their language proficiency. They were then asked to "write down as much of the exact story as you can." It was emphasized that they needed to write down every bit of the story that they could remember.

When they finished writing, they were given five minutes to do a spatial puzzle test. Then they were asked to recall the story a second time.  Half were told to do some from the same perspective that they used in the first recall; half were told to use the other perspective.

The researchers found that:

• Burglar information was better recalled than homebuyer information.  This may be because students do not have a well-developed homebuyer schema. In other words, it is likely that the schema influenced encoding.
• The group that had the burglar perspective recalled more burglar information and the group that had the homebuyer perspective recalled more homebuyer information. In other words, it is likely that the schema influenced retrieval.
• Participants who changed their perspective recalled an additional 7.1% of the information relevant to their new perspective. The group that did not change perspective recalled 2.9% less information relevant to their perspective.

Borland et al's (1987) replication of Anderson and Pitchert's (1978) study used a sample of 214 introductory psychology students at the University of Melbourne. They used an extended version of the original text.  It was 760 words and had 27 ideas related to burglary, 26 relevant to homebuying, and 30 "other."

Like in the original study, there were four conditions:

• Read as burglar; recall as burglar; recall again as burglar.
• Read as burglar; recall as burglar; recall again as homebuyer.
• Read as homebuyer; recall as homebuyer; recall again as homebuyer.
• Read as homebuyer; recall as homebuyer; recall again as burglar.

Instead of giving them the vocabulary test, they gave them a distractor task of asking them to provide antonyms (opposites) for 12 words - several of which had no clear opposite.  They were given 15 minutes to recall the story.  Then instructions were read for the second half of the study - and half were told to switch their perspective.  They were given 8 more minutes to try again to recall the story.

The researchers found that there was a slight (but significant) tendency for homebuyer information to be recalled less well than the burglar information.

In the first recall task, they found that schema appeared to assist with encoding and recall.

Note that there was no significant difference in the mean number of ideas recalled in the two conditions.  This indicates that, most likely, participant variability with regard to memory skills did not play a significant role.

In the second recall task, the researchers found that over there was a 10.8% increase in the number of burglar ideas recalled, a 15.5% increase in homebuyer ideas, and an 8.7% increase in neutral ideas.  However, the increase was significantly greater when the participant was asked to change their perspective than when they were asked to redo the task from the same perspective.

The experiment is highly controlled and rather artificial.  This means that it has high internal validity and low ecological validity.

It is possible that the participants understood the hypothesis of the study and engaged in expectancy effect - that is, it is possible that when asked to redo the task from the other perspective, they did not write down details from the first recall of the story because they did not think that they were relevant.  This means we do not know if they were forgotten or simply excluded.

Order effects could also play a role. A greater period of time between the two recall conditions may have led to different findings.  In addition, fatigue effects may have played a role in the amount of detail recalled.

It is assumed that the homebuyer details were recalled less frequently because students do not have the schema for home buying as they do for burglaries.  It could be, however, that the story was written in such a way that the burglary ideas were more engaging than the homebuyer details - and this could be the reason for the higher recall rates.