Our life’s memories are filed away in our long-term memory and our mind categories these into clusters, which we call schema. These schema then affect how we process new information and remember old information. At least, these are the claims of schema theory. In this post we’ll look at how these claims are supported by a classic study.
Because of their effects on reading comprehension and memory, most schema research in the 1960s and 1970s was based on school-like procedures. For instance, studies would look at how schemas’ affect classroom-type learning like remembering facts and reading comprehension. Brewer and Treyens wanted to see how schemas would affect memory in a memory in a more “ecologically valid situation” – an office.
Previous studies had shown that saliency (how noticeable something is) and schema expectancy (schema consistent information) have an influence on memory: people are more likely to remember schema consistent information and salient information. The effect of schema-congruent (i.e. schema-consistent) information is known as the congruency subsequent memory effect.
The following study tested these effects of schema on episodic memory.
This study was recommended for the IB Psychology IA in an old textbook. But it’s fraught with dangers and is a nightmare to conduct. I do not recommend this study to be replicated for the IA.
Role of Schemata in Memory for Places
(Brewer and Treyens, 1981)
The aim of this experiment was to see how schema affects memory in an everyday life situation, like the memory of places. The researchers proposed five hypotheses about how schemas are used in such memories:
- Framework hypothesis: schemas make it easy to comprehend new information by connecting it to our existing schemas.
- Encoding hypothesis: schemas affect what information is the focus of our attention and what is not (e.g. how long we look at something), thus affecting the encoding of memory from short-term to long-term memory.
- Retrieval hypothesis: schemas influence memory in the process of retrieving the information from our long-term memory to our short-term memory.
- Integration hypothesis: schemas help us combine (integrate) new episodic memory with existing schemas (episodic memories are memories of events).
- Communication hypothesis: schemas influence what information is communicated during recall (e.g. what we want to talk about remember and what we decide isn’t important).
A room was created to look like a regular office. However, there were two anomalies:
- Some random materials that you wouldn’t expect to see were included, like a skull, a frisbee and a wine bottle)
- Some things you would expect to see in an office weren’t there, like books or a lamp.
Before the experimental phase, a group of volunteers ranked the items based on their saliency and their schema expectancy.
In the next step of the study, the researchers studied 86 participants one-by-one. They were led to believe they were about to take part in an experiment but first had to wait in the office while someone else finished. They were left alone in the office for 35 seconds. They didn’t expect that they would later have to remember what was in the office.
Afterwards they were given a memory test. Some participants were asked to write down what they remembered, others to draw and some to write down and also recall from a list of possible answers.
A positive correlation was found between schema expectancy and memory – participants were more likely to remember items if they were regular office items (e.g. a desk, a pencil).
Participants were more likely to have false memories of items not actually in the office but were rated high on schema expectancy (e.g. books, a lamp, a window, etc.).
There was almost no correlation between salience and memory.
The findings as they relate to the hypotheses about how schemas influence memory can be summarized as follows:
- Retrieval: The results suggest “…part of the general positive correlation of schema expectancy and recall was due to office-schema information serving as a retrieval mechanism in recall.” In other words, our schemas make it easier to retrieve information from our long-term memory.
- Integration: The false memories of high-schema-relevant objects (e.g. books) supports this hypothesis as there was no episodic memory of these items (as they weren’t in the room) so the false memory “…must be due to old schema knowledge” which is integrated (combined) with the episodic memory of being in the room.
- Communication hypothesis: In the written recall, participants generally didn’t give additional details of items if they were “normal” (i.e. schema-consistent) but they would if they were peculiar. For example, they would not say “a glass wine bottle” but would just say “a wine bottle,” but the chair was made of plastic so they were more likely to say “a plastic chair” (since most office-chairs aren’t made of plastic). This suggests they use schema in how they communicate information based on their understand of what others would expect in an office (i.e. their “office schema”).
- Encoding hypothesis: According to prior research, memory for non-schema related items (e.g. the skull) should be better since they will be the focus of attention more than schema-consistent information (e.g. a typewriter in 1981 or a computer today). The results showed that schema-consistent information was remembered more than schema-inconsistent information. Thus, not providing support for the encoding hypothesis.
- Framework hypothesis: The better memory for schema consistent information could support this hypothesis, but it could also support the retrieval and integration hypotheses, too. Thus, the researchers concede that “…the results of this experiment allowed no unique test of the hypothesis that schemata can serve as a framework which preserves schema-relevant episodic information.”
Conclusions and Applications:
The researchers conclude that “the results of this experiment suggest that place schemata play an important and complex role in place memory.” We’re more likely to remember information that is consistent with our memories and less likely to remember information that is inconsistent.
- Is it actually a false memory of books, or just a false answer? In other words, how can we tell for sure they actually remember there being books in the office and they’re not just writing down things they would expect because they want to get the “right” answer?
- The author’s claim that this office is “an ecologically valid” situation. To what extent do you agree with this claim? Are there reasons to suspect the same results might not occur in other situations?
- This study shows schemas affect episodic memory. What about other memories, like semantic memory (memory of information) or procedural memory (memory of how to do things)?
- This study is over 40 years old. Do you think we’d expect the same results today? Why/why not?
Travis Dixon is an IB Psychology teacher, author, workshop leader, examiner and IA moderator.