Memory of syntax and semantics

October 2, 2011

Jean Aitchison’s The Articulate Mammal describes and evaluates many interesting psycholinguistic experiments, one of which I want to draw attention to here:

a number of psychologists have found that all memory of syntax and vocabulary normally fades very fast indeed, unless subjects are specifically told that they will be asked to recall the sentence. Memory for syntax of any kind is near to chance approximately half a minute after a sentence has been spoken (Sachs 1967). In normal circumstances, it seems, people remember only the gist of what has been said, and they often confuse this with a number of extra beliefs and expectations about the topic under discussion (Fillenbaum 1973).

Jacqueline Strunk Sachs, speaking subsequently (PDF) about her experiment, said it showed that we forget “the specific wording of an utterance . . . within seconds”, though we might retain its meaning for a very long time.

Her 1967 paper (“Recognition memory for syntactic and semantic aspects of connected discourse”) was based on her doctoral dissertation and can be downloaded here.* It’s a short and clear account of a smartly designed study, well worth reading if you’re into this sort of thing and you don’t mind the Chomskyan terminology.

The abstract concludes:

The results suggest that the original form of the sentence is stored only for the short time necessary for comprehension to occur. When a semantic interpretation has been made, the meaning is stored. Thus the memory of the meaning is not dependent on memory of the original form of the sentence.

I imagine this rings true for most people. What say you? Have you noticed the rapid divergence between memory of what is said and memory of exactly how it is said?

* Typo fans will enjoy SpringerLink’s mangled paper title, “Recopition memory”, and the suggestion that some of the data were presented at a meeting held in April, 1066.