False recall serial position effects

Abstract

A series of 5 experiments investigated whether false memory in associated word lists present with serial position effects (SPE) and how any such effects behave in response to manipulations of true recall SPE. Recall for a series of events is typified by SPE such that items nearer the beginning, primacy effect, and end, recency effect, of a series are remembered better than middle items. Recall is also typified by the intrusion of falsely remembered information. Word-lists segmented into trimesters of either semantically (e.g., hot, snow, warm.../ bed, rest, awake.../ looking, lens, shatter...) or phonologically (e.g., code, called, fold.../ sweep, sleet, steep.../ class, grass, glad...) associated words produced false recall (e.g., cold, sleep, glass), allowing for the simultaneous investigation of SPE for true and false recall. Typical SPE for true recall were observed for each of the five Experiments. For immediate free recall, semantic false recall declined from early to late study trimesters whereas phonological false recall displayed a false primacy and recency effect similar to true recall SPE. Phonological false recall was significantly reduced when a 15 second distractor task was implemented during the retention interval. Dividing attention during study using a concurrent handwriting task reduced true recall whereas semantic false recall increased at primacy and phonological false recall increased at recency. This suggests distinct processes underlying the two forms of false recall. Dividing attention using an articulatory suppression task produced less true recall and less false recall than using concurrent handwriting. This research indicates that false recall SPE exist and that the semantic and phonological forms of false recall SPE are distinct. Current theories of false memory and of true recall SPE are considered.