Recognition for word triplets in complex networks.

Abstract

To explore the strategy use in associative recognition, we constructed two word-triplet lists to represent the information networks in the real world featured by repetition, co-occurrence, and change. We predicted that word-triplet recognition would depend upon the co-occurrence of repeated context words and nonrepeated unique words within a list, and the word change between two lists. In Experiment 1, we compared the probability of accepting the triplet test trials that consisted of: (a) different numbers of word links between context words and unique words, and (b) context words from same or different lists, and we found that recognition judgments only relied on the retrieval of word links. In the follow-up experiments, we increased participants' awareness of list-membership cues by explicitly informing them of the word change between lists prior to triplet encoding (Experiment 2), and by using self-generated context words from two lifetime periods (Experiment 3). The results suggested that participants might use a strategy based on both the retrieval of word links and list-membership cues, but only if they perceived the between-list word change during encoding. The present research provides new evidence for Transition Theory using the approach of word-triplet recognition. (PsycInfo Database Record (c) 2022 APA, all rights reserved).