Differential Effects of Association and Semantics on Priming and Memory Judgments.

Semantic memory is the storage of world knowledge or facts, while associative memory contains information about how words are related in context of speech or writing (Tulving, 1993; Nelson, McEvoy, & Dennis, 2000). For example, the word pair CAT-DOG is semantically related because cats and dogs have many of the same features, such as legs, tails, fur, house pets, etc. CAT-DOG is also associatively related because the words appear together frequently in text such as “it’s raining cats and dogs”. Research examined to see if these two types of memory are separable or if all context and dictionary knowledge are contained in one memory system (Lucas, 2000). Currently, results are mixed on the separability of these two memory systems, where effects are seemingly dependent on stimuli used. Previous studies had difficulty controlling for both relationships for word pairs, meaning that word pairs were dually related despite claims for single relationships. Several studies were designed using large semantic and associative word norm databases (Maki, McKinley, & Thompson, 2004; Nelson, McEvoy, Schreiber, 2004). These databases made it possible to create separate word lists; so only one relationship (semantic or associative) was present. From there, priming for both memory types was tested using a rapid serial visual presentation task (RSVP) and judgment task. In the RSVP task, participants watched a very fast presentation of symbols, which were either distractors or target words. Participants were required to name the target word they saw among the distractors. Priming occurred when target words that were related were named more than unrelated target words. In a judgment task, participants were asked to read two words and rate how much they thought the words belonged together based on their feature overlap (how many features they share) or their associative relationship (how many people out of 100 would put them together). Judgments were part of the related word pair for the RSVP task to create priming. Priming for both relationships was found, which indicated that these two systems were separate. Judgments showed that people could separate the two memory systems when making judgments. Associative judgments were processed by associative information but not semantics, while semantic judgments required both associative and semantic information. However, attentional differences in judgments found previously did not transfer to priming results (Buchanan, Maki, & Patton, 2007). These studies, including priming tasks, showed that information appeared to be readily accessible (including semantic information) but in a different order of processing. Associative information appears to be processed earlier at a word or lexical level while semantic information is processed later and in a separate store. From there, judgment processes might occur even later than information processing used for priming.