FREE RECALL AND RECOGNITION MEMORY ESTIMATES FOR 532 CONCRETE NOUNS

Abstract

THE SUCCESSFUL RETRIEVAL OF WORDS IS LIKELY TO BE INTRICATELY LINKED TO THEIR LEXICAL AND SEMANTIC PROPERTIES. AS SUCH, VARIATIONS IN THESE PROPERTIES DETERMINE HOW THE WORD IS ENCODED, STORED, AND RETRIEVED. WE COLLECTED BOTH RECALL AND RECOGNITION MEMORY ESTIMATES FOR 532 CONCRETE NOUNS (MCRAE, CREE, SEIDENBERG, & MCNORGAN, 2005) USING THE MEGASTUDY APPROACH AND DEMONSTRATED THE VALIDITY OF THE DATA. COLLECTING MEMORY ESTIMATES USING THIS APPROACH ALLOWS RESEARCHERS TO MOVE AWAY FROM THE CONSTRAINTS OF FACTORIAL DESIGNS. THIS RESOURCE COULD BE USED TO CARRY OUT MORE FINE-GRAINED INVESTIGATIONS ON LEXICAL-SEMANTIC INFLUENCES ON MEMORY OR TO TEST NEW HYPOTHESES. TWO USES OF THIS DATASET WERE ILLUSTRATED. FIRST, WE REGRESSED RECALL AND RECOGNITION PERFORMANCE ON A NUMBER OF LEXICAL-SEMANTIC VARIABLES. IN FREE RECALL, THESE ACCOUNTED FOR 26.0% OF THE VARIANCE, WHILE IN RECOGNITION THEY ACCOUNTED FOR 32.3% OF THE VARIANCE IN HIT RATES, 16.5% OF THE VARIANCE IN FALSE ALARM RATES, AND 35.0%