Failures to Replicate Hyper-Retrieval-Induced Forgetting in Arithmetic Memory

Abstract

Campbell and Phenix (2009) observed retrieval-induced forgetting (RIF) (slower response time) for simple addition facts (e.g., 3 + 4) immediately following 40 retrieval-practice blocks of their multiplication counterparts (3 × 4 = ?). A subsequent single retrieval of the previously unpracticed multiplication problems, however, produced an RIF effect about twice as large for their addition counterparts. Thus, a single retrieval of a multiplication fact appeared to produce much larger RIF of the addition counterpart than did many multiplication retrieval-practice trials. In subsequent similar studies, however, this hyper-RIF effect was not observed (e.g., Campbell & Thompson, 2012). The current studies further investigated hyper-RIF in arithmetic. In Chapter 2 (Experiment 1), composition of operands (unique vs. common) and amount of multiplication practice (6 vs. 20 repetitions of each problem) were manipulated. Participants solved multiplication problems (4 × 7 = ?) and then were tested on their memory for the addition counterparts (4 + 7 = ?) and control additions. Chapter 3 (Experiment 2) attempted an exact replication of Campbell and Phenix. In both studies, hyper-RIF was not observed. The results confirm the basic RIF effect of multiplication retrieval practice on addition counterparts, but cast doubt on the on the reality of the hyper-RIF effect observed by Campbell and Phenix. It is concluded that the hyper-RIF effect reported by Campbell and Phenix is an elusive or non-existent phenomenon; consequently, it cannot at this time be considered an important result in the RIF literature.