Disambiguating the Effect of Interpolated Testing on the Build-up of Proactive Interference During Learning

Document Type


Publication Date



Medicine and Health Sciences


Students frequently report difficulty learning information presented at the end of long study sequences. Prior research has shown that information presented earlier in a study sequence interferes with learning of information presented later in a study sequence, but that interpolating study with testing can serve to protect against such buildup of proactive interference (PI). Nonetheless, it remains debated whether interpolated testing improves encoding or retrieval of information presented in later portions of study. In the present study (N =96), students learned five interrelated lists of words. Students were always tested on the fifth list and then again during a final cumulative test. We used a 2 x 2 design to vary whether or not students were tested on the first four lists (i.e., testing versus control) and also the rate of presentation of individual words (4s versus 10s per word). We found that learning of the fifth list was impaired due to PI in the absence of previous testing (i.e., for lists 1 through 4) at both short and long presentation rates. However, during a final cumulative test that required participants to recall all study words, non-tested participants recalled words from the fifth list as well as tested participants, but only if they had studied words for 10s. Taken together, our results indicate the PI during learning arises largely due to constraints imposed by initial memory tests (i.e., during retrieval), and that an encoding manipulation can serve to overcome PI during later tests that do not impose similar constraints on retrieval.


Copyright 2017 all authors

This document is currently not available here.