Two routes to memory benefits of guessing.

Abstract

Attempting to guess an answer to a memory question has repeatedly been shown to benefit memory for the answer as compared to merely reading what the answer is, even when the guess is incorrect. In this study, we investigate two potential explanations of this effect in a single experimental procedure. According to the semantic explanation, the benefits of guessing require a clear semantic relationship between the cue, the guess, and the target, and arise at the stage of guessing. The attentional explanation places the locus of the effect at the stage of feedback presentation and ignores the issue of semantic relatedness. To disentangle the two mechanisms, we used homograph cues with at least two different meanings (e.g., arms) and asked participants to either study an intact cue-target pair or guess a word related to each cue before being presented with the target. This allowed us to compare memory performance meaning of the cue as the later presented target (e.g., a guess legs for a pair arms-hug), versus a different meaning (e.g., weapons). In four experiments, we demonstrate that both the semantic and the attentional mechanism operate in the guessing task, but their roles are different: semantic relatedness supports memory for cue-to-target associations, while increased attention to feedback benefits memory for targets alone. We discuss these findings in the context of educational utility of errorful learning.

Selective effects of errorful generation on recognition memory: the role of motivation and surprise

“…The participants might not have had any great expectation that their guesses would be correct on Generate trials, and this could have reduced the potential to detect differences in surprise for guessed and non-guessed facts. It is possible that we would have obtained greater differences in surprise with an unsuccessful retrieval paradigm, where participants generate guesses for familiar materials that they have prior knowledge of (although see Zawadzka & Hanczakowski, 2018). While this would be an interesting avenue for future research, it would not, of course, shed light on the mechanisms that underlie errorful generation effects (which was the purpose of the present experiments).…”

“…Support for this view comes from studies showing that generating errors during learning sometimes impairs memory on subsequent tests (Baddeley & Wilson, 1994;Forlano & Hoffman, 1937;Kessels & De Haan, 2003;Squires, Hunkin, & Parkin, 1997). The more frequent finding in recent years, however, is that errors aid learning (Cyr & Anderson, 2015;Grimaldi & Karpicke, 2012;Hays, Kornell, & Bjork, 2013;Huelser & Metcalfe, 2012;Kane & Anderson, 1978;Knight, Ball, Brewer, DeWitt, & Marsh, 2012;Kornell, 2014;Kornell, Hays, & Bjork, 2009;Richland, Kornell, & Kao, 2009;Slamecka & Fevreiski, 1983;Tanaka, Miyatani, & Iwaki, 2019;Vaughn & Rawson, 2012;Yan, Yu, Garcia, & Bjork, 2014;Yang, Potts, & Shanks, 2017;Zawadzka & Hanczakowski, 2018). In these cases, failed tests are beneficial, and they can even be as beneficial as successful tests (Kornell, Jacobs Klein, & Rawson, 2015).…”

Learning from errors: students’ and instructors’ practices, attitudes, and beliefs

“…Such research might occur in authentic educational environments (e.g., Geller et al, 2017), and for the case of pretesting, address the specificity of learning that has repeatedly been observed in some experiments (e.g., James & Storm, 2019;cf. Pan, Lovelett, et al, 2019), the role of surprise (Butterfield & Metcalfe, 2001), the finding that generating errors that are semantically related to the correct answers yields more potent learning (Cyr & Anderson, 2018;Zawadzka & Hanczakowski, 2019), various types of pretest and criterial test questions (e.g., St. Hilaire et al, 2019), and the absence of pretesting effects for materials that lack strong cue-target associations (e.g., Grimaldi & Karpicke, 2012;cf.…”

Pretesting boosts recognition, but not cued recall, of targets from unrelated word pairs

“…Search set theory, for example, suggests that the process of generating guesses at encoding covertly activates a semantic network of related concepts, including the correct target. This prior activation is then suggested to improve the encoding of that target when it is subsequently revealed (Grimaldi & Karpicke, 2012;Hays, Kornell, & Bjork, 2013;Kornell et al, 2009;Zawadzka & Hanczakowski, 2018). Thus, search set theory emphasises the importance of the target already being partially activated when it is revealed.…”

“…Search set theory predicts this result because unrelated targets should not be activated in the search set for a cue, and so should not benefit from prior activation. More recently, Zawadzka and Hanczakowski (2018) used homograph cues that had two possible targets (e.g., arms-hug and arms-nuclear), although participants only ever saw one target. Pretesting was only beneficial when the participants' guesses were related to the target (e.g., guess shoulder for the pair arms-hug).…”