Variability Across Subjects in Free Recall Versus Cued Recall (Talk, WPA 2021)
Abstract
Two common memory tasks are free recall (in which participants study a list of items then later attempt to recall those items) and paired-associate learning/cued recall (in which participants study a list of cue-target pairs then later attempt to recall targets when presented with cues). Typically, researchers studying memory using these tasks in young adults emphasize comparisons of mean performance on some measure (accuracy, confidence, latency) as a function of experimental conditions. Here, we focus on individual differences in performance variability in young adults performing free and cued recall tasks. In two experiments, undergraduate participants studied and were tested on lists of English words, either animal and object names (Experiment 1) or concrete nouns (Experiment 2). In these experiments, participants completed two study-test cycles of free and cued recall and answered post-test self-report questions about strategy use and subjective ease of remembering. In Experiment 1 (N = 101), we found incidentally that inter-individual variability in performance was greater for cued recall than for free recall. Experiment 2 (target N = 120) will serve as a preregistered direct test of the hypothesis that variability in cued recall performance is greater than variability in free recall performance. Though data collection for Experiment 2 is still ongoing, preliminary results support our hypothesis. To our knowledge, no published research has directly tested this hypothesis, and no major theory of memory (e.g., ACT-R, Anderson et al., 1998; SAM, Raaijmakers, 1980) makes explicit predictions about differences in performance variability across the memory tasks. Using both accuracy and post-test self-report data, we consider and evaluate several possible explanations for this finding, including the possibility that 1) cued recall promotes more varied strategy use among participants, 2) cued recall task instructions may be more likely to be interpreted different (e.g., “guessing” vs. “omitting”) and 3) cued recall draws on a greater number of underlying cognitive processes than free recall, and it is simply in virtue of these additional processes (themselves varying across individuals) that cued recall variability is higher. We argue that comparisons of variability across subjects can yield insights regarding the mechanisms underlying task performance.
