Enhancing the production effect in memory: Singing, underlying mechanisms, and applications

Abstract

The production effect is the finding that memory performance is better for words that are produced (i.e., read aloud) compared to words that are not produced (i.e., read silently). This dissertation aimed to expand previous research by investigating: 1) whether alternate forms of vocal production have a greater impact on the production effect than reading aloud in a normal voice and 2) the possible mechanisms underlying any influence of alternate forms of vocal production on the production effect. In Experiments 1-3, we found evidence of a graded pattern of memory performance: Both reading items aloud loudly and singing items at study resulted in greater subsequent memory performance than did reading items aloud in a normal voice, with singing items at study resulting in even greater memory performance than reading aloud loudly. In Experiments 4 through 6, we examined possible mechanisms underlying the greater production effect for singing versus reading aloud. Our results provided evidence against three potential explanations for the greater production effect for singing versus reading aloud including a bizarreness explanation, differences in production duration, and differences in trace memory strength. Taken together, the findings from this dissertation provide evidence that alternative forms of vocal production, such as singing and reading aloud loudly, have a greater impact on memory performance than reading aloud in a normal voice. Our findings also provide strong support for a distinctiveness account of the production effect, emphasizing that the number and type of potential distinct elements available at encoding is likely associated with subsequent memory performance at test and consequently with the magnitude of the production effect (i.e., the greater the number and type of distinct elements available at encoding, the greater the magnitude of the production effect).