Aging and the inhibition of competing hypotheses during visual word identification: evidence from the progressive demasking task

Abstract

Two experiments used the progressive demasking (PD) task to examine age differences in the ability to inhibit higher frequency competitors during the process of identifying a visually degraded word. In Experiment 1, older adults exhibited a larger inhibitory neighborhood frequency effect (i.e., slower identification of words with many higher frequency competitors) than younger adults, but additional analyses indicated that this difference could be explained by general slowing rather than a deficit in inhibitory abilities. In Experiment 2, a primed version of the PD task was used to promote hypothesis testing by semantically priming the target word (e.g., cry–weep) or a higher frequency competitor of the target (e.g, day–weep) prior to the onset of the demasking sequence. Although older adults were more likely to make identification errors consistent with an inhibitory deficit (e.g., identifying weep as week), these errors were infrequent overall and there was no corresponding evidence of a larger interference effect in the older adults’ identification latencies. Taken together, performance in these two tasks provides little evidence of reduced inhibitory functioning in older adults. The implications for the inhibitory deficit hypothesis of cognitive aging and directions for future are discussed.

Keywords:

• aging
• inhibition
• progressive demasking task
• semantic priming
• hypothesis testing

Acknowledgments

We thank three anonymous reviewers for their excellent feedback and suggestions. We also thank Cindy Lahar for her support in the early stages of this research.

Notes

^1. The larger groups in Experiment 2 (N = 30 vs. N = 20 in Experiment 1) led to an increase in the statistical power to detect group differences on some of the measures and, as a result, a few discrepancies between the experiments. For the backward span test, the effect size was not large (d =.59), and the power to detect this difference was 61% in Experiment 2 versus 44% in Experiment 1. Similarly, for years of education (d =.81), power was 87% in Experiment 2 versus 70% in Experiment 1.