ABSTRACT
Background: Previous studies on perceptual letter-matching have found that younger and older adults showed “fast-same” effects for response time and “false-different” effects for errors but the effects were more pronounced for older adults. According to the Noisy Operator Theory, internal noise in visual processing distorts “same” trials into appearing different whereas distortion for “different” trials does not affect performance. Older adults have a “noisier” representation of items within perceptual processing which can impact perceptual matching. However, EEG measures may provide a more direct measure of letter-matching decisions.
Methods: We measured the P300 event-related potential (ERP) amplitude, an index of familiarity in stimulus categorization, and behavioral measures (response time and accuracy) to assess letter-matching performance.
Results: Individuals responded faster to “same” trials than to “different” trials but were less accurate. Older adults showed similar P300 amplitudes across trial type whereas younger adults produced a larger amplitude for “same” than “different” trials, suggesting that older adults showed less familiarity for “same” trials than did younger adults – a prediction of the Noisy Operator Theory.
Conclusions: These ERP results are consistent with the Noisy Operator Theory – suggesting that an age-related increase in internal noise affected letter-matching performance.
Notes
1. Jackknifing involves finding peak amplitude using a grand averaging function. Jackknifing is sensitive to subtle amplitude changes that may normally be missed using a fixed window. Because younger and older adults have different peak latency windows, it is important to pinpoint the peak using a different time window. If this is done at an individual level, it could lead to misleading results (Ulrich & Miller, Citation2001). However, using jackknifing that inter-individual variability is greatly reduced as it is assessing a grand average wave form for each individual. This process finds precise fluctuations in amplitude and locate the latency of the wave’s peak. A potential drawback of this averaging is that it can “smear” the waveform, specifically after peak latency for bump-like components such as P300 and N2 (Miller, Ulrich, & Schwarz, Citation2009). This limitation is minimized by jackknifing each group (younger and older adults) individually as there would be far less temporal “jitter” which would exacerbate grand-average smearing. The resulting benefit is a more precise measure of the onset peak for both groups. In situations where groups show natural differences in temporal processing, the cost of the flattening after the peak is outweighed by the benefit of finding the precise peak measurement.