Modulating the Focus of Attention for Spoken Words at Encoding Affects Frontoparietal Activation for Incidental Verbal Memory
Figure 1
Experimental design for studying incidental encoding and memory. During the Encoding scan [39], listeners were presented with a list of single nouns (animals or foods). Half of the words were recorded in a female voice, and half in a male voice, thus yielding multiple combinations of item and voice context. Throughout the experiment, listeners’ attention was focused on only one subset of words: animal words presented in the female voice (Target words). This was accomplished first by giving them explicit instructions to select and remember these words during the Encoding scan, and later, by instructing them to select these items by responding “yes” during the Memory scan. However, as predicted, some of the Nontarget words were also encoded unintentionally, with food words presented in the female voice carrying greater salience (the High-Attention condition) than food words presented in the male voice (the Low-Attention condition). During the Memory scan, participants were presented once again with both High- and Low-Attention food words, along with novel food words (NEW) as a control category representing the baseline attention condition. Following the Memory scan, a Surprise memory test was used to confirm the extent to which High-Attention, Low-Attention, and NEW food words were successfully encoded. The chart at right shows the group-averaged memory scores (; means ± SEM) for these three word categories, along with ten foils to measure false alarm rate. Bars capped by different letters are significantly different at (two-tailed ANOVA).
