The latest experiment by a different group shows that this cuing during sleep can modify bad attitudes and habits. The test involved counter stereotype-training of certain biased attitudes during wakefulness, and investigators reactivated that counter-training during sleep by playing a sound cue that had been associated with the wakefulness training.
In the experiment, before a 90-minute nap 40 white males and females were trained to counter their existing gender and racial biases by counter-training. A formal surveyed allowed quantification of each person's level of gender or racial bias before and after counter-training. For example, one bias was that females are not good at math. Subjects were conditioned to have a more favorable attitude about women and math with counter-training that repeatedly associated female faces with science-related words. Similarly, racial bias toward blacks was countered by associating black faces with highly positive words. In each training situation, whenever the subject saw a pairing that was incompatible with their existing bias they pressed a "correct" button, which yielded a confirmatory sound tone that was unique for each bias condition. Subjects were immediately tested for their learning by showing a face (female or black) and the counter-training cue, whereupon they were to drag the appropriate bias-free face on to a screen with the positive word. For example, if the first test screen was that of a woman, accompanied by the sound cue, the subject dragged a woman's face onto a second screen that said "good at math." Results revealed that this conditioning worked: both kinds of bias were reduced immediately after counter-conditioning.
Then during the nap, as soon as EEG signs indicated the presence of deep sleep, the appropriate sound cue was played repeatedly to reactivate the prior learning. When subjects re-took the bias survey a week later, the social bias was reduced in the sound-cued group, but not in the control group that was trained without sound cues.
Experimenters noted that the long-term improvement of bias was associated with rapid-eye-movement (REM) (dream) sleep which often followed the deep sleep during early stages of the nap. That is, the beneficial effect was proportional to the amount of nap time spent in both slow-wave sleep and REM sleep, not either alone. It may be that memories are reactivated by cuing during deep (slow-wave) sleep, but that the actual cell-level storage of memory is provided by REM sleep.
Implications of this approach to enhancing learning and memory show a great deal of promise. Can it be used for enhancing learning in school? Can it be used in rehabilitation of addicts or criminals? But there is a dark side. Now might be a good time to re-read Huxley'sBrave New World wherein he actually described conditioning values in young children while they slept. Sleep is a state where people are mentally vulnerable and without conscious control over their thoughts. Malevolent people could impose this kind of conditioning and memory enhancement on others for nefarious purposes. These techniques may have valid social engineering applications, but they must be guided by ethical considerations.
