Abstract
It is often assumed that implicit learning of skills based on predictive relationships proceeds independently of awareness. To test this idea, four groups of subjects played a game in which a fast-moving “demon” made a brief appearance at the bottom of the computer screen, then disappeared behind a V-shaped occluder, and finally re-appeared briefly on either the upper-left or upper-right quadrant of the screen. Points were scored by clicking on the demon during the final reappearance phase. Demons differed in several visible characteristics including color, horn height and eye size. For some subjects, horn height perfectly predicted which side the demon would reappear on. For subjects not told the rule, the subset who demonstrated at the end of the experiment that they had spontaneously discovered the rule showed strong evidence of exploiting it by anticipating the demon's arrival and laying in wait for it. Those who could not verbalize the rule performed no better than a control group for whom the demons moved unpredictably. The implications of this tight linkage between conscious awareness and implicit skill learning are discussed.
Highlights
A critically important aspect of skill acquisition is learning to take advantage of the various predictive relationships that exist within the relevant domain
The question posed in the present article is: do people learn to exploit predictive relationships without showing any conscious awareness of the relationship that they are exploiting?
A small proportion of subjects were unable to report the most common flanker-character pairing, and for these “unaware” subjects the flanker effect on response latencies was stronger, rather than weaker, than for other subjects. While these results would appear to suggest that unconscious implicit learning is probably ubiquitous, the research designs represent only a rather narrow set of behavioral changes compared to the typical real-world skill acquisition challenge that people face
Summary
A critically important aspect of skill acquisition is learning to take advantage of the various predictive relationships that exist within the relevant domain. A small proportion of subjects were unable to report the most common flanker-character pairing, and for these “unaware” subjects the flanker effect on response latencies was stronger, rather than weaker, than for other subjects ( the difference was not statistically significant.) While these results would appear to suggest that unconscious implicit learning is probably ubiquitous, the research designs represent only a rather narrow set of behavioral changes compared to the typical real-world skill acquisition challenge that people face. All subjects were told that if they were able to score a point on every single play within a given block of 30, their participation would be complete at the end of the block, and they would be paid as soon as they answered a few final questions (“exit interview”) In this exit interview, subjects in all groups (except the Full Instruction group) who reached the performance threshold were asked if they had any hunches enabling them to predict which way the demon would go. The opportunity to be excused from the study after a perfect-scoring block served two purposes: it motivated the subjects to do as well as they could, and it insured that exit interviews took place only a short time after the moment at which the subject first demonstrated mastery of the game
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call