People with Autism are Still Superb at Learning Things Implicitly
Children with autism spectrum condition do a fine job learning things implicitly
Posted July 26, 2010
The general consensus in cognitive science is that implicit learning involves "learning that proceeds from practice with any structured environment, in the absence of an intention to learn, and results in knowledge that improved performance even when it is difficult to verbalize" .
On first blush, it may seem as though individuals with autism spectrum condition (ASC), who are typically characterized by social, communicative, and motor impairments, would have deficits in implicit learning mechanisms that would cause such overt impairments. Research recent suggests this isn't the case, however. In fact, research is converging on the fascinating conclusion that children with autism can actually do a superb job implicitly learning things.
Overcoming some prior limitations of earlier studies, I conducted a study with a team of researchers led by Jamie Brown at the Laboratory for Research into Autism at the University of Cambridge [read full article in The Quarterly Journal of Experimental Psychology here].
We compared the performance of 26 children with high-functioning autism spectrum condition to 26 typically developing children on four implicit learning tasks, each one designed to tap into a different domain of implicit learning. We even administered a measure of implicit social learning which researchers have assumed would show an impairment in people with autism. We minimized reliance on explicit strategies. We matched the two groups on IQ. We administered a measure of explicit learning for comparison purposes and we administered a reliable index of autistic symptomatology that relates to real-world implicit learning performance [Read "Can People with Autism Learn Implicitly?" for a more detailed summary of the methods and results of our study.]
We found no difference in implicit learning ability between the autism spectrum condition group and the typically developing control group. We found that this difference was not a consequence of compensation by explicit learning ability or IQ; in fact we found there was a difference between the two groups in explicit learning, but not implicit learning. Finally, we found no relationship between implicit learning ability in the laboratory and real-world autism spectrum condition symptomatology. Our conclusion: whatever is causing the social, communicative, or motor impairments found in those with autism spectrum condition, it is not implicit learning.
Fast forward to a more recent study. Today in my inbox I get an email from Dezso Nemeth, project director of the Memory and Language Lab which is part of the Cognitive Science and Neuropsychology group at University of Szeged. He sent me an article  that replicates and extends our findings.
Whereas our study administered implicit learning tasks altogether at one point in time, they looked at the effect of a 16-hour delay on learning a difficult 4-element implicit probabilistic sequence learning task. Their task was constructed in such a way as to allow them to differentiate between general skill (assessed by overall reaction time) and sequence-specific learning (assessed by taking the difference between the reaction time to predictable, sequence events and the reaction time to less predictable random ones).
13 children with autistic spectrum disorder were compared to 13 IQ-matched and 14 age-matched children. Since they also were interested in the effects of consolidation, the children went through two testing sessions: a learning phase and then a testing phase 16 hours later.
They found that children with autism spectrum condition demonstrated similar levels of general skill learning and implicit learning of probabilistic sequences compared to the two control groups, one matched in IQ and the other matched in age. The groups did not differ from one another in consolidation over a 16 hour period, there were no differences among the groups in forgetting the sequence-specific learning, and there were no differences among the groups in offline improvements in general skill.
In our paper, we argued that our findings overcame prior study limitations by administering an implicit learning task that minimizes explicit strategies. It looks like the Nemeth et al. study , also using a complex regularity that minimizes explicit strategies, discovered similar findings. [See "Can People with Autism Learn Implicitly?" for other potential discrepancies between our findings and prior studies.]
Nemeth et al.'s result concerning consolidation are consistent with two prior studies [3, 4]. All of these studies show no difference between healthy young and older adults and those with autism spectrum disorder in their ability to remember the sequence between sessions. Also, these studies point to the ability of children and adults, typically developing and those with autism spectrum condition, to demonstrate offline enhancement of general skill by starting their second session at a faster response rate than at the end of the first session.
It's also important to note that neither  nor  found that sleep aided in either general skill learning or in sequence-specific learning. As Nemeth et al.  note, this is important since autism spectrum disorder has been linked to sleep difficulties in prior research. Therefore, the consolidation found in those with autism spectrum disorder is probably not due to sleep disturbance.
Taken together, these latest studies suggest that individuals with autism spectrum disorder can learn implicitly just as well as typically developing individuals, over short and longer periods of time.
© 2010 by Scott Barry Kaufman
 Brown, J., Aczel, B., Jiménez, L., Kaufman, S.B., & Grant, K.P. (2010). Intact implicit learning in autism spectrum conditions. The Quarterly Journal of Experimental Psychology. doi: 10.1080/17470210903536910. [pdf]
 Nemeth, D., et al. (2010b). Sleep has no critical role in implicit motor sequence learning in young and old adults. Experimental Brain Research, 201, 351-358.
 Song, S., Howard, J.H., Jr., & Howard, D.V. (2007). Sleep does not benefit probabilistic motor sequence learning. Journal of Neurscience, 27, 12475-12483.