Tuesday, April 21, 2009

What. The. HECK?

Recently (within the last week or two), I got into something of a debate with Michelle Dawson regarding the internal validity of single-subject designs. As a result, I've been spending some time reviewing the literature and looking for a design to pick apart. While doing so, I came across a paper entitled "Functional analysis of stereotypical ear covering in a child with autism."

The abstract -- the entire abstract -- is as follows:
We studied stereotypical ear covering in a child with autism. Results of a descriptive analysis were inconclusive but revealed a correlation between ear covering and another child's screaming. An analogue functional analysis showed that ear covering was emitted only when the screaming was present.
Way to avoid citing other literature, people.

The rest isn't much better. To quote the last paragraph of the paper:
These findings suggest that ear covering was maintained by negative sensory reinforcement (noise attenuation) and illustrate the importance of linking descriptive and analogue functional analyses when idiosyncratic events are implicated in behavioral maintenance. Whether a similar relation between ear covering and noise occurs for other children with autism awaits further investigation. However, the current data set implicates a previously unidentified source of reinforcement as one possible cause of stereotypical behavior.
Umm... that is most certainly not a "previously unidentified source of reinforcement". Documentation on what we politely call "sensory issues" in autism goes back a biiiiit further than 2003.


  1. So someone got published by noticing that one kid covers his/her ears when another kid screams? WTF? I wonder if they got a grant for that, or did it self-funded for a master's degree?

    This makes me feel better about my master's project. I'm doing my master's project on the effects of sudoku on problem solving ability in college students. Common sense says there ought to be an effect, but working with human subjects when you don't have a regular class of students / captive audience is like herding cats! But at least I'm showing something that has worthwhile application (a reliable way of improving cognitive fitness in college students) and is not, um, duh.

  2. As I said above, however, "sensory issues" in autism have been documented for a pretty long time.

    Then again, the study was published in JABA -- the Journal of Applied Behavior analysis... and this is the first time, to my knowledge, that this sort of thing has been mentioned in the ABA literature... at least as a factor with autistics.

    Of course, it's come to their attention in other contexts (see http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1284083 ; I suspect, given the descriptions, that at least one of the "participants" was on the spectrum), but not explicitly for autism.

    Then again, behavior analyists can be a bit thick (cf. http://aspieperspective.blogspot.com/2009/03/pace-of-aba-research.html , http://aspieperspective.blogspot.com/2009/03/aba-establishment.html , http://aspieperspective.blogspot.com/2009/03/functional-analysis.html , or http://aspieperspective.blogspot.com/2009/03/latest-and-greatest.html ).

    As for your project, it sounds pretty interesting. Could you tell me how it turns out?