A-not-B task
This year I've been working with infants on a task called the A-not-B task. 8-12 month-olds often do this interesting thing where if you impose a brief delay between hiding and searching they keep reaching back to the first spot they found a hidden toy, even though they've seen it moved to a new hiding spot in the meantime. There's all sorts of research out there providing different evidence for various theories as to why infants do this, and when I started this stuff and had just started reading this stuff, it seemed like it fit in well with inhibitory control, in the sense of being able to inhibit inappropriate actions (in this case, the argument would be that your reaches to the original hiding location are very rewarded, and so it's hard to later inhibit reaching to that location, even though you've seen the toy move somewhere else). There's some good research for this explanation, and I had a bit of a bias to like it, because it ties in with research I've done in the past, and some of my more general interests.
However, as I did more reading and thought about it more, an explanation of mental representations (basically short term memory) seemed more convincing to me. This explanation is basically that you have a representation of the toy being hidden in the original location, which is strengthened by a few hiding events at that location. You then see the toy hidden in a new location, so you make a new mental representation of the new location. However, during the delay, this new representation doesn't stand up too well to the strengthened old representation, and so by the end of the delay, the old representation is back in your head, and you reach back to where the old representation indicates, namely the now incorrect location. With 9 month olds, it can take as little as about three seconds before that old representation is the one that wins. There are very cool computational models of this stuff, which I don't entirely understand yet, but find fascinating as a concept.
So, here's the part that prompted me to post this today. I've kept my toothbrush in a stand to the right of the sink all year. Two weeks ago I got a new toothbrush (thanks to Seth, who had a spare when my luggage got delayed) which has a base that didn't fit into the same stand. Instead, I put the toothbrush into the cabinet to the left of the sink. Given my frequent travels, I only had a couple of days' worth of opportunities to find the toothbrush in the new location before I left again. However, in those few days, I did just fine getting the toothbrush from the cabinet when I wanted to brush my teeth. After those few days of learning, I went away for a full week. When I came back after the week, I walked into the bathroom, and my thought process was something like "why isn't my toothbrush in the stand? did someone take it? who would want my toothbrush" etc, etc in coming up with reasons for my toothbrush wasn't in the stand. It took me quite a few moments before I came up with the thought that I had moved the toothbrush to the cabinet. And even then when I opened the cabinet I couldn't remember where in the cabinet it was. At first, I figured this was compounded by being sleepy, etc, but it happened again the next day, though not quite as extreme. It's not just that I was automatically reaching for the old spot, but that at least briefly, I really thought that's where I would find the toothbrush. So, basically, this story has become personal experience for me in support of my current thoughts on why kids make the error they do. I had the old representation of where the toothbrush was, which was well rehearsed. I then had a new location, and when I was able to keep that new representation updated and current, I continued to successfully find the toothbrush in the new location. However, given the delay of a week while I was gone, that old representation overpowered the new one and I went back searching for my toothbrush in the old location. :)
At least it took a week's delay, rather than 3 seconds. Some evidence that I have more working memory than a 9 month old.