Memory can be biased - here's how...
As an aside, I always like to see the authors of the papers - that way, if I should ever see them at a conference, I can capitalize on my terrible source memory and approach them saying "don't I know you from somewhere?". It also helps attribute, recognition where it is due.
|
|
|
I attended a talk given by Katherine Duncan at the Rotman Research Institute (where I work). The talk, was largely based on the findings from the Science paper that came out in July. The talk was great - and just as importantly, so is the paper and its findings.
The idea comes from neuroscience, yet the paper itself is entirely behavioral - something some of the senior faculty liked - that it was still possible to get a psychology/neuroscience paper into Science without resorting to newfangled methods like fMRI.
Even without fMRI, the paper is pretty neat.
The basic premise is that since neuromodulators such as acetylcholine work over the span of a few seconds, it is possible that their effects form a "penumbra" or period of influence that might extend into a subsequent task.
Using this idea, the authors adapted Craig Stark's Behavioural Pattern Separation Object task (I was pleasantly surprised to find this is freely available on the author's website), which consists of a familiarization phase, followed by a task where new objects, old objects and objects that were similar to some of the other objects were presented (see figure below).
The critical manipulation was whether the object preceding the similar trial was new or old. The logic is that an "old" object would activate memory and bias the respondent towards pattern completion which makes it harder to distinguish a similar object from it's predecessor. Alternately, a "new" object should bias the respondent towards pattern separation making it easier to distinguish the similar object from it's predecessor. Thus the critical measure was accuracy on the similar trial...
There is a second experiment in this paper (where the authors prove the inverse, namely that a preceding "old" trial helps with pattern completion, but I will not cover that here save to say that it nicely complements the initial experiment.
From a practical, real world point of view, what does this mean? I guess it means that if you were just asked to recall something and then saw someone who looked very similar to someone you knew, you might mistake that person for your acquaintance - at least for a few seconds. Conversely, if you were asked to examine a new object analytically, the opposite might hold true.
This paper worth the read and is rigorous from a theoretical and scientific view suggesting many ways to expand on the basic findings. Coming from an inhibitory theory background, my first thought is can this phenomenon be extinguished by inhibition? If so, what are the consequences? I have a few ideas I may want to try out...
Using this idea, the authors adapted Craig Stark's Behavioural Pattern Separation Object task (I was pleasantly surprised to find this is freely available on the author's website), which consists of a familiarization phase, followed by a task where new objects, old objects and objects that were similar to some of the other objects were presented (see figure below).
The critical manipulation was whether the object preceding the similar trial was new or old. The logic is that an "old" object would activate memory and bias the respondent towards pattern completion which makes it harder to distinguish a similar object from it's predecessor. Alternately, a "new" object should bias the respondent towards pattern separation making it easier to distinguish the similar object from it's predecessor. Thus the critical measure was accuracy on the similar trial...
 |
| Adapted from Duncan et al., Science, 2012, |
From a practical, real world point of view, what does this mean? I guess it means that if you were just asked to recall something and then saw someone who looked very similar to someone you knew, you might mistake that person for your acquaintance - at least for a few seconds. Conversely, if you were asked to examine a new object analytically, the opposite might hold true.
This paper worth the read and is rigorous from a theoretical and scientific view suggesting many ways to expand on the basic findings. Coming from an inhibitory theory background, my first thought is can this phenomenon be extinguished by inhibition? If so, what are the consequences? I have a few ideas I may want to try out...
Duncan, K., Sadanand, a., & Davachi, L. (2012). Memory’s Penumbra: Episodic Memory Decisions Induce Lingering Mnemonic Biases. Science, 337(6093), 485–487. doi:10.1126/science.1221936
Comments
Post a Comment