neurodudes

This study provides fMRI evidence that, after forgetting some memories, the brain has to work less hard.

Brice A Kuhl, Nicole M Dudukovic, Itamar Kahn and Anthony D Wagner. Decreased demands on cognitive control reveal the neural processing benefits of forgetting. Nature Neuroscience – 10, 908 – 914 (2007)

Pop sci article.

Subjects memorized some cue-associate word pairs (i.e. a query word, and a response that they are supposed to give when asked that particular query). Each cue had multiple alternative associates (i.e. each query word had several alternative responses). We call alternative associates for the same cue “competing memories”.

Then, subjects practiced retrieving only some of the associates for the cues (I’ll call those the “selected associates”). This made them get better at retrieving those associates when presented with those cues, but made them partially forget the other alternative (unselected) associates which were associated with the same cues.

fMRI activation of parts of the anterior cingulate cortex (specifically, dorsal ACC, also known as Brodmann’s area 32) and prefrontal cortex (specifically, right ventrolateral PFC, also known as right Brodmann’s area 47) during the remembering task decreased as subjects practiced remembering the selected associates. The main new result was that this decrease correlated with how much the subjects forgot the unselected associates. In other words, this suggests that the reason that the brain is programmed to forget the unselected associates is because this allows it to, in some sense, work less hard in the future when it recalls a competing selected associate.

Interestingly, reaction time also improved as selected associates were practiced, but (unlike fMRI activation of dACC and vlPFC) this did not correlate with forgetting.

The researchers speculate that the ACC detects conflicting memories and generates “conflict signals”, and that the right ventrolateral PFC is triggered by the ACC and “bias[es] retrieval toward task-relevant representations”.

They note that if, as some believe, the ACC’s function is not just that it “detects conflict” but rather that it “predicts errors”, then that explanation also explains this data because conflicting memories might lead to errors in recall.

This entry was posted on Tuesday, July 3rd, 2007 at 9:17 pm by Bayle Shanks and is filed under Cognitive science, Memory systems. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.