EPS/Psychonomics, Edinburgh, 2007

The conference was fun. Some notes to help me remember what I want to follow up!

Geraint Rees. Gave an outline of multivariate regression work which essentially allows higher resolution imaging, and then went on to give applications of this work to detect representations in early visual cortex. The from-memory-gist of the theory of why it works: recording each voxel individually gives an average of the activity within the voxel, so interesting effects can be averaged out. The situation gets worse if you use smoothing kernals and average across participants to try remove noise. Suppose the representation you’re interested in has a BOLD correlate in more than one voxel. Further suppose that some of your voxels contain more of that representation correlate than others. If in the stats you allow the voxels to relate temporally to each other, then you can take advantage of this latter fact to detect the smaller-than-voxel representation correlate. Yukiyasu Kamitani and Frank Tong (2005) explain the technique properly, as far as I can tell.

Bennett Scwartz. Tip-of-the-tongue phenomena. Interesting example demonstrating nicely the relationships between cognition and the feeling of what one’s cognitive machinery is capable of doing. Got me thinking again about the various effects like:

  • Being able to communicate things you’re not consciously aware you’re communicating.
  • Feeling things you can’t communicate.
  • Having processes, the state of which you can neither be aware nor communicate.
  • Having a feeling of being able to do something, e.g. recall something, which in reality you may not be able to.

Anyway, Bennett gave an interesting theory of cognitive control and some evidence for how it relates to tip-of-the-tongue phenomena.

Russell Devlin. Taught people how to use Euler circles and found that they weren’t very good, being worse at classical logic with their use than without. He concedes that perhaps the training wasn’t sufficient. But also interesting was that a dual task for suppressing the phonological loop tended to stop people drawing inference based on long term memory—or to use the traditional terminology, reduced belief bias effects.

Liz Robinson. Children’s learning and lack of insight into—or at least the inability to communicate—the explanation for why they knew something.

Jules Davidoff. Fascinating stuff on how the Himba culture seem to be extremely field independent. They don’t, however, have a individualist culture, as would be predicted by Nisbett. They don’t have traits associated with autism, perhaps predicted by weak central coherence. Though I guess autistic traits implies field independence but not the converse; if so, then the Tamarin model of the “perceptual difficulties” in autism proposed by Julie Neiworth is really a model of field independence and not of autism.