Monday, March 26, 2012

Google Brain: Inductive Thinking and Curiosity

In a Scientific American blog post Deep thought is dead, Long live deep thought, a bioinformatics analyst broods on the question, ‘Where are these jobs that will require such rapid “searching, browsing, assessing quality, and synthesizing the vast quantities of information?" and decides quiet a lot of information can be gained by this type of superficial processing of large quantities of material.

"Our ability to produce data is outstripping our ability to understand it. In fact, the need to make sense of these mountains of information is so great that it’s given rise to one of the hottest interdisciplinary fields on the market: data mining and predictive analytics."

Perhaps it's a trade-off. A lot can be gained from slowly and deeply reading a dense but wise text, but a different sort of knowledge (and equally legitimate) can be arrived at by superficial processing of large quantities of material. This more superficial processing may be particularly well suited to inductive problems where principles may be extrapolated from different examples or instances.

Recently researchers in Neuroimage found that the striatal-thalamic regions (blue left) were important for the extrapolation step in inductive problem solving. This is all very interesting because of the association of striatal structures with curiosity and novelty.

One wonders whether strong caudate learners should be considered as a distinct learning style - novel, curiosity driven, inductive learners who learn best by engaging primary or direct experiences- then reasoning back to first principles.

We see many of these types of learners in high tech / computer engineering fields - and that probably also jives with the video gamers have bigger brains (caudates) research.

Monday, March 19, 2012

Reading Metaphors and Individual Differences in the Brain

Researchers at the University of Washington and Carnegie Mellon found that reading and comprehending metaphors had very different patterns of brain activation depending on whether statements were provided in context. In the right context, metaphor comprehension was an easy - required much less brain work- but with neutral or opposite context, look how much effort was expended. In the opposite context, the metaphor was ironic or sarcastic.

- If students aren't 'getting it', they may need more context
- Irony needs a brain workout - no wonder irony is more common among brainy types

Other interesting findings:
- right temporoparietal was active for all figurative vs. literal conditions
- the metaphor area largely overlap an area associated with verbal analogical reasoning
- the benefits of context was stronger among individuals with stronger working memory
- increased right hemispheric activation was noted among lower vocabulary individuals under all processing conditions - (could some have mild dyslexia?)
- not all students are alike - researchers noted fMRI differences among students with different working memory and vocabulary - simple reading fMRI studies may need to consider more cognitive variables in their design of tests

All the subjects were students at Carnegie Mellon University

Reading Metaphors and Individual Differences Brain fMRI

Monday, March 12, 2012

Metacognition, Math, and the Brain

Researchers from Carnegie Mellon found that students solving 'regular' problems based on an example showed overlapping, but distinct patterns of brain activation when 'exception' problems were presented. 'Cognitive' pathways were activated whether a problem was hard or easy; 'metacognitive' pathways were activated especially when challenging problems were presented, and they were activated for a much longer time after problems were already solved. Regular problems involved small positive number values and a single unknown, whereas exception problems used fractions, negative numbers, or repeated variables.

The study was a good reminder that the real work of learning takes place when a student gets something wrong - and that's when the higher network (the 'A' team) gets called into work - and keeps ruminating even after the problem has been solved.

If our students are up to the challenge, it's important we give them a chance to work on very hard problems. If they aren't getting anything wrong, then they're probably not getting any workouts to their metacognitive network. It's not just students, of course. If you're not making and analyzing you're mistakes, then maybe you're coasting too much on cognitive auto-pilot.

The metacognitive regions included sites like the superior prefrontal gyrus, angular gyrus, and frontopolar regions.