Week 3: Insight Article, Lab notes

Article: Facilitate Insight by Non-Invasive Brain Stimulation

• its hard to have insight because people get into a rut. Experts are even worse because they are further in their ruts.
• In what way are we hypothesis driven?
• We agree with evidence that fits with our preconceived notions. This is a way of bias.
• The part of our brain that makes us hypothesis driven: May we be able to damage or inhibit this area to help you see outside the box?
• such as artistic talent (left anterior lobe) damage there may help artistic ability.
• to not be able to think outside the box is to not have insight
• what is a mental set?
• The rut - once we've solved a problem in a certain way, we want to keep doing it that way.
• Restructuring: seeing something in a new way.
• ATL on right is associated with insight, and if the left part is inhibited, it produces more novelty and insight.
• lateralization: the two sides of the brain are different (more lateralized = more differences in hemispheres).
• handedness is important because people who are right handed have more lateralization (more differences in hemispheres). So for right handers, language etc is usually in the left.
• anterior temporal lobe (on left and right) were stimulated.
• cathodal (negative) stimulation: decrease cortical excitability
• anode (positive) stimulation: increases cortical excitability.
• 3 groups
• Control
• cathode on the left and anode on the right
• cathode and the right and anode on the left
• The method was for them to do a lot of the same kind of problems, and then try to do a different type.
• Then, you would get TCDs before you start the new problem
• So, does getting tCDs make a difference?
• Yes. People do a lot better with left inhibition and right excitation. (especially for type 2 problems.)
• Why might tDCS help? 4 possibilities:
• making us less driven by the hypothesis we already have
• interrupting the mental set
• improving set-switching ability
• direct enhancement
• Those who are less right handed (with left hemispheric dominance) are less stuck in their mental sets
• maybe there is a trade off between the two hemispheres in cognitive functioning?

A few Lab Notes:

• Variance Test/Analysis
• Use when you have more than one independent variable
• use when there is more than one condition (such as congruent/incongruent/baseline)
• In a within subjects design:
• order is important because of carry over effects (practice effects or fatigue)
• To control for carry over effects, we counter balance or randomize the order in which subjects do the tasks.
• In an anova:
• The degrees of freedom is usually the number of groups -1
• The Mean Sq is the variance of components (sum/degrees of freedom)
• Pr (>F) is asking, is the P value greater than F?
• Differences in means could be due to random differences of people, or due to differences in the IV.
• Variance between groups (top row) is due to random variation. Variance within groups (bottom row) is just random variation.
• The F value is finding the ratio between the 2 mean squares to show how much the independent variable is the cause of the effect.
• If F is bigger than 1, we are likely to have a big effect; more than what you would expect with chance alone.
• To reject a hypothesis, check to see which is bigger or smaller by looking at the "pairwise" box in variance.
• The more tests you do, the more likely you are to conclude that there is something more than random variation when there isn't.
• Type 1 error: saying effect of IV when there isn't.
• more and more tests make you more likely to make a type 1 error.Tukey test tries to control for these errors.
• confidence intervals: ex. I am 95% confident that the value of the difference between A and B is between x and y. It is important if it doesn't include zero.
• This type of test doesn't work for a within subjects approach.
• Interaction between 2 (or more) IV - the condition that exists when the effect of 1 independent variable differs at different levels of the other independent variables.
• within subjects is the same as repeated measure designs.