+ Behavioral Psychology and Technologies for Education Marti Hearst March 10, 2012.

+Behavioral Psychology and Technologies for EducationMarti HearstMarch 10, 2012

+This Lecture

I present psychological results

Collectively we decide how these impact education and technology + education.

+The Elephant and the Rider

Your brain isn’t of one mind: The elephant and the

rider.

Jonathan Haidt, University of Virginia, Psychology

+

Your Brain Isn’t of One Mind

The rider is the conscious controller

Perched on top and seems to hold the reins.

Thinks long term

But also overthinks; analysis paralysis

The elephant is the emotional side: much more powerful

Often wins the argument about which way to go.

Instant gratification.

Sometimes for the benefit

Instinct to protect your child

Provide the energy behind motivating major efforts.

+

Willpower is an exhaustible resourceExperiment:

• Can eat radishes but not cookies

• Afterwards, radish eaters have less energy to engage in a puzzle.

The rider is too tired to work any longer.

+How to Change Behavior

To change behavior Must engage both the

rider and the elephant

Three main steps: Direct the rider Motivate the elephant Shape the path

Chip Heath, Stanford Business School, Organization Behavior

+

Example: changing purchasing habitsEngage the elephant

+

Example: change hospital practicesDefect rate in hospital practice analyzed to be 1 in 10.

The methods to improve were clear, but nothing was changing.

How to fix? Engage the elephant:

• Brought in mother of patient who didn’t make it

• Set a tangible goal:• Save 100,00 lives in 18 months.

Shaped the path:

• Step-by-step instructions

• Support groups, mentors

• Peer pressure

+

Example: how to save malnourished children with no budget?The rider lists the dozens of impossible problems

• Water quality, sanitation, poverty

• TBU (true but useless)

Strategy:

• Look for bright spots

• Engage the elephants of the population to make these work

Eventually engaged 2.2M people

+

Example: how to sell a new product?A few salespeople were successful

The Rider discounts this, says they had an unfair advantage

Instead, see what they do differently.

They provided a path for success

The drug required an unfamiliar administration method, so they focused on tutorials for this part.

+Habit

100 years ago, few people brushed their teeth A Pepsodent marketer

changed this.

Cue, routine, reward

Charles Duhigg, journalist

+

HabitRats in maze:

• Brains highly active when first learning

• Then as rats got faster, mental activity decreased.

• At this point, there are two spikes of mental activity: at the start when they heard the click of the maze opening, and at the end when they got the reward of the chocolate.

The brain is ready to change at those points (no chocolate, new maze)

+

HabitDuhigg always got a cookie at 3pm

Gained 20 pounds!

Habit psychology helped him solve it

Three part loop:

• Cue (trigger)

• Routine

• Reward

+

HabitHow to change it:

• Change the cue

• Change the reward

• The routine gets reinforced over time

+The Paradox of Choice

Contrary to popular opinion, more choices can be negative Can lead to more anxiety

Barry Schwartz, Swarthmore, psychologist

+

Jam StudyLepper & Iyengar

24 flavors of jams available

Showed either 24 or 6 varieties

• When 24 shown, attracted more customers, but 3% bought

• When 6 shown, 30% bought

+

Decision paralysisIf there is one drug left to try before surgery, doctors try it.

• But if there are 2, they go for surgery

401k investments:

• For every 10 options offered, participation rate decreases 2%

Speed daters who meet 8 people make more matches than those who meet 20

Shopping makes you tired:

• The rider has to make lots of choices

+Nudges

Decision making: No setting is neutral Even small and seemingly

insignificant details can matter

“Libertarian paternalism” People retain their right to

make choices But are “nudged” in the

“right” direction

Richard Thaler & Cass Sunstein, University of Chicago, Law & Behavioral Economics

+

Settings Influence BehaviorHow much someone eats is influenced by the size of the serving

+

Default settings matter

+

Choices are presented in non-neutral frames• Cafeteria offerings

• 401(k) investments

• Organ donation

+Predictably Irrational

Anchoring effects People exposed to one

number carry it over to others

Dan Ariely, Behavioral Economist, MIT

+Cognitive Biases

Conjunction Falacy: Conjunction of 2 events

seen as more probably then either in isolation

Hindsight bias Tendency to revise history

of one’s beliefs in light of what actually happened

Halo effect: Tendency to like or dislike

everything about an entity based on a few salient traits

Thanks to Arthur Suermondt, Jeff Zych, and Evan Smith for this slide

+How To Live in a World We Don’t Understand Very Well

People underestimate low-probability events

People are biased not to notice negative information

Nassim Taleb, Former professor, financial advisor, “epistemologist of randomness”

+

Survivor bias• WWII: placement of armor on

returning airplanes• Looked at where the bullet

holes appeared on the plane

• They don’t see the planes that were shot down!

• Lessons learned from business successes: I had 3 successful startups so now I’m the guru.

• Or am I just the lucky one statistically?

+

Low Probability Events

• In applying probabilities there is an asymmetry between safety and danger.

• Some decisions require vastly more caution than others:• You don’t need evidence that

the waster is poisonous to not drink it.

Taleb: http://www.edge.org/3rd_culture/taleb08

+

Survivor Bias and Low Probability Events• If the same study is done 20 times, and by

chance it’s positive once time in 20, we only see the published paper with the low-probability positive event.

• Someone in a crowd of 1000 people will get heads more often than chance.

• Someone in a large enough set of people will appear to have ESP.

• Some of the companies in “Good to Great” by Jim Collins in 2001 are now defunct

• If you invested in the 11 “great” companies you would have underperformed the market

• Did some of the not “great” companies also practice a “culture of discipline?

• Among these “greats” were Fannie Mae and Circuit City.

• http://www.freakonomics.com/2008/07/28/from-good-to-great-to-below-average/

+Discussion