Brain Teasers

1
Brain Teasers

• Brought to you by
• Yifat Tamir
• David Anderson
• Kaylie Dienelt
• Mirelle Phillips

Are Dartmouth students this anxious under time
pressure?
2
Introduction

• Getting admitted to Dartmouth College requires
  the combination of high SAT scores, strong
  grades, and involvement in extra-curricular
  activities.
• Performing well on a test such as the SAT
  correlates with an ability to maintain high
  performance under time pressure.
• But does time pressure help or hinder cognitive
  performance?

3
The Test Nervous yet?

• Choosing the Brain Teaser Task
• Somewhat cerebral, but still fairly simple
• Short and sweet
• Lack of associated prior sense of anxiety
• Two groups one low-pressure (untimed) and one
  high-pressure (time limit)
• HP group was told the mean time needed to
  complete brain teaser task for participants in LP
  scenario.
• Does this foster a sense of competition or
  anxiety?

• Selection of Participants
• Two experimenters assigned to each subject group
• 30 LP, 30 HP subjects
• Random selection of participants across a number
  of campus locations (collis, hop, baker, novack)
• Each participant tested individually so as not to
  increase any group-oriented competitiveness
• Results would be analyzed both for time and
  response accuracy

4
The Test Exposed
0________________  Ph.D. B.Sc.
B.A.

• 123safety456

5
Hypotheses Will they freak out??

• Null Hypothesis There will be no difference in
  the mean amount of time it takes for each group
  to complete the task.
• Alternate Hypothesis There will actually be a
  difference in the mean amount of time it takes
  for each group to complete the task.

6
Our Beautiful Thoughts Can they handle the
PRESSURE?

• We hypothesized two possibilities for the effects
  of pressure
• Subjects may take more time to complete the task
  due to anxiety induced by time-pressure
• Subjects may take less time to complete the task
  because they are already motivated to perform
  well in high-pressure, competitive situations (by
  virtue of being Dartmouth students).
• These effects could have confounded each other,
  confusing the difference between our means and
  increasing variance.

7
Holy Platypus! What Brilliance! But are there
any significant effects?

• Results
• The average time for the LP group was 198 seconds
  (3 min. 18 sec.) with a standard deviation of 89
  seconds.
• Average time for HP group (once told the average
  for LP) was 221 seconds (3 min. 41 sec.) with a
  standard dev. of 144 seconds.
• But what about accuracy?!

8
Mathematical Shizzy Graphical Comparison of
Performance across Subject Groups
9
Shooting in the Dark Accuracy across
experimental situations
10
Accuracy Issues Did some punks like to hit it
and quit it, or did they like to stay and play?

• Average number of correct answers in the
  low-pressure group was 5.13 (out of 7) with a
  standard deviation of 1.5.
• Average number of correct answers in the high
  pressure group was 5 with a standard deviation of
  1.5.
• No real statistically significant difference
  between group accuracy

Our friend Sir Mix-a-Lot

• Had to remove one subject from LP group only
  answered two teasers
• Removed three subjects from HP group two only
  answered a limited number of teasers, and one was
  a huge outlier (4 standard deviations away from
  mean)
• Subject pool became 29 for LP and 27 for HP.

11
Significance means and sds- F-test and t-test

• To calculate the t-score for the difference
  between two means, we used t (x1 x2)/ (s12/n
  1 s2.52/ n2.5)
• Our critical region required that t 1.671 or t
  -1.671
• t 0 No significance
• To calculate the difference in variance between
  the two samples, we used F s12/ s22
• To be significant at a 5 level, F would have to
  be in the range of 1.87-1.91
• F 1.27 No significance

12
Discussion Robbed of rightful results with the
rebus

• The problem with the 3 degrees below zero
  rebus.
• How the rebus might have skewed the data in two
  directions
• students who spent too much time on one rebus
  (more time)
• students who just gave up (less time)
• Perhaps we need a cognitive measure that would
  make time much more of a salient factor than
  accuracy.

0________________  Ph.D.
B.Sc. B.A.
Ridiculous Rebus
13
Discussion Part Deux Faster, Hotter, and More
All-Encompassing

• Sampling Problems
• Self-Selection
• After seeing the task, people could choose not to
  participate and it is likely that people who are
  not familiar with or do not like brain teasers
  would opt not to participate
• Location
• Public area, Presence of other students
  increased sense of anxiety or competition
  (respectively)
• Ideally, subjects would be isolated in a small
  room

14
Conclusion Overly verbose suggestions for
further, equally-awesome projects

• The results of the study indicate that applying
  pressure in the form of an implied time
  constraint does not significantly affect
  cognitive performance.
• Potential mplications for the SAT and other
  high-stakes testing, suggesting that there is not
  much truth to the claim that cognitive
  performance is negatively affected by time
  constraints. On the other hand, no one was
  really invested in our little task, so who really
  knows what would happen under real pressure.
• However, certain common characteristics of our
  participants should be noted. As we might be able
  to assume at Dartmouth, the participants in such
  an elite academic environment tend to have been
  trained to perform well under timed conditions.
• Therefore, we would recommend that future
  projects be performed on a large population of
  students from many different grade-levels and
  institutions, that a more time-oriented cognitive
  task be chosen, and that the subjects be truly
  isolated in the testing situation.