Authors: Martine Baldassi, Ye Li, Jing Qian, Eric J. Johnson, Elke U. Weber

Contact Email: mb3195@columbia.edu

 

Method: Detailed description of the measures used in this study

Decision Making measures

1) Framing (Kahneman & Tversky, 1979; Tversky & Kahneman, 1981)

Two different framing scenarios were used in this study. We used the classic Asian disease problem (with the situation framed either as saving or losing lives) and a layoff problem (with the situation either framed as the number of factories kept open or shut down) to assess the effect of framing on the choices of these two situations. In both scenarios the sure response option was associated to a 100% chance to occur (i.e., 100% chance that 200 people will be saved) and the risky option was associated with two outcomes having different chances to occur (i.e., 33% chance that 600 people will be saved or 70% chance that 0 people will be saved). DV = participants are consistent (1) or not (0) in their choice preference independent of the frame (consistency with Expected Utility).

2) Loss Aversion (Tversky & Kahneman, 1992)

Our loss aversion task consisted in a titration method asking participants if they were willing to play different gambles that involve both gains and losses of different monetary amounts. For all the gambles there was an equal odd (50% chance) of winning and of losing money. We have used two mixed gambles, one with a smaller (gain amount = $6; loss amount varies from $0.5 to $7 in $0.5 increments) and one with a larger payoff (gain amount = $20; loss amount varies from $2 to $24 in $2 increments). DV = loss aversion coefficient l (l=gain amount/loss amount).

3) Anchoring (Chapman & Johnson, 1999; 2002; Kahneman, Ritov, & Schkade, 1999; Wilson, Houston, & Brekke, 1996)

This task provides an individual level measure of susceptibility to anchoring, which is the tendency for estimations to be influenced by the presence of arbitrary and irrelevant numerical values. In our study, each participant made three numeric estimations on (1) the distance between New York and Cairo, (2) the number of gallons of gas used by an average American family in one month, (3) the height of the Washington Monument in Washington DC. Those estimation questions were preceded either by a binary choice question, which asked whether the estimation object is higher or lower than a given value (high or low anchor), or not preceded by such a binary choice question (no anchor). DV = percentile of high anchor minus percentile of low anchor.

4) Intertemporal Choice (Weber, Johnson, Milch, Chang, Brodscholl, & Goldstein, 2007)

In our study we have used two different versions of this task, delay and acceleration, which differ in the way the choice situation is framed. In the delay condition, participants were told that they could choose between having a smaller gift certificate today or delay the consumption and have a larger gift certificate in the future. In the accelerate condition, participants were told that they could choose between a larger gift certificate in the future and a smaller gift certificate today by accelerating the consumption. Besides the two framing conditions, our intertemporal choice items differed in regards to the outcome used (e.g., Amazon gift certificate or Paypal payment), the time period between ‘today’ and the ‘future’ (e.g., 3 and 4 months) and the amounts of the gift certificates. Like the loss aversion task, the intertemporal choice task presented a series of choices in a titration format (11 choices per item). DV = discounting is quantified as d = (x1/x2)^(1/t) (e.g., discount factor), where x1 is the amount received today that was seen equivalent to the amount x2 received in time period t.

Cognitive measures

     Fluid Intelligence

1) Alphabet Span (Craik, 1986): adapted from CREATE: Common Core Battery of Measures (Czaja, Charness, Dijkstra, Fisk, Rogers, & Sharit, 2006; Czaja, Charness, Fisk, Hertzog, Nair, Rogers, et al., 2006)

This task is used as a measure of verbal working memory capacity. In our adapted version of this task, words appeared sequentially on the computer screen. Participants were required to recall and type the words in alphabetical order (different from the order the words appeared in on the screen). Initially, only two words (e.g., level 1) had to be recalled correctly but the number of words presented increased with the number of levels a participant reached (e.g., level 2 = series of 3 words, level 3 = series of 4 words, etc.). Each level consisted of 3 trials and participants had to remember all the words of at least 2 trials in the right order to go on to the next level. DV = level a participant was able to reach.

2) Cognitive Reflection Task (Frederick, Loewenstein, & O'Donoghue, 2002)

This test contains three mathematical questions that yield intuitive, impulsive erroneous responses, which need to be overridden/inhibited to get the correct answer. According to Frederick, the “three items on the CRT are ‘easy’ in the sense that their solution is easily understood when explained, yet reaching the correct answer often requires the suppression of an erroneous answer that springs ‘impulsively’ to mind” (Frederick, 2005, p. 27). DV = number of correct responses.

3) Numeracy (Lipkus, Samsa, & Rimer, 2001).

Numeracy is the ability to understand probability and mathematical concepts. The numeracy task we have used in our study consisted of 11 questions that test comprehension and manipulation of proportions, percentages and probabilities. DV = number of correct responses.

    Inhibitory control

 

All measures developed by Lumos Labs (“lumosity, reclaim your brain”, www.lumosity.com)

4) Spatial 1-back

This task is frequently used to measure the capacity to update and actively manipulate working memory contents (Del Messier, Maentylae, & Bruine de Bruin, 2010). In contrast to these authors, the version of the task we employed in our study is based on visuo-spatial information processing, not verbal material. Participants were presented with three circles, two white and one blue, which formed a triangle on the screen. In each trial they saw the same triangle but the blue circle changed its location. From trial to trial participants had to evaluate if the location of the blue circle was the same or different from the trial before (1-back). DV = reaction time.

5) Stroop (Stroop, 1935).

 

In the Stroop task we have used in this study, participants saw two words written on the screen. In each trial, the word on the left was written in black and indicated the name of a color (i.e., blue, green, red, etc.). On the right side of the screen, another name of a color was printed in the same color as the semantic value of the color name on the left of the screen (e.g., congruent trials) or in a different color than the semantic value of that word (e.g., incongruent trials). Participants had to judge as quickly as possible if the semantic value of the word on the left side corresponded or not to the actual color the word on the right side of the screen was printed in. DV = reaction time of incongruent trials - reaction time of congruent trials.

6) Flanker (Eriksen & Eriksen, 1974)

In the Eriksen Flanker task participants are expected to respond to a centered and directed item surrounded or flanked by distracting symbols. In our version of the task, participants saw five birds in a row on their computer screen. They had to pay attention to the bird in the middle (target) and indicate, by pressing one of the four arrow keys, in which direction the target bird was flying. In some of the trials, the additional four birds (two on the left and two on the right) were flying in the same direction as the target/middle bird (e.g., congruent trials) and in other trials they were flying in a different direction than the target/middle bird (e.g., incongruent trials). DV = reaction time of incongruent trials - reaction time of congruent trials.

 

    Fluid intelligence and Inhibitory control

7) Fragment Completion Task

Participants were given a combination of three letters representing the beginning of a word. They had to complete this letter-fragment with legitimate English words coming to their minds. Participants were not allowed to type two words of the same length. There were a total of three trials, each showing a different ‘3-letter’ combination and lasting 60 seconds. DV = number of correct words generated.

 

 

 

[1] Johnson, E. J., Häubl, G., & Keinan, A. (2007). Aspects of endowment: A Query Theory of value construction. Journal of Experimental Psychology: Learning, Memory, and Cognition, 33(3), 461-474.

[2] Weber, E. U., Johnson, E. J., Milch, K. F., Chang, H., Brodscholl, J. C., & Goldstein, D. G. (2007). Asymmetric discounting in intertemporal choice: A Query Theory account. Psychological Science, 18(6), 516-523.

[3] Dinner, I., Johnson, E. J., Goldstein, D., & Liu, K. (under review).

 

[4] Hardisty, D. J., Johnson, E. J., & Weber, E. U. (2010). A dirty word or a dirty world? Attribute framing, political affiliation, and Query Theory. Psychological Science, 21(1), 86-92.

 

[5] Del Missier, F., Mäntylä, T., & Bruine de Bruin, W. (2010). Executive functions in decision making: An individual differences approach. Thinking & Reasoning, 16(2), 69-97.

[6] Salthouse, T. A. (2010). Major issues in cognitive aging. NY: Oxford University Press.