Decision Psychology

Reposted from lensofliterature.com

Here’s another example of using personal motivation for the greater good: BBC News reports that Carnegie Mellon has started using samples from old books for CAPTCHAs.

But the CMU research team, based in Pittsburgh, Pennsylvania, has devised an ingenious system to put the time used interpreting CAPTCHAs to good use.

Text files

The team is involved in digitising old books and manuscripts supplied by a non-profit organisation called the Internet Archive, and uses Optical Character Recognition (OCR) software to examine scanned images of texts and turn them into digital text files which can be stored and searched by computers.

But the OCR software is unable to read about one in 10 words, due to the poor quality of the original documents.

The only reliable way to decode them is for a human to examine them individually - a mammoth task since CMU processes thousands of pages of text every month.

To solve this problem the team takes images of the words which the OCR software can’t read, and uses them as CAPTCHAs.

These CAPTCHAs, known as reCAPTCHAS, are then distributed to websites around the world to be used in place of conventional CAPTCHAs.

What a nifty side effect! I love this because it takes something people have to do anyway, validate themselves as humans, and turns it into historical restoration without the person even knowing it. I am endlessly fascinated by these little backchannel ways of getting people to do good.

Wow, it has been two months since my last update!  The last day of class for this quarter is tomorrow and I hope to get back on the habit a bit.  I have lots of articles to write up and will also be posting my thesis by the end of the summer!  I just turned in my first draft and got it approved, the first revision is due next week.  Not so far anymore *fingers crossed*

Anyway, here is an interesting abstract about blood sugar and heuristics courtesy of Decision Science News:

This experiment used the attraction effect to test the hypothesis that ingestion of sugar can reduce reliance on intuitive, heuristic-based decision making. In the attraction effect, a difficult choice between two options is swayed by the presence of a seemingly irrelevant “decoy” option. We replicated this effect and the finding that the effect increases when people have depleted their mental resources performing a previous self-control task. Our hypothesis was based on the assumption that effortful processes require and consume relatively large amounts of glucose (brain fuel), and that this use of glucose is why people use heuristic strategies after exerting self-control. Before performing any tasks, some participants drank lemonade sweetened with sugar, which restores blood glucose, whereas others drank lemonade containing a sugar substitute. Only lemonade with sugar reduced the attraction effect. These results show one way in which the body (blood glucose) interacts with the mind (self-control and reliance on heuristics).

Read more here.

• Energy use cut if savings are apparent, study finds [seattle p-i]
• Electric grid meets Web 2.0, savings result [ars technica]
• enerjar DIY energy display [information aesthetics]
• saverclip electricity usage pin [information aesthetics]

Hoffman et al. (1993) review existing psychological research in color. Most color research follows a simple methodology: arrays of colored or achromatic symbols are presented and participants are asked to recall data seen or complete some task under time pressure. Reaction time and accuracy are measured to see if color helps. A few studies:

• Christ (1975) concluded that color is easier to search for than symbols.
• Carter (1982) and Hitt (1961) both found that it is easier to locate colored symbols than grayscale ones that vary in brightness or size.
• Ware and Beatty (1998) have shown that information about multidimensional data can be conveyed using color.

On page 507 the authors provide comprehensive guidelines to use of color as well as some qualifications to its use. Although color can enhance the “findability” of symbols, hue and brightness interact, so some colors may distract from others (such as yellow vs. blue). He also cautions against dependence on such research, as research shows that discriminability breaks down when symbols come in various sizes and convoluted shapes, as in scientific displays.

The authors also point to research on what colors should be used. They note that although the design community prefers bold primary colors, research actually shows that low-saturation colors attenuate memory deficiencies and size misjudgments (Cleveland et al. 1983).

This is an excellent review of literature, if an outdated one.  It’s a great place to start with reviewing color literature, and since that’s not the focus of my thesis probably all the review that I need.  Much of this echoes what I learned in Information Dashboard Design (review coming soon), particularly the research on low-saturation colors.  I plan to use these guidelines in designing the visualizations for my project, possibly replicating one or more of these studies.  In particular I’d like to retest the primary color vs. low-saturation colors since that interests me.  The information on what colors to use echoes Brewer 1997.

——————————-
Hoffman, R.R., Detweiler, M., Conway, J.A., Lipton, K. (1993). Some Considerations in Using Color in Meteorological Displays. Weather and Forecasting, 8, 505-518.

My chaya at decisionpsychology dot com email isn’t working, I am currently working on fixing it.  In the meantime, please email me at chaya at lensofliterature dot com.

In this paper, Gigerenzer et al. (2005) surveyed 750 participants on the street in New York, Amsterdam, Berlin, Milan, and Athens. As the authors write, “probabilities of rain were introduced into mass media weather forecasts in New York in 1965, in Amsterdam in 1975, and in Berlin in the late 1980s; in Milan, they have been introduced only on the Internet; and in Athens, they are not reported in the mass media at all.”

Participants were asked to imagine a weather forecast that states “there is a 30% chance of rain tomorrow” and then asked which of the following are the most appropriate and least appropriate interpretation:

1. It will rain tomorrow in 30% of the region.
2. It will rain tomorrow for 30% of the time.
3. It will rain on 30% of the days like tomorrow.

Participants were then asked to provide their own interpretation in a free-response format, and finally were asked at what probability of rain they would take an umbrella. On the first question, 2/3 of New York participants favored days, 1/4 chose time, and a minority chose region; in the European cities, the preferred interpretation was time. The authors attribute this to differences in familiarity with probabilistic forecasts. In Athens in particular the first choices were uniformly distributed, reflecting a lack of familiarity with probabilistic forecasts.

However, I wonder if there’s a language problem here. It took me a while to figure out the difference between the days and times answer. I thought of them as identical, because I would say that it would rain 30% of the time, meaning that 3 out of 10 times it would rain on a particular day. Perhaps others faced the same confusion.

———————————-
Gigerenzer, G., Hertwig, R., van den Broek, E., Fasolo, B., Katsikopoulos, K.V. (2005). “A 30% Chance of Rain Tomorrow”: How Does the Public Understand Probabilistic Weather Forecasts?. Risk Analysis, 25(3), 623-629. DOI: 10.1111/j.1539-6924.2005.00608.x

Reposted from lensofliterature.com

As a follow-up to “what do bestsellers tell us about society?”, I thought I’d share a couple of relevant posts by Tim O’Reilly in the last week. Somehow I often find his posts highly relevant to my perspective and sometimes even my research. First, “The Techmeme Leaderboard: The Enduring Appeal of the Bestseller List” is an alternate angle on what bestsellers tell us about society, and precisely why they can be used as such a lens. Or at least that’s one thing to take from it!

Second, today he posted a lovely example of the motivation provided by stories I spoke of at the end of my piece:

Few business leaders appreciate the power of stories to connect with their audiences. A few weeks ago I was working with one of the largest producers of organic food in the country. I can’t recall most, if any, of the data they used to prove organic is better. But I remember a story a farmer told. He said when he worked for a conventional grower, his kids could not hug him at the end of the day when he got home. His clothes had to be removed and disinfected. Now, his kids can hug him as soon as he walks off the field. No amount of data can replace that story. And now guess what I think about when I see the organic section in my local grocery store? You got it. The farmer’s story. Stories connect with people on an emotional level. Tell more of them.

We need more farmer’s stories, especially in the non-profit world! Who is going to go out and write them?

• The future of data and analysis [decision strategist]
• The gift of information transmission [decision strategist]
• The hardest decisions of all [decision strategist]
• Two new papers on natural rationality [natural rationality]
• Reducing relative risk reduction: can psychologists help doctors make better decisions? [decision science news]

Last month Decision Science News had a lovely post on decision strategies, adapted from Reid Hastie and Robyn M. Dawes’s Rational Choice in an Uncertain World.

Strategy: RECOGNITION HEURISTIC
Mental Effort Compensatory vs. Noncompensatory? Whole vs. Part Exhausive?
LOW NON-COMPENSATORY ALTERNATIVE NO
“In some choices, people are so poorly informed about the alternatives that they simply rely on ‘name recognition.’ They choose the first alternative that they recognize … in many realistic choices and judgments the ‘fast and frugal’ recognition choice heuristic behaves surprisingly well.”

The post lists eight strategies in total. I am thinking of distributing this list to participants in the troubleshooting study (I’ll have to ask my advisor if that is okay) alongside my in-progress Sources of Power summary as guidelines for explaining their decision-making process.

Davis and Keller (1997) begin with a discussion of uncertainty. There are two generally accepted terms for uncertainty, “error” and “accuracy.” They point out that accuracy, unlike error, is more easily quantifiable since error measures deviations from some “true” value that is generally unknown. The discussion is then narrowed to uncertainty in the measurement of natural resource data, which can be subdivided into position, attribute, and temporal uncertainty. It is pointed out that uncertainty is not necessarily a product of measurement or modeling, but can also be the result of the changing nature of the environment itself.

The visualization of uncertainty data is key to linking geographic visualization into analytical pools. Research shows several options, including changing focus, color, fog, or texture for static displays as well as duration and toggling for animated displays.

A slope-stability scenario was chosen as a platform to develop an uncertainty model. This model requires soil type, slop, and forest cover as source data. There are five key types of uncertainties associated with this data: soil or forest type classification, data gathering, spatial polygons, envelopes around quantified items, and envelopes around elevation values.

Techniques for visualizing uncertainty can be split into static and animation techniques. Static visualizations can include pairing a static map with an uncertainty map or combining data on one map, the most successful variables for display including value, color, and texture.

Advantages of static methods include that they are easy to produce and reasonably easy to understand. Map comparisons are one of the most popular methods of representing uncertainty, but there is little evidence to support this. (Note: MacEachren did a study in 1998 supporting the use of static comparison, summarized earlier on this blog.) Bivariate maps allow immediate correlation between uncertainty and other variables but are often complex. By contrast dynamic visualization allow the addition of another variable, time, which can help ease the complexity of bivariate maps.

This is a highly technical cartography paper with lots of mathematical discussion, but its summary of visualization techniques is useful. It constitutes a miniature literature review (much like Aerts) which despite being slightly out of date is still edifying. For example, the citation of Goodchild, Buttenfield, and Wood (1994) quickly sums up the study as identifying value, color, and texture as the most important variables in a static map. Unfortunately there is no experimental portion to this paper, limiting its usefulness. Still the literature review section provides excellent references to experimental papers that is very useful.

——————————
Davis, T.J., Keller, C.P. (1997). Modelling and Visualizing Multiple Spatial Uncertainties. Computers & Geosciences, 23(4), 397-408.