By Michael Caruso
Judges generally believe they are objective and impartial and can ignore irrelevant information when they make decisions. Research, however, has shown that many legally irrelevant factors may influence legal decision-making.
The “anchoring effect” may be the most well-known example. The anchoring effect is a type of cognitive bias where numbers that act as a reference point influence a person. These numbers may or may not be utterly irrelevant. In 2006, Prof. Dan Ariely asked students at MIT to bid on items in an arbitrary auction using social security numbers as their anchor. Students were asked how much they would pay for two types of wine, a cordless mouse, a cordless keyboard, a design book, and chocolates. The professor instructed students to write down the last two digits of their Social Security number at the top of the page and then write them again as a price next to each item. So, if the last two digits were four and five, the student would write $45. When they finished that task, students were asked to indicate for each item “yes” if they would pay that price or “no” if they would not. As the last step, students wrote down the maximum amount they would be willing to pay for each item. Ariely found that students’ Social Security numbers influenced the amount they were willing to pay. Students with the highest last two digits of their Social Security number (80-99) bid the highest, and those with the lowest last two digits (1-20) bid the lowest. When students were debriefed on the experiment and asked if they thought their Social Security numbers influenced the prices they would pay, they stated no.
In the federal criminal legal system, the Sentencing Guidelines have, post-Booker, remained the essential starting point in all federal sentences. Because the Guidelines produce a numerical value, they create a cognitive “anchoring effect” bias that some say exerts a disproportionately strong impact on a judge’s decision-making that may undermine the fairness of sentencing decisions. Researchers have other ways that anchoring biases purportedly impact judicial decision-making.
But more decidedly irrelevant factors and circumstances may influence a judge’s imposition of sentence. With the disclaimer that I’m not vouching for any of the conclusions made by the authors of these studies (and note contrary studies), here are a few examples where judges purportedly impose longer sentences or render more severe decisions:
—Bad weather (higher sentences when raining) Cf. https://hal.science/hal-03864854/document with https://eprints.whiterose.ac.uk/137050/8/sentencing%25
—Lack of sleep (higher sentences on the Mondays after the start of Daylights Savings Time) See https://hbr.org/2017/02/sleep-deprived-judges-dole-out-harsher-punishments
—Hunger (judges are more severe right before a meal break)
Cf. https://www.pnas.org/doi/10.1073/ pnas.1018033108
with https://www.scientificamerican.com/article/lunchtime-leniency/ and https://www.annieduke.com/no-judges-dont-give-harsher-sentences-hungry-annies-newsletter-october-5-2018/
—Sports (move to continue when the judge’s football team loses) https://www.theatlantic.com/education/archive/2016/09/judges-issue-longer-sentences-when-their-college-football-team-loses/498980/
—Attractiveness (the more unattractive the defendant, the higher the sentence) See
https://www.thelawproject.com.au/insights/attractiveness-bias-in-the-legal-systemhttps://www.thelawproject.com.au/insights/attractiveness-bias-in-the-legal-system
Whatever the import of these and other studies may be, they certainly try to scientifically test Jerome Frank’s belief that a judge’s decisions are but a part of their total behavior and that the process of making decisions is, in reality, a composite of the psychological, environmental, and socioeconomic factors that go into the development of the personality of the individual judge.
The bottom line is if you have a sentencing set for 11:30 the day after the Dolphins play Kansas City on Monday night and the forecast calls for rain, you may want to move to continue.
