Using Research Into Intuitive Fallacies to Guide Advocacy
Source: For The Defense
The Monty Hall Paradox
You are a contestant on a game show with three doors. Behind one door is $10,000 cash; behind the other two are goats. The host asks you to choose the door you want, and you select door number one. The host opens door number three, revealing a goat, and then gives you the opportunity to keep what's behind door number one or switch to door number two. What do you do? (Stop and think about this problem for a moment before continuing.)
Most attorneys would answer that the decision makes no difference, because there is a 50–50 chance that the cash is behind door number one. That answer is wrong. From the moment you chose door number one, you had a one-third chance of having selected the door with the cash behind it. The other two doors, combined, had a two-thirds chance of hiding the cash. When it is revealed that door number three does not have the cash, then door number two, alone, has a two-thirds probability of hiding the cash. Thus, when given the choice, you should always switch doors, because you are twice as likely to win the game.
The inability to perceive that switching is more likely to result in a win is the genius of the current television game show hit, Deal or No Deal. On that show, the odds that a contestant has picked the briefcase with $1 million in it are always 1 in 100. As more and more unpicked briefcases are opened without revealing the $1 million, the contestant believes that the odds he or she picked the right briefcase at the start must be increasing, resulting in the rejection of increasingly high-dollar "deals." In fact, the exact opposite is true: the odds are increasing—up to 99 percent—that one of the unpicked and unopened briefcases actually contains the $1 million, and the contestant will be left with very little.
The hypothetical game contestant's blindness to statistical reality demonstrates the danger of relying on intuitive judgments. It is one thing when such blindness infects our own judgment; the situation is much more dire when it is the blindness of an appellate judge that causes a court to reach an illogical result for a client. The purpose of this article is to explore some of the science regarding the most common cognitive illusions, and to consider strategies for shaping appellate advocacy that take advantage of or overcome those illusions.
Introduction to the Psychology of Mistakes
All of us are easy prey to various "cognitive illusions," that is, to the illusion of "knowing" something we actually don't. As noted researcher Massimo Piattelli-Palmarini has explained it, these are "errors we commit without knowing that we do so, in good faith, and errors that we often defend with vehemence, thus making our power of reasoning subservient to our illusions." Massimo Piattelli-Palmarini, Inevitable Illusions 18 (1994). Researchers and Nobel Prize recipients Amos Tversky and Daniel Kahneman describe cognitive illusions as "neither rational nor capricious." Id. Rather, they are spontaneous, unconscious mental strategies that serve as a proxy, or shorthand, for solving problems. Id.
For example, when we consider whether a number is even or odd, we don't look at the entire number; we analyze only the last digit, because the even- or odd-ness of the number can be determined without considering all of the available information. Such problem-solving proxies are essential for an appellate judge who must read tens of thousands of pages a week and decide multiple cases. Unfortunately, as the Monty Hall Paradox demonstrates, not all of our mental proxies produce solutions as reliable as whether a number is even or odd. The key is to recognize those decision-making proxies that may adversely affect the outcome of an appeal, and then combat them.
A Preliminary Note Regarding Intuition and Accessibility
"A defining property of intuitive thoughts is that they come to mind spontaneously, like percepts. To understand intuition, then, we must understand why some thoughts are accessible and others are not." Kahneman, Maps of Bounded Rationality: A Perspective on Intuitive Judgment and Choice, Les Prix Nobel: The Nobel Prizes, 2002 452 (2003) [hereinafter "Maps"]. Certainly, physical salience can have a large impact on accessibility. For example, quickly shout out the color of the word printed below:
If you said "red," you fell victim to a decision-making proxy that relied on the accessibility of the written word "Red." If you said "blue," then you successfully overcame both accessibility and a natural proxy by giving careful attention to the actual color of the word.
The importance of accessibility suggests strongly that an appellate brief be kept as simple as possible. Uncluttered and simple language, a narrow set of issues, frequent headings, and a logical argument progression aid accessibility rather than hinder it. Indeed, the more complex the case, the more important it is to keep things simple, particularly if you represent the appellant. Depending on the court and the type of case, affirmance rates may range anywhere from 60–90 percent. Thus, the initial proxy for answering any complex appeal issue is "affirm." If an appeal brief is in any way difficult to penetrate, you've lost the accessibility battle and likely the appeal. Resist the temptation to raise every possible issue in a 60-page brief, even when you're appealing from an adverse verdict in a monthlong trial.
We assume that irrelevant features of options or outcomes do not affect preferences. This phenomenon has been called "invariance." See generally Tversky & Kahneman, Rational Choice and the Framing of Decisions, 59 Journal of Business S251 (1986). Though invariance is an essential part of rational problem solving, it is frequently violated in demonstrations of "framing effects," such as the disease problem described in Tversky & Kahneman, The Framing of Decisions and the Psychology of Choice, 211 Science 453 (1981):
Imagine that the United States is preparing for the outbreak of an unusual Asian disease, which is expected to kill 600 people. Two alternative programs to combat the disease have been proposed. Assume that the exact scientific estimates of the consequence of the programs are as follows:
If Program A is adopted, 200 people will be saved.
If Program B is adopted, there is a one-third probability that 600 people will be saved and a two-thirds probability that no people will be saved.
Which of the two programs would you favor?
In this version of the problem, a substantial majority of respondents favor program A, the most risk-avoiding decision. A different group of subjects is given the same background story, but different choices:
If Program A' is adopted, 400 people will die.
If Program B' is adopted, there is a one-third probability that nobody will die and a two-thirds probability that 600 people will die.
This time, a clear majority favors program B', the most risk-seeking option, even though the options are substantively identical. The difference results from the fact that the "certainty of saving people is disproportionately attractive, and the certainty of deaths is disproportionately adverse." Maps at 457.
Another famous study where issue framing induced different outcomes involved a choice between surgery and radiation therapy to cure cancer, where the surgery option was described alternatively as 90 percent likely to result in short-term survival, and 10 percent likely to cause immediate death. Maps at 457. The "survival" frame of reference resulted in a substantially higher preference for surgery, despite the lack of a substantive difference in result. Most disturbing, the "framing effect was no less pronounced among experienced physicians than it was among patients." Id.
These research studies suggest that no matter how dispassionate and experienced the appellate judge, issue framing will change outcome preferences. It is therefore imperative to give careful consideration to issue framing in appellate briefing. Consider the following two issues presented:
Ex 1: Did the trial court correctly grant summary judgment in Defendant's favor on Plaintiff's claim of minority shareholder oppressions?
Ex 2: An appraisal is the exclusive remedy for a minority shareholder claiming unfairness in a corporate merger. Plaintiff failed to exercise his exclusive appraisal remedy after his shares were purchased in a lawful statutory merger, and he does not seek appraisal in this litigation. Did the court err in granting Defendant summary judgment on Mr. Smith's claim for shareholder oppression?
Example 1 correctly identifies the procedural posture of the appeal and explains what the trial court did. Procedurally, it is very accessible. But Example 2 maintains that simplicity and goes further by conveying the substantive information that frames the issue presented in the light most favorable to the defendant. This is a Bryan Garner "deep issue" one that, in 75 words or less, gives an appellate judge everything needed to make a decision before even a single page of argument has been read. Garner, The Redbook: A Manual on Legal Style §22.1(c) (2d ed. 2006). The "framing effect" of Example 2 is more informative and more persuasive.
Sometimes, persuasive framing will take place in the context of a dominant judicial philosophy, e.g., "Did the trial court err when it relied on legislative history in derogation of the plain language of the statute?" Other times, it may involve non-merit context that colors a party or a position, rather than the issue presented, e.g., "This is Plaintiff's third attempt in a third forum to state a claim for breach of contract. This Court should follow the decisions of the two previous courts and likewise dismiss." Or, "This dispute arises out of a former employee's attempt to extract hundreds of thousands of dollars from his former company because he cannot find gainful employment." Each of these examples gives the court a frame of reference that makes a favorable decision more likely. Given the considerable impact framing effects can have on outcome preferences, it is crucial to spend substantial time crafting the issues presented and appellate themes.
"A judgment is said to be mediated by a heuristic [proxy] when the individual assesses a specified target attribute of a judgment object by substituting a related heuristic attribute that comes more readily to mind. This definition elaborates a theme of the early research, that people who are confronted with a difficult question sometimes answer an easier one instead." Maps at 466. Consider a study that involved the following character summary:
Linda is 31 years old, single, outspoken and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice and also participated in antinuclear demonstrations.
Respondents were shown this description and a list of eight possible outcomes describing Linda's present employment and activities. The two critical items listed were #6 ("Linda is a bank teller") and #8 ("Linda is a bank teller and active in the feminist movement"). One group of respondents was asked to rank the outcomes in the order to which Linda resembled a typical person in that field. Another group was asked to rank the outcomes in the order of probability that the statement was true. The percentage of respondents who ranked #8 higher than #6 was the same in both groups (between 85–89 percent), even though the conjunctive item, #8, was far less probable than outcome #6. (It is far less likely that a given individual is both a bank teller and active in the feminist movement than that the person is simply a bank teller, regardless of how she uses her personal time!)
The outcome of this study shows not only an attribute substitution but also a basic misapplication of statistics, a topic that has many applications to appellate advocates. Say you are a juror, and the key fact in a negligence case is whether the taxi that hit a pedestrian was blue or green. The evidence shows that there are only two taxi companies in town; the whole fleet of one company is green, the whole fleet of the other, blue. Eighty-five percent of all taxis on the road the night of the accident were green, and only 15 percent were blue. A single witness testifies that the pedestrian was struck by a blue taxi. Vision tests have demonstrated that she can successfully distinguish a blue taxi from a green one 80 percent of the time. What is the probability that the taxi was blue?
Many participants in a clinical test say the probability is 80 percent, and a majority consider the probability to be at least higher than 50 percent. But the actual probability that the pedestrian was struck by a blue taxi is only 41 percent, no higher. The formula, known as Bayes' Law, is as follows:
WA / (WA + OE)
WA is the weighted average: the product of the probability that the witness testimony is correct (0.8) times the base probability that the taxi was blue, regardless of any witness testimony (0.15). OE is the optimistic error: the product of the probability that the witness testimony is incorrect (0.2) times the base probability the taxi was not blue, regardless of any witness testimony (0.85). In this example, the weighted average is 0.12 (0.8 × 0.15), and the optimistic error is 0.17 (0.2 × 0.85). 0.12 / (0.12 + 0.17) = 41 percent. Inevitable Illusions at 83–84.
The wide disparity in perceived probability and actual probability triggered by attribute substitution might warrant a jury instruction or, at a minimum, a different focus in closing arguments. And though the probability disparity is unlikely to be sufficient to overcome an adverse verdict, it is still relevant when setting the stage within the appropriate framing context, as noted above.
"There is compelling evidence for the proposition that every stimulus evokes an effective evaluation, which is not always conscious." Maps at 470 (citing various studies). This evidence has strong implications for how a fact section is ordered. For example, a two-question survey was given to college students: "How happy are you with your life in general?" and "How many dates did you have last month?" When the questions were asked in this order, the correlation between the two questions was negligible, but when the questions were reversed, their correlation rose to a very significant 0.66. This result "suggests that the dating question automatically evokes an affectively charged evaluation of one's satisfaction in that domain of life, which lingers to become the heuristic attribute when the happiness question is subsequently encountered." Maps at 470.
Fact patterns in an appeal brief can also be used to evoke an automatic and biased evaluation of case outcomes. This is not to suggest that an appellate brief's factual statement should be written like an argument, drawing conclusions for the court. To the contrary, the factual statement should be a completely objective presentation of the facts. But, the statement should be structured in such a way that the court draws the desired correlations.
For example, consider a case involving the breach of a contract for the sale of goods. You represent the seller, and it is critical to the result that your client's standard terms and conditions became part of the parties' contract. Unfortunately, although the client referenced the standard terms and conditions in its quote, it admittedly never transmitted the terms to the buyer, and the buyer never saw them. There is solid case law supporting the proposition that in these circumstances the burden is on the buyer to ask for the terms and review them. But, an appellate court's willingness to use that law will likely increase if you first explain in the factual background that: (1) the parties had a written agreement, (2) the seller spent significant amounts of time and money to create and prepare the part to be sold, (3) the buyer then precipitously terminated the contract to avoid paying for the seller's investment, and (4) unless the seller's standard terms are incorporated, the buyer will be able to get away with its gambit. In other words, a compelling storyline, told in a sequence that promotes correlation, increases the likelihood that an appellate panel will accept your legal conclusion. See generally Bursch, Storytelling in Brief Writing, For The Defense 42 (April 2004).
The affective heuristic is strengthened by the suitable-script theory. "Offering a 'plausible' sequence of events that are causally linked one to another has the effect of immediately raising our estimate of probability." Inevitable Illusions at 135. It is a natural tendency to calculate the probability of the last link in the chain by combining the probabilities of each link in the chain, from the first to the last, even though the probability of the entire chain being true is "always and without exception less probable than the probability of the least probable link in the chain." Inevitable Illusions at 135. By focusing on how to order the chain of events in the factual background, it is possible to create the perception that an outcome is more ineluctable than the sum of its parts. And although storytelling as a form of advocacy is traditionally advocated only in the personal injury context, it applies equally well to commercial disputes. For example, the fact that a defendant breached an agreement seems more likely if you start by explaining the chain of events that motivated the termination, rather than beginning with an alleged act that may or may not have constituted a breach in the strictest sense.
A group of subjects is asked to multiply the following numbers in their heads in five seconds or less: 2 × 3 × 4 × 5 × 6 × 7 × 8. Another group of similarly situated individuals is then asked to do the same thing, but in reverse order: 8 × 7 × 6 × 5 × 4 × 3 × 2. The correct answer is exactly the same in both cases, and the vast majority of people, if confronted with both equations, would recognize that fact (i.e., the calculations operate commutatively). But that's not the response that the questions generate. The average of the replies to calculation one, from a group of many respondents (students at Stanford University, the University of British Columbia, or Tel Aviv University) is 512. The average of the replies to calculation two is 2,250. The exact answer is 40,320. Inevitable Illusions at 202.
This experiment shows two things. First, we tend to underestimate the effects of multiplication. Second, we tend to stay "anchored" to the first figures that we see in the equation. In other words, we are trapped by the first information that we process. This cognitive illusion has many implications for appellate advocacy.
To begin, it is absolutely critical to convey the key appellate theme early and often in a brief. The theme should be readily apparent in the table of contents, the issues presented, the summary of argument, and even the standard of review.
In addition, the anchoring effect counsels in favor of presenting the strongest argument first. This advice is particularly true in the brief, which might contain three or four issues (though hopefully no more than that). But it is even more pointed at oral argument, where an advocate may have only 10 or 15 minutes of presentation time. If your client lost a summary judgment motion in the trial court, and your appellate theme is that a material question of fact exists that mandates a trial by jury, then hit that point voraciously and repeatedly in the opening 30 seconds of argument, and continue to come back to that theme in response to questions. The goal is to ingrain the concept in the minds of the panel to such a degree that "factual dispute" is the first and last thing the panel members think about and discuss when they confer following the argument. The most effective advocacy is rarely the most eloquent or lengthy argument, but the argument that stresses primacy and recency.
Ease of Representation
"The easier it is to imagine an event or a situation, and the more the occurrence impresses us emotionally, the more likely we are to think of it as also objectively frequent." Inevitable Illusions at 128. A perfect example of this cognitive illusion is the commonly held belief that great basketball players develop a "hot hand," where seemingly every shot they take results in a basket. If you've ever seen an NBA player make four, five, or six shots in a row, you may have been taken in by this theory. But a statistical analysis of the Philadelphia 76ers team during the 1980–81 season disproved it. Contrary to fans' expectations, players were not more likely to make a shot after making their last one, two, or three shots than after missing their last one, two or three shots. In fact, "there was a slight tendency for players to shoot better after missing their last shot." Gilovich, The Fallibility of Human Reason in Everyday Life 12 (1991). Moreover, there was no tendency by any individual player (including, for example, Julius Erving) to have streaks of lengths exceeding the number to be expected if individual shots were statistically independent. Id. at 12–13. In sum, a player's performance on a given shot is independent of his performance on previous shots, notwithstanding the common fan's belief that streaky shooting is common. Fans maintain that belief because they believe they have seen it. Streak shooting has an ease of representation.
A simpler illustration is that the vast majority of people asked believe that there are more words beginning with the letter "r" than words that have an "r" in the third position. Inevitable Illusions at 42. That is because it is much easier to imagine words that begin with a letter than to imagine words that have the same letter in the third position. The ease of representation again drives an incorrect outcome preference.
The ease of representation illusion suggests that an essential appellate tool is linking a case with something simple and common to which anyone (or, more specifically, any judge) can relate. Thus, one could argue, a seller's failure to give notice under UCC section 2-607 bars the seller's claims for breach of warranty not because of the strict construction of an unforgiving rule, but because the rule was designed to prevent unfairness to a seller who has no reason to believe anything is wrong and fails to maintain key documents or preserve witness testimony. The latter is a much more persuasive appellate theme because it is memorable and persuasive, revolving around a simple theme of morality that is easily represented in the experiences of many people.
One final example:
A certain town is served by two hospitals. In the larger hospital, about 45 babies are born each day, and in the smaller hospital, 15 babies are born each day. As you know, about 50 percent of all babies are boys. The exact percentage of baby boys, however, varies from day to day. Sometimes it may be higher than 50 percent, sometimes lower.
For a period of one year, each hospital recorded the days on which (more/less) than 60 percent of the babies born were boys. Which hospital do you think recorded more such days?
The average response from a large number of subjects is: 24 percent larger clinic, 20 percent smaller clinic, 56 percent no difference. Inevitable Illusions at 204. But, there is a difference. While it is true that on average a baby boy will be born 50 percent of the time, the subject of the question is the statistical fluctuation, and such a fluctuation is more likely to occur in a small sample. We must be careful not to "confuse the cause (in this case, biological) of a particular phenomenon with the probability of a causal fluctuation in the consequences thereof." Inevitable Illusions at 79.
The way to reveal the flaw in causal reasoning is to create an extreme hypothetical that makes the point more clear. What if the question had instead been to predict which hospital had more days in which all births were of the same sex? With this small change, more than 80 percent of respondents gave the correct answer (the smaller hospital). Inevitable Illusions at 204. Simply by making the question more extreme, accuracy in selecting the correct outcome increased dramatically.
The same tools can be used effectively in an appeal brief. Where an intuitive judgment is likely to carry the day for an opponent, create an extreme hypothetical or devise a way to remove the outcome from the assumed paradigm. For example, following a lengthy jury trial, the automatic (nearly unchangeable) intuitive judgment of the appellate court is to affirm. One way to diminish the impact of that intuition is to create the extreme hypothetical: "even assuming that every witness testified that X was true, there was insufficient evidence in the record to support the finding that Y was also true." Another way is to change the paradigm: "The trial court should never have given the jury the opportunity to decide this question because the plain language of the parties' contract dictated that judgment be entered in favor of the Defendant as a matter of law." In both examples, the advocate anticipated the intuitive judgment, then dealt with the misplaced causality. Either is effective.
A complete survey of judgment heuristics is beyond the scope of this article, though Daniel Kahneman's Nobel Prize lecture of 2002 is an excellent place to begin further study. See http://nobelprize.org/nobel_prizes/economics/ laureates/2002/kahnemann-lecture.pdf.
At bottom, effective appellate advocacy demands attentiveness to an appellate court's most likely intuitive values and judgments, and the adoption of an argument strategy that maximizes the likelihood intuition will favor your client's position, while defusing intuitive judgments that may run contrary to the position for which you advocate. In the words of the great Judge Ruggero J. Aldisert:
We know by now that court decisions are not necessarily a precise barometer of the beliefs and demands of society. Always present are the jurisprudential idiosyncrasies of the men and women in black robes who sit on our tribunals . . . . Whatever be the judge's view of his or her court, whether as a passive institution or a force for change, the judge must adhere to the canons of logical order in deciding a case and therefore present a reasonable, and therefore, acceptable, 'performative utterance.'
Confirming or rebuking cognitive illusions is one important way to ensure that the judicial utterance at the end of an appeal favors your client.