The Invisible Gorilla: And Other Ways Our Intuitions Deceive Us (38 page)

1.
Details of this case are drawn from a variety of sources, including several excellent, in-depth investigative articles written by award-winning journalist Dick Lehr for the
Boston Globe
. Lehr has written a book,
The Fence
(New York: HarperCollins, 2009), that discusses the case and the larger issues surrounding it. Our sources also include the following articles by Dick Lehr in the
Globe:
“Boston Police Turn on One of Their Own,” December 8, 1997, p. A1; “Truth or Consequences,” September 23, 2001; “Free and Clear,” January 22, 2006; “Witness in ’95 Brutality Case Offers New Account,” September 17, 2006. Other sources included the opinions of the U.S. district and circuit courts in the case, especially
United States v. Kenneth M. Conley
, 186 F.3d 7 (1st Cir. 1999); and
Kenneth M. Conley v. United States
, 415 F.3d 183 (1st Cir. 2005); as well as a brief filed by Conley in U.S. District Court for the District of Massachusetts
(Kenneth M. Conley v. United States
, No. 01-10853-WGY, No. 01-97-cr-10213-WGY, June 26, 2003). When any sources provided discrepant details, we have regarded
The Fence
as definitive because it was written most recently and incorporated the most research.

2.
Biographical information about Michael Cox is from a profile prepared for his participation in a conference on “Race, Police, and the Community” at Harvard Law School, December 7–9, 2000,
law.harvard.edu/academics/clinical/cji/rpcconf/coxm.htm
(accessed May 18, 2009).

3.
S. Murphy, “A Settlement Is Reached in Beating of Police Officer,”
The Boston Globe
, March 4, 2006, p. B3.

4.
Lehr, “Boston Police Turn on One of Their Own.”

5.
The juror quotes are from Lehr, “Truth or Consequences.” The widespread belief that police officers are superior to civilians at observing and remembering relevant information appears to be inconsistent with the scientific evidence; e.g., P. B. Ainsworth, “Incident Perception by British Police Officers,”
Law and Human Behavior
5 (1981): 231–236.

6.
Perjury is the crime of making a false statement while under oath in a legal proceeding. Each individual false statement can lead to a separate charge of perjury. Conley was accused of perjuring himself by claiming (1) that he did not see Cox (or any other police officer) chase Brown to the fence, and (2) that he did not see the attack on Cox. He was
acquitted of the second charge but convicted of the first. His conviction for obstruction of justice, which is the more general crime of interfering with law enforcement, in essence flowed automatically from the jury’s finding that he had committed perjury, and it did not reflect any additional malfeasance.

7.
All four suspects from the gold Lexus were arrested that night. The victim at the hamburger restaurant had been shot multiple times in the chest, allegedly because he’d witnessed another shooting at a nearby bar earlier that same night. He died several days later. The next year, two of the suspects were convicted of first-degree murder; Smut Brown, who wasn’t accused of pulling the trigger himself, was acquitted. Michael Cox eventually recovered from his physical injuries and returned to work after a six-month absence. He went on to become a deputy superintendent of police in Boston. Two of those accused by Cox of being involved in the beating were later found civilly liable and lost their jobs when Cox sued the Boston Police Department.

8.
Our study was reported in the following article: D. J. Simons and C. F. Chabris, “Gorillas in Our Midst: Sustained Inattentional Blindness for Dynamic Events,”
Perception
28 (1999): 1059–1074. Dan first learned about Neisser’s 1970s experiments when he was a college student. Neisser’s experiments used a complicated mirror apparatus to create ghostlike images of people who appeared to walk through one another. He designed those videos to test whether subjects could pay attention to one set of people while ignoring others who occupied exactly the same areas. That is, he asked whether people focus their visual attention on individual objects rather than on individual regions of space, and when they focus on objects, how selectively they focus. The most detailed description of Neisser’s earlier studies that inspired our experiment is in U. Neisser, “The Control of Information Pickup in Selective Looking,” in
Perception and Its Development: A Tribute to Eleanor J. Gibson
, ed. A. D. Pick, 201–219 (Hillsdale, NJ: Erlbaum, 1979).

9.
The term
inattentional blindness
comes from the title of a 1998 MIT Press book by Arien Mack of the New School for Social Research in New York and the late Irvin Rock of the University of California at Berkeley, two psychologists who did pioneering work in this area. In their original experiments, subjects stared at a point on a computer screen until a large cross appeared. One arm of the cross—either the horizontal or the vertical—was always longer than the other, and subjects tried to judge which was longer. The cross was visible for only a fraction of a second before it disappeared, so this was not an easy judgment to make accurately. After a few trials of this task, an additional, unexpected object appeared along with the cross. The object could be a geometric figure like a small square, or a simple picture, or even a word. In most cases about one-quarter of subjects claimed not to have seen the unexpected object. Neisser’s original selective-looking studies and our gorilla experiment provide a somewhat more dramatic demonstration of inattentional blindness because they presented a large, central, moving object for several seconds, rather than a briefly flashed static image, but the conclusion is consistent: It is surprisingly easy to not notice what is in plain view.

10.
We hired SurveyUSA to ask a nationally representative sample of fifteen hundred adults a series of questions designed to probe how people think about the workings of their own minds. The respondents matched the entire U.S. population in gender, age, and regional distribution. SurveyUSA used a prerecorded voice to read a set of sixteen statements, and after each one, respondents used their telephone keypad to indicate whether they
strongly agreed, mostly agreed, mostly disagreed, strongly disagreed, or weren’t sure. We also collected demographic information about each person’s age, sex, income level, and race. Finally, we asked people how many psychology classes they had taken and how many books about psychology they had read over the past three years. This sort of prerecorded survey provides a level of control that is ideal for scientific research because each person hears exactly the same questions, in the same order, and in the same voice. SurveyUSA has been one of the most accurate political polling firms over the past few election cycles. The entire poll was completed over the course of one week in early June 2009. The percentages of agreement we give represent the sum of respondents who answered “strongly agree” or “mostly agree” to the question. If 75 percent either strongly or mostly agree with a statement, this means that the other 25 percent either strongly or mostly disagree, or are not sure. However, it is important to keep in mind that all of the statements we presented are almost certainly false, so the rate of agreement in a world without everyday illusions should be close to 0 percent!

11.
Our colleague Daniel Levin, a psychology professor at Vanderbilt University, along with Bonnie Angelone of Rowan University, described the gorilla experiment to over one hundred undergraduate students, but without actually showing them the video or asking them to perform the task. After hearing about the experiment, including the appearance of the gorilla—but not hearing about the results—they were asked whether they would have noticed the gorilla if they had participated in the experiment themselves. Fully 90 percent of them predicted that they would have seen it. When we originally conducted the study, though, only 50 percent actually did. See D. T. Levin and B. L. Angelone, “The Visual Metacognition Questionnaire: A Measure of Intuitions About Vision,”
American Journal of Psychology
121 (2008): 451–472.

12.
Simons and Chabris, “Gorillas in Our Midst.”

13.
CSI: Crime Scene Investigation
, Season 2, Episode 9, “And Then There Were None” (originally broadcast on CBS, November 22, 2001).

14.
Lehr,
The Fence
, 270.

15.
O. Johnson, “Fed Court: Convicted Hub Cop’s Trial Unfair,”
The Boston Herald
, July 21, 2005, p. 28. Ironically, the witness, Officer Robert Walker, had initially claimed that he saw Conley at the fence. Later he recanted, saying that he had not actually seen Conley but said that he had because he was at the scene and
should have seen him
. Another victim of the illusion of attention! The appeals court found that the problem was not Walker’s faulty intuition about how the mind works, but the fact that the defense was never told about an FBI memo that documented his later requests for hypnosis and a polygraph (lie detector) test, information that would tend to cast further doubt on the credibility of his memories.

One more interesting twist in the case of Kenny Conley deserves mention. In 2006, months after Conley rejoined the police force, Smut Brown was interviewed by Dick Lehr while Brown was in jail in Maine for a drug conviction (“Witness in ’95 Brutality Case Offers New Account”). Brown told Lehr about a crucial misrepresentation in the original trial eight years earlier. Brown had testified that he had seen a white cop on the other side of the fence, and he identified Conley as the white cop who had eventually caught him. The way this information was presented in court gave the impression that Conley was the white cop Brown had seen standing next to the beating. But Brown did not specifically
identify Conley as the cop he had seen next to the beating. The prosecution never asked him to, and the defense did not cross-examine him on this specific point. Brown later said that he had gotten a good look at the officer on the other side of the fence, but not at the one who caught him, and he had just assumed they were the same person. Speaking of Conley, Brown told Lehr, “When I seen him sitting at the defense table I didn’t have no clue, like, why they were using me for that—because I didn’t recognize him.” In fact, Brown claimed that just before he testified, he spotted the cop he had seen at the site of the beating standing in the courthouse hallway and that he told this to the FBI agent in charge of the case. If true, Brown’s jailhouse claim would further undermine the legal case against Conley, by subtracting one witness who placed him at the scene of the attack on Cox. But as we will discuss in Chapter 2 of this book, this sort of sudden recollection is easily distorted, and trusting a memory like this can be dangerous, even when the person doing the remembering does not have self-serving motives for changing his previous story.

16.
C. Ross, “2 Embattled Cops Welcomed Back to Force,”
The Boston Herald
, May 20, 2006, p. 6; Lehr, “Free and Clear.”

17.
D. Wedge, “Two Officers Cleared in ’95 Beating Get Back $$$,”
The Boston
Herald
, November 20, 2007, p. 4.

18.
Lehr,
The Fence
, 328.

19.
This quote is from p. 100 of R. Pirsig,
Zen and the Art of Motorcycle Maintenance
(New York: William Morrow, 1974).

20.
Except as noted, all of the quotes and facts about this incident are drawn from the wonderfully detailed and illustrated National Transportation Safety Board (NTSB) Marine Accident Brief for Accident # DCA-01-MM-022 (
www.ntsb.gov/publictn/2005/MAB0501.htm
). Other sources include M. Thompson, “Driving Blind,”
Time
, February 18, 2001 (
www.time.com/time/magazine/article/0,9171,99833,00.html
); T. McCarthy and J. McCabe, “Bitter Passage,”
Time
, April 15, 2001 (
www.time.com/time/magazine/article/0,9171,106402-1,00.html
); and S. Waddle,
The Right Thing
(Nashville, TN: Integrity Publishers, 2003).

21.
This quote is used by permission from the transcript of a portion of an interview of Scott Waddle by Stone Phillips for
Dateline NBC
.

22.
For a recent analysis of “looked but failed to see” accidents, see A. Koustanaï, E. Boloix, P. Van Elslande, and C. Bastien, “Statistical Analysis of ‘Looked-But-Failed-to-See’ Accidents: Highlighting the Involvement of Two Distinct Mechanisms,”
Accident Analysis and Prevention
40 (2008): 461–469.

23.
D. Memmert, “The Effects of Eye Movements, Age, and Expertise on Inattentional Blindness,”
Consciousness and Cognition
15 (2006): 620–627. Memmert’s subjects were children with an average age of about eight years, but the rate of noticing the gorilla was virtually the same as in our studies of college students: 8 out of 20, or 40 percent. Psychologists use many different devices for tracking a subject’s eye movements. A typical design involves a small, lightweight helmet with one or two cameras directed at the subject’s eyes. Harmless infrared light is bounced off the subject’s eyes and detected by the cameras. Because the cameras are in a fixed position relative to the subject’s head (they’re attached firmly to the helmet, which is attached firmly to their head), experimenters can use these reflections to determine which way subjects are looking. Many systems use a
second camera to determine where the subject’s head is relative to the scene being viewed, providing the necessary additional information to calculate exactly where in an image the subject is fixating their eyes. Current eye-tracking systems can measure the focus of gaze with exceptionally high spatial and temporal precision.