The Effects Of Eyewitnesses On Jurors

Published Date: 02 Nov 2017

Christina Scroggs

Louisiana Tech University

Abstract

Eyewitness testimony is often the most influential part of a trial. Misplaced confidence in eyewitness testimony is a major cause of wrong verdicts due to the fact that eyewitness testimony is heavily weighted by jurors. This experiment was conducted in order to compare three conditions (no eyewitness, discredited eyewitness, and credible eyewitness). Participants were presented a crime scene scenario and then a mock trial as if they were members of a jury. Group one was presented with no eyewitness, group two was presented with a discredited eyewitness and group three with a credible eyewitness. Then participants were asked to decide upon guilt using a Likert scale ranging from 1 (not guilty) to 7. The results of the experiment show that a jury presented with a credible eyewitness and no eyewitness have a higher guilty verdict decision than a jury presented with an eyewitness that is discredited. Other research has resulted in the same conclusion and the results of this study do not support the hypothesis that states no witness results in the highest amounts of not-guilty verdicts.

The Effects of Eyewitnesses on Jurors

A serious issue in the criminal justice system that has been extensively studied in the field of psychology is the effect of eyewitness testimony on a juror. The use of eyewitnesses by both the prosecution and the defense is calculated and intentional; however, may not produce the outcome desired. Over-confident witnesses can influence jurors’ confidence in their assignment of guilt. Jurors tend to believe witnesses who are confident in their own testimony. Due to the misinformation effect, a witness may have a very false sense of confidence. The way jurors perceive the credibility of an eyewitness can greatly affect the outcome of a verdict in a trial. A main purpose for the attention to this issue is that many times, guilt (or innocence) is determined by the jury according to whether or not eyewitnesses are presented and, if so, are credible. "Eyewitness testimony in general, and visual identification of a defendant by a victim of witness in particular, often provides what is apparently the most persuasive, and certainly the most dramatic, evidence at trial" (Greene & Loftus, 1984, p. 396).

The jury-eyewitness study is designed to determine this effect. To compare the effects of these conditions, researchers present each of three groups with a separate independent variable; a credible witness, a discredited witness, and no eyewitness at all. Whether or not juries are presented with an eyewitness may affect how confident they are in their decision to assign guilt. It is important for the prosecution as well as the defense to determine the weight of an eyewitness’s testimony as there can be negative consequences to putting a witness on the stand. For example, Siegler and Couch (2002) found that it’s better to have no eye-witness at all than to present a witness to the jury who will be discredited. Witnesses for the prosecution that are credible can greatly increase the likelihood of a guilty verdict. Likewise, a discredited witness can make jurors less confident in their testimony which can result in a ‘not guilty’ verdict. According to Siegler and Couch (2002), jurors presented with credible eyewitness testimony, are more likely to return guilty verdicts; conversely, when those eyewitnesses are discredited, jurors are more likely to return not guilty verdicts than if there had been no witness at all. The weight in which jurors give eye-witness testimony may be directly related to whether or not the jurors find the witness reliable. "Laypeople and even some judges are often unaware of the potential unreliability of eyewitness accounts. As a result, they may give more weight to such testimony than is warranted" (Greene & Loftus, 1984, p. 396). Further, if a juror feels intentionally deceived, the result may be opposite of what was intended.

Critics of eyewitness expert testimony also argue that expert testimony may be prejudicial because it will simply induce skepticism about eyewitness testimony generally. They contend the testimony will induce jurors to give insufficient and therefore inappropriate weight to such testimony. (Penrod & Cutler, 1992)

Method

Participants

The participants in this study were male and female graduate students enrolled in classes at a university in the South-Eastern U.S. The participants in this study were not offered any compensation; although some participated in this research as part of a class assignment. There were a total number of 108 participants, of which 68 percent were female. Their ages ranged from 20-55 with a mean age of 24.5 years.

Materials and Apparatus

The materials and apparatuses used for this experiment were computer terminals and the digital experiment itself accessed via the web at http://www.psych.uni.edu/psychexps/Exps/labexperiments.htm.

Procedure

The researchers conducted an experiment by devising a scenario of a crime and a trial and presented the information to participants. The independent variables were three conditions: no eyewitness, the presentation of a discredited eyewitness, and that of a credible eyewitness. Verdict and confidence in assignment of guilt were expected to be influenced by the independent variables. After giving informed consent, and being presented with information such as the name and the description of the experiment, they were then presented with a crime scenario followed by a trial scenario. After reading all of the information, participants were asked to decide, based upon the evidence and the arguments of both sides, if the defendant was guilty or innocent.

Measures

Data was collected using an interval data scale to measure the participants’ confidence in their assignment of guilt or innocence. The scale was presented to the participants directly following all arguments and it ranged from 1 (one) to 7 (seven), or not guilty to guilty. Each anchor point on the scale was given a different level of certainty with one and seven being labeled as "absolutely sure", two and six were labeled "quite sure", three and five were assigned the value of "somewhat sure", and four was the equivalent to "not sure". Participants were to choose a number corresponding to their confidence of guilt or innocence.

Results

The means and standard deviations for groups one, two, and three were M = 4.36, SD = 1.496, M= 3.67, SD = 1.195, and M = 4.67, SD = 1.287 respectively (see table one). A one-way ANOVA was used and the results for the mock juror’s confidence of guilt were F(2, 105) = 5.328, p= .006 (see table 2). The Tukey HSD determined that Group 2 (M = 3.67, 95% CI [2.475, 4.865]) was significant with Group 3 (M = 4.67, 95% CI [3.383, 5.957]), p= .005. The results of Group 1 (M = 4.36, 95% CI [2.864, 5.856]) compared with Group 2 were p= .074. Group 1 and Group 3 resulted in p= .595. An analysis of variance showed that the presence of a discredited eyewitness did increase the likelihood and confidence of a not guilty verdict, thus the hypothesis is not supported.

Discussion

The results of this research found that jurors are more likely to deliver a guilty verdict when there is a credible eyewitness; however, are less likely to find guilt when presented with a discredited eyewitness than if there had been no eyewitness at all. Previous research supports the theory that a discredited eyewitness results in more not guilty verdicts than no eyewitness at all. This is in contrast to my hypothesis which states that with the presence of an eyewitness, even a discredited one; a juror is more likely to return a guilty verdict than in a case where there is no eyewitness at all. The hypothesis in this study is clearly not supported. Groups one and two were significantly different with group two having greater guilty verdicts than group one. In other words, those presented with a discredited eyewitness actually voted not guilty more often than those who were not presented with an eyewitness at all. This finding is supported by previous research done by Siegler and Couch (2002) which found that jurors are more likely to vote for a not-guilty verdict in the case that the eyewitness is discredited. There are; however, findings that support the hypothesis that any eyewitness is better than no eyewitness at all. For example, in a study done by Loftus (1974), of the group presented with the credible witness testimony, seventy eight percent voted guilty. In the group with no eyewitness, only eighteen percent of the participants found guilt; however, even with a discredited witness presented, sixty eight percent of the group still voted for a guilty verdict (Seigler & Couch, 2002).

One way in which this study was limited is that it was done online and you can’t control for who is allowed to participate; and participants can participate more than once skewing the results. Also, level of education isn’t asked, and because of the high level of college/graduate- student participation, isn’t representative of a random sample. The fact that graduate students are above average and jurors are selected from the community may result in a bias because there is a greater chance that the students are more informed on matters of this nature.