Gregory Christophe appeals his conviction, following a jury trial, of one count of unarmed bank robbery in violation of 18 U.S.C. § 2113(a). He claims the trial court erred in refusing to admit the testimony of an expert on eyewitness identification, admitting the testimony of an FBI agent as an expert on fingerprints, and refusing to declare a mistrial when he claimed prosecu-torial misconduct.

We conclude that expert testimony on the unreliability of eyewitness identification was properly excluded. Despite additional psychological studies on this subject since 1973, when we held it was not reversible error to exclude such testimony in United States v. Amaral, 488 F.2d 1148 (9th Cir.1973), this testimony is still properly excluded if it does not meet the four criteria listed in Amaral: qualified expert, proper subject, conformity to a generally accepted explanatory theory, and probative value compared to prejudicial effect. Id. at 1153.

We also conclude it was not error for the FBI agent to testify, or for the court to refuse to declare a mistrial.

The conviction is affirmed.

FACTS

On February 14, 1986, a black male entered the Los Angeles Federal Savings and Loan Association at 6399 Wilshire, Los An-geles, California. Deanda Williams, a black teller, noticed him because of his nice looks and athletic physique. She testified later that not many blacks are customers of the Los Angeles Federal Savings and Loan.

The man approached Williams and announced a robbery. When Williams gave him $150, he demanded more. Williams[*1298] had to open her cash drawer repeatedly to produce enough money to satisfy him.

The robbery lasted sixty to ninety seconds. Most of that time, Williams looked directly over the counter into the robber’s face. He did not have a weapon or threaten her physically. She admitted she was afraid, but not extremely so.

The robber took a total of $1,742 from Williams, which he placed in an attache case. As he left the bank, he walked toward, and passed, the desk of Assistant Manager Meagan Patton. Patton was seated at the desk, about twenty-five feet from Williams’ teller window.

Patton, who is white, first saw the robber standing by Williams’ station when she was tipped off by another teller that a robbery was in progress. Five to ten seconds later, the robber turned and walked toward her. She remained seated to conceal her attention, but looked at him as he passed her desk at a distance of about eight feet. She could see the front and side of his face as he approached her. After he passed her, Patton followed him into the lobby of the bank, all the while observing him. When he left the bank, he was still visible through its large plate glass windows. Patton followed him outside and saw him leave the area in a car. She observed him for a total of sixty to ninety seconds, and testified that she had gotten “a very good look” at him. She also testified that visibility in the bank was very-good because of the plate glass windows and interior lighting. Later, teller Williams described the robber as a black male, about five feet nine inches tall, with black hair styled in a “Gerri curl,” a mustache, facial hair under the chin, glasses, and a baseball cap. Assistant Manager Patton described him as a black male, late twenties to thirty years of age, 200 pounds, six feet tall; having an athletic build, broad shoulders, a mustache, “an appearance of a small beard under his throat”; wearing glasses, a plaid shirt, blue jeans, and a baseball cap; and carrying a small bag.

When Christophe was arrested on April 1, 1986, the police described him as a black male, twenty-three years old, five feet ten inches tall, 155 pounds, with a mustache. A photograph of Christophe taken two weeks before the robbery shows a mustache and facial hair under his chin. A photograph taken six weeks after the robbery depicts him with a mustache and a “Gerri curl” hair style similar to the robber’s.

On February 14,1986, exactly one month after the robbery, Williams was shown a photospread of six individuals. She identified Christophe as the robber. At trial she testified she was sure her identification was accurate because she remembered the robber’s facial characteristics.

On May 12, 1986, approximately three months after the robbery, Patton viewed the same photospread. Patton selected two photographs which she thought resembled the robber. She selected Christophe’s photograph as the one most resembling the robber. However, Patton stated that she could not be 100 percent positive that any of the photographs depicted the robber.

At trial, both Williams and Patton identified Christophe as the robber. Patton stated she was “absolutely certain” and Williams stated she was “very certain” the defendant was the robber.

No evidence other than these eyewitness identifications and the bank’s video camera recording connected Christophe with the robbery of the Los Angeles Federal Savings and Loan. The Federal Bureau of Investigation (FBI) search of his home was unproductive. No fingerprints were obtained at the bank. Christophe’s alibi witness, his sister, testified that he was at her house from 11:00 a.m. until dinnertime on the day of the robbery. On cross examination, she testified that Christophe caught the last bus home at 11:00 p.m. Chris-tophe’s main defense was that the eyewitnesses misidentified him.

I. EXPERT TESTIMONY ON EYEWITNESS IDENTIFICATION

Christophe filed an ex parte application for the appointment of an expert on eyewitness identification. The district court[*1299] denied Christophe’s request. Christophe now claims the court committed reversible error in excluding this expert testimony.

Standard of Review

On review, we reverse only if the district court abused its wide discretion or committed manifest error in excluding expert testimony. United States v. Marabelles, 724 F.2d 1374, 1381 (9th Cir.1984). Moreover, the defendant must show, by clear and convincing evidence, the “prejudice ... caused by the court’s failure to appoint an expert.” United States v. Sims, 617 F.2d 1371, 1375 (9th Cir.1980) (citations omitted).

Discussion

Fed.R.Evid. 702 sets the standard for the admissibility of expert testimony. Such testimony is admissible “[i]f scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue.” The main inquiry is whether the jury will receive “appreciable help” from expert testimony. Amaral, 488 F.2d at 1152 (citations omitted). No federal authority requires the admission of testimony on the unreliability of eyewitnesses. United States v. Langford, 802 F.2d 1176, 1179 (9th Cir.1986) (quoting United States v. Moore, 786 F.2d 1308, 1312-13 (5th Cir.1986)), ce rt. denied, — U.S. —, 107 S.Ct. 3235, 97 L.Ed.2d 740 (1987). We have repeatedly upheld the exclusion of such testimony. See Langford, 802 F.2d at 1180; United States v. Brewer, 783 F.2d 841, 843 (9th Cir.), cert. denied, — U.S. —, 107 S.Ct. 118, 93 L.Ed.2d 64 (1986); Amaral, 488 F.2d at 1153.

In Amaral, we set forth four criteria to determine the helpfulness of expert testimony: (1) qualified expert; (2) proper subject; (3) conformity to a generally accepted explanatory theory; and (4) probative value compared to prejudicial effect. Id. at 1153. A jury will not be helped unless the witness is an expert in the field. See Fineberg v. United States, 393 F.2d 417, 421 (9th Cir.1968) (a witness must have such knowledge or experience so as to aid the jury with his opinion or inference). If the matter testified to is within the knowledge of jurors, it is not a proper subject for expert testimony. See United States v. Winters, 729 F.2d 602, 605 (9th Cir.1984). Unless the testimony conforms to a generally accepted explanatory theory, it will not be the best available approximation of truth in the judgment of the majority of scientists who work in the particular specialty involved. See id. (testimony is admissible when the state of scientific knowledge permits the assertion of a reasonable opinion); Frye v. United States, 293 F. 1013, 1014 (D.C.Cir.1923) (expert testimony is admissible if the scientific principle is sufficiently established to have gained general acceptance in the field to which it belongs). Moreover, a jury is misled when a theory presented as scientific truth is in fact not generally accepted. Finally, a jury is helped appreciably when testimony is probative rather than prejudicial, confusing, or misleading. See Amaral, 488 F.2d at 1152.

Amaral left to the broad discretion of the trial court whether, using these criteria, expert testimony on the unreliability of eyewitnesses should be admitted. In Amaral, we held that such expert testimony was properly excluded, because cross examination is sufficient to expose any deficiencies in eyewitness testimony. Id. at 1153.

The appellant claims that studies after our decision in Amaral demonstrate the unreliability of eyewitnesses and the unquestioning acceptance of their testimony by juries. He claims that expert testimony is necessary to counterbalance the jury’s tendency to rely too heavily on eyewitness testimony.

However, the proffered expert testimony does not conform to a generally accepted explanatory theory. Psychologists do not generally accept the claimed dangers of eyewitness identification in a trial setting. See McCloskey & Egeth, Eyewitness Identification: What Can A Psychologist Tell A Jury? 38 Am.Psychologist 550, 551 (May 1983) (stating that “there is virtually no empirical evidence that [jurors] are unaware of the problems with eyewitness tes[*1300] timony”). ^[1] Consequently, this criterion set forth in Amaral for the admission of expert testimony is not met. The trial court neither abused its discretion nor prejudiced Christophe in excluding the proffered expert testimony. We adhere to the position that skillful cross examination of eyewitnesses, coupled with appeals to the experience and common sense of jurors, will sufficiently alert jurors to specific conditions that render a particular eyewitness identification unreliable. Cross examination was sufficient to bring to the jury’s attention any difficulties in Williams’ or Patton’s identification of Christophe as the robber.

II. EXPERT TESTIMONY ON FINGERPRINTS

The government asked the FBI agent who investigated the case why latent fingerprints are obtained in only ten percent of the bank robbery cases. Christophe objected that the agent was not qualified as an expert on fingerprints. The trial court allowed the agent to testify. Christophe claims the court abused its discretion. He also claims it was error to allow the agent to testify to the contents of a police report.

Standard of Review

A trial court has broad discretion to admit or exclude expert testimony and is reversed only for manifest error. United States v. Gann, 732 F.2d 714, 724-25 (9th Cir.) (quoting Wood v. Stihl, Inc., 705 F.2d 1101, 1104 (9th Cir.1983)), cert. denied, 469 U.S. 1034, 105 S.Ct. 505, 83 L.Ed.2d 397 (1984). We review trial court rulings on whether evidence is supported by a proper foundation for abuse of discretion. United States v. Benny, 786 F.2d 1410, 1419 (9th Cir.), cert. denied, — U.S. —, 107 S.Ct. 668, 93 L.Ed.2d 720 (1986).

Discussion

The case agent testified he had seven years of experience with the FBI. He spent the last four years investigating 800 to 1000 robberies. He also received specialized law enforcement and investigative training at the FBI Academy in Quantico, Virginia.

The trial court did not abuse its discretion in admitting the testimony. The agent was qualified as a “skilled” witness. Fed. R.Evid. 702 advisory committee’s note.

Our ruling in United States v. Booth, 669 F.2d 1231 (9th Cir.1981), is not to the contrary. In Booth, we held that the trial court did not abuse its discretion in excluding a criminologist’s testimony as to why no fingerprints were found in a get-away vehicle following a bank robbery. Id. at 1240. There, the government failed to show the criminologist’s training qualified him as an expert on why no fingerprints were found on the vehicle. The jury was capable of inferring the fingerprints were wiped from the vehicle without the assistance of an expert. Id. Here, it is not so easy to infer why no fingerprints were found at the bank. Further, the government laid a foundation as to the agent’s expertise in investigating bank robberies. It was within the discretion of the court to admit the testimony.

The court did not err in refusing to exclude as hearsay the agent's statement that a police report contained no mention of fingerprints, so far as he knew. The agent did not testify that the police report was accurate; he merely testified as to what he saw in the report and how it affected his knowledge that no fingerprints were found at the Los Angeles Federal Savings and Loan. The testimony as to why none were found was also properly admitted.

III. THE MOTION FOR A MISTRIAL FOR PROSECUTORIAL MISCONDUCT

Christophe claims the court erred in denying his motion for a mistrial. He contends that the prosecutor improperly elicited a reference to his pretrial incarceration, which resulted in jury prejudice against him.

Standard of Review

A claim of prosecutorial misconduct must be viewed in the entire context of the trial. Reversal is justified only[*1301-1307] when such misconduct denies the defendant a fair trial. United States v. Ochoa-Sanchez, 676 F.2d 1283, 1289 (9th Cir.), cert. denied, 459 U.S. 911, 103 S.Ct. 219, 74 L.Ed.2d 174 (1982). Even if misconduct occurred, reversal is warranted only if it is more probable than not that the misconduct materially affected the verdict. United States v. Flake, 746 F.2d 535, 541 (9th Cir.1984), cert. denied, 469 U.S. 1225, 105 S.Ct. 1220, 84 L.Ed.2d 360 (1985); United States v. Nadler, 698 F.2d 995, 1001 (9th Cir.1983).

Discussion

For reversible error on prosecutorial misconduct, a defendant must establish: (1) the existence of prosecutorial misconduct; (2) that the issue was preserved for appeal; and (3) that defendant was prejudiced by the misconduct. United States v. Berry, 627 F.2d 193, 197 (9th Cir.1980), cert. denied, 449 U.S. 1113, 101 S.Ct. 925, 66 L.Ed.2d 843 (1981).

The transcript shows that the witness volunteered the information regarding the pretrial incarceration. ^[2] The government did nothing to prompt the witness’s response. Defense counsel did not object to the response, nor did he request a curative instruction.

The trial court properly denied the motion for a mistrial.

CONCLUSION

The trial court did not abuse its discretion in refusing to admit expert testimony regarding eyewitness identification, and in allowing the FBI agent to testify as an expert on fingerprints. No prosecutorial misconduct occurred when a witness volunteered information about Christophe’s pretrial incarceration.

AFFIRMED.