— Sharon Hanson appeals from her conviction of murder in the second degree, contending that the court erred in excluding testimony regarding the battered woman syndrome. We affirm.
Hanson and Mel Cochran had a stormy relationship, which frequently resulted in the police being summoned. During these confrontations, Hanson testified that Cochran had pinned her against a wall, had threatened to injure her, and had struck her.
On March 7, 1988, Hanson and Cochran quarreled about Cochran's share of their living expenses. Hanson testified that Cochran became angry and that she became frightened. She went to her bedroom and returned with a handgun, which she had previously displayed to convince Cochran to leave. Hanson testified that Cochran started laughing when he saw the handgun, and then started yelling at her while he grabbed her arms and began shaking her. She testified that during this shaking, the handgun *506discharged. She stated that Cochran said he was going to call the police and ran out the front door, and that she did not realize he had been hit.
Sometime later, Hanson said she noticed blood on the porch. Expecting the arrival of the police, Hanson called a friend, Claudia Nooney, and told her, "Mel and I got into it tonight and I shot him with my .38." While making arrangements for the care of her pets, Hanson testified that she went out onto the porch, saw Cochran's body at the bottom of the stairs, and went back inside. She then called 911 to summon help. During this conversation, Hanson told the dispatcher that, "[h]e came at me and I shot at him." Police found Cochran dead at the foot of the stairs and took Hanson into custody. During the trip to the police station, Hanson (who appeared to be intoxicated to the officer) volunteered that, "[h]e was jacking me off so I shot him."
The State charged Hanson with murder in the second degree based on alternative theories of intentional murder and felony murder, with assault (the display of the handgun) as the underlying felony. Hanson sought to introduce testimony from Karil Klingbeil, an expert on the battered woman syndrome. The State objected. Hanson offered the testimony to assist the jury in understanding why she retrieved the gun and why she stayed in the relationship. The court reserved its ruling. The State presented Hanson's statements made the night of the shooting, and introduced ballistics evidence to show the amount of force necessary to pull the trigger (11V2 pounds) and to show the distance between Hanson and Cochran when the shot was fired (more than 1 foot). At the end of the State's case, the issue was addressed again. Rather than accept the court's offer to have Klingbeil testify in the absence of the jury, Hanson's counsel made the following offer of proof:
I would offer . . . that Dr. Klingbeil would also testify that the actions of the defendant in going to get the gun are part of a learned state of helplessness, which is a critical element of the *507battered woman's syndrome and consistent with that and reasonable, which again would put us, I believe, within the law which would hold Miss Hanson not culpable should an assault be found.
The court excluded the testimony, ruling that battered woman syndrome testimony would only be relevant to a claim of self-defense, which Hanson did not allege.
After both parties rested, the court ruled that there was insufficient evidence of assault to allow a jury instruction on felony murder. Therefore, the jury was instructed only on intentional murder and manslaughter. The jury convicted Hanson of murder in the second degree, and she appeals.
Hanson contends that the court erred in excluding expert testimony on the battered woman syndrome, even though there was no claim of self-defense. She contends that the testimony would have "enhanced [her] general credibility",1 and would have shown that her introduction of the handgun was a defensive measure rather than an act of aggression. She argues that if the jury believed that the handgun was retrieved out of fear for her safety, the jury would more likely have accepted Ms. Hanson's testimony that the gun discharged accidentally. She also contends that the exclusion of the testimony deprived her of her constitutional right to testify in her own defense and to present witnesses in her defense.2
The claimed error in rejecting Ms. Klingbeil's testimony is not properly before us because the offer of proof is insufficient. It mentions only the assault and makes absolutely no effort to connect the battered woman syndrome to the issue eventually submitted to the jury: Was the shooting accidental or intentional? When the felony murder charge was eliminated, the trial judge was not offered any remaining basis for admission of the testimony.
*508This view of the offer is supported by the prosecutor's affidavit, which was submitted in connection with Hanson's posttrial motions. That affidavit asserted that in a chambers conference, defense counsel acknowledged that a dismissal of the assault-based felony murder count would remove this testimony as an issue on appeal.3 When significant matters are agreed upon in chambers without a reporter, the proper procedure is for the court or counsel to summarize the agreement on the record when the court next convenes. However, there is no controverting affidavit from defense counsel, and the hearing was held before the trial judge shortly after the trial. Accordingly, the State's assertion is properly before us, and Hanson has not preserved this issue for appeal.
Even if the merits of admission were properly before us, the result would be the same because the trial court was correct in ruling that the testimony was not relevant. The scientific basis and relevancy of such testimony in proper cases is now well established.4 On the facts before us, if Hanson had claimed self-defense, the testimony would have been appropriate and admissible as supportive of her apprehensions and mental state in firing the gun. However, such testimony is not supportive of the claim of accident presented here. Indeed, it would be a strange anomaly if the same evidence could support two diametrically opposed theories of how the death ensued, by intentional self-defense or by accident.
Battered woman syndrome testimony is not admitted for purposes of "general credibility" as asserted by the defendant. It is admitted for the purpose of helping the *509jury understand and evaluate conduct which otherwise would seem strange or bizarre, such as staying in a battering relationship. Indeed, in the analogous use of expert testimony to explain the behavior of young children in reporting sexual abuse, as in the cases cited by Hanson, the courts are extremely careful to insure that the testimony does not amount to an endorsement of the child's account of what happened but is used solely to help the jury understand why the child failed to act in what would seem a more normal way to the average juror.5
Finally, Hanson's reliance on Rock v. Arkansas, supra, is misplaced. That case dealt with limitations on the defendant's right to testify, which right was found to be superior to an evidentiary rule. It does not stand for the proposition that the Rules of Evidence do not apply to other aspects of a defendant's case. No limitations were imposed on Hanson's testimony.
Affirmed.
Winsor, J., concurs.
Ibn-Tamas v. United States, 407 A.2d 626, 634 (D.C. 1979).
Rock v. Arkansas, 483 U.S. 44, 97 L. Ed. 2d 37, 107 S. Ct. 2704 (1987).
"(4) I recall Mr. Kimberly acknowledging that a dismissal would remove that issue as an appellate issue, and articulating that tactically that is how he chose to proceed."
Such testimony has been held admissible to explain a defendant's perception of threat and the reasonableness of the force employed in self-defense against that threat, State v. Allery, 101 Wn.2d 591, 597, 682 P.2d 312 (1984), and to explain a delay in reporting abuse and a failure to leave the abusive environment, State v. Ciskie, 110 Wn.2d 263, 278-79, 751 P.2d 1165 (1988).
State v. Petrich, 101 Wn.2d 566, 576-76, 683 P.2d 173 (1984).