. OPINION OF THE COURT BY
JUSTICE ABRAMSONJared Futrell and Kayla Lord appeal as of right from Judgments of the Wayne Circuit Court convicting each of them of wanton murder and sentencing each, in accord with the jury’s recommendation, to a maximum term of twenty-five years in prison. Lord and Futrell (Appellants) were found guilty of having participated, as principal or as accomplice, in the wanton killing of Lord’s seventeen-month-old son. Appellants were tried jointly, and because their cases thus overlap to a large extent, both factually and procédurally, we have consolidated the two appeals for consideration in this single Opinion.
Also overlapping are the issues raised, because Appellants make the identical allegations of error. Each contends that he or she is entitled to be acquitted because the Commonwealth failed to prove his or her guilt. If that relief is denied, each further *265contends that for a number of reasons the case should be retried. The trial court erred, Appellants maintain: (1) by failing to excuse two potential jurors for cause; (2) by allotting them too few peremptory juror challenges; (3) by recognizing a child-abuse pediatrician as an expert witness; (4) by allowing that witness to'opine that the injuries suffered by the child, in this case were not the result of an accident; (5) by admitting evidence of both parties’, prior bad acts; (6) by admitting into, evidence gruesome autopsy photographs; (7) by denying Appellants a full opportunity, to cross-examine one of the Commonwealth’s witnesses; (8) by giving a wanton murder jury instruction that incorporated unproved theories of the .crime; (9) by giving a combination “principal or accomplice” jury instruction; and (10) by refusing to give jury instructions on.the lesser included offenses of first-degree manslaughter, and reckless homicide. We agree with Appellants that in both cases the trial court abused its discretion, by refusing to remove for cause two unqualified prospective jurors and that under Gabbard v. Commonwealth, 297 S.W.3d 844 (Ky. 2009), it is necessary to reverse and remand for additional proceedings. Other issues will be addressed only to the extent that they could recur upon a retrial.
RELEVANT FACTS
The Commonwealth’s proof tended to show that at about 7:55 a.m. on July 16, 2011 Lord and her boyfriend,.. Futrell, brought Lord’s seventeen-month-old son, Staten Stephenson, to the Wayne County Hospital emergency room. An emergency-room nurse, Tabitha Watters, testified that Futrell, distraught and hugging the child against his chest, carried the child through the lobby area directly into an examining room. He was soon followed by Lord, although it was the nurse’s impression that Lord had stayed behind momentarily to attend to her makeup. The child was dressed in. only a diaper, and the nurse testified that he was “covered in bruises.” She charted at the- time six bruises, but she testified that there were more than that1 — on the front and back of the head, on the cheek, on the right arm, on both sides of the abdomen extending around to the back, and on the right thigh.
The emergency room physician, Dr. Glenn Proudfoot, testified that initially' the child was not' breathing, had a ghastly pallor, and appeared to be dead. The child also had a grossly distended and taut abdomen," a sign, according to the doctor, that the abdomen was full of air. Concerned that pressure from the abdomen would interfere with efforts to breathe artificially for the child, Dr. Proudfoot attempted to release the air by inserting an “NG” (naso-gastrial) tube down the child’s esophagus and into his stomach.1 .The doctor found, however, that the tube would not go all the way down. When the doctor withdrew the tube he observed what appeared to be (and what later was determined to be) chewing gum on the end of it. Dr. Proudfoot then, inserted (after three tries to get the right fit) an endotracheal breathing tube into one of the child’s lungs.
Notwithstanding the setback with the NG tube, Dr. Proudfoot eventually succeeded in restoring Staten’s vital signs, and arrangements were made to airlift him to the University of Kentucky Medical Center. The child was placed in the helicopter, but before the helicopter could depart, he again lapsed into cardiac arrest *266and was returned to the emergency room. Dr. Proudfoot then determined that relieving the pressure in the child’s abdomen was essential. Although he had never before performed the procedure, he inserted a large IV needle into the abdominal cavity. He testified that air immediately escaped through the needle with an audible rush. As soon as the abdominal pressure was relieved, the doctor testified, the child’s vital signs stabilized so as to allow his evacuation to UK. Dr. Proudfoot testified that the discovery of the gummy substance on the NG tube suggested at the time that Staten may have choked on chewing gum, but he could not know that to any degree of certainty, and he deliberately did not tell the family that that was the case. Asked by the Commonwealth how gum could have migrated from the airway to the esophagus, the doctor admitted that he did not know and had never heard of such a case, but, he testified, he did not think it impossible.
At the UK Medical Center, Staten was initially treated by Dr. Marion Turner. Dr. Turner testified that the child “looked horrible,” that he was pale, cold, and was covered with bruises. His condition was highly unstable and required full life support for several hours. Even before he could be stabilized, surgery was required to repair a stomach rupture revealed by a CT scan. During the surgery, doctors discovered that Staten’s small intestine was also damaged, an injury that required additional surgeries to repair. The CT scan and a later MRI revealed numerous traumatic injuries to the child’s head, torso, and abdomen, including brain injuries severe enough to cause the cardiac arrest, which in turn, according to the doctor, caused injuries to other organs.
Once the child had been stabilized, he was photographed. At trial, Dr. Turner used the photographs to point out for the jury numerous traumatic bruises to the child’s head, ears, back, and right arm. Dr. Turner testified that it was soon apparent that the brain injury was irreversible. Not long after that determination Lord and the child’s father, Johnny Stephenson, agreed to have the child removed from life support.
Dr. Glenn Elmore, who in July 2011 was a resident at the UK Medical Center, testified that the removal from life support took place at about 10:00 pm on July 26, ten days after the child’s appearance at the Wayne County Hospital emergency room. Dr. Elmore testified that Lord requested and was allowed to hold the child until his breathing ceased at about 1:40 am on July 27. A few minutes after the child’s passing, according to Dr. Elmore, he overheard Lord say to the friend accompanying her, “I killed him.” Later, during her testimony, Lord claimed that her remark was a reference to her consent to the removal of life support, not a confession to homicide.
About eight hours after Staten Stephenson’s death, Dr. Victoria Graham, a forensic pathologist and an assistant Kentucky state medical examiner, performed the autopsy. Illustrating her findings with photographs taken in the course of her examination, Dr. Graham distinguished for the jury between more than thirty bruises which covered the child’s body and which the doctor believed were the result of traumatic injury, and several other bruises which the doctor thought were likely the result of medical treatment. Dr. Graham illustrated abrasions and bruises on the child’s scalp and additional bruises that appeared only beneath the scalp. She also illustrated an all-the-way-through fracture of the child’s occipital bone — the back, lower portion of the skull. Dr. Graham testified that she also observed both subdural and subarachnoidal hemorrhages — bleed*267ing, that is, outside the brain but beneath the dural and arachnoidal membranes. The cause of death, the doctor concluded, was hypoxic (lack of oxygen) ischemic (lack of blood flow) encephalopathy (brain injury), secondary to blunt force injury to the head. The blunt force injuries, the doctor explained, caused the brain to swell, and the swelling, cut off the flow of blood and oxygen. Dr. Graham testified she was unaware of any case in which CPR had caused a ruptured stomach or ruptured intestines. She also testified that the skull fracture did not show signs of healing and so had to have occurred withih days, not weeks, of the child’s presentation at the emergency room. She testified that she did not examine the fracture microscopically, which would have enabled her to date the fracture somewhat more precisely, because signs of hemorrhage were still associated with the fracture, indicating that it could not be very old, and she believed that the cause of death was apparent without the more precise dating.
In addition to Dr. Graham’s causation testimony, the Commonwealth also introduced testimony by Dr. Melissa Currie, the director of the Univérsity of Louisville’s department of forensic medicine and a board certified child abuse pediatrician. Dr. Currie testified that taken together the number, nature, and severity of the child’s injuries — in particular the widespread bruising in areas -not easily bruised; the ruptures to the child’s stomach and intestine; the severe fracture of the thick, not easily struck occipital-bone; and the subdural hematoma with its associated brain swelling — ruled out all but the most extreme sorts of accident scenarios — high speed automobile crashes, for example, or falls from high places. Dr. Currie seconded Dr. Graham’s view that CPR, however ineptly, performed, does not result in ruptured organs; she testified that choking does not cause subdural hematoma; and she explained .that the head and abdominal injuries were so severe that they would immediately have rendered the child symptomatic and probably unconscious. All of this made it virtually certain, in Dr. Currie’s view, that the fatal injuries in this case (where no traumatic accident was alleged,, much less an extreme one) had been inflicted, that the child had been violently shaken and that his head had either been slammed against a hard surface or forcibly struck with a hard, blunt object.
As to who might have inflicted Staten’s injuries, the Commonwealth’s proof followed two related tracks: proof, on , the one hand, that. Lord and Futrell were the only ones with the opportunity to cause the child’s injuries, and, on .the- other, hand, proof that the catastrophic injuries of mid-July 2011 were the culmination of a series of lesser injuries, a series that began soon after Futrell came into the picture. With respect to the first track, the Commonwealth showed that Staten was bom in early February 2010,. by which time Lord and Johnny Stephenson, the baby’s father, had already split. At some point after the baby was bom, they reunited for a couple of months, but later in 2010 they again broke up, and Lord moved to Texas, where her . family lived. ;In early 2011 she returned to Monticello, Kentucky. She and Stephenson briefly gave their relationship a last try, but when that did not work out, Lord spent a few months moving among different friends. She began seeing Fut-rell in about mid-May 2011, about sixty days before Staten’s fatal injuries. Early in July of that year, apparently while waiting for their mobile home to be ready, she moved in with him at his father’s home on Highway 789 outside Monticello.
Futrell’s father, Rick Futrell, testified that his son and Lord were staying with him on July 15, 2011. His . daughter and son-in-law were- also staying with him at *268the time, and that evening the- five adults and Staten had dinner together in the elder Futrell’s kitchen. According to Rick Futrell, the child was fine.' He had spent time early in the' evening playing outside with the dogs, and later had played in the kitchen while Rick prepared the meal. Rick remembered Staten bumping his head hard on the underside of the kitchen table, but he was his usual self again in a short while.
After dinner Rick Futrell went fishing with' his son-in-law; Lord and Jared Fut-rell drove into town to run errands; and Jared’s sister, Sarah, and Sarah’s friend, Madison Daffron, watched the child. Daf-fron testified that at least until Lord and Futrell returned, Staten was normal. He had no trouble moving about, responded to his ñamé, was in good spirits, and had'no noticeable bruises. Her impression later, when she came in to go to bed, was that Staten was in bed already and was sleeping normally.
Rick Futrell testified that at about 7:00 the next morning' (July 16, 2011), he knocked on the wall of his son’s room, as he had promised to do (the two bedrooms were adjoining). He then heard Lord get up and go into the bathroom.. He estimated that she was in the bathroom for about thirty minutes, during which time, he testified, he heard nothing out of the ordinary. He heard Lord return to the room she was sharing with Futrell; and only moments later heard her scream that the .baby was not breathing. He dressed as quickly as he could and came out into the hallway. He saw his son carry Staten out of the house toward the carport, and there he stopped him. He took the child from his son, checked for and could not find a pulse, reached his finger down the baby’s throat, and, finding no obstruction, began applying CPR. "Rick Futrell testified that during his employment ás a coal miner he had served on a rescue unit and had received CPR training. He had not been trained in CPR for babies, but he knew that babies required much gentler treatment than adults. He showed his son how to press with three fingers on the child’s sternum and how, every fifteen compressions or so, to breathe into the child’s mouth and nose. The elder Futrell then drove his son, Lord, and the child to the emergency room, with his son in the back seat performing CPR on the child. ’
■ Rick Futrell’s account of that morning was substantially in accord with the accounts Lord and Futrell gave to the inyes-tigators. The lead investigator, Detective Derrick Lester of the Monticello Police Department, testified that he had been called to the Wayne. County Hospital emergency room not long after Staten was admitted. He later traveled to the UK Medical Center, where he took brief statements from Lord and Futrell. Both described being awakened by Futrell’s father at about 7:00 that morning; both claimed to have seen Staten reaching for his sippy cup as Lord was getting up to go into the bathroom; and both stated, that upon Lord’s return .from the bathroom Futrell had discovered the child limp, not breathing, and with bluish lips.
The detective testified that he attended the autopsy during the early morning of July 27, 2011, and that having heard the medical examiner’s findings, he asked Lord and Futrell that same day to come to the police department’s headquarters for questioning. The couple willingly complied, and at the police station, after having waived their Miranda rights, each was interviewed separately two times. Video recordings of the statements were played for the jury. Throughout the interviews both repeatedly denied having injured the child or having seen the other do so. They expressed surprise at the autopsy findings, *269and they suggested that the child’s bruising and traumatic injuries may have occurred as a result of the child’s throwing himself backwards during temper tantrums, as a result of rough handling by the emergency medical providers, or as the result of an accident sometime previously while the child was in the custody, of his father or was being watched by someone else. Lord denied having given the child any chewing gum, and Futrell described performing CPR gently with his fingers, as his father had instructed him. The detective’s insistence that their stories did not jibe with the medical examiner’s report elicited only Futrell’s statement that he wished he could help but could not and Lord’s angry outburst that she was “going through” something the investigators could not comprehend. At the conclusion of the interviews the detective arrested the pair, and on September 12, 2011 a Wayne County grand jury indicted them on'one count each of murder and first-degree criminal abuse.
The Commonwealth’s second line of proof regarding who inflicted Staten’s injuries included testimony by a number of the couple’s acquaintances, who described instances of Lord or Futrell’s striking the child or expressing animosity toward him. This line of proof also included testimony to the effect that prior to mid-May 2011, when Lord began spending time with Fut-rell, the child had not been subject to unusual bruising, but that after Futrell became involved the child was seen with unusual bruises a number of times. For example, the child’s father, johnny Stephenson, and Johnny’s mother, Betty Lyons,2 both testified that they saw Staten fairly regularly and that in late May or early June 2011 they began noticing more bruising. Chelsea Humble, Fufrell’s former girlfriend, testified that on June 9, 2011 she ran into Futrell at Wal-Mart. Futrell was carrying the baby, Humble testified, and, in addition to his appearing very diriy the baby1 had bruises on his lower face suggestive of finger marks, as though someone had squeezed his face hard; Whitney Bell, a friend of Futrell’s from high-school, testified that in mid June 2011 she saw Futrell with the baby. Staten was crying, and. Futrell, obviously exasperated, shook him and told him to hush..-
Johnny’s sister, Lydia Stephenson, and Lydia’s friend, Chancie Pyles, both testified that Lord lived with Pyles in May and June 2011. Futrell was a frequent visitor. Lydia and Pyles testified that Lord would curse at the baby for misbehaving, and that to discipline him she would slap his mouth, strike the bottom of his feet, and pinch his ears. They also testified about an incident that occurred in late June. Pyles testified that on June 27, 2011 Lord called hqr and asked her to come home, because she and Futrell were arguing. Pyles was with Lydia at the time, and when the two friends got to Pyles’s house they found that Futrell was angry with Lord for being drunk. At one point, Pyles heard Futrell scream at Lord that he was “tired of your f ... king kid crying all the time.” A little later Lord was in the bathroom with the baby when Futrell came in and pushed her, knocking her into the child. Concerned, Pyles and Lydia offered to watch Staten for a while, and they took him to a nearby lake. Although a little dingier than usual, Pyles testified, the baby was fine during that outing and did not have bruises an hour or two later when she and Lydia returned him to Lord. The next morning, however, Staten “was all bruised up.” There were bruises on his face, on the back of his head, and on his back.
*270Lydia testified similarly about the incident on June 27, and stated that two days later, June 29, she saw Staten at the emergency room with bruises on his head and back and a scratch on his cheek. It was less than a week after that when Lord and Staten moved in with Futrell at the home of Futrell’s father.
The Commonwealth’s theory of the case was that Staten’s catastrophic injuries could not have occurred accidentally and that Lord and Futrell were the only people with access to the child when the injuries must have been inflicted. The defense countered with expert testimony by Dr. Donald Jason, an anatomical pathologist on the faculty at the Wake Forest University School of Medicine in Winston-Salem, North Carolina. Dr. Jason testified to the effect that the child choked on chewing gum, that the choking caused the cardiac arrest and the subdural hematoma, that Futrell’s overly energetic CPR both inflated the child’s stomach and then popped it, and that most of the bruising and the damaged intestine all occurred after the child entered the emergency room, either as a result of medical treatment or as a result of lack of oxygen. The skull fracture, according to the defense expert, was a “red herring,” an injury the child could have been walking around with for days.
Both Lord and Futrell testified at trial, and for the most part they reiterated the statements they had given previously. In accord with their expert’s testimony, however, Lord changed her statement about never having given her son chewing gun, and claimed that indeed she had given him' gum. Futrell claimed that the CPR he applied was not as gentle as he originally described, but was more like what would be appropriate for an adult.
At the close of proof, the Commonwealth moved to dismiss the charge of first-degree criminal abuse. It also moved to amend the indictment so as to charge wanton murder as an alternative to intentional murder' and to charge complicity as an alternative to principal liability. ‘ In support ' of its motion to add complicity charges, the Commonwealth argued that Lord could be thought to have breached her duty to protect her son, and that, if the jury believed that Lord was the principal, Futrell could be thought to have aided and abetted the killing. The trial court granted the Commonwealth’s motion to amend and thus ultimately instructed the jury on six different theories of the crime as to each defendant: intentional murder-principal, intentional murder-accomplice, or intentional murder-principal or accomplice; wanton murder-principal, wanton murder-accomplice, or wanton murder-principal or accomplice. The jury found both defendants guilty under the last alternative, wanton murder-principal or accomplice. Appellants take issue with these instructions on a number of grounds, beginning with their claim that the Commonwealth failed to prove any theory of the crime thereby entitling them to a directed verdict. We begin our analysis with the directed verdict claim.
ANALYSIS
I. Neither Appellant Was Entitled to a Directed Verdict.
Lord' and Futrell both maintain that the trial court erred by denying their respective motions for a directed verdict. A directed verdict is required if, but only if, the evidence, construed favorably to the Commonwealth, would not permit a reasonable juror to find the defendant guilty beyond a reasonable doubt. Jackson v. Commonwealth, 392 S.W.3d 907 (Ky. 2013) (citing Commonwealth v. Benham, 816 S.W.2d 186 (Ky. 1991)). .
*271Under the pertinent portions of Kentucky Revised Statute (KRS) 507.020(1), “[a] person is guilty of murder when: (a) With intent to cause the death of another person, he causes the death of such person ... or (b) ... under circumstances manifesting extreme indifference to human life, he wantonly engages in conduct which creates a grave risk of death to another person and thereby causes the death of another person.” “Wantonjessj” here means “aware[ness] of and conscious[ ] disregard” of the grave risk of death engaging in the conduct creates. KRS 501.020(3).
Clearly, construed favorably to the Commonwealth, the evidence that the child suffered catastrophic traumatic injury at a time when Appellants were the only people with access to him permitted the jury to find that at least one of them either killed the child deliberately or did so with wanton disregard for a grave risk of death under circumstances manifesting extreme indifference to human life. Appellants maintain, however, that that is not enough.
Each of the Appellants argues that the evidence did not show that she (or he) was the guilty party. It showed only that she (or he) may have been present at the time of the killing, and mere presence, both Appellants insist, does not make her (or him) either a principal or an accomplice. Hayes v. Commonwealth, 175 S.W.3d 574 (Ky. 2005) (citing Houston v. Commonwealth, 975 S.W.2d 925 (Ky. 1998)). While we agree that mere presence at the scene of a crime is not enough to prove involvement in the crime, we do not agree that the Commonwealth failed to prove anything more than mere presence. In addition to the proof that Appellants had access to Staten during the time the injuries were inflicted, the jury heard proof that both of them had mistreated the child and expressed contempt for him.' It also heard that Lord had expressed a sense of responsibility for the child’s death. The weight of that evidence with respect to either Lord or Futrell was for the jury to assess. The jury also had before it the Appellants’ police statements and their trial testimonies (which differed as to the chewing gum and the CPR), the credibility of which was again for the jury to determine. Depending on those assessments of weight and credibility, a reasonable juror could have been convinced that either Appellant or both of them inflicted the fatal injuries. Because either Appellant could have- been found guilty as the principal, neither was entitled to a directed verdict.
II. The Trial Court’s Failure to Remove For Cause Two Disqualified Potential Jurors Requires That Appellants’ Convictions Be Reversed.
In addition to their claims that they should have been granted directed verdicts, Appellants contend on several grounds that errors during the original proceedings entitle them to new trials. The first of these contentions is that the trial court abused its discretion when it denied defense motions to strike for cause two prospective jurors whose responses during voir dire raised sufficient doubts about their impartiality to disqualify them. We agree that the trial court’s refusal to strike these two panel members was an abuse of discretion. The issue was preserved, in accord with Gabbard v. Commonwealth, 297 S.W.3d 844 (Ky. 2009), by virtue of Appellants’ motions to strike, by their use of peremptory strikes to remove the objected-to persons from the panel, by their exhaustion of their peremptory strikes, by their indication on the strike sheet of other panel members they wished to strike, and by service on the deciding jury of one of those would-be strikes. In these circumstances, we have held that the *272trial court’s failure to remove the challenged panel member deprives the party of a substantial right — the unhindered use of his or her full allotment of peremptory strikes — which deprivation is presumptively prejudicial and so amounts to a reversible error. Id.
As Appellants correctly note, an impartial jury . is a right guaranteed a criminal defendant by Section 11 of our Kentucky Constitution as well as the Sixth and Fourteenth Amendments to the Constitution of the United States. Ordway v. Commonwealth, 391 S.W.3d 762 (Ky. 2013) (citing Fugett v. Commonwealth, 250 S.W.3d 604 (Ky. 2008)). This right is implemented in part by Criminal Rule 9.36, which provides, among other things, that “[wjhen there is reasonable ground to believe that a prospective juror cannot render a fair and impartial verdict on the evidence, that juror shall be excused as not qualified.” Kentucky Rule of. Criminal Procedure (RCr) 9.36(1). This Court has recently emphasized the importance of juror impartiality by urging caution on trial courts as a “fundamental rule” of jury selection: “When there is uncertainty about whether a prospective juror should be stricken for causes, the prospective juror should be stricken. • The trial court should err on the. side of caution by striking the doubtful juror.” Ordway, 391 S.W.3d at 780.
' [5-12] Doubts about a prospective juror’s ability to “render a fair and impartial verdict on the evidence" can arise for a host of reasons, but they often arise from a juror’s having prejudged the' defendant based on information, or supposed information, acquired outside of court; or from the juror’s having some personal reason, such as a relationship with a trial participant or personal experience of a crime like the one alleged, to lean one way or the other. See, e.g., Gabbard v. Commonwealth, 297 S.W.3d at 844 (discussing prejudgment and bias as reasons for striking a prospective juror). Among jurors with some knowledge about the case the trial court ■ must distinguish between those whose minds remain open and those whose outside knowledge is apt to make them unable to base their decision on the evidence presented at trial. As for jurors with some relationship to- the case, the trial court must distinguish between those whose objectivity, whose “indifference,”3 remains intact and those so closely related to the case ór so susceptible to the relationship as to bé predisposed to be more (or less) critical of one side’s evidence than the other’s. In all cases these distinctions are to be based on the totality of the voir dire circumstances: the juror’s demeanor, the context of any questions, and the entirety of the juror’s responses. Shane v. Commonwealth, 243 S.W.3d 336 (Ky. 2007). Where the juror’s responses and the rest of the circumstances have created a genuine doubt as to the juror’s impartiality, further questioning meant to resolve the doubt by eliciting further information is certainly appropriate, Paulley v. Commonwealth 323 S.W.3d 715 (Ky. 2010) (regretting, in a close case, the lack of followup questioning), but leading questions calling for “impartial” answers do not “cure” or “rehabilitate” prospective jurors whose relationship to some important aspect of the case is so close as to be presumptively disqualifying, Marsch v. Commonwealth, *273743 S.W.2d 880 (Ky. 1987) (citing Ward v. Commonwealth, supra), or who in some other way have already made their disqualification apparent. Montgomery v. Commonwealth, 819 S.W.2d 713 (Ky. 1991). Again, “where questions about the impartiality *of a juror cannot be resolved with certainty, or in marginal cases, the questionable juror should be excused.” Ordway, 391 S.W.3d at 780.
Appellants maintain that the trial court abused its discretion in this case by refusing to excuse potential jurors 27 and 75. Both of these jurors acknowledged during voir dire a significant relationship with Lee Tobbe, an attorney of long standing in Wayne County, it appears,''and the assistant prosecutor for this case. Both jurors had been represented by attorney Tobbe in the past, and both had a more immediate connection with him. Juror 27, who owned and managed residential rental properties in Monticello, stated that attorney Tobbe was then representing his, Juror 27’s, son. We discuss Juror 27 below, but begin with Juror 75 for obvious reasons.
Juror 75, a former principal at one of the local schools, stated that, in addition to having been attorney Tobbe’s client “a long time ago,”, he had more recently had contact with attorney Tobbe on social services committees and through the school system. Juror 75 stated that attorney Tobbe “represented me or the school system in a juvenile case.” When asked during the prosecutor’s voir, dire whether his relationship with attorney Tobbe would “cause you to automatically give the Commonwealth’s case or witnesses more weight than you would anything else?” Juror'75 replied, “I think so.” The prosecutor then asked whether Juror 75’s relationship with attorney Tobbe would make it difficult for the juror to vote to acquit the defendants even if he felt the Commonwealth had failed to prove its case, and Juror 75 responded, “I really can’t answer that. I’m trying to be honest with you.” These exchanges occurred before the entire venire. At that point the prosecutor declared himself at a loss as to what else to ask Juror 75, said that he was sure defense counsel would have some questions for the juror and then moved on to other topics. Inexplicably, Appellants declined to question Juror 75 further regarding attorney Tobbe, so there the matter -stood until the very end of voir dire. Only then, after the close of questioning and just before the parties made their peremptory'strikes, did Futrell move to strike for cause some six or seven potential jurors, including Juror 75 “because of his relationship with Mr. Tobbe.” Lord joined Futrell’s motion, which the- trial court summarily denied. As noted above, the defense then used a peremptory strike to have Juror 75 removed and indicated panel members it would have struck instead had its .motions for “for cause” strikes been granted. Appellants now contend that the trial court abused its discretion when it denied their motion to remove Juror 75 for cause.4
*274Our law, not surprisingly, has long been that a potential juror’s close relationship with the prosecutor trying the case is presumptively disqualifying. Ward v. Commonwealth, 695 S.W.2d 404 (Ky. 1985) (holding that Commonwealth Attorney’s ex-brother-in-law was not so closely related as to be presumptively excluded, but that the Commonwealth Attorney’s uncle should have been removed for cause). By itself, trial counsel’s prior representation of a potential juror does not imply so close a relationship as to require the juror’s removal, Grubb v. Norton Hospitals, Inc., 401 S.W.Sd 483 (Ky. 2013), but any suggestion of an on-going relationship with the prosecutor, such as the potential juror’s intent to make use of his professional services again, is disqualifying. Fugate v. Commonwealth, 993 S.W.2d 931, 938 (Ky. 1999) (stating that “a trial court is required to disqualify for cause prospective jurors who had a prior professional relationship with a prosecuting attorney and who profess that they would seek such a relationship in the future.”); Ordway v. Commonwealth, supra (noting that prospective juror whose sister was the Victim’s Advocate in that very case and thus closely associated with the prosecutor, should have been removed for cause). But see Clay v. Commonwealth, 291 S.W.3d 210 (Ky. 2008) (holding that a secretary for the former Commonwealth’s Attorney was not so closely related to the current Commonwealth’s Attorney, even though she had a friend who worked in his office, as to be presumed biased).
Although the record regarding this matter cannot be characterized as well developed, Juror 75’s significant associations in several capacities with assistant prosecutor Tobbe were, under our case law, sufficient to impute bias to Juror 75 and support his removal for cause. However, to the extent there was any question, removal for cause became the only acceptable course of action once Juror 75 acknowledged actual bias. Cf. Paulley v. Commonwealth, supra (holding that to the extent that the scantiness of the record cast doubt on potential juror’s uncertainty regarding her freedom from bias, the benefit of that doubt would be afforded the defendant). We agree with the Appellants, therefore, that the trial court’s failure to remove Juror 75 for cause was an abuse of discretion.
Unlike prospective Juror 75, prospective Juror 27 indicated that neither attorney Tobbe’s prior representation of him, nor Tobbe’s then current representation of Juror 27’s son would have any bearing on the juror’s ability to weigh the evidence. The cases cited above, however, make clear that the prosecutor’s on-going representation of potential Juror 27’s son gives that juror a close and presumptively disqualifying relationship with .the prosecutor, a relationship so apt to produce bias that even confident assurances to the contrary by the juror cannot erase significant doubts about his impartiality. See Fugate, 993 S.W.2d at 938, 939 (holding that two jurors who had been represented by the prosecutor in the past and remained on good terms with him should have been removed for cause notwithstanding juror assurances that the “association with the prosecutor would [not] cause him [the juror] to be biased in favor of the prosecution” and that “the contact [with the prosecutor] would not affect his [the juror’s] judgment whatsoever.”). Doubtful jurors, as noted above, should be removed, not papered over.
*275The impartiality of Juror 27 was doubtful for other reasons as well. He knew the child-victim’s biological father (Johnny Stephenson) through having rented property to him, and he had known and done business with Johnny Stephenson’s father (the victim’s paternal grandfather) “for years.” He had also, through his rental business, witnessed several instances of what he believed was child abuse and had reported instances to the authorities. Some of his reports had even led to prosecutions. These “connections to the case” — presumably his relationship with prosecutor Tobbe as well as his acquaintance with the victim’s family and awareness of child abuse — not only prompted a defense motion, to strike Juror 27 for cause, but prompted Juror 27 himself to ask to be excused.
It is certainly true, as the Commonwealth notes, that at the bench conference following Juror 27’s request, the juror was not able to articulate what it was about his “connections to the case” that made him doubt his suitability as a juror, and it is true that in response to the trial court’s prompting he denied being biased or prejudiced against anyone with whom he was acquainted who was involved in the case. He also replied to defense counsel’s concern that, if he, Juror 27, were on the jury, the defense would be at a disadvantage, by emphatically stating that he “would have to hear the evidence first.” If the record contained only Juror 27’s bench conference questions and responses, the trial court’s refusal to remove him for cause likely would not have been an abuse of discretion. Cf. Derossett v. Commonwealth, 867 S.W.2d 195 (Ky. 1993) (holding that mere acquaintance with victim’s family did not presumptively imply bias). The bench conference did not directly address, however, Juror 27’s then-current relationship through his son with assistant prosecutor Tobbe. Given the presumptive bias created by Tobbe’s ongoing legal representation of Juror 27’s son, the juror’s numerous expressions of discomfort throughout the course of voir dire_ and at the beginning of his bench conference assume a more troubling aspect.
All of Juror 27’s “connections with the case” were in fact connections with the prosecution, and while none of those connections by itself appears to have given Juror 27 much pause, as the number of connections became increasingly apparent Juror 27 expressed increasing discomfort. He acknowledged the relationship with prosecutor Tobbe; he recalled renting a residence to Johnny Stephenson the infant victim’s biological father and possibly to the father’s sister, the victim’s, aunt (also a witness for the Commonwealth); .he acknowledged that he had in effect helped to prosecute child abuse cases by watching out for child abuse among his tenants and reporting it whenever he suspected it. At each revelation of a connection, Juror 27 seemed, surprised that the questioner was not more concerned that the juror was perhaps too connected. Finally, when it was brought home to him that the infant victim’s paternal grandfather was a man whom he had known and done business with for twenty years, Juror 27 had “heard enough” to feel compelled to bring his concerns to the court’s attention and to ask to be excused.
Based on-questioning at' the bench, the trial court’-was apparently satisfied that potential Juror 27 clearly understobd the need for juror impartiality and clearly hoped, at least, that he would be able to be impartial. As the court saw it, those factors ' outweighed the potential juror’s concerns that somehow, in spite of himself, his several “connections to the case” would cause him to be biased. Neither the Appellants nor the Commonwealth shared that belief, it seems, because all parties *276used a peremptory strike on Juror 27. In fact, the trial court’s error was in failing to perceive that Juror 27⅛ other connections to the prosecution and his own misgivings, vague as they were, added to and confirmed the bias presumptively inherent in Juror 27’s close relationship to the assistant prosecutor, the attorney who was representing his son. Consideration of the totality of the circumstance should have eliminated any lingering doubt that Juror 27 needed to be removed for cause.
Because the trial court did not allow the parties any 'extra, error-correcting per-emptories, see Dunlap v. Commonwealth, 435 S.W.3d 537 (Ky. 2013) (approving the use of extra peremptories as1 a way of ■rendering harmless that same number of erroneous failures to remove for cause), and because Appellants established, in accord with Gabbard, supra, that at least one of the trial court’s two erroneous failures to remove for cause was prejudicial (one of the “otherwise stricken” jurors on Appellants’ list having sat on the jury), we must reverse Appellants’ convictions and remand the matter to the Wayne Circuit Court for additional proceedings. We limit our discussion of the other allegations of error to questions likely- to recur on retrial. ...
III. The Jury Instructions Raise Certain Unanimous Verdict Concerns.
A. Although There is Sufficient Evidence to Retry: Lord on a “Legal Duty” Complicity Theory, There is Not Sufficient Evidence that Futrell Aided or Abetted Lord, if the Jury Deems Her the Principal, To Retry Him For Complicity.
Appellants raise several issues with regard to the jury instructions. Two of their contentions — that the instructions in two ways deprived them of unanimous verdicts — warrant consideration in case of a new trial. Indeed, Appellants contend that the instructions included inadequately supported theories of the crime, and if that contention is" correct then double jeopardy principles would preclude the Commonwealth from reasserting those theories if the case were tried again. The more serious of Appellants’ concerns, in our view, is their objection to the combination principal/aecomplice instruction, and we begin with that objection.
As noted above, Appellants were not entitled to a directed verdict because there was sufficient evidence, allowing for the jury’s responsibility to make credibility determinations, for a rational jury to believe beyond a reasonable doubt that either of them (or both of them together) was guilty as a principal of murdering the child. That determination does not end the analysis, however, because the jury did not find either Appellant guilty simply as a principal. As noted above, the jury found both Appellants guilty of wanton murder under a so-called combination prineipal/ac-complice instruction. The wanton murder instructions directed the jury as follows:5
INSTRUCTION NO. 6 WANTON MURDER
If you do not find the Defendant guilty [of intentional murder], you will find the Defendant guilty of Wanton Murder under this Instruction if you believe from the evidence beyond a reasonable doubt all of the following:
A. That .on or about July 16, 2011 in Wayne County, Kentucky and before the finding of the Indictment herein, the defendant killed [the child] by inflicting *277blunt force trauma to his head, torso and/or other part's ‘ of his body. AND
B.That in so doing, she was wantonly engaging in conduct which created a grave risk of death to [the child] and thereby caused the death of [the child] under circumstances manifesting extreme indifference to human life.
INSTRUCTION NO. 6-B COMPLICITY TO WANTON MURDER
If you do not find the Defendant guilty of Murder under Instruction No. 6, you will find the Defendant guilty of Complicity to Wanton Murder under this Instruction if you believe from the evidence beyond a reasonable doubt all of the following:
A. That on or about July 16, 2011 in Wayne County, Kentucky and before the finding of the Indictment herein, Jared Futrell Killed [the child] by. inflicting-blunt force trauma to his head, torso and/or other parts of his body. AND
B. The Defendant was the mother of [the child]. AND
C. The Defendant failed to make a proper effort to protect [the child]. AND
D.- The failure by the defendant was in conscious disregard of a substantial and unjustifiable risk that [the child] would be killed, and that her disregard of that risk constituted a gross deviation from the standard of care that a reasonable person would have observed in the same situation under circumstances manifesting an extreme indifference to human life.
INSTRUCTION NO. 6-C WANTON MURDER — PRINCIPAL OR ACCOMPLICE
If you believe from-the evidence beyond a reasonable doubt that the Defendant is guilty of either Wanton Murder under Instruction No. 6 or Complicity to Wanton Murder under Instruction No. 6-B, but you are unable to determine from the evidence whether the Defendant committed the crime as Principal under Instruction No. 6 or Accomplice under Instruction No. 6-B, then you will find her guilty of Wanton Murder, Principal or Accomplice, under this Instruction and so state in your verdict.
The jury found both Appellants guilty under Instruction 6-C. Apparently, the jury believed them guilty of wanton murder as principal and accomplice, but could not agree which one was the principal and which was the accomplice. Combination instructions — instructions incorporating alternative, theories of a single crime — are appropriate, we have many times held, when, but only when, the evidence supports all of the alternatives. Caudill v. Commonwealth, 120 S.W.3d 635 (Ky. 2003) (citing Wells v. Commonwealth, 561 S.W.2d 85 (Ky. 1978)). This is so, we have explained, because “no matter which theory they believed, all the jurors convicted under a theory supported by the evidence and all the jurors convicted the defendant of the same offense.” Smith v. Commonwealth, 366 S.W.3d 399, 403 (Ky. 2012).
Complicity is not a separate offense. It is rather an alternative theory of the charged offense. K.R. v. Commonwealth, 360 S.W.3d 179, 186 (Ky. 2012) (“KRS 502.020 does not create a new offense, known as complicity_ Rather than being a separate 'crime, complicity is simply the means of committing another crime.”) (citations and internal quotation marks omitted); Smith v. Commonwealth, 370 S.W.3d 871, 873 n.l (Ky. 2012) (“[U n-der oür penal code “complicity” is not a separate crime; rather, it is a means by which a crime may be. committed.”). Accordingly, where both alternatives are supported by the evidence, combination princi*278pal/accomplice instructions, such-as those given in these cases, are proper, Halversen v. Commonwealth, 730 S.W.2d- 921 (Ky. 1986); Pate v. Commonwealth, 243 S.W.3d 327 (Ky. 2007).6 If one of the instructed upon theories is not supported by sufficient evidence, however, and there is a reasonable possibility that one or more jurors relied on the unsupported theory, then, as Appellants correctly assert, “a unanimous verdict has been denied and the verdict must be overruled.” Travis v. Commonwealth, 327 S.W.3d 456, 463 (2010).7
As discussed above, there was sufficient evidence to support a verdict finding either Appellant guilty as a principal. It remains to consider whether the evidence also supported finding either of them guilty of wanton murder as an accomplice.
Under the pertinent parts of the complicity statute, KRS 502.020,
[w]hen causing a particular result is an element of an offense, a person who acts with the kind of culpability with respect to the result that is sufficient for the commission of the offense is guilty of that offense when he
(a) Solicits or engages in a conspiracy with another person to engage in the conduct causing such result; or
(b) Aids, counsels, or attempts to aid another .person in planning.or engaging .in the conduct causing such result; or
(c) Having a legal duty to prevent the conduct causing the result, fails to make a proper effort to do so.
KRS 502.020(2). Given the severity of Staten’s injuries, given the evidence that the child was fine or normal just hours before he was taken to the emergency room, and given the evidence that doctors managed to revive him, albeit ultimately they could not save him, a rational juror could have believed (and most likely did believe)' that “the conduct causing the result” — the death — was some sort of severe beating inflicted on the morning of July 16, 2011, a beating that produced the blunt force head injuries identified as the cause of death.
*279The Commonwealth’s theory of Lord’s complicity in the murder was derived from subpart (c) of the statute, ie., having a legal duty as his parent to protect the child from being beaten, Lord failed to make a proper effort to do so. See Instruction No. 6-B. Certainly the jury would have believed that Lord was Staten’s mother with a parent’s duties, there being no evidence to the contrary. Construed favorably to the. Commonwealth, the proof reflected that virtually from the beginning of his relationship with Lord, Futrell was prone to angry outbursts against Lord’s child, outbursts that included verbal abuse and rough handling such as shaking and pinching the child’s face. On several occasions the treatment was rough enough to leave bruises on Staten, and since Lord had to have been aware of the bruises she was likely to have been aware of their source. The child’s paternal grandmother and aunt, Lydia Stephenson, testified that the bruises started appearing when Lord began seeing Futrell in May. ’ Lydia Stephenson and Chancie Pyles testified to ah incident on June 27, where Futrell expressed rage at and contempt for the child. In the course of that argument, Futrell pushed Lord into the then sixteén-month old child who fell against a wall. That episode in conjunction with the more extensive and more serious bruising appearing on the child the very next day, injuries that prompted Lord to take him to the emergency room, could reasonably be thought to have put Lord on notice that Futrell’s growing impatience and temper posed a grave risk of death to her small child. Disregarding that risk, however, Lord not only continued to see Futrell, but she even moved out of the home she shared with Pyles and moved in with Futrell thereby increasing the risk to the child, a risk she, as the child’s mother, was duty bound to decrease if she could. That decision was so at odds with her child’s safety as to be reasonably deemed a manifestation of extreme indifference to the value of the child’s life. , There being thus sufficient evidence of Lord’s complicity for the jury to base its conviction of her on a “legal duty” theory, the trial court did not err by incorporating that theory in the jury instructions, including the combination principal/accomplice instruction upon which1 the jury ultimately relied in finding Lord guilty of wanton murder. At retrial, Lord can be retried for complicity to murder based on her failure to fulfill her legal duty, KRS 502.020(2)(c). '
Futrell’s complicity conviction is a different matter. The Commonwealth’s theory of Futrell’s complicity was that if Lord was the killer then Futrell aided and abetted her. The jury instruction meant to convey that theory provided as follows:
INSTRUCTION NO.. 6-B COMPLICITY TO MURDER
If you do not find the Defendant guilty of Murder under Instruction No. 6, you will find the Defendant guilty of Complicity to Wanton Murder under this Instruction if you believe from the evidence beyond a reasonable doubt all of the following:
A. That on or about July 16, 2011 in Wayne County Kentucky and before the finding of the Indictment herein, Jared Futrell killed [the child]' by inflicting blunt force trauma to his head, torso and/or other parts of his body.
B. Prior to the killing, the Defendant solicited, commanded, aided, attempted
to aid, planned, counseled or engaged in a conspiracy with Kayla Lord to engage in the conduct, or aided, counseled, or attempted to aid Kayla Lord in planning or committing discipline or abuse of [the child]; AND
C. The defendant was wantonly engaging in conduct which created a grave *280risk of death to' [the child] and thereby caused the death of [the child] under circumstances manifesting an extreme indifference to human life.
(emphasis supplied). This instruction is patently erroneous as a matter of law.
Part A of Futrell’s .complicity instruction requires for- a guilty verdict a finding that “Jared Fldrell killed [the child.]” As the Commonwealth acknowledges, Futrell’s complicity instruction should- have required a finding that Lord killed the child. Futrell could not be complicit in his own killing of the child. The Commonwealth contends that this error was so obvious that the jury can be relied upon to have recognized it and corrected for it. We reject that proposition. The misdrafted instruction allowed for a finding of complicity where Futrell himself killed the child, having previously aided, assisted or conspired with Lord in mistreatment of the child, and we cann'ot accept the Commonwealth’s argument that the jury would readily recognize this as an erroneous statement of the law of complicity to murder.
Next, part B of the instruction ostensibly presents the Commonwealth’s theory of Futrell’s complicity, but in fact it' simply lists the theories included in subsections (a) and (b) of KRS 502.020(2):
When causing a particular result is an element of an offense, a person who acts with the kind of culpability with respect to the result that is sufficient.for the commission of the offense is guilty of that offense when he:
(a) Solicits or engages in a conspiracy with another person to engage in the conduct causing the result; or
(b) Aids, counsels, or attempts to aid another-person in planning, or engaging in the conduct causing such result.
As discussed above .with regard to Lord’s alleged complicity, the proof was that, if the result, Staten’s death, was not an accident, then “the conduct causing the result,” had to have been an intense assault upon the child not long before he was admitted at the emergency room. There is no evidence’ in this case to support an instruction under subsection (a) — that Fut-rell solicited Lord to commit such an assault on her child or conspired with her to do so. While under Travis these irrelevant, unsupported theories might not require reversal, their inclusion was nevertheless erroneous.
The Commonwealth’s real complicity theory as to Futrell was under subsection (b) — that Futrell’s prior abusive treatment of the child aided somehow Lord’s fatal assault. The instruction fails to make clear, however, that to amount to complicity the aid had to pertain to the “conduct that caused the result,” that is, to the ultimate attack. Instead, the instruction merely asks the jury to find that Futrell “aided, counseled, or attempted to aid Kayla Lord in planning or committing discipline or abuse of [the child].” Aside from what again amount to extraneous theories — there was no evidence of planning, counseling, or attempting — the instruction erroneously directed the jury to focus not on the “conduct that caused the result,” (the blunt force trauma referenced in part A of the instruction) but on “planning or committing discipline- or abuse.” ■Futrell could not be found guilty of wanton murder by complicity if the jury merely believed that on other occasions he countenanced Lord’s disciplinary practices of slapping the child’s mouth and feet and pinching his ears.
The complicity theory lurking in the instructions is that Futrell’s own mistreatment of Staten aided the commission of the murder by encouraging Lord’s mis*281treatment of the child up to and including Lord’s ultimately beating him to ■ death. Even assuming ■ that Futrell’s mistreatment of the child could be deemed “aiding” for the purposes of KRS 502.020(2)(b), however, the problem with this theory is that the jury would have to find that he was aware that his encouragement ’of Lord’s disciplinary practices created a grave risk of death and that, under circumstances manifesting extreme indifference to human life, he disregarded that risk. The evidence simply does not support such a finding. Whatever one might, think of Lord’s alleged manner of disciplining her child (slapping his mouth and feet and pinching his ears), nothing the jury was told that she did leading up to July 16, 2011, could reasonably be thought to imply that she posed a grave risk of death to her son. Futrell, therefore, could not reasonably be convicted of murder by complicity for having disregarded such a risk. In addition to its other faults, therefore, the complicity instruction in Futrell’s case was erroneous because-it did not have adequate evidentiary support. Because there'is no assurance that Futrell’s conviction was not based at least in part on the erroneous instruction, his conviction has this alternative ground of reversal.
On remand both Appellants are subject to retrial as principals in the murder (intentional or wanton) of the child. Bowling v. House, 2011 WL 4072849 (E.D. Ky. 2011) (denying double-jeopardy based ha-beas petition where retrial as principal was allowed following virtually identical instructional error); Commonwealth v. Zanetti, 454 Mass. 449, 910 N.E.2d 869 (2009) (holding that conviction as accomplice does not imply acquittal or bar retrial as principal); United States v. Garcia, 938 F.2d 12 (2nd Cir. 1991) (holding that retrial on sufficiently proved theory was proper after conviction based on alternative instruction was reversed because one of the alternative theories had not' been proved). Lord is also subject to retrial on a complicity theory but Futrell is not, the Commonwealth having failed to present any evidence that he aided and abetted Lord in committing a fatal attack on her son.
B. The Murder Jury Instructions’ Cause of Death Provision Should he Corrected on Retrial.
Appellants also contend that the jury instructions regarding murder ran afoul of the unanimous verdict requirement by directing the jury to base a guilty verdict against either of them on a finding that he/she “killed [the child] by inflicting blunt force trauma to his head, torso and/or other parts of his body.” -This instruction ■ is faulty, Appellants complain, because there was no evidence of fatal blunt force trauma to any part of the child’s body but his head. We agree. There was certainly evidence of trauma to other parts of the child’s, body but no evidence that other, trauma caused his death. At a retrial, the instruction should conform to the evidence, which-focused on the blunt force trauma to the child’s head as the cause of death.
IV. The Trial Court Correctly Deemed Admissible Expert Testimony By the Child-Abuse Pediatrician.
Another issue apt to recur in the event of a retrial concerns the competence of the Commonwealth’s evidence that the child suffered inflicted, as opposed .to accidental, injuries. Undoubtedly the strongest evidence on that point was the testimony of Dr. Melissa Currie, who, as noted above, is a board certified child-abuse pediatrician. Dr. Currie testified that, to a virtual certainty, the child’s fatal injuries were inflicted. Both Appellants contend that the trial court erred two-fold — under both KRE *282702 and KRE 403 — by admitting that testimony. We disagree.
In our courts, the admissibility of expert testimony is governed by Kentucky Rule of Evidence (KRE) 702. That rule provides as follows:
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if
(1) the testimony is based upon sufficient facts or data,
(2) the testimony is the product of reliable principles and methods, and
(3) the witness has applied the princi-pies and methods reliably to the facts of the case.
Our rule is identical to its federal counterpart, Rule 702 of the Federal Rules of Evidence.8 Both rules incorporate guidance provided by the Supreme Court in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). Under Daubert, a trial court’s task in assessing proffered expert testimony is to determine whether the testimony “both rests on a reliable foundation and is relevant to the task at hand.” Daubert, 509 U.S. at 597,113 S.Ct. 2786. In making its reliability determination, the trial court must consider “ “whether the reasoning or methodology underlying the testimony is scientifically valid and whether that reasoning or methodology properly can be applied to the facts in issue.’ ” Toyota Motor Corp. v. Gregory, 136 S.W.3d 35, 39 (Ky. 2004) (quoting Daubert, 509 U.S. at 592-93, 113 S.Ct. 2786). As the United States Court of Appeals for the Sixth Circuit has noted, “Daubert attempts to strike a balance between a liberal admissibility standard for relevant evidence on the one hand and the need to exclude misleading ‘junk science’ on the other.” Best v. Lowe’s Home Ctrs., Inc., 563 F.3d 171, 176-77 (6th Cir. 2009). The court’s role is not to judge the correctness of the expert’s conclusions; that assessment is for the jury. The court’s ga-tekeeping role, rather, is to “focus ... solely on [the] principles and methodology” employed to generate the conclusions, and to ensure that those principles and methods are reliable. Daubert, 509 U.S. at 595, 113 S.Ct. 2786.
There is no “definitive checklist or test” for determining reliability, but in Daubert the Court recognized a number of factors bearing on the inquiry. These include whether the principle, theory, or method in question “can be (and has been) tested,” whether it “has been subjected to peer review and publication,” whether it has a “known or potential rate of error,” and whether it enjoys acceptance within “a relevant scientific community.” 509 U.S. at 593-94, 113 S.Ct. 2786. Appellants contend that all of these “Daubert factors” weigh against the admissibility of Dr. Cur-rie’s inflicted-injury testimony, and that the trial court therefore abused its discretion by allowing that testimony to be introduced. Appellants, however, have failed to appreciate the basis of Dr. Currie’s opinion.
To begin, Daubert concerned scientific expertise, but the Court subsequently explained that Rule 702 applies to other types of expertise as well and that “[t]he factors identified in Daubert may or may not be pertinent in assessing reliability, depending on the nature of the issue, the *283expert’s particular expertise, and the subject of his testimony.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 150, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). Citing Kumho Tire, this Court has upheld the admission of a forensic pediatrician’s opinion to the effect that burns on a child had been inflicted. The opinion was adequately supported, we held, by case reports documenting burns inflicted by means of cigarette lighters and by the doctor’s own experience with patients who had suffered such burns. “The cause of an injury may be within the ambit of an expert witness’s specialized knowledge,” we held, “and is properly admissible subject to the trial judge’s KRE 702 determination.” Ratliff v. Commonwealth, 194 S.W.3d 258, 270 (Ky. 2006). At the Daubert hearing in this case, Dr. Currie testified that her opinion was in part based on her familiarity with a voluminous body of case reports concerning injuries like those suffered by the child in this case and on her experience as a consultant in more than 3000 cases of potential child abuse, in most of which she had herself examined the child.
Even with regard to the Daubert factors, moreover, Dr. Currie’s testimony easily passes muster. As Dr. Currie explained, in cases of potential child abuse whether an injury was inflicted or accidental is of medical as well as legal significance — it can bear on decisions concerning what tests to perform and what treatments to pursue — and accordingly it has become a standard medical- diagnosis, arrived at differentially, like other diagnoses. That is, a child-abuse ■ pediatrician presented with an injured child will attempt to eliminate from a list of potential causes of the injury those causes not likely in the given case until arriving at the cause that is most likely. For example, Dr. Currie testified that in cases of subdural hematoma blood disorders are a potential cause that must be considered, and in cases of bone fracture bone disorders must be considered. In most cases, trauma turns out to be the likeliest cause, and at that point the question becomes whether the trauma was inflicted or accidental.
As Dr. Currie explained, since the middle of the twentieth century, when advances in radiology gave rise to the discovery that child abuse was far more common than had previously been suspected, see Narang, A Daubert Analysis of Abusive Head Trduma/Shahen Baby Syndrome, 11 Houston Journal of Health Law & Policy 505 (2011) (Narang) (recounting the development of the inflicted injury diagnosis), study after study has been devoted to distinguishing accidental traumatic injury from inflicted traumatic injury. Dr. Cur-rie testified that by comparing cases of known accident with cases of known abuse researchers have established with a high degree of statistical significance that certain injuries and certain patterns of injury are far more likely to have been inflicted than accidental. By the 198Qs, indeed, according to Dr. Currie, the literature in this area had become so extensive that a general pediatrician could not keep up with it. General pediatricians then began to rely on advice from pediatricians who made child abuse the focus of their practice. In 2009, the American Board of Pediatrics recognized this development and endorsed it by offering certification in the sub-specialty' of child-abuse pediatrics. Dr. Curry is so certified, she is the author of a chapter in one of the standard pediatric textbooks on child abuse, and she lectures and presents frequently on that topic at medical conferences.
This Court, as have many others, has recognized differential diagnosis . as a “widely-used technique in the medical community to identify and isolate causes of disease and death,” Hyman & Armstrong, P.S.C. v. Gunderson, 279 S.W.3d 93, 107 *284(Ky. 2008), andj when properly performed, as “a reliable method of ascertaining medical causation.” Id. See also, Best v. Lowe’s Home Ctr., Inc., 568 F.3d at 178-180 (noting the wide-spread acceptance of differential diagnosis as a valid medical technique). In Best, the Sixth Circuit adopted the following test for assessing the reliability of a differential diagnosis in a given case: did the doctor (1) objectively ascertain the nature of the .patient’s injury; (2) “rule in” one or more potential causes of the injury using a. valid methodology; and (3) reach a conclusion by engaging in standard diagnostic techniques to rule out alternative causes. Dr.- Currie’s diagnosis followed this procedure.
We have also recognized .that an expert’s causation testimony may be based fully or primarily on his' or her review of the pertinent scientific literature, provided that the expert is qualified “to review the literature and render an opinion” on the specific matter at issue, Burton v. CSX Tmnsp., Inc., 269 S.W.3d 1, 8 (Ky. 2008), and further provided that there be “objective, verifiable evidence that the testimony is based on scientifically valid principles.” Id. at 9. “One means of showing this is by proof that'the research and analysis supporting the proffered conclusions have been subjected to normal scientific scrutiny through peer review and publication.” Id. As Dr. Currie testified, especially since the advent of computerized databases made available large volumes of medical literature and made possible comprehensive reviews of that literature, literally thousands of1 peer-reviewed studies have been published on whether particular types of pediatric traumatic injury are apt to have been inflicted or accidental. See, for example, Narang, supra, and its companion article, Narang, et al., A'Daúbert Analysis of Abusive Head Trauma/ Shaken Baby Syndrome — Part II: An Examination of the Differential Diagnosis, 13 Houston Journal of Health Law & Policy, 203 (2013) (reviewing the literature — some 700 studies — pertaining to just two types of head injury, subdural hematoma and retinal hemorrhage).' Dr. Currie’s specialized training qualifies her to review and apply those studies, and she testified that appropriate studies informed and strongly supported her diagnosis in this case.
Thus, even with respect to the Daubert factors (testing, peer review and publication, error rate, and acceptance within a relevant community ' of practitioners) an iriflicted pediatric injury diagnosis can be, and in this case was, admissible.- While it is true, as Appellants note, that'the diagnosis generally has not been and' cannot be tested through randomized, controlled trials, it has been tested by repeated scrutiny in peer-reviewed observational studies conducted in accord with well-established statistical principles. Dr, Currie testified that error rates generally apply only to tests meant to establish the existence or not of a single factor, and not to diagnoses that must take multiple, varying factors into consideration. Nevertheless, differential diagnosis is generally recognized. as “ ‘not frequently lead[ing] to incorrect results,’ ” Best v. Lowe’s, 563 F.3d at 179 (iquoting In re Paoli R.R. Yard PCB Li-tig., 35 F.3d 717, 758 (3rd Cir. 1994)); many of the studies underlying the inflicted injury diagnosis have shown a highly predictive association of certain injuries or the- concurrence of certain injuries with inflicted injury, Narang at 579; and Dr. Currie testified that her own practice is to diagnose inflicted injury only conservatively, when the evidence for it;is compelling. On cross-examination in response to a question by defense 'counsel, he stated she was not aware of a single case in which she had diagnosed abuse and was later deter*285mined to be wrong.9 Finally, the fact that child-abuse pediatrics has become a certified pediatric sub-specialty recognized by the American Board of Pediatrics, plainly indicates that the abuse diagnosis is accepted in the most relevant scientific community. In sum, the trial court , did not abuse its discretion under KRE 702 by deeming Dr. Currie’s testimony reliable.
Reliability is not the end of the Daubert analysis. In addition to requiring that an expert’s testimony be reliable, KRE 702 also requires that it “assist the trier of fact to understand the evidence or to determine a fact in issue.” The testimony, that is, must be relevant, it must relate to a material issue in the case and it must “fit” the case in the sense that there must be “a valid scientific connection” between the expert’s specialized knowledge and the pertinent inquiry confronting the trier of fact. Daubert, 509 U.S. at 591-92, 113 S.Ct. 2786. Even if relevant, moreover, expert testimony is subject to KRE 403 and may be excluded “if its probative value is substantially outweighed by the danger of undue prejudice, confusion of the issues, or misleading the jury. — ” Because of the difficulty a lay jury is apt to have in evaluating it, expert testimony pose’s a significant risk of being misleading, a risk that should inform the trial court’s KRE 403 analysis. See, Daubert, 509 U.S. at 595, 113 S.Ct. 2786 (because expert testimony can be “both powerful and quite misleading,” the trial court “in weighing possible prejudice against probative force ... exercises more control over experts than over lay witnesses.”) (citation and internal quotation marks omitted). Appellants contend that even if reliable for the purposes of KRE 702, Dr. Currie’s testimony should still have been excluded because it was not relevant and because it was unduly prejudicial. We disagree.
Evidence is relevant under KRE 401 if it has “any tendency to make the existence of any fact that is of consequence to the determination- of the action more probable or less probable than it would be without the evidence.” Dr. Currie’s testimony that the child’s injuries were not accidental but had been inflicted was obviously relevant and helpful to the jury in a case where the defense was accident. Also relevant and beyond the jury’s lay- knowledge was Dr. Currie’s testimony that both'the head injury and the abdominal injury were so severe that upon receiving them the child would likely have 'been rendered unconscious and would certainly have immediately been rendered symptomatic. That testimony tended to' refute defense claims that the injuries may have been' sustained sometime prior to the early morning of July 16, 2011.
In the face of this obvious relevance, the Appellants’ assertion that Dr. Currie’s testimony was not relevant is puzzling, but the claim seems to be that Dr. Currie’s opinion about an “ultimate”. issue in the case was not relevant and should have been left to the jury. This claim is absolutely meritless. Dr. Currie did not opine that Appellants were guilty (the ultimate issue) ‘or that Appellants were the perpetrators. See State v. Sanchez-Alfonso, 352 Or. 790, 293 P.3d 1011 (2012) (holding that identity of the' perpetrator not a valid part of the abuse diagnosis). Otherwise, as the Appellants acknowledge, this Court aban*286doned the so-called ultimate issue rule in Stringer v. Commonwealth, 956 S.W.2d 883 (Ky. 1997). If the Appellants’ “relevancy” argument is meant to be an invitation to reconsider Stringer, we decline the invitation.
Equally meritless is the contention that Dr. Currie’s testimony ought to have been excluded under KRE 403. Among other things, that rule permits, the exclusion of evidence which, though relevant, poses a substantial risk of undue prejudice, ie., a risk that the evidence will induce the jury to base' its decision on emotion or some other improper ground. Wilson v. Commonwealth, 438 S.W.3d 345 (Ky. 2014). Appellants assert that Dr. Currie’s testimony was unduly prejudicial because the doctor expressed confidence in her diagnosis (“This is a clear case of abuse.”), and because of her testimony, noted above, that she was not aware of ever having made an incorrect abuse diagnosis. Since the latter testimony was presented to the jury only because it was specifically elicited by the Appellants, they will not be heard now to complain about it. Furthermore, Dr. Currie explained in detail her findings and why she thought this a clear casé of abuse. Her confidence did not render her explanations unduly prejudicial.
In sum, the trial court did not abuse its discretion under KRE 702 by allowing Dr. Currie to opine that the injuries suffered by the child in this case were inflicted and not accidental. Many other, courts, have similarly held. See, e.g., Commonwealth v. Roderiques, 462 Mass. 415, 968 N.E.2d 908 (2012); McFolley v. State, 289 Ga. 890, 717 S.E.2d 199 (2011); O’Brien v. United States, 962 A.2d 282 (D.C. 2008); State v. Kuehn, 273 Neb. 219, 728 N.W.2d 589 (2007); State v. Boyer, 741 N.W.2d 749 (S.D. 2007); People v. Martinez, 74 P.3d 316 (Colo. 2003); State v. Struzik, 269 Kan. 95, 5 P.3d 502 (2000). And see, Estelle v. McGuire, 502 U.S. 62, 112 S.Ct. 475, 116 L.Ed.2d 385 (1991) (holding that admission of shaken-baby-syndrome testimony did not violate the Due Process Clause); Cavazos v. Smith, — U.S. ——, 132 S.Ct. 2, 181 L.Ed.2d 311 (2011) (reversing grant of habeas corpus and holding that expert shaken-baby-syndrome testimony was sufficient to support conviction of assault on child resulting in death). In short Dr. Currie’s testimony did not taint the trial in any way.
V. The Trial Court Erred By Allowing the Appellants too Few Peremptory Juror Challenges.
Other possibly recurring issues raised by the Appellants can be addressed more succinctly. Clearly, the trial court granted Appellants too few peremptory juror challenges. The trial court allowed Lord and Futrell a total of eleven such challenges. As construed in Springer v. Commonwealth, 998 S.W.2d 439 (Ky. 1999), RCr 9.40 allows two defendants to be tried jointly before a jury with one or two alternate jurors a total of thirteen peremptory challenges: “nine to be exercised jointly pursuant to- RCr 9.40(1) and (2); one each to be exercised independently pursuant to RCr 9.40(3); and an additional one each to be exercised independently pursuant to RCr 9.40(2).” 998 S.W.2d at 444.
VI. The Trial Court Erred By Disallowing Diversion-Agreement Impeachment Cross-Examination.
Appellants also correctly contend that they should have been allowed to impeach Johnny Stephenson with, the fact that at the time of his testimony he-was on pre-trial diversion following his guilty plea to burglary and theft charges. The, diversion agreement was subject to revocation, *287and the United States Supreme Court has held that similarly revocable deals between witnesses and the prosecution implicate the defendant’s Confrontation Clause right to effective cross-examination. Delaware v. Van Arsdall, 475 U.S. 673, 106 S.Ct. 1431, 89 L.Ed.2d 674 (1986); Davis v. Alaska, 415 U.S. 308, 94 S.Ct. 1105, 39 L.Ed.2d 347 (1974). Such a deal implicates that right, the Court has explained, because “a jury might reasonably have found [that the threat of revocation] furnished the witness a motive for favoring the prosecution in his testimony.” Van Arsdall, 475 U.S. at 679, 106 S.Ct. 1431. As the Commonwealth points out, in Holt v. Commonwealth, 250 S.W.3d 647 (Ky. 2008), we held that in the circumstances of that case the trial court had not erred by disallowing diversion-agreement cross-examination, but this ease we think falls squarely within the Supreme Court’s holdings.
VII. The Trial Court Did Not Err By Allowing the Admission of Autopsy Photographs.
The trial court did not err by allowing the introduction during the medical examiner’s testimony of autopsy photographs illustrating the examiner’s findings. The court took care to exclude redundant photos and photos otherwise with little probative value. We have upheld the admission of such autopsy photographs many times. Staples v. Commonwealth, 454 S.W.3d 803 (Ky. 2014) (collecting cases).
VIII. The Trial Court Did Not Err By Allowing the Admission of Prior-Act Evidence Relevant to a Material Issue Other Than Appellants’ Characters.
Nor did the trial court err by allowing friends and acquaintances of Appellants to testify about seeing bruises on the child after Futrell came to be involved with him; about seeing Appellants physically disciplining the child and verbally abusing him; and about seeing Appellants engage in an argument during which Fut-rell expressed exasperation with and contempt for the child, and during which the couple’s conflict directly endangered the child. Appellants maintain that all of this testimony should have been excluded under KRE 404(b) as evidence calculated only to show bad character and action in conformity therewith. That rule provides, however, that evidence of other acts may be offered for such non-character purposes as proof of motive, opportunity, identity, or absence of mistake or accident. All of the challenged evidence here was relevant to rebut Appellants’ claim that the child’s fatal injuries were the result of a series of accidents, ie., swallowed chewing gum and then improper CPR. Noel v. Commonwealth, 76 S.W.3d 923, 931 (Ky. 2002) (holding in child-sex-abuse case that “evidence of similar acts perpetrated against the same victim are almost always admissible for [the purposes enumerated in the rule].”); Dant v. Commonwealth, 258 S.W.3d 12 (Ky. 2008) (citing Noel and upholding in a physical abuse case the admission of prior acts of physical abuse against the same victim). And see, Estelle v. McGuire, 502 U.S. at 62, 112 S.Ct. 475 (holding in a physical abuse case that due process did not require the exclusion of evidence of prior acts of abuse against the same victim because the prior acts tended to prove that the ultimate injury was not accidental).
Finally, Appellants also contend that the trial court erred by denying their requests for lesser-included-offense instructions. The particular errors complained of may or may not recur at a new trial, where the evidence bearing on lesser offenses is apt to be different. We decline, therefore, to address these contentions.
*288 CONCLUSION
The trial court erred in failing to strike two' jurors for 'cause and, given that all elements of the Gabbard inquiry are satisfied, the Appellants’ convictions must be reversed. Kayla Lord may be retried on both principal and complicity theoriés of murder (intentional or wanton) but, for the reasons outlined Jared Futrell is not subject to retrial on a complicity theory, the Commonwealth having presented no evidence that he aided and assisted Lord as she committed a fatal attack on her child. Futrell may, of course, be retried as a principal in the murder (intentional or wanton) of Staten Stephenson. Accordingly, in both 2013-SC-000184-MR (Futrell) and 2013-SC-000200-MR (Lord), we reverse the Judgment of the Wayne Circuit Court and remand for additional proceedings consistent with this Opinion.
Minton, C.J.; Barber; Keller, Noble, and Venters, JJ., concur. Cunningham, J., concurs in part and dissents in part by séparate opinion.. The tube was inserted through the child's mouth and apparently in that circumstance it is often referred to as an "OG” (oral-gastrial) tube rather than an "NG” tube. The medical records in this case have it both ways, so we have followed the parties’ practice.
. She is also referred to in the record as Betsy Lyons.
. “A properly qualified juror must be impartial, which former United States Supreme Court Chief Justice Charles Evans Hughes described as comprising a ‘mental attitude of appropriate indifference.’ ” Paulley v. Commonwealth, 323 S.W.3d 715, 720-21 (Ky. 2010) (quoting United States v. Wood, 299 U.S. 123, 145-46, 57 S.Ct. 177, 81. L.Ed. 78 (1936)).
. The Commonwealth notes that while both Appellants argue that potential Juror 75 should have been removed for cause, neither of them in the portion of his or her brief explaining how the alleged error was preserved, "makes [an] allegation that he [or she] ... used a peremptory strike on 75.” Both Appellants expressly make that allegation with respect to potential Juror 27, and we agree with the Commonwealth that the lack of a similar express allegation with respect to Juror 75 seems odd. Odd or not, however, we reject the Commonwealth's unsupported suggestion that the failure to expressly assert in their brief the use a peremptory strike on potential Juror 75 as well as on potential Juror 27 amounts somehow to an abandonment of their claim with respect to 75. The record makes clear that potential Juror 75 (as was potential Juror 27) was in fact removed *274by way of defense peremptory. Whatever the reason for the failure of the Appellants’ briefs to allege that fact expressly, the briefs adequately imply it and proceed to address the issue.
. We quote the instructions from, Lord’s case. With exceptions, discussed below, with respect to Instruction 6-B regarding Complicity to Wanton Murder, the instructions in Fut-rell’s case were identical aside from name and gender changes.
. Appellants’ arguments on this issue conflate a basic claim — that in this particular case a rational juror could not-have made a complicity finding and thus the combination instruc- ' tion was improper because it included an unsupported theory of the offense — with more general claims — that combination murder .instructions are inherently improper because they invite the jury to impute the principal’s mental state, and hence degree of culpability, to the accomplice and because complicity should be deemed a separate offense. We discussed the specific claims below. We reject the more general claims as we have before. See Halvorsen v. Commonwealth, supra. The instructions here plainly required the jury to find Lord herself aggravatedly wanton before it could find her guilty of wanton murr der, either as principal or as accomplice. Even considering the mental state element, therefore, we are not persuaded that, where both theories are supported by the evidence, a combination principal/accomplice instruction is materially different from any other combination instruction. Appellants also request, in effect, that we reconsider our holdings to the effect that complicity is not a separate ' offense.' This we decline to do. Although, we acknowledge that under our current approach complicity instructions can assume a troubling complexity. This precise issue was not preserved. KRS 502.020, moreover, clearly provides that “complicity” is not itself an offense; it is rather a basis for imputing liability to the complicitor for "an offense committed by another.”
. Travis’s rule of mandatory reversal presumes that the instructional error was preserved. The issue was preserved here. At the close of proof, when the Commonwealth moved to amend the indictment so as to charge both Lord and Futrell as accomplices as well as principals; they both objected to the amendment expressly on the ground that the evidence of complicity was not sufficient.
. Our Rule was amended in 2007 to “follow the development [of] and adopt[] exact language set by the Federal Rules.” KRE 702, Review Commission Notes.
. Appellants complain that this testimony was improperly self-bolstering. Clearly though this testimony was proper during the Daubert hearing, when the court was trying to assess the reliability of Dr. Currie’s methods. At trial, Dr. Currie did not mention her track record until defense counsel, apparently recalling the Daubert hearing testimony, asked her what it was. .Dr. Currie then repeated her Daubert-hearing remarks. That completely invited response was not improper self-bolstering.'