Glass, J.
On February 18, 1982, the defendant, Michael Joly, was indicted by a grand jury and charged with the crime of murder in violation of General Statutes § 53a-54a (a).1 He was tried by different juries on [236]*236three separate occasions between 1984 and 1988. After his first trial in 1984 before Judge Quinn, the jury found the defendant guilty as charged, but a new trial was ordered on the ground of juror misconduct. The defendant’s second trial in 1986 before Judge Kline resulted in a deadlocked jury and ended in a mistrial. His third trial in 1988 before Judge Kaplan concluded in a guilty verdict and a sentence of sixty years imprisonment. This appeal followed.
On appeal, the defendant challenges several of the trial court’s evidentiary rulings. He claims that the court should have: (1) suppressed Anthony Pasquarelli’s allegedly hypnotically refreshed testimony describing and identifying the defendant; (2) suppressed Anthony Pasquarelli’s testimony on the alternative ground that the defendant was prejudiced by the state’s failure to disclose statements made by Anthony to the police before the hypnosis session; (3) excluded Detective Charles Miller’s testimony recounting statements made by the defendant during the search of his apartment because Miller was incompetent to testify, and Miller’s testimony was irrelevant, or alternatively, more prejudicial than probative; (4) excluded Daniel Pasquarelli’s testimony describing the defendant’s “silence” after Daniel accused him, while incarcerated, of committing the crime; and (5) admitted evidence offered by the defendant to rebut the inference of consciousness of guilt arising from certain evidence favorable to the state. We affirm.
The jury could reasonably have found the following facts. On July 17, 1981, between 5:26 and 5:55 p.m., the defendant made three emergency “911” telephone calls to the Bristol police department under an assumed name. He informed the police that a girl in need of medical attention was in the woods near a path along the railroad tracks by the Pequabuck River in the For[237]*237estville section of Bristol. Considering these to be crank calls, the police conducted a cursory search of the area and found nothing. At approximately 3:15 the next morning, however, Michael Mattei notified the police that his fifteen year old daughter, Diane Mattei, the victim, was missing. The police again searched the woods and discovered the victim’s body. She was lying on her back, unclothed from the waist up, and her face was unrecognizable. The chief medical examiner testified that the victim had died of a fractured skull with subdural hemorrhage resulting from contusions of the brain and/or blunt force injuries to the head and neck. The time of death was placed between 2 and 6 p.m. on July 17, 1981.
At trial, the defendant stipulated that he had made the 911 calls. He defended against the charge of murder, without testifying, by presenting thirty-three witnesses to establish that he was a concerned citizen who had merely reported his discovery of the injured victim, and that the crime was more likely to have been committed by one of a number of persons who had been suspects in the case. The state presented circumstantial evidence to the contrary through the testimony of thirty-four witnesses and numerous exhibits. Several witnesses, including Anthony and Daniel Pasquarelli, testified that they had seen the defendant alone with the victim in the immediate area of the woods between 2:30 and 3:45 p.m. on the date of her death.
I
The defendant first claims that the court should have suppressed as the unreliable product of hypnosis all testimony by Anthony Pasquarelli describing or identifying him. The state counters that the trial court properly determined that Anthony had not been hypnotized. We provide the following facts to place the defendant’s specific claims in their proper context.
[238]*238On July 21, 1981, at an interview with Detective Thomas Killiany of the Bristol police department, Anthony revealed that while walking along the tracks with his brother, Daniel Pasquarelli, at approximately 3:15 p.m. on the date of the crime, he had observed and conversed with a man accompanied by the victim. He described the man, signed a written statement incorporating the description and assisted in the making of a composite drawing of the man’s face. Although the police eventually lost this statement, Anthony testified that he had described the man as follows: dark hair, dark eyes, approximately 5’8” tall, one hundred sixty pounds, wearing blue jeans and a tee-shirt with a pocket. Anthony gave an “identical” or “fairly identical” description to the police on July 22, 1981, which also was reduced to writing and thereafter lost. On one or both of the above occasions, Anthony told the police that he had recognized the man but could not remember his name. He later testified, however, that he had remembered the man’s name as that of the defendant on July 21 or 22, 1981, but had deliberately withheld it on those dates and at subsequent interviews because he had been displeased with his treatment by the police. It was not until August 11, 1981, that Anthony disclosed the defendant's name to the police.
In the meantime, at Killiany’s request, Anthony had agreed to submit to hypnosis as an aid to recall the name. John Haksteen, a psychologist and practicing hypnotist, conducted the hypnosis session on July 23, 1981.2 Haksteen testified that in his opinion, Anthony at some point had gone into a “hypnotic trance,” during which time he had disclosed the following information that Haksteen had later compiled in a report: [239]*239Young man of his age (twenty), dark hair, dark eyes, two days growth of beard, black or navy blue polo shirt, blue jeans, brown leather belt with tarnished oval buckle, a little taller than he, with a French name that was very hard to pronounce. Haksteen also testified, however, that although he had regarded Anthony’s behavior during the session as consistent with that of a hypnotized person, he had not been trained to “spot malingerers” or subjects who might be intentionally feigning hypnosis. Haksteen admitted that Anthony “could have faked it.”
Immediately after the session, Anthony again described the man to the police.3 Killiany testified that “the only thing that was new” in the July 23, 1981 description was the “brass buckle on the belt of the person he had seen.” Similarly, Anthony testified that “[t]he statements] that I gave before I saw the hypnotist and after I saw the hypnotist are the same,” “except for I might have . . . said . . . different words or described maybe something that they didn’t press on the day before or the day before that.”
Contrary to Haksteen’s perception that Anthony had achieved a hypnotic state, Anthony testified that at “no time” during the hypnosis session had he been unaware of “what was going on.” He described the tests Haksteen had performed in detail that matched Haksteen’s testimony, and stated that he had told Haksteen “pretty much exactly what I told the police the first time I talked to them.” Anthony also testified that he had known the man’s name as the defendant’s name before the session, but when Haksteen “got to the part about who I saw ... I would always come back with the same answer ... I would just keep telling him I saw faces.” Eventually, Anthony had given Haksteen [240]*240“a different name” that he had made up, and he stated that “it was a French name and I was trying to think of French names when I was sitting there.” Anthony explained that he had withheld the name at and after the session “[because] of the attitude problem we were having with the police department.”
On these facts, the court ruled that Anthony had not been hypnotized and denied the defendant’s motion to suppress. The court thus declined to consider the effect of any alleged “hypnosis” on the reliability of Anthony’s subsequent descriptions and identification of the defendant.4 Following this ruling, the state called Anthony as a witness and he testified to his recollections in full.
The defendant argues that the court incorrectly admitted Anthony’s testimony for several reasons. He claimed in his brief that the “majority American rule” “presumes that anyone who has undergone hypnosis has in fact been hypnotized,” and only in “the rare case” can this presumption “be overcome by expert opinion to the contrary.” At oral argument, the defendant contended that the question of whether a witness has in fact been hypnotized is “not a proper inquiry [241]*241for the court” because “the court cannot tell whether the [witness] has been affected or not by hypnosis.” He further asserted at argument that “the last person who is competent to testify as to whether he was influenced by the hypnosis session is the person who was subjected to the procedure,” and that in the absence of expert testimony that the witness had not been hypnotized, “the court has no discretion to bypass the state’s burden of showing that [the witness’] testimony was recalled and related prior to the hypnosis process.”
Initially, we note that while the defendant refers us to a number of cases that purportedly follow a “majority rule” of presumptive hypnosis, only one of those cases involves a situation where the occurrence of hypnosis was disputed. Where no one disputes that hypnosis occurred, a court might reasonably presume that hypnosis in fact occurred and go on to consider whether the proffered hypnotically refreshed testimony accurately reflects the witness’ true memory, or rather, the distorted and unreliable product of hypnotic suggestion, confabulation, or memory hardening. See State v. Pollitt, 205 Conn. 61, 78-81, 530 A.2d 155 (1987); see also Rock v. Arkansas, 483 U.S. 44, 59-61, 107 S. Ct. 2704, 97 L. Ed. 2d 37 (1987). In this case, however, the state argued and the court decided that hypnosis had not occurred. Our review of that decision is not directly aided, much less controlled, by cases that do not confront this distinct issue.
The single apposite case on which the defendant relies undermines, rather than supports, his position. In People v. Johnson, 47 Cal. 3d 1194, 1232-33, 767 P.2d 1047, 255 Cal. Rptr. 569 (1989), the California Supreme Court did not disapprove of the trial court’s inquiry into whether a witness had achieved a hypnotic state as alleged by the defendant in that case. The court instead [242]*242rejected the defendant’s claim that the trial court improperly relied on the opinion of an expert who had not participated in the hypnosis session in reaching its decision that the witness had not in fact been hypnotized. The court’s disposition of that claim, moreover, expressly condoned the trial court’s consideration of the witness’ testimony that he had not been hypnotized in determining whether hypnosis had occurred. Id. Furthermore, the court implicitly sanctioned the trial court’s authority to reject the testimony of the hypnotist who had conducted the hypnosis session and who had been “of the opinion that [the witness] had been hypnotized.” Id., 1232.5
The defendant also points out that authorities on hypnotism caution that “very often hyponotic subjects have refused to believe they actually went into a trance. Some claim they were wide awake during the whole experience, others that nothing unusual happened, still others that they were only pretending to be hypnotized.” B. Diamond, “Inherent Problems in the Use of Pretrial Hypnosis on a Prospective Witness,” 68 Cal. L. Rev. 313, 334 (1980). Such authorities nonetheless acknowledge that “even the best experts cannot consistently distinguish between actual and pretended hypnosis” and that “no reliable and truly objective criteria [for detecting] the state of hypnosis have yet been discovered”; B. Diamond, supra, 337. Despite the pos[243]*243sibility of conflicting evidence of an alleged hypnotic state, we do not agree with the defendant’s contention that the weighing of such evidence is a prospect so formidable that our courts should be divested of authority to decide whether hypnosis has in fact occurred. Neither do we agree that that decision, as the defendant essentially argues, should be placed in the hands of experts who cannot consistently distinguish between an actual and feigned hypnotic state. The law traditionally recognizes the trial court as a tribunal equipped to resolve disputed evidentiary issues, and as the “final judge of credibility.” Clark v. Haggard, 141 Conn. 668, 674,109 A.2d 358 (1954). Moreover, as we have often noted, the fact that a witness testifies as an expert does not compel the acceptance of his or her testimony as true. See, e.g., Aspiazu v. Orgera, 205 Conn. 623, 634, 535 A.2d 338 (1987).
We conclude, rather, that the determination of whether a witness has in fact been hypnotized is appropriately committed to the sound discretion of the trial court. In resolving this issue, the court may be aided, but is not bound, by expert opinion. Id. Such testimony must be “considered, weighed and tested like any other evidence”; id.; and assessed “ ‘in relation to the other circumstances in evidence bearing on the question in issue’ ”; Blake v. Blake, 207 Conn. 217, 225, 541 A.2d 1201 (1988); including, if offered, the testimony of the allegedly hypnotized witness. The trial court “ ‘is privileged to adopt whatever testimony [it] reasonably believes to be credible’ ”; (emphasis in original) Eichman v.J&J Building Co., 216 Conn. 443, 451-52, 582 A.2d 182 (1990); and expert testimony may be rejected in favor of other evidence found more persuasive. Transportation Plaza Associates v. Powers, 203 Conn. 364, 377, 525 A.2d 68 (1987). The trial court’s preliminary determination of whether a witness has in fact [244]*244been hypnotized will not be disturbed on appeal except for an abuse of discretion. Cf. State v. Walker, 215 Conn. 1, 6, 574 A.2d 188 (1990); State v. Stange, 212 Conn. 612, 617, 563 A.2d 681 (1989).
In the present case, the record discloses that Anthony had described the events at the hypnosis session in a manner that diverged little from Haksteen’s description, thereby substantiating Anthony’s claim of awareness during the session. His explanation of a conscious attempt to supply Haksteen with French names further suggests that he was in sufficient control of his faculties to feign hypnosis if he so desired. In addition, Anthony had a plausible motive for refusing to cooperate with Haksteen. He had expressed displeasure with his treatment by the police, and the circumstances surrounding his dealings with the police suggested, as the trial court surmised, that he had been “trying to position himself and maybe cut a deal for himself on unrelated charges . . . .” Notwithstanding Haksteen’s opinion that Anthony had appeared cooperative, Anthony’s refusal to cooperate with Haksteen, whom he had visited at the behest of the police, was consistent with his refusal to cooperate fully with the police for reasons of his own. Under these circumstances, we conclude that the court was well within its discretion in finding Anthony’s testimony the more credible and in determining that he had not, in fact, been hypnotized. Anthony’s testimony was therefore properly admitted.
II
The defendant next claims that the trial court should have suppressed Anthony’s testimony because he was prejudiced by the nondisclosure of Anthony's July 21 and 22, 1981 statements to the police.6 As discussed [245]*245in Part I above, these statements contained Anthony’s descriptions, before the hypnosis session, of the man he claimed to have seen on the tracks. The defendant moved to suppress Anthony’s testimony after learning that the police could not locate the statements and that the state consequently could not produce them pursuant to Practice Book § 752.7
After hearing the defendant’s motion, the court found the loss of the statements attributable to “some form of negligence as opposed to some intentional act.” The court also determined that the defendant had not been unduly prejudiced by the nondisclosure in light of his access to Anthony’s July 23 and August 11,1981 statements, Anthony’s prior testimony “with all his discrepancies,” Killiany’s testimony regarding the content of the statements and Haksteen’s testimony recounting the details of the hypnosis session. Accordingly, the court denied the motion.
The defendant attacks this ruling as “clearly incorrect.” He argues that the nondisclosure of the statements severely prejudiced his ability to: (1) demonstrate which portions of Anthony’s testimony should have been excluded as the unreliable product of hypnosis; (2) impeach Anthony’s credibility; (3) establish dissimilarities between Anthony’s descriptions and his own [246]*246appearance; and (4) establish similarities between Anthony’s descriptions and the appearances of other suspects in the case. The defendant imputes particu-. lar significance to the statements for impeachment purposes because, in his view, Anthony was a “critical” witness whose testimony was “confirmed” by hypnosis and thus “may well have had special effect.” For these reasons, the defendant maintains that he is “entitled to a new trial at which [Anthony’s] testimony is stricken.” We disagree.
The record amply supports the court’s finding that the loss of the statements resulted from negligence rather than “a deliberate act done with intent to deprive the defense of information.” State v. Williamson, 212 Conn. 6, 16, 562 A.2d 470 (1989). Since the state was negligent in the nondisclosure, it was the defendant’s burden to demonstrate that he was so prejudiced by lack of access to the statements that Anthony’s testimony should be suppressed notwithstanding the state’s minimal culpability. Id., 18; State v. Kelly, 208 Conn. 365, 383-84, 545 A.2d 1048 (1988); State v. Palmer, 206 Conn. 40, 59, 536 A.2d 936 (1988); see United States v. Sommer, 815 F.2d 15, 17 (2d Cir. 1987); United States v. Miranda, 526 F.2d 1319, 1328 (2d Cir. 1975), cert. denied, 429 U.S. 821, 97 S. Ct. 69, 50 L. Ed. 2d 82 (1976). The trial court, in turn, was vested with broad discretion in balancing the prejudice accruing to the defendant against the state’s culpability; State v. Belle, 215 Conn. 257, 268, 576 A.2d 139 (1990); and in realistically appraising the significance of the statements in view of their nature, their bearing upon critical issues in the case, and the strength of the state’s untainted proof. State v. Williamson, supra, 23; see United States v. Grammatikos, 633 F.2d 1013, 1019-20 (2d Cir. 1980).
The primary source of prejudice identified by the defendant was his inability to use the statements to pin[247]*247point discrepancies between Anthony’s recollections before and after the hypnosis session, and thus to demonstrate which portions of his testimony should be excluded as unreliably rooted in his “posthypnotic” memory. As we concluded in Part I above, however, the trial court reasonably determined that Anthony had not in fact been hypnotized. It was reasonable, therefore, for the trial court to reject the defendant’s contention that the statements given prior to the hypnosis attempt would illuminate the alleged unreliability of the witness’ “posthypnotic” statements. Similarly unavailing is the defendant’s contention that Anthony’s testimony, “confirmed” by hypnosis, was so critical to the state’s case that he suffered great prejudice from his inability to impeach Anthony with the statements. Four witnesses besides Anthony placed the defendant in the vicinity of the crime scene on the date of the victim’s death;8 see State v. Johnson, 214 Conn. 161, 174, 571 A.2d 79 (1990); compare State v. Williamson, supra; and, by eliciting the fact of the hypnosis session on cross-examination, the defendant was responsible for any resulting “confirmation” of Anthony’s testimony.
Moreover, the impeachment materials available to the defendant were ample in supply. See State v. Johnson, supra, 175; State v. Myers, 193 Conn. 457, 469, 479 A.2d 199 (1984). Anthony’s July 23 and August 11,1981 statements to the police as well as his former testimony were riddled with inconsistencies that defense counsel effectively explored on cross-examination, eliciting [248]*248Anthony’s express admission that he had repeatedly and deliberately lied to the police about the details of his observations as well as the name of the man he had seen. Also in evidence was the composite drawing made with Anthony’s assistance contemporaneously with the July 21, 1981 statement, as well as the list of “foils” that Anthony had chosen as consistent with the man’s features and that had been used in the creation of the drawing. Accord State v. Santangelo, 205 Conn. 578, 588, 534 A.2d 1175 (1987). The defendant could have used the composite and the foils to identify distinctions, if any, between his own features and those depicted in the drawing, and to point out any similarities to the features of other suspects.
The nondisclosure of Anthony’s July 21 and 22,1981 statements, therefore, hardly left the defendant bereft of means to attack Anthony’s credibility respecting the matters he claimed to have observed and recalled both before and after the hypnosis session. See United States v. Sommer, supra; State v. Kelley, supra, 384. In light of the minimal prejudice resulting from the nondisclosure, we conclude that the trial court properly admitted Anthony’s testimony.
Ill
The defendant’s next claim is in two parts, each challenging the court’s ruling admitting the testimony of Detective Charles Miller as to his recollection, refreshed by his police report, of certain unsolicited statements made by the defendant in his presence during the execution of a search and seizure warrant at the defendant’s apartment on July 31, 1981.9 The defendant maintains that Miller’s testimony should [249]*249have been excluded because: (1) Miller was incompetent to testify as to the defendant’s statements; and (2) Miller’s testimony was irrelevant, or in the alternative, more prejudicial than probative.10 Although the defendant invokes the due process clauses of the federal and state constitutions in support of his claims, they are properly viewed as evidentiary in nature. See State v. Castonguay, 218 Conn. 486, 496-97, 590 A.2d 901 (1991). The standard for review of his claim is therefore a determination of whether the court abused its discretion in admitting Miller’s testimony.
A
The defendant first argues that Miller was not competent to testify regarding his statements because Miller was unqualified to determine whether they had been rationally made, and could give no reason for his professed belief that they related to the case other than his “police instincts.” We disagree.
In asserting this claim, the defendant appears to have mixed his trials. Miller testified at the defendant’s second trial that the statements were rational and relevant, but Miller gave no such testimony at his third trial that is the subject of this appeal. At the defendant’s third trial, Miller confined his testimony to an account of his role in the search and a recitation of his recollection, as refreshed by his report, of the defendant’s statements during the search. The defendant has not explained why Miller lacked sufficient powers of observation, recollection, narration and truthfulness simply [250]*250to recite the statements before the jury, without offering his personal opinion as to their significance or the defendant’s capacity. See State v. Weinberg, 215 Conn. 231, 244, 575 A.2d 1003, cert. denied, U.S. , 111 S. Ct. 430,112 L. Ed. 2d 413 (1990). Consequently, he has not demonstrated that Miller’s testimony failed to meet the threshold requirement of “minimum credibility” that would have warranted its exclusion as incompetent. Id., 243-44.
B
The defendant next argues that Miller’s testimony should have been excluded as irrelevant, or alternatively, as more prejudicial than probative. He asserts that Miller’s testimony was irrelevant because the defendant’s statements were “psychotic babblings” that the jury could not interpret or logically link to the case except by resort to “speculation” and “conjecture” that they referred to the victim. The defendant contends that Miller’s testimony, if relevant, should nonetheless have been excluded due to its “explosively prejudicial” effect. He maintains, in particular, that “[ljittle else could be as damaging as an open admission of guilt, combined with sneering disrespect for a fifteen year old girl’s death.” We are unpersuaded.
“ Tn a criminal trial, it is relevant to show the conduct of an accused, as well as any statement made by him subsequent to the alleged criminal act, which may fairly be inferred to have been influenced by the criminal act.’ ” State v. Reid, 193 Conn. 646, 655, 480 A.2d 463 (1984). The probative value of such statements stems from “ ‘our knowledge of the human mind and its workings, [for] we expect, with almost positive certainty, that when it is the sole repository of so dreadful a secret it will affect the conduct and sayings of the person.’ ” 3A J. Wigmore, Evidence (Chadbourn Rev. [251]*2511979) § 273. “No one doubts that the state of mind which we call ‘guilty consciousness’ is perhaps the strongest evidence . . . that the person is indeed the guilty doer . . . .” Id., § 273 (1); see State v. Moynahan, 164 Conn. 560, 596, 325 A.2d 199, cert. denied, 414 U.S. 976, 94 S. Ct. 291, 38 L. Ed. 2d 219 (1973).
Given the circumstances under which the defendant made the statements as well as their content, the trial court could quite properly have concluded that the statements were relevant manifestations of his consciousness of guilt. See State v. Maturo, 188 Conn. 591, 598, 452 A.2d 642 (1982). The police had previously interviewed the defendant with regard to his whereabouts on the date of the crime, and, prior to the defendant’s making of the statements, they had announced the purpose of the search in his presence and subjected him to a body search. At the time of the statements, therefore, the defendant was aware that the investigation of the victim’s death at that time was focused upon him, and he had reason to anticipate that he was to be accused of the crime. See State v. Moynahan, supra, 596-97; State v. Cronin, 64 Conn. 293, 305, 29 A. 536 (1894) (“manner in which [defendant] conducted himself when accounts by others in respect to the subject were made in his hearing, may always be shown”).
The content of the statements further demonstrates their relevance. Specifically, the defendant’s words betrayed his belief that a female had died, deservedly so, and his description of that female’s anatomical features was consistent with those of a female of the victim’s age. The jury could fairly have inferred that he had indeed spoken of the victim of the crime under investigation, thereby circumstantially corroborating the testimony of witnesses who had observed the defendant with the victim on the date of her death; see [252]*252State v. Alvarez, 216 Conn. 301, 309, 579 A.2d 515 (1990); State v. DeMatteo, 186 Conn. 696, 703, 443 A.2d 915 (1982); and making it more probable that he was responsible for her death. State v. Moody, 214 Conn. 616, 628, 573 A.2d 716 (1990).
Contrary to the defendant’s assertions, neither his failure to identify the female subject of his statements as the victim nor his alleged psychotic state renders Miller’s testimony irrelevant. Evidence need not be conclusive to be relevant; State v. Greene, 209 Conn. 458, 478, 551 A.2d 1231 (1988); and “[t]he fact that evidence is susceptible of different explanations or would support various inferences does not affect its admissibility, although it obviously bears upon its weight. So long as the evidence may reasonably be construed in such a manner that it would be relevant, it is admissible.” State v. Reid, supra, 656 n.12; see State v. Morrill, 197 Conn. 507, 508, 498 A.2d 76 (1985). A reasonable construction of the statements, for the reasons previously stated, would support an inference that the defendant had been referring to the victim. The defendant has not claimed that he was precluded from presenting evidence or argument suggesting that his statements were the product of an irrational mind and hence could not have referred to the victim.11 Accordingly, the trial court properly concluded that Miller’s testimony was relevant.
[253]*253Nonetheless, “ ‘[tjhere are situations where the potential prejudicial effect of relevant evidence would suggest its exclusion.’ ” State v. Holliman, 214 Conn. 38, 51, 570 A.2d 680 (1990). One of those situations is where the evidence offered may unduly arouse the jury’s emotions or hostility; id.; thereby unduly tending to suggest a decision on an improper, emotional basis. D. Louisell & C. Mueller, Federal Evidence § 403. Evidence is not prejudicial in this sense, however, merely because it tends to incriminate a defendant. See State v. DeMatteo, supra, 702-703; see also United States v. Figueroa, 618 F.2d 934, 943 (2d Cir. 1980). The prejudicial effect that the rules of evidence are designed to alleviate is that which “results from an aspect of the evidence other than its tendency to make the existence of a material fact more or less probable, e.g., that aspect of the evidence which makes conviction more likely because it provokes an emotional response in the jury or otherwise tends to affect adversely the jury’s attitude toward the defendant wholly apart from its judgment as to his guilt or innocence of the crime charged.” United States v. Bailleaux, 685 F.2d 1105, 1111 (9th Cir. 1982); see State v. DeMatteo, supra, 703. The trial court’s discretionary determination that the probative value of evidence is not outweighed by its prejudicial effect will not be disturbed on appeal unless a clear abuse of discretion is shown. State v. King, 216 Conn. 585, 603, 583 A.2d 896 (1990).
The defendant correctly points out that Miller’s testimony would have been “damaging” if the jury had believed that his statements referred to the victim. This aspect of Miller’s testimony, however, is precisely what infused it with considerable probative value. On the [254]*254other hand, the nonprobative and prejudicial aspects of Miller’s testimony included the remorseless tenor of the defendant’s statements and his use of expletives and sexually explicit language. These aspects of Miller’s testimony unquestionably had the tendency adversely to affect the jury’s attitude toward the defendant, and were we sitting as the trial court in this case, we might well have concluded that the testimony was too prejudicial for admission. Nevertheless, broad discretion is reposed in the trial court in conducting the inherently difficult process of balancing prejudicial effect against probative value. Id. We cannot say that the court abused its discretion in reaching the opposite conclusion.
IV
The defendant next claims that the trial court should not have permitted the state to introduce certain testimony by Daniel Pasquarelli in its case-in-chief. In connection with the state’s offer of proof, Daniel testified on voir dire that while he and the defendant had been incarcerated in the same facility before the defendant’s second trial, he had asked the defendant: (1) “if he would deny to my face. I was there, he was there, my brother was there, [the victim] was there. Deny to my face when me and my brother left you didn’t do it,” to which the defendant responded “no, I’m not going to”; and (2) “why [the defendant] did it. I said between me and you, we know we were there, why?” to which the defendant did not respond, and then responded that “he was on drugs that day. He was doing a lot of drugs.” Defense counsel objected to Daniel’s testimony due to his uncertainty whether the defendant’s response concerning drugs was a “direct response” to Daniel’s second question or whether “more questioning” occurred between the question and response. Find[255]*255ing the defendant’s second response a “direct response” after a “momentary lapse,” the court admitted Daniel’s testimony.
On the witness stand, however, Daniel gave a somewhat different account of his conversation with the defendant. At one point, he described the defendant’s response to his first question in the same manner as on voir dire. He then testified, however, that the defendant “didn’t answer” the first question, he had asked it again, and the defendant “waited a couple seconds and then said he was doing drugs. A lot of drugs.” As for the second question, Daniel testified that he had asked it a “couple times” before the defendant had responded that “he was doing a lot of drugs.” The court overruled the defendant’s objections to this testimony. No immediate curative instruction was requested.
Relying on Doyle v. Ohio, 426 U.S. 610, 96 S. Ct. 2240, 49 L. Ed. 2d 91 (1976), the defendant contends that the admission of Daniel’s testimony violated his federal and state constitutional rights to due process.12 [256]*256He argues that his refusals and hesitation to respond to Daniel’s inquiries constituted an invocation of his right to remain silent in reliance upon the “implicit assurance” that he would not be penalized for remaining silent.13 He also contends that his ultimate responses to Daniel’s questions were no less insolubly ambiguous than absolute silence, and thus were equivalent to silence and/or affirmative assertions of his right to remain silent. Consequently, he maintains that by introducing Daniel’s testimony, the state indulged in a “fundamentally unfair” use of his silence against him in contravention of the implicit assurance that his “silence” would not be so used.14 We do not agree.
The factual predicate of a claimed Doyle violation is the use by the state of a defendant’s postarrest and postMiranda silence either for impeachment or as affirmative proof of his guilt. In this case, however, the defendant did not remain silent, nor were his statements so fraught with ambiguity that they could be equated with silence. Despite his initial disinclination, the defendant made unequivocal statements about his “involvement in the crime.” Anderson v. Charles, 447 U.S. 404, 407, 100 S. Ct. 2180, 65 L. Ed. 2d 222, reh. denied, 448 U.S. 912, 101 S. Ct. 27, 65 L. Ed. 2d 1173 (1980).15 Specifically, he stated without evident com-
[257]*257pulsion that he would not deny “it,” that is, having committed the crime. While this statement, standing alone, might be considered ambiguous, the defendant’s second explanatory statement regarding his use of drugs both refers to his “involvement in the crime” and exposes the first statement as a reluctant acknowledgement rather than a rebuff. “[A] defendant who voluntarily speaks after receiving Miranda warnings has not been induced to remain silent. As to the subject matter of his statements, the defendant has not remained silent at all.” Id., 408; see State v. Talton, 197 Conn. 280, 295-96, 497 A.2d 35 (1985); compare State v. Morrill, 197 Conn. 507, 534-38, 498 A.2d 76 (1985). Rather than relying on the implicit assurance that silence would carry no penalty, therefore, the defendant failed to heed the warning that “anything said can and will be used against [him] in court.” Miranda v. Arizona, 384 U.S. 436, 469, 86 S. Ct. 1602, 16 L. Ed. 2d 694 (1966).
Furthermore, the state offered Daniel’s testimony for the permissible purpose of presenting the defendant’s statements, not his refusals to speak, as evidence of his guilt. Cf. United States v. Traitz, 871 F.2d 368, 401 (3d Cir.), cert. denied, 493 U.S. 821, 110 S. Ct. 78, 107 L. Ed. 2d 44 (1989) (prosecutor did not “manifestly intend” to comment on defendant’s silence). When Daniel unexpectedly deviated from his voir dire testimony, moreover, the court’s overruling of the defendant’s objections did not permit the state to inquire as to his silence, but rather, allowed the state to inquire [258]*258as to the statements previously related by Daniel. The state’s questions to Daniel on the stand, similarly, were designed to elicit the statements and not to draw meaning from his refusals to respond. See Anderson v. Charles, supra, 409; State v. Casey, 201 Conn. 174, 185, 513 A.2d 1183 (1986).
Consistent with this purpose, the state thereafter focused its closing remarks upon the statements, without mentioning the refusals to respond, as circumstantial evidence of the defendant’s guilt. The court did not refer to the defendant’s refusals in its charge; Greer v. Miller, 483 U.S. 756, 765, 107 S. Ct. 3102, 97 L. Ed. 2d 618 (1987); see State v. Shashaty, 205 Conn. 39, 50 n.4, 529 A.2d 1308 (1987), cert. denied, 484 U.S. 1027, 108 S. Ct. 753, 98 L. Ed. 2d 766 (1988); State v. Talton, supra; and, significantly, the defendant’s sole request to charge regarding Daniel’s testimony pertained to the relationship between the fact that he had been “on drugs” to the state’s burden of proving specific intent. See United States v. Ivey, 546 F.2d 139, 144-45 (5th Cir.), cert. denied sub nom. Taglione v. United States, 431 U.S. 943, 97 S. Ct. 2662, 53 L. Ed. 2d 263 (1977); accord State v. Silano, 204 Conn. 769, 782, 529 A.2d 1283 (1987). We conclude, accordingly, that the admission of Daniel’s testimony did not impair the defendant’s constitutional rights to due process of law.
V
Finally, the defendant claims that he was deprived of his federal constitutional right to present a defense by the court’s refusal to allow him to introduce circumstantial evidence of a beating that he allegedly had suffered at the hands of the Bristol police following his arrest for an unrelated incident in 1978.16 The evidence [259]*259proffered by the defendant consisted of a photograph, the testimony of two witnesses primarily describing the defendant’s appearance and behavior after the alleged beating, and the testimony of one witness corroborating that the defendant had commenced a civil suit against the city of Bristol and certain members of the Bristol police department in connection with the alleged beating.17 According to the defendant, the purpose of this evidence was to rebut the state’s argument that his lies to the police about his identity and location during the 911 calls, and his nervousness when questioned by the police one week after the crime occurred, evinced his consciousness of guilt. The court excluded the evidence, in part, because it would open the door to distracting evidence concerning the defendant’s arrest and the alleged beating.
The defendant argues that his evidence should have been admitted because it could logically be inferred from the fact of the beating that he had feared the police “strongly enough to conceal his identity to them,” which in turn could support a logical inference that his concealment and nervousness were consistent with innocence rather than consciousness of guilt. He [260]*260contends that “[wjhile the trial judge was not persuaded that the beating itself created a sufficient motive for the defendant to conceal his identity from the Bristol police . . . the fact remains that as a purely logical matter, it would not have been unreasonable or irrational for a juror to do so.”18 In response, the state argues that the court properly concluded that an excursion into the facts and circumstances surrounding the defendant’s arrest and the alleged beating would have unnecessarily complicated the case and unduly distracted the jury’s attention from the principal issues before them. We agree with the state.
The defendant’s argument in favor of the logical relevance of his evidence assumes that the fact of the beating was undisputed. Had the beating occurred, it would doubtless have provided a sufficient predicate for the logical chain of inferences that the defendant sought to draw. It is not logical relevance alone, however, that secures the admission of evidence. Logically relevant [261]*261evidence must also be “legally relevant”; see Sanderson v. Steve Snyder Enterprises, Inc., 196 Conn. 134, 145 n.8, 491 A.2d 389 (1985); that is, not subject to exclusion for any one of the following prejudicial effects: “(1) where the facts offered may unduly arouse the jury’s emotions, hostility or sympathy, (2) where the proof and answering evidence it provokes may create a side issue that will unduly distract the jury from the main issues, (3) where the evidence offered and the counterproof will consume an undue amount of time, and (4) where the defendant, having no reasonable ground to anticipate the evidence, is unfairly surprised and unprepared to meet it. McCormick, Evidence (2d Ed.) § 185, pp. 439-40.” State v. DeMatteo, 186 Conn. 696, 702-703, 443 A.2d 915 (1982); see Hamling v. United States, 418 U.S. 87, 127, 94 S. Ct. 2887, 41 L. Ed. 2d 590, reh. denied, 419 U.S. 885, 95 S. Ct. 157, 42 L. Ed. 2d 129 (1974). Where the prejudicial effect of logically relevant evidence outweighs its probative value, the trial court has wide latitude to exclude the evidence as legally irrelevant, and its decision will not be disturbed absent manifest abuse. See State v. Fritz, 204 Conn. 156, 169, 527 A.2d 1157 (1987); Farrell v. St. Vincent’s Hospital, 203 Conn. 554, 563, 566, 525 A.2d 954 (1987); State v. Sebastian, 81 Conn. 1, 4, 69 A. 1054 (1908); see also Hamling v. United States, supra. No such abuse occurred in this case.
In view of the critical importance of the beating as the factual foundation for the defendant’s proposed chain of inferences, the court would have been hard pressed to allow him to present unchallenged circumstantial evidence of the beating. The court would have been obliged, in fairness, to permit the state to dispute the defendant’s evidence of the beating by probing the circumstances attendant to his arrest and his injuries and by presenting counterproof. See State v. Wade, 96 [262]*262Conn. 238, 247-48, 113 A. 458 (1921); see also United States v. Renfro, 620 F.2d 497, 501 (5th Cir.), cert. denied, 449 U.S. 921, 101 S. Ct. 321, 66 L. Ed. 2d 149 (1980). Due to the prejudicial tendency of the defendant’s evidence to divert the trial from its primary course and provoke a barrage of distracting counterproof, we conclude that the trial court acted well within its discretion in excluding the evidence notwithstanding its logical relevance.19 See State v. Fritz, supra, 169-70; State v. Mastropetre, 175 Conn. 512, 517, 400 A.2d 276 (1978); accord State v. Talton, supra, 284-85.
The judgment is affirmed.
In this opinion the other justices concurred.