The state appeals a pretrial order granting defendant’s motion to suppress evidence and dismissing a one-count indictment against defendant for possession of methamphetamine. ORS 138.060(l)(c). The evidence was obtained after the police observed defendant and a companion, Jacobs, “bent over a bag” in front of a restaurant; initiated contact with them; implicitly denied defendant’s request to go to the bathroom; asked defendant and Jacobs for identification; said “hang on there, or hang on a second”; ran a records check that revealed an outstanding warrant for defendant’s arrest; arrested defendant on the warrant; and searched defendant incident to that arrest. We conclude that the evidence was obtained after defendant was unlawfully seized in violation of his rights under Article I, section 9, of the Oregon Constitution, and that the state failed to prove attenuation — that is, the state failed to prove, under the totality of the circumstances, that the violation of defendant’s rights under Article I, section 9, had such a tenuous factual link to the disputed evidence that the unlawful police conduct cannot be properly viewed as the source of that evidence. Accordingly, we affirm.
We are bound by the trial court’s findings of fact as long as there is constitutionally sufficient evidence to support them. State v. Ehly, 317 Or 66, 74-75, 854 P2d 421 (1993). In the absence of express factual findings, we presume that the trial court decided the disputed facts in keeping with its ultimate conclusion. Id. at 75. On appeal, “[o]ur function is to decide whether the trial court applied legal principles correctly to those facts.” Id. We state the facts consistently with those standards.
Officer Lemons, who worked for the East Precinct street crimes unit, was on daytime patrol in Portland when he saw defendant and Jacobs, “bent over a bag” in front of a restaurant. Lemons was “not real sure what was going on there,” so he “[c]ircled back around” in his patrol vehicle, “intend [ing] to contact the two.”
As soon as defendant and Jacobs saw Lemons circling back around, “they started walking away.” In Lemons’s *185experience, when people see the police and “avoid police contact, something is * * * usually up.” Thus, Lemons’s suspicion was raised — he wondered, “why leave the area just because of me?” However, at that point, Lemons did not believe that he had “reasonable suspicion to actually stop [defendant] for a crime.” In fact, Lemons “definitely [did not] think [he had] a crime.” Rather, he had a “hunch,” and he wanted “to know who these folks [were] and why they [were] wanting to leave the area because [Lemons was] there.”
Lemons parked his patrol vehicle, got out of the vehicle, and began walking toward defendant and Jacobs. During his approach, Lemons said something to the effect of, “‘Hey, what’s going on, guys?”’ As Lemons continued to walk toward defendant and Jacobs, defendant and Jacobs came back toward Lemons. Lemons then asked the men “what they were up to” and “asked what was in the bag.” The men responded that the bag contained cans, and, although Lemons did not open the bag, he had “no reason to doubt” that the bag contained cans. Defendant then told Lemons that “he ha[d] to go to the bathroom.”
Lemons did not directly respond to defendant’s request to go to the bathroom; instead, Lemons asked defendant and Jacobs for identification. Defendant, who did not have an identification card, told Lemons his name and date of birth, and Lemons apparently wrote that information down in a notebook. Jacobs handed Lemons an identification card, which Lemons retained. Lemons then told defendant and Jacobs to “hang on there, or hang on a second.” Lemons returned to his patrol vehicle to run a records check. While Lemons was at his patrol vehicle running the records check, Officer Slyder and another officer arrived as backup. At some point, Officer Edwards arrived as backup, apparently along with his partner, Officer Strawn.
The records check revealed an outstanding warrant for defendant’s arrest. Lemons did not know about the warrant before he ran the records check, and he did not later recall what the warrant was for. If Lemons had not discovered the outstanding warrant, he would have “let [defendant] go on his way” — that is, he “would have let him know that he was free to leave” because he “had no reason *186to arrest him.” Instead, Slyder informed defendant that he would be taken into custody.
Defendant was asked whether there was anything on him that Lemons was going to find, and defendant said that he “might have” marijuana or “crystal” on him. Lemons advised defendant of his Miranda rights, searched a “coin pocket” on defendant’s jeans, and found a folded piece of paper containing a substance that later tested positive for methamphetamine. Defendant was subsequently charged with unlawful possession of methamphetamine, in violation of ORS 475.894(1).
As the case proceeded to trial, defendant moved to suppress the “evidence discovered subsequent to the illegal detention [,]” arguing that the evidence was obtained in violation of his rights under Article I, section 9, and the Fourth Amendment to the United States Constitution. Specifically, defendant argued that he was unlawfully seized when Lemons conducted a records check on him and that the discovery of the outstanding warrant did not “ ‘purge the taint’ ” of the unlawful seizure.
At a hearing on the matter, the state responded that defendant was not seized and alternatively argued that, even if defendant was seized, suppression was not warranted because the discovery of the outstanding warrant “create [d] an attenuation.” To support its attenuation argument, the state relied upon State v. Dempster, 248 Or 404, 408, 434 P2d 746 (1967), in which the Oregon Supreme Court concluded that the lawful arrest of the defendant on an outstanding warrant “purged the search incident thereto of the taint of any illegality in the detention of defendant prior to that time.”
Relating to seizure, the trial court ruled:
“[U]nder these facts and circumstances, the officer did not have a reasonable suspicion or probable cause. And I would find that walking up to these gentlemen, making this inquiry, taking identification, going back to the car leads me to a reasonable assumption with him saying hang on there, or hang on a second that he is being stopped and he is not free to leave.
*187“There’s other officers showing up. No reasonable person, I would find under the totality of circumstances, would feel free to leave. And, in fact, I don’t think he was free to leave because had he started walking off, I’ll bet you the officer would have grabbed him.
“So I’m comfortable this was an unlawful stop by the police officers and he was stopped.”
The trial court then reviewed both state and federal case law, commenting on both state and federal constitutional theories of suppression. The court stated:
“Nothing stops a police officer from having contact with citizens and talking to them and asking questions. It only becomes an issue when the police use some illegality by requiring a person to do something that they have a lawful right not to do. And then when they intend to use that evidence as — as a result of exploitation.
“So the question for me is do I think that the Courts are moving towards excluding tainted evidence or moving towards allowing more tainted evidence in so that people that have warrants out or commit a crime should be punished for what they did?”
After noting that the “trend of the law is moving towards excluding this evidence [,]” the court ruled:
“And I would find that, based on the evidence and circumstances that I’ve heard, that the State has not met its burden and convinced me that this evidence is not tainted and I think because of the illegality, they exploited an illegal stop of the defendant and I am going to suppress the evidence.”
The state now appeals.
On appeal, the state essentially reprises its arguments under both Article I, section 9, and the Fourth Amendment. Thus, we must first determine whether defendant was seized in violation of his constitutional rights. If we conclude that a constitutional violation occurred, we must then consider whether the state proved attenuation — that is, whether the state proved that the violation of defendant’s rights had such a tenuous factual link to the disputed evidence that the unlawful police conduct cannot be properly *188viewed as the source of that evidence. We consider questions of state law first. See State v. Kennedy, 295 Or 260, 262-63, 666 P2d 1316 (1983) (court considers and disposes of questions of state law before reaching federal law claims).
Article I, section 9, provides that the people have the right “to be secure in their persons, houses, papers, and effects, against unreasonable search, or seizure!.]” When construing Article I, section 9, the Oregon Supreme Court “has identified three categories of encounters along the continuum of meetings between law enforcement officers and citizens!.]” State v. Holmes, 311 Or 400, 407, 813 P2d 28 (1991). Those categories are:
“(1) a mere conversation or noncoercive encounter that involves no restraint of liberty and, therefore, is not a seizure that requires any justification; (2) a stop, which is a type of seizure that occurs when an officer temporarily restrains a person’s liberty or freedom of movement, that must be justified by reasonable suspicion of criminal activity; and (3) an arrest, which also is a type of seizure, that must be justified by probable cause to believe that the person arrested has committed a crime.”
State v. Toevs, 327 Or 525, 534-35, 964 P2d 1007 (1998) (citing Holmes, 311 Or at 407).
The Oregon Supreme Court has also acknowledged that, “in practice, the line between a mere encounter and something that rises to the level of a seizure does not lend itself to easy demarcation.” State v. Backstrand, 354 Or 392, 399, 313 P3d 1084 (2013) (internal quotation marks omitted). On the one hand, police officers are “free to approach persons on the street or in public places, seek their cooperation or assistance, request or impart information, or question them without being called upon to articulate a certain level of suspicion in justification if a particular encounter proves fruitful.” Id. at 400 (citing Holmes, 311 Or at 410). Thus, “[t]he fact that the citizen is discomforted by an officer’s approach and request for assistance or information — either because / the officer is a known police officer, or because the encounter otherwise involves ‘inconvenience or annoyance’ — does not make the contact a seizure.” Id. (quoting Holmes, 311 Or at 410).
*189On the other hand, a seizure occurs only if the officer’s conduct would be “reasonably perceived as coercive in the sense that it would cause [a] citizen to reasonably believe that the officer is intentionally restraining the citizen’s liberty or freedom of movement in a significant way — that is, in a way that exceeds the bounds of ordinary social encounters between private citizens.” Id. In a “show of authority” that gives rise to a seizure in the constitutional sense, a person must have “a reasonable perception that [the] officer is exercising his or her official authority to restrain” — that is, a seizure occurs when an officer conveys, either explicitly or implicitly, “to the person with whom he is dealing, either by word, action, or both, that the person is not free to terminate the encounter or otherwise go about his or her ordinary affairs.” Id. at 401 (internal quotation marks omitted). The test is an objective one, and is determined under the totality of the circumstances: “Would a reasonable person believe that a law enforcement officer intentionally and significantly restricted, interfered with, or otherwise deprived the individual of his or her liberty or freedom of movement.” Id. at 399.
In State v. Hall, 339 Or 7, 10, 115 P3d 908 (2005), an officer saw the defendant walking along a street, stopped his vehicle next to the defendant, and then motioned with two fingers for the defendant to approach him. When the defendant approached, the officer got out of his vehicle and asked the defendant if he had any personal identification. Id. The defendant handed the officer an identification card, which the officer used to run a warrant check. Id. The officer returned the identification card to the defendant “before he had received back any information [.]” Id. Noticing that the defendant appeared to be carrying something inside his jacket, the officer then asked the defendant if he was carrying any weapons, knives, or illegal drugs, and the defendant, after replying in the negative, consented to a patdown search, during which the officer discovered a controlled substance. Id. at 10-11.
The Oregon Supreme Court determined that the meeting at issue in Hall began as a noncoercive encounter between the defendant and the officer but then evolved into *190a seizure. Id. at 19. As the court explained, the officer’s “initial actions of stopping his vehicle next to defendant and then gesturing for defendant to approach him did not intrude upon defendant’s liberty of movement, because, even if [the officer] inconvenienced defendant, his actions did not constitute a show of authority involving conduct significantly beyond that accepted in ordinary social intercourse.” Id. (internal quotation marks omitted). But the court concluded that “the consensual nature of that encounter dissipated” when the officer took the defendant’s identification card and radioed for a warrant check. Id. The court acknowledged that the officer promptly returned the defendant’s identification card, but maintained that, at that point, the defendant was aware that he was the subject of a pending warrant check and, because of that fact, it was “difficult to posit” that a reasonable person would have felt free to leave “when that person is the investigatory subject of a pending warrant check.” Id.
The court in Hall further observed that the officer “did nothing to dispel what would have been an objectively reasonable belief that defendant was restrained from leaving until [the officer] had received the results of the warrant check.” Id. When later commenting on its seizure analysis in Hall, the court stated that “none of the officer’s actions (hailing defendant, asking for identification, checking that identification, asking about weapons and drugs, asking for consent) individually was sufficient to amount to a stop. In combination, however, *** those actions crossed over the line and transformed what began as a mere encounter into a stop.” State v. Highley, 354 Or 459, 473, 313 P3d 1068 (2013).
In Backstrand, a deputy was monitoring a “ ‘triple-X’ ” store that sold adult sexual materials while the defendant and his girlfriend were inside shopping. 354 Or at 394-95. The deputy thought that the defendant looked “‘pretty young’” and believed he might be under the posted 18-year minimum age to be in the store, and so he approached the two and asked their ages. Id. at 395. After the defendant answered that he was 22, the deputy asked both the defendant and his girlfriend if they had any identification, and they gave him their driver licenses. Id. The deputy called *191dispatch to verify the validity of the licenses, and, after having the licenses for 10 to 15 seconds, returned the licenses to the defendant and his girlfriend. Id. The deputy then left the store and continued to monitor the store from outside, while the defendant and his girlfriend continued to shop inside. Id. The deputy, who had not asked dispatch to check on anything other than the validity of the licenses, was then advised by dispatch that the defendant’s license was suspended and the defendant was on probation in another county. Id. After the defendant later drove away from the store, the deputy initiated a traffic stop and arrested the defendant for driving with a suspended license. Id.
The Oregon Supreme Court determined that the encounter at issue in Backstrand was not a seizure. Id. at 413-16. Regarding a request for identification, the court stated that “[a]sking a citizen to identify himself or herself and to show police a formal piece of identification is a form of cooperation and involves the kind of information that, as a general proposition, police are free to request.” Id. at 412. However, the court noted that “when the content of the questions, the manner of asking them, or other actions that police take (along with the circumstances in which they take them) would convey to a reasonable person that the police are exercising their authority to coercively detain the citizen, then the encounter rises to the level of a seizure, the lawfulness of which must be analyzed as such.” Id. Regarding verification of identification, the court stated that the verification of a citizen’s identification does not, in and of itself, elevate a mere encounter into a seizure, because there was “no principled basis for concluding that, when an officer checks the validity of a proffered identity or piece of identification, such an action per se conveys to a reasonable person — who is not otherwise restrained and who has willingly tendered the information to the officer that the officer is now exercising his or her authority to coercively restrain the person’s liberty or freedom of movement.” Id.
Applying those principles to the circumstances presented in Backstrand, the court stated that asking a person his or her age in an age-restricted store “would not cause a reasonable person to believe that the officer had significantly *192restricted his or her liberty.” Id. at 414 (emphasis in original). The court then stated that “there was nothing distinctive about the content of [the deputy’s] questions” or the manner of his request that caused his mere inquiries to amount to a seizure. Id. at 415-16. The court further stated that there was nothing in the record “to suggest that [the deputy] was overbearing, intimidating, or coercive in his demeanor or behavior” and that the deputy “merely asked for, and defendant complied with, his request for identification.” Id. at 416.
With the circumstances of those cases in mind, we return to the facts of this case. Under the principles articulated in Backstrand, the mere fact that Lemons requested defendant’s identification does not, by itself, indicate that defendant was seized, nor does the fact that Lemons ran a records check. However, that conclusion does not end our analysis; under Backstrand, we must analyze whether “the content of the questions, the manner of asking them, or other actions that police [took] (along with the circumstances in which they [took] them) would convey to a reasonable person that the police [were] exercising their authority to coercively detain the citizen [.]” Id. at 412.
When analyzing the circumstances in this case, we first note that Lemons’s initial conduct — approaching defendant and Jacobs while saying something to the effect of, “Hey, what’s going on, guys?” and then asking “what they were up to” and “what was in the bag” — did not effectuate a seizure because, even if Lemons inconvenienced defendant by changing his direction of travel, Lemons’s initial conduct would not cause a citizen to reasonably believe that Lemons was exercising his authority to coercively detain the citizen or intentionally restrain the citizen’s liberty or freedom of movement in a significant way — that is, in a way that exceeded the bounds of ordinary social encounters between private citizens. See Hall, 339 Or at 19 (concluding that the officer’s “initial actions of stopping his vehicle next to defendant and then gesturing for defendant to approach him did not intrude upon defendant’s liberty of movement, because, even if [the officer] inconvenienced defendant, his actions did not constitute a show of authority involving conduct significantly beyond that accepted in ordinary social intercourse”) (internal quotation marks omitted).
*193However, under all the circumstances surrounding the encounter as it developed in this case, we conclude that defendant was unlawfully seized by the time that Lemons told defendant and Jacobs to “hang on there, or hang on a second.” Arguably, the trial court reasonably could infer that the encounter had lost its consensual nature even before then — that is, when Lemons requested defendant’s identification. That is so because, by that point, defendant had sent Lemons a signal that he did not wish to have a police encounter by walking away from Lemons; returned and complied with Lemons’s request for information about contents of the bag; and then sent Lemons a more explicit signal that he wanted to leave by stating, directly, that he needed to go to the bathroom. Because Lemons did not address defendant’s request to go to the bathroom but, instead, responded by asking defendant for identification, perhaps the trial court could reasonably have inferred from those circumstances that Lemons had restrained defendant’s liberty by means of a show of authority — that is, by implicitly denying defendant’s request to go to the bathroom.
We need not determine whether defendant was unlawfully seized when Lemons requested defendant’s identification, however, because — even if he was not — the trial court reasonably could infer that the encounter evolved into a seizure when Lemons took additional actions that culminated in a communication that the trial court described as “hang on there, or hang on a second.” As we understand the trial court’s finding, it is that Lemons conveyed, through his words and actions, that defendant and Jacobs should remain until Lemons could return after checking the identification. By that point in time, the police conduct (approaching defendant and Jacobs while they were walking away from Lemons in an attempt to avoid police contact; asking for information about the contents of the bag; after receiving information about the contents of the bag, implicitly denying defendant’s request to go to the bathroom; asking defendant for identification after defendant had communicated a need to go to the bathroom; recording defendant’s identifying information in a notebook and retaining a companion’s identification card; telling defendant and Jacobs to “hang on there, or hang on a second”; and running a records check while as many as two *194additional patrol vehicles and four additional police officers were arriving on the scene) would have communicated to a reasonable person that the person was not free to terminate the encounter or decline when Lemons communicated that the person should “hang on there, or hang on a second.” In other words, we conclude that a reasonable person in defendant’s position would not feel free to leave and that defendant was seized in violation of his rights under Article I, section 9, because the actions of the police, when viewed in light of the trial court’s findings, “crossed over the line and transformed what began as a mere encounter into a stop.” Highley, 354 Or at 473.
When the state has obtained evidence following the violation of a defendant’s rights under Article I, section 9, we presume “that the evidence was tainted by the violation and must be suppressed.” State v. Jackson, 268 Or App 139, 151, 342 P3d 119 (2014) (citing State v. Unger, 356 Or 59, 84, 333 P3d 1009 (2014)). The state may rebut that presumption by proving, as relevant here, that the police did not exploit the unlawful police conduct to obtain the challenged evidence— that is, that the unlawful police conduct was “independent of, or only tenuously related to” the disputed evidence. Hall, 339 Or at 35; see Unger, 356 Or at 84 (adhering to that requirement, as stated in Hall). Stated another way, the evidence in this case must be suppressed unless the state proves attenuation — that is, that the violation of defendant’s rights had such a tenuous factual link to the disputed evidence that the unlawful police conduct cannot be properly viewed as the source of that evidence.
At the suppression hearing, when arguing that the evidence need not be suppressed, the state relied upon the per se rule of attenuation set forth in Dempster, 248 Or at 404. In Dempster, the Oregon Supreme Court concluded, under the Fourth Amendment, that the lawful arrest of the defendant on an outstanding warrant “purged the search incident thereto of the taint of any illegality in the detention of defendant prior to that time.” Id. at 408. We later applied Dempster in State v. Snyder, 72 Or App 359, 695 P2d 958, rev den, 299 Or 251 (1985), and the “Dempster/Snyder rule” became the basis for determining, under Article I, section 9, the legal effect of the discovery and execution of an *195outstanding arrest warrant in cases involving prior unlawful police conduct. See, e.g., State v. Langston, 223 Or App 590, 594-95, 196 P3d 84 (2008) (“ [a]pplying the Dempster/Snyder rule to [the] case,” and concluding, under Article I, section 9, that “the discovery and execution of the outstanding warrant for defendant’s arrest purged the taint of, any prior unlawful stop”); State v. Allen, 222 Or App 71, 78-79, 191 P3d 762, rev den, 345 Or 503 (2008) (citing Dempster and Snyder and concluding, under Article I, section 9, “that the intervening and independent event of the discovery of the outstanding arrest warrants operated to attenuate the taint of the prior unlawful arrest”); State v. La France, 219 Or App 548, 557-59, 184 P3d 1169 (2008), rev den, 349 Or 664 (2009) (setting forth the Dempster and Snyder principle “that the existence of an outstanding arrest warrant can serve to attenuate the link between a police illegality and evidence discovered thereafter [,]” and concluding, under Article I, section 9, that the record was “inadequate for us to apply the rule oí Dempster and Snyder”).
However, after the trial court granted defendant’s motion to suppress in this case, the Oregon Supreme Court disavowed its holding in Dempster, stating that “Dempster’s per se rule is inconsistent with the subsequent development of the Fourth Amendment attenuation exception set out in Brown v. Illinois, 422 US 590, 95 S Ct 2254, 45 L Ed 2d 416 (1975), where the United States Supreme Court rejected such an approach.” State v. Bailey, 356 Or 486, 488, 338 P3d 702 (2014). Rather than applying Dempster’s per se rule of attenuation, the Oregon Supreme Court in Bailey applied Brown’s three-factor test to determine, under the Fourth Amendment, whether the causal connection between the unlawful police conduct and the challenged evidence was sufficiently attenuated so as to purge the taint of the illegality. Id. at 505-06. Those factors are (1) the temporal proximity between the unlawful police conduct and the discovery of the challenged evidence; (2) the presence of intervening circumstances; and (3) the purpose and flagrancy of the unlawful police conduct. Id.
To resolve this case under Article I, section 9, we must first decide whether our holding in Snyder can still be considered good law. In Snyder, the defendant appealed the *196trial court’s denial of his motion to suppress evidence that was obtained after the defendant was taken to the police station and arrested on an outstanding warrant. 72 Or App at 363. The trial court had denied the motion, in part, on the ground that the discovery of the outstanding warrant “purged the evidence of taint from prior illegality, if any.” Id. We agreed with the trial court on that ground and stated:
“Under State v. Dempster, 248 Or 404, 434 P2d 746 (1967), [the officer’s] discovery of the arrest warrant and his arrest of defendant under the warrant’s authority purged the evidence from the taint of prior illegality. The Supreme Court in Dempster assumed that the police may have unlawfully taken the defendant to the police station. While the defendant was there, an officer discovered a bench warrant for his arrest. The officer then conducted a search incident to the arrest and found narcotics and related paraphernalia on the defendant’s person. The court held that the intervening discovery of the bench warrant dissipated the taint of the illegal detention because ‘when the sergeant found the warrant he was bound to obey its command and arrest defendant.’”
Id. at 364 (quoting Dempster, 248 Or at 407). We then rejected the defendant’s attempts to distinguish Dempster. Id. at 364-65.
Our review of Snyder demonstrates that in Snyder we did not consider or evaluate Dempster’s animating principles, and we did not separately analyze whether Dempster’s per se attenuation rule should be applied in the context of Article I, section 9. Rather, we merely cited Dempster and, without stating whether we were conducting a state or federal constitutional analysis, agreed with the trial court’s conclusion, under Dempster, that the discovery of an outstanding “warrant purged the evidence of taint from prior illegality, if any.” Id. at 363-64. Accordingly, Snyder does not rely on its own analysis, but relies solely on Dempster, which has now been disavowed. Because our holding in Snyder cannot stand without Dempster, its decisional precedent, we conclude that our holding in Snyder can no longer be considered good law.
Having concluded that our holding in Snyder can no longer be considered good law, we must now address the *197parties’ arguments as to whether it is appropriate to apply Bailey or Unger in the context of Article I, section 9. The state argues that, although Bailey is a Fourth Amendment case, the attenuation analysis set forth in Brown and adopted by Bailey is similar or identical under Article I, section 9, apparently suggesting that we should apply the three-factor Brown test to resolve this case: the temporal proximity between the unlawful police conduct and the discovery of the challenged evidence; the presence of intervening circumstances; and the purpose and flagrancy of the unlawful police conduct. Defendant argues that the Article I, section 9, attenuation analysis should be conducted under the exploitation test established by Hall and modified by Unger, contending that we should consider the totality of the circumstances, with reference to five considerations: the temporal proximity between the unlawful police conduct and the discovery of the challenged evidence; the presence of mitigating circumstances; the presence of intervening circumstances; the purpose and flagrancy of the unlawful police conduct; and the nature and extent of the constitutional violation. See Unger, 356 Or at 86 (summarizing considerations that are relevant to the exploitation inquiry). For the reasons that follow, we conclude that it is appropriate to apply the exploitation analysis portion of the Unger test to determine, under the totality of the circumstances, the legal effect of the discovery and execution of an outstanding warrant on prior unlawful police conduct under Article I, section 9.
First, as the Oregon Supreme Court has explained, “the overarching inquiry” in Unger was “whether the evidence that the state [sought] to introduce must be suppressed because that evidence was obtained in violation of the defendant’s constitutional rights.” Id. at 85. Thus, although Unger involved the question whether evidence must be suppressed following a defendant’s consent to search, rather than the question whether evidence must be suppressed following the discovery and execution of an outstanding arrest warrant, the overarching inquiry in Unger is the same as the overarching inquiry in this case.
Second, the totality of the circumstances test, as expressed in Hall, has long served as the legal foundation for determining whether the discovery of an outstanding *198warrant operates to attenuate the taint of the prior illegality under Article I, section 9, and the Dempster!Snyder rule has been applied within its framework. Allen is exemplary. In Allen, “[t]he state appeal [ed] the trial court’s pretrial order under Article I, section 9, * * * granting defendant’s motion to suppress all evidence obtained by the police resulting from a contact with defendant that ended with his arrest on outstanding arrest warrants, including items seized after the search of his jacket.” 222 Or App at 73. When addressing the suppression issue in Allen, we first noted that, “ [i] n Hall, the [Oregon Supreme C]ourt explained how Oregon’s exclusionary rule operates in these circumstances.” Id. at 77. We then framed the issue as whether “under the totality of the circumstances, the state ha[d] carried its burden under Hall to prove that the preceding violation of defendant’s rights under Article I, section 9, ha[d] such a tenuous factual link to the disputed evidence that the unlawful police conduct cannot be viewed properly as the source of that evidence.” Id. at 78. When conducting that totality of the circumstances inquiry, we considered the state’s “contention that the discovery of the arrest warrants purged the taint of the unlawful arrest” based on Dempster and Snyder, “where we and the Supreme Court held that the existence of an outstanding arrest warrant can serve to attenuate the link between a police illegality and evidence discovered thereafter.” Id. at 78-79. We ultimately concluded, “as we did in Snyder, that the intervening and independent event of the discovery of the outstanding arrest warrant operated to attenuate the taint of the prior unlawful arrest.” Id. at 79.
Although the test that we apply in this case must be modified to reflect recent changes in the law, including Unger’s partial disavowal and modification of Hall and our conclusion that our holding in Snyder can no longer be considered good law, as well as the specific context of this case, we see no reason to deviate from our established analytical framework, which was developed under Article I, section 9, and designed to serve its aims. See Hall, 339 Or at 24 (stating, while developing its two-part suppression test, that “the aim of the Oregon exclusionary rule is to restore a defendant to the same position as if the government’s officers had stayed within the law” (internal quotation marks omitted)).
*199We thus apply the exploitation analysis portion of the Unger test to determine whether the state proved, under the totality of the circumstances, that the violation of defendant’s rights under Article I, section 9, had such a tenuous factual link to the disputed evidence that the unlawful police conduct cannot be properly viewed as the source of that evidence.1 The underlying question in the exploitation inquiry is “whether police ‘exploited’ or ‘took advantage of or ‘traded on’ their unlawful conduct” to obtain the challenged evidence, or — stated another way — whether the challenged evidence was “‘tainted’ because it was ‘derived from’ or was a ‘product of the unlawful conduct.” Unger, 356 Or at 80. In this case, we focus on the temporal proximity between the unlawful police conduct and the discovery of the challenged evidence; the presence of mitigating circumstances; the presence of intervening circumstances; the purpose and flagrancy of the unlawful police conduct; and the nature, extent, and severity of the constitutional violation.
We first consider temporal proximity. See id. at 86 (identifying temporal proximity as a consideration in the exploitation analysis). In this case, although the state presented no direct evidence as to how much time elapsed between the unlawful seizure of defendant and the discovery of the challenged evidence, it appears that those two events occurred in close proximity to one another. After Lemons unlawfully conveyed, through his words and actions, that defendant and Jacobs should remain until Lemons could return, Lemons walked back to his patrol vehicle, ran the records check, and returned with Slyder, who placed defendant under arrest. Thus, the first consideration — temporal proximity — suggests that there was no extended temporal break that may have served to attenuate the factual link between the unlawful seizure and the discovery of the evidence, and the first consideration weighs in favor of suppression. See State v. Clemons, 267 Or App 695, 701, 341 P3d 810 *200(2014) (concluding under the Fourth Amendment that the first factor — temporal proximity — weighed in favor of suppression because little or no time had elapsed between the unlawful extension of a traffic stop and the discovery of the evidence at issue).
We next consider the existence of any mitigating circumstances. See Unger, 356 Or at 86 (identifying mitigating circumstances as a consideration in the exploitation analysis). In this case, the state does not argue that any mitigating circumstances occurred between the unlawful police conduct and the discovery of the evidence and presented no evidence of any mitigating circumstances. For example, defendant was not told that he was not required to remain while Lemons ran a records check or that he was free to go to the bathroom, as he had requested. See State v. Kuschnick, 269 Or App 198, 211, 344 P3d 480 (2015) (applying Unger and stating that the parties agreed that there were no mitigating circumstances to separate the defendant’s “consent from the unlawful police conduct: Defendant was not told that he could simply remain in the car or that he could refuse to consent to the search”). Thus, the second consideration — the absence of any mitigating circumstances — weighs in favor of suppression.
We next consider the existence of any intervening circumstances. See Unger, 356 Or at 86 (identifying intervening circumstances as a consideration in the exploitation analysis). The posited intervening circumstance here is the discovery of the arrest warrant, which provided a lawful basis for defendant’s arrest and subsequent search incident to arrest and, under the Dempster/Snyder rule, would have independently operated to purge the taint of the unlawful police conduct. Now that the Dempster I Snyder rule is no longer good law, we are left to consider the significance of the discovery of the warrant as an intervening circumstance case by case.
When considering how to view the strength of the causal connection between an unlawful seizure and the challenged evidence in the context of Article I, section 9, we apply the reasoning of the Oregon Supreme Court, in the context of the Fourth Amendment, that “the weight assigned *201to the discovery of the arrest warrant depends on the degree to which it was the direct consequence or objective of the unlawful detention.” Bailey, 356 Or at 505-06. We reiterate the facts that defendant initially attempted to walk away from Lemons — that is, to “avoid police contact” — and then later, after returning to answer Lemons’s question about the contents of the bag, stated that he had to go to the bathroom. Thus, viewing the record in keeping with the trial court’s findings, if Lemons had not taken actions that culminated in a communication that the trial court described as “hang on there, or hang on a second,” defendant would have terminated the encounter with Lemons. Accordingly, the discovery and execution of the outstanding warrant was the direct consequence of the unlawful detention.2 Further, although it does not appear that the discovery of the warrant was a specific objective of the unlawful detention, as Lemons did not know of the existence of the warrant before running the records check, the discovery of the arrest warrant was a general objective of the unlawful detention because, as we discuss below, the nature of the unlawful detention was investigatory. Thus, the third consideration — the presence of intervening circumstances — weighs in favor of suppression.
We next consider the purpose and flagrancy of the unlawful police conduct. See Unger, 356 Or at 86 (identifying “the purpose and flagrancy of the misconduct” as a consideration in the exploitation analysis). Our task is to objectively evaluate the circumstances, focusing on “statements made by the [officers] and the undisputed facts surrounding the contact with defendant,” rather than their “subjective intent or motivations.” Unger, 356 Or at 90.
In this case, the trial court made no express factual finding as to the purpose of the unlawful police conduct. *202However, while the prosecutor was arguing that defendant had not been seized, the trial court and the prosecutor engaged in the following exchange:
“[PROSECUTOR]: And then the next question is whether or not the officer’s request that he stay there, would that convert this to a stop. And I don’t believe that, as the officer has termed this, that would amount to a significant enough show of authority that a reasonable person would feel that they were stopped.
“THE COURT: What other purpose would there be to
getting his name and address and going back to his car? If you can think of one, I’d appreciate it.
“ [PROSECUTOR]: I don’t know — I think I’d be speculating. I don’t know. I’d have to ask the officer. We could call him back if you’d like to hear more about that issue.
“THE COURT: No, no. I was just thinking maybe you thought there was another reason he might do that. The only reason I can think of is that he was going to run a records check and that’s what the defendant probably thought. I’m just assuming that.”
(Emphases added.) Thus, the trial court inferred that the purpose of the unlawful police conduct was to run a records check — a process that is, by nature, investigatory. That inference is reasonable in this case, because Lemons told defendant and Jacobs to “hang on there, or hang on a second,” conveying, through his words and actions, that defendant and Jacobs should remain until Lemons could return after checking the identification. We therefore conclude that the purpose of the unlawful seizure was investigatory — that is, it was a “‘shot in the dark’” to see what might turn up. See State v. Musser, 356 Or 148, 159, 335 P3d 814 (2014) (as part of an analysis leading to the conclusion that “the police improperly exploited their unlawful stop of defendant to obtain her consent to [a] search,” applying the purpose factor under Article I, section 9, and deciding that the purpose of the unlawful stop “apparently was a ‘shot in the dark’ to check for criminal activity”).
Further, Lemons’s unlawful seizure of defendant may be characterized as flagrant, in the sense that it was a fishing expedition that occurred after defendant had *203communicated a need to go to the bathroom and while Lemons felt, as he later affirmed, that there was no basis “to investigate an actual crime.” Lemons’s conduct here is analogous to the conduct of the officer in Clemons, who “was engaged in a purposeful and flagrant fishing expedition” in violation of the defendant’s rights under the Fourth Amendment, when he unlawfully extended a traffic stop without probable cause or reasonable suspicion to believe that a crime had been committed. 267 Or App at 702. Thus, the fourth consideration — the purpose and flagrancy of the unlawful police conduct — weighs in favor of suppression.
Finally, we consider the nature, extent, and severity of the constitutional violation. See Unger, 356 Or at 86 (identifying “the nature, extent, and severity of the constitutional violation” as a consideration in the exploitation analysis). In Unger, the Oregon Supreme Court explained that “ [i] f the conduct is intrusive, extended, or severe, it is more likely to influence improperly a defendant’s consent to search. In contrast, where the nature and severity of the violation is limited, so too may be the extent to which the defendant’s consent is ‘tainted.’” Id. at 81. As we recently explained in Kuschnick, three recent cases decided by that court— Unger, State v. Lorenzo, 356 Or 134, 335 P3d 821 (2014), and Musser — “illuminate [a] qualitative, fact-intensive distinction” involved in the consideration of the “nature, extent, and severity of the constitutional violation”:
“In both Unger and Lorenzo, the court concluded that the police misconduct was limited rather than severe. In Unger, four officers, who were investigating a complaint about drug activity, unlawfully trespassed on the defendant’s property when they followed a path around to the defendant’s back door, where the defendant eventually consented to their entry. The court reasoned that, although the officers had conducted an unlawful ‘search’ to reach the back door, they had interacted with the defendant ‘just as they would have at the front door,’ their conduct ‘did not rise to the level of an unlawful arrest or stop,’ and they ‘did not unlawfully enter [the] defendant’s home or ignore any gates or no trespassing signs.’ Unger, 356 Or at 89, 92. In Lorenzo, an officer, who was concerned for the defendant’s safety, opened the defendant’s apartment door and reached in to knock on a bedroom door in order to contact the *204defendant, who later consented to the officer’s entry. The officer’s unlawful search, the court explained, ‘was limited in time and severity’ — it ended before the officer requested consent to enter the apartment — and ‘did not demonstrate any effort to control or direct [the] defendant.’ Lorenzo, 356 Or at 143-44.
“The court contrasted those minimal intrusions with more severe police misconduct in Musser. There, an officer saw the defendant walking at night in a high-crime area behind a shopping center, and the officer, without reasonable suspicion, called out to her, stopped her, and later requested to search her purse during the stop. The court observed that, in Musser, ‘the police order to [the] defendant to return and talk to the police, rather than to continue in the direction she was heading, clearly indicated to [the] defendant that she had no choice but to respond to the order, bringing her significantly under the control of the police.’ Musser, 356 Or at 157. That unlawful stop ‘was a more severe violation of [the] defendant’s rights than the violation in Unger, which was a daytime trespass onto the defendant’s property that allowed the police to contact the defendant at his back door, or the similar conduct in Lorenzo, where the officer reached into the defendant’s apartment to knock on the defendant’s bedroom door in an effort to contact him because of concern for his safety.’ Id. at 156.”
269 Or App at 213-14 (some internal quotation marks omitted).
Based on our analysis of Unger, Lorenzo, and Musser in Kuschnick, we believe that the unlawful police conduct in this case is more accurately characterized as intrusive or severe, rather than limited. A court, when analyzing the nature of unlawful police conduct preceding a defendant’s consent to search, may consider whether that conduct was merely incidental to the later consent, as in Unger, where the unlawful trespass to the defendant’s back door “allowed the police to interact with the defendant and request consent in the same way they would have if they acted lawfully by using the front door[.]” Id. at 215. Furthermore, a court may consider whether the unlawful police conduct may have occurred because the officer “was concerned for the defendant’s safety”; whether that conduct “did not demonstrate *205any effort to control or direct [the] defendant”; and whether that conduct had ended before the request for consent. Id. at 214 (analyzing those considerations when discussing the nature, extent, and severity of the constitutional violation in Lorenzo (internal quotation marks omitted)).
None of those considerations are present in this case. First, the unlawful police conduct was not incidental to the subsequent discovery and execution of the warrant, but rather the cause of it. Further, there is no evidence that Lemons or another officer was concerned for defendant’s safety; the unlawful police conduct demonstrated an effort to exercise control over defendant’s physical person by directing defendant’s movements; and that conduct had not ended by the time that the evidence was discovered. Indeed, the police control over defendant was increasing as the encounter developed; as the trial court found, other officers were “showing up” as Lemons was running the records check— that is, as many as two additional patrol vehicles and four additional police officers were arriving on the scene. Rather, the unlawful police conduct in this case is more akin to the more severe conduct in Musser, where the officers’ actions reasonably conveyed to the defendant “that she had no choice but to respond to the order, bringing her significantly under the control of the police.” Musser, 356 Or at 157. Thus, the fifth consideration — the nature, extent, and severity of the constitutional violation — favors suppression.
To conclude our analysis, we briefly review the totality of the circumstances presented by this case. As noted, Lemons approached defendant and Jacobs after they had started walking away from Lemons in an apparent attempt to “avoid police contact.” After defendant and Jacobs returned to answer Lemons’s question about the contents of the. bag, defendant told Lemons that “he ha[d] to go to the bathroom.” Lemons did not respond to defendant’s request to go to the bathroom; rather, he asked defendant and Jacobs for identification, which they provided. Lemons then told defendant and Jacobs to “hang on there, or hang on a second,” conveying, through his words and actions, that defendant and Jacobs should remain until Lemons could return after checking the identification. While Lemons was running the *206records check that revealed the outstanding warrant, as many as two additional patrol vehicles and four additional police officers were arriving on the scene. Defendant was then arrested on the warrant, and the search incident to arrest yielded evidence of methamphetamine.
When analyzing those circumstances, we note that Lemons unlawfully seized defendant without reasonable suspicion and that the discovery of the challenged evidence occurred shortly after the unlawful seizure. Further, despite defendant’s attempt to avoid and then to terminate the encounter, Lemons did not inform defendant that he was free to leave. Further still, the unlawful police conduct placed defendant at a distinct disadvantage — because defendant was required to remain in a place that he had twice attempted to leave and apparently would have left if not for the police show of authority, the police were able to achieve the investigatory purpose of discovering and then executing the warrant, which directly led to the discovery of the challenged evidence. Finally, the circumstances do not suggest that Lemons was concerned for defendant’s safety; the police exercised control over defendant’s physical movements; and police control over defendant increased as the encounter developed.
Based on the foregoing, we conclude that the state failed to prove attenuation — that is, the state failed to prove, under the totality of the circumstances, that the violation of defendant’s rights under Article I, section 9, had such a tenuous factual link to the disputed evidence that the unlawful police conduct cannot be properly viewed as the source of that evidence. Rather, the totality of the circumstances demonstrates that the police “‘exploited’ or ‘took advantage of or ‘traded on’ their unlawful conduct” to obtain the challenged evidence. Unger, 356 Or at 80. Thus, the state failed to rebut the presumption that the evidence of methamphetamine must be suppressed, and the trial court did not err in granting defendant’s motion.
Affirmed.
In Unger, the Oregon Supreme Court explained that, when unlawful police conduct “preceded a consent to search,” a court’s “inquiry has two prongs.” 356 Or at 85. “First, the court must assess whether the consent was voluntary.” Id. “Second, even if the consent is voluntary, the court must address whether the police exploited their prior illegal conduct to obtain the evidence.” Id. at 86. Because, in this case, the unlawful police conduct did not precede a consent to search, we need not consider the first prong.
During oral argument, the state noted that Lemons had already obtained defendant’s identifying information by the time that he told defendant and Jacobs to “hang on there, or hang on a second,” and suggested that Lemons could have discovered the warrant even if defendant had walked away at that point. However, the state did not argue to the trial court that, because Lemons obtained defendant’s identification part way through the encounter that became a seizure, the court should consider only the circumstances that existed when Lemons obtained defendant’s identification to evaluate whether the discovery of the relevant evidence — the methamphetamine in defendant’s pocket — was attenuated from the unlawful seizure. Accordingly, any such argument is unpreserved, and we do not address that possibility.