The state appeals from an order suppressing the results of field sobriety tests (FSTs) in a prosecution for driving under the influence of intoxicants (DUII), ORS 813.010, and from an ensuing order of dismissal. ORS 138.060. We conclude that there was objective probable cause to support defendant’s arrest for DUII that preceded the administration of the FSTs. State v. Nagel, 320 Or 24, 880 P2d 451 (1994). We further conclude that all of the FSTs administered here were “nontestimonial” for purposes of State v. Fish, 321 Or 48, 893 P2d 1023 (1995). Accordingly, we reverse and remand.
On review of a motion to suppress, we are bound by the trial court’s findings of historical fact if there is sufficient evidence to support them. State v. Ehly, 317 Or 66, 75, 854 P2d 421 (1993). At about 11:00 on the night of May 21, 1999, Portland Police Officer Taylor was on duty in northeast Portland when he saw defendant’s car turn north onto N.E. 122nd Avenue. Taylor began to follow defendant’s car, and saw the car weave several times within its lane and fluctuate between 35 and 45 miles an hour in a 35 mile per hour zone.1 At one point, defendant’s car made a lane change, which Taylor perceived to be unsafe, in that it left insufficient space for a following vehicle, causing the driver of that car to brake. Shortly thereafter, defendant stopped at a red light and, after the light turned green, waited three seconds to proceed. Taylor regarded that pause as evincing impairment.
Taylor stopped defendant’s car for the unsafe lane change. When Taylor approached the driver’s side of the car, the window was down, and Taylor smelled a strong odor of alcohol, which he believed was coming from defendant. The strong odor persisted even when defendant was not talking. Taylor asked defendant for his driver’s license and proof of *132insurance, and defendant produced them without any noticeable difficulty or fumbling. Taylor asked defendant whether he had been drinking, and defendant responded that he had been drinking earlier in the evening and had recently been at The Refectoiy (an establishment in northeast Portland), but did not have anything to drink there. Defendant’s speech was not slurred, but Taylor characterized it as “halting,” with unnatural pauses. Defendant’s eyes were watery but not bloodshot.
Taylor asked defendant if he would perform some FSTs, and defendant refused. Taylor then asked defendant to step out of the car and arrested defendant for DUII. Taylor then took defendant to Portland’s east precinct for DUII processing.
At the precinct, Portland Police Officer Hedges read defendant his Miranda rights and asked him if he would perform some FSTs. Again, defendant refused. Hedges then asked defendant if he would perform tests, which Hedges characterized as “nonverbal/non-testimonial tests,” including the horizontal gaze nystagmus (HGN) test, the nine-step walk-and-turn test, and the one-leg stand, modified so that defendant would count to himself, rather than out loud. See OAR 257-025-0020 (describing FSTs). Hedges explained to defendant that, if he refused to do the tests, his refusal would be used against him in court. ORS 813.136. Defendant then performed, and failed, each of those tests.
Defendant was subsequently charged with one count of DUII. Thereafter, defendant moved to suppress the results of the FSTs, arguing: (1) The arrest for DUII was unlawful in that it was not supported by objective probable cause; (2) similarly, given the lack of probable cause, Hedges’s “request” that defendant perform FSTs was unlawful under Nagel; and (3) various aspects of the walk-and-turn and one-leg stand tests — and, particularly, those evincing defendant’s ability to process and follow instructions — were “testimonial” under Fish.
The trial court suppressed the FST results, holding that, while Taylor had subjective probable cause, he lacked objective probable cause to arrest defendant for DUII. See *133State v. Owens, 302 Or 196, 204, 729 P2d 524 (1986) (probable cause has both subjective and objective components). The trial court further determined that, given the absence of objective probable cause, neither Taylor nor Hedges could lawfully “even request the driver” to perform FSTs. That is, the court determined that the lack of objective probable cause precluded the officers from seeking and obtaining even voluntary and consensual performance of the FSTs. Finally, the court determined that aspects of the nonverbal walk-and-turn and one-leg stand FSTs administered by Hedges were “testimonial” under Fish:
“As to the [walk-and-turn] test, the officer can testify about the performance of the [walk-and-turn] test, and if he observed the defendant fall off the line or if he observed the defendant have difficulty by swaying, or et cetera, he can certainly testify as to that.
“Now, the question here is whether he can say anything about whether the defendant performed nine steps or ten steps or seven steps.
“My reading of these cases is that he cannot say anything about the number of steps because that really involves the counting part, even though it was not done out loud, and it was done internally. It makes no difference, at least in the mind, of the appellate courts, since I read the cases.
“So that involves a communication, the number of steps, the officer can’t testify about the number of steps that were made or done or not done, but the rest he can describe what he observed and that is my reading on that.
‡ ifs ifs
“Focusing simply on what occurred at the station, I think my ruling stands that, to the extent that there is a verbal component, whether it is out loud or internally, since the courts define communication as being communication by words or conduct, that that part does not come in.
“But the officer can testify about his observations with respect to the tests and so the horizontal gaze nystagmus test, the officer could testify about that and about the way the [walk-and-turn] was performed.
*134“I have not yet gotten to the point of the probable cause. I am dealing with that issue separately.
“Now, holding one’s leg up for a certain length of time, Nielsen[, 147 Or App 294, 936 P2d 374, rev den 326 Or 68 (1997)] actually specifically dealt with that particular test and said that the counting itself is testimonial, arguably, but the fact that the person cannot hold their leg up is not. It is an observation.
“So the length of time and the counting is not anything that is admissible. However, the fact that the person cannot hold their leg up without falling over, that is an observation. That is a physical observation, and so it is a physical task that we’re talking about.”
The trial court consequently suppressed the FST results. The state appeals from that order of suppression and from the ensuing order of dismissal. See State v. Robinson, 158 Or App 494, 500, 974 P2d 713 (1999) (appellate court lacks jurisdiction to consider appeal from order granting motion to suppress where trial court entered a valid order of dismissal and state failed to appeal from that order).
On appeal, the state challenges each component of the trial court’s reasoning: First, the state asserts, there was objective probable cause for the DUII arrest. That, in turn, coupled with exigent circumstances (eg., dissipation of defendant’s blood-alcohol content) would permit the officers, without a warrant, to lawfully direct defendant to perform the FSTs. See Nagel, 320 Or at 33 (risk that evidence of the defendant’s intoxication may have dissipated provided the exigent circumstances justifying the warrantless administration of FSTs). Second, even if the officers lacked objective probable cause, they were entitled to request defendant’s consent to voluntarily perform FSTs, and the trial court’s assumption to the contrary — that the officer must have “probable cause to even request the driver to do field sobriety tests” — was incorrect. See State v. Barber, 151 Or App 84, 86, 949 P2d 1236 (1997) (“ ‘[C]onsent is a valid exception to the warrant requirement and * * * probable cause is not a necessary prerequisite to asking a defendant for consent to perform field sobriety tests in the context of a DUII stop.’ ” (emphasis added) (quoting State v. Ramos, 149 Or App 269, 272, 942 P2d 841 (1997)). See also State v. Maddux, 144 Or *135App 34, 41, 925 P2d 124 (1996) (upholding constitutionality of the defendant’s voluntary consent to perform FSTs). Third, the state argues that the trial court erred in its determination that nonverbal aspects of the walk-and-tum and one-leg stand FSTs were “testimonial.” In particular, the state argues that that determination, although consonant with the dissenting view in State v. Nielsen, 147 Or App 294, 936 P2d 374, rev den 326 Or 68 (1997), is irreconcilable with the en banc majority in that case.
As explained below, we conclude that Taylor’s arrest of defendant for DUII was supported by objective probable cause and that, given the exigent circumstances, Hedges could lawfully direct defendant to perform the FSTs. We further conclude that the results of the FSTs here, which involved no verbal component and were entirely nontestimonial, were admissible as evidence. Consequently, the trial court erred in suppressing the FST results.
We begin with the issue of probable cause. In Nagel, the court held that the administration of FSTs constitutes a search — and usually, as here, a warrantless search. 320 Or at 31, 36-37. As the proponent of the warrantless search, the state must establish both that the search was supported by probable cause that the suspect/defendant was driving under the influence of intoxicants, and that a recognized exception to the warrant requirement — generally, exigent circumstances — applied. Id. at 31-32.
Here, there is no dispute as to exigent circumstances, viz., dissipation of defendant’s blood-alcohol content. See, e.g., Nagel, 320 Or at 33 (noting exigency in context of DUII). Moreover, the trial court explicitly determined that Taylor had subjective probable cause to believe that defendant had driven under the influence. Thus, the issue under Nagel reduces to whether, at the time Taylor arrested defendant— which, in turn, led to Hedges’s administration of the FSTs— there was objective probable cause to believe that defendant had driven under the influence of intoxicants. Unlike the trial court, we conclude that there was objective probable cause. State v. Gilmour, 136 Or App 294, 901 P2d 894, rev den 322 Or 360 (1995).
*136Gilmour is similar. There, an officer stopped the defendant after seeing him commit a traffic infraction while leaving a tavern parking lot at 2:00 a.m. The officer, in his discussions with the defendant, observed that the defendant’s speech was not slurred and that he had no difficulty producing his driver’s license, but that his eyes were watery and bloodshot and there was a strong odor of alcohol emanating from his vehicle. We held that those facts were sufficient to establish objective probable cause. Id. at 300.
Hére, the facts provide a stronger basis for probable cause than do those in Gilmour. Before making the stop, and (Taylor observed defendant’s car turn onto a road near The Refectory, weave within its lane, accelerate and decelerate between 35 and 45 miles per hour, and make an unsafe lane change. After stopping defendant’s car for the unsafe lane change, Taylor further observed that there was a strong odor of alcohol coming from defendant’s car, that defendant’s eyes were watery, and that his speech, while not slurred, was halting. Finally, when Taylor asked defendant if he had been drinking, defendant denied having any drinks at The Refectory but acknowledged that he had been drinking earlier. Given those facts, we conclude that Taylor’s arrest of defendant and the ensuing FSTs were supported by objective probable cause. See also State v. Wetzell, 148 Or App 122, 125-26, 939 P2d 106, rev den 325 Or 621 (1997) (the defendant’s erratic driving “and the odor of alcohol that emanated from his vehicle when the officer first approached constituted objective probable cause”).2 Given the exigent circumstances, Hedges’s request that defendant perform the FSTs in this case did not impinge upon defendant’s constitutional right to *137be free from unreasonable searches and seizures. See Nagel, 320 Or at 33.
Defendant nevertheless contends — and the trial court held — that the results of the FSTs here were inadmissible because the “testimonial” aspects of those tests violated defendant’s rights to be free from “testifying] against himself’ under Article I, section 12, of the Oregon Constitution. See Fish, 321 Or at 53-59 (discussing the application of Article I, section 12, in the FST context). We disagree.
In Fish, the Supreme Court considered whether the admission of evidence of the defendant’s refusal to perform FSTs violated his right against self-incrimination under Article I, section 12. The court noted that, given the circumstances of that case — and particularly that the defendant had been informed, as required by ORS 813.136, that refusal to perform FSTs would be admissible as evidence in any civil or criminal proceeding arising out of the DUII allegations— the defendant was faced with a Hobson’s choice of either refusing to perform the tests or of performing them, and that either choice contained “testimonial” components that could be used as evidence against him. Fish, 321 Or at 58-59. Thus, the count concluded, “[b]ecause a refusal to perform field sobriety tests and the performance of such tests are both ‘testimonial,’ defendant was compelled to testify against himself.” Id. at 60. In so holding, the court noted that tests, or aspects of tests, that involved “counting; answering questions relating to the individual’s residence and date of birth; estimating a period of time; and reciting the alphabet” were testimonial under Article I, section 12. Id. (citations omitted).
In Nielsen, an en banc decision, we considered the content and contours of “testimony” for purposes of the Fish analysis:
“[F]rom Fish, we understand that ‘testimony’ is the communication by words or conduct of an individual’s thoughts, beliefs or ‘state of mind.’ Thus, purely verbal answers to purely verbal questions (‘On an intoxication scale of 1 to 10, I think I rank a 2.’) are testimony, as are answers by conduct to the same question (holding up two fingers to self-rank intoxication on a 1 to 10 scale). From the physical evidence cases, we also understand that, as a general rule, *138tests that produce physical evidence of an individual’s intoxication are not testimonial. Thus, a test that reveals an individual’s intoxicated state, without requiring the individual to reveal his or her thoughts, beliefs or ‘state of mind,’ is not testimonial.” Id. at 306.
We then addressed whether the specific tests administered by the arresting officer in Nielsen — the walk-and-turn, one-leg stand, and “modified” Romberg tests — had testimonial components, thus bringing them within Fish’s proscription. We concluded that: (1) the walk-and-turn test was not testimonial; (2) the one-leg-stand test was not testimonial if the counting components of the test were not considered; and (3) the “modified” Romberg test, by virtue of the fact that it required the defendant to recite the alphabet, contained some testimonial components. Id. at 308-09.
Two judges dissented, arguing that the majority’s interpretation of Fish was overly narrow and that any test that compelled a defendant to act in a manner that might permit inferences about a defendant’s “memory, perception, cognitive processes and * * * ability to communicate” was subject to Article I, section 12’s protection against self-incrimination. Id. at 312 (Edmonds, J., dissenting). The dissent went on to note that while some FSTs, like the HGN test, are not “testimonial” because they “measure solely physical reactions,” id. at 312 n 2, the walk-and-turn and one-leg stand FSTs were “testimonial” because they revealed something about the defendant’s state of mind — specifically, the defendant’s ability to process and follow directions:
“[T]he officer gave defendant multiple instructions regarding the walk-and-turn test, in part to test his ability to follow directions. Whether he followed those directions revealed something about his state of mind, i.e., whether he was able to perceive the directions, remember them and translate them into the requested actions. The instructions for the one-leg test and the ‘Romberg Balance Test’ make the same inquiries. Thus, more than the mere performance of a physical exercise is at stake. The tests probed defendant’s mental faculties as well. In that sense, the walk-and-turn test, the one-leg stand test and the ‘Romberg Balance Test’ are no different from the tests like counting backwards that require cognitive responses, and that the court *139found subject to section 12 in Fish.” Id. at 312 (footnote omitted).
Here, the trial court essentially adopted the reasoning of the Nielsen dissent — i.e., that nonverbal performance of task-specific FSTs was innately “testimonial” as evincing the subject’s unarticulated mental processes. See 174 Or App at 133-34. That position is certainly plausible, and was ably and forcefully propounded by the Nielsen dissent. But we have rejected it, and we decline to revisit our analysis and holding in Nielsen.
Here, as noted, Hedges administered three FSTs. The HGN test was, as all parties properly agree, nontestimonial. The second FST, the walk-and-turn, was indistinguishable from the walk-and-turn test in Nielsen that we held was nontestimonial. Nielsen, 147 Or at 308. Finally, the one-leg-stand FST administered here differed fundamentally from the standard one-leg-stand administered in Nielsen, because defendant here did not count from 1 to 30 out loud. See OAR 257-025-0020(l)(c) (describing standard one-leg-stand test). Rather, Hedges modified the test so defendant could count to himself. That distinction, eliminating any verbal expression by defendant while he attempted to perform the physical tasks of that FST, rendered the results of that test nontestimonial.3
In sum, the administration of the FSTs here comported with both Nagel and Fish. Accordingly, the trial court erred in suppressing the FST results.
Reversed and remanded.