concurring in part and dissenting in part.
(39 I agree with the majority's recitation of Colorado law governing consent searches. The trial court misapplied that law in finding Munoz-Gutierrez's oral consent to search invalid based on the troopers' failure to give the statutorily required advisement set forth in section 16-8-310(1), C.R.S. (2014). Seetion 16-3-810(8) plainly states that an officer's failure to give the statutory advisement is simply "a factor" in determining the volun-tariness of a consent search; it is not disposi-tive. Should these officers have given the advisement? Yes. (And the majority certainly does not suggest otherwise.) Does their omission of it alone, however, vitiate Munoz-Gutierrez's oral consent? No. This is where the trial court took the wrong turn. But having identified the proper construction of the statute, I believe we should return this case to the trial court for further findings and to weigh the evidence-evidence that it saw and heard, and that we did not-because the record and the trial court's findings are ambiguous in important respects.
1140 As the majority correctly suggests, remand is required if there is any evidence that the police conduct overbore Munoz-Gutierrez's will, Maj. op. 114. Indeed, People v. Magallenes-Aragon, 948 P.2d 528 (Colo.1997), which the majority finds instructive, sheds light on this point as well. There, we stated we "may decide, on appellate review, whether a consent was voluntarily given if the record clearly contains no evidence to support the trial court's determination." Id. at 533 (emphasis added); see also People v. Castro, 159 P.3d 597, 600 (Colo.2007) (remand unnecessary because there was "no evidence" of involuntary consent). If, however, the record does contain evidence to support the trial court's determination, "we must remand for additional findings and a re-evaluation of voluntariness in light of those findings." Magallenes-Aragon, 948 P.2d at 533. After all, "[iJt is the function of the trial court and not the reviewing court to weigh the evidence and determine the credibility of witnesses," and thus "we will not substitute our judgment for that of the trial court...." People v. Mendoza-Balderama, 981 P.2d 150, 157-58 (Colo.1999).1
*450T41 As the majority acknowledges, the voluntariness of consent to search is highly fact-driven. It requires the court to correlate a defendant's personal characteristics and police conduct. See maj. op. 129 ("In determining that the oral consent was voluntary, we assess the troopers' conduct in relation to Munoz-Gutierrez's characteristics and cireumstances by evaluating Munoz-Gutierrez's age, education, knowledge, and perceptions in relation to the location, duration, and environment of the police interaction."). Thus, a defendant's extremely poor comprehension can heighten the significance of even relatively innocuous police conduct. See 4 Wayne R. LaFave, Search and Seigure: A Treatise on the Fourth Amendment § 8.2(e) (5th ed. 2014) ("[TJhe propriety of the investigative and interrogation techniques used must be judged in light of what the police knew or should have known about defendant's ability to comprehend the events and cireumstances surrounding him or her."). In Magallenes-Aragon, for example, we explained that a trial court's finding that police conduct was "proper," lawful, or based upon legitimate concerns is not dispositive of whether consent was voluntary. 948 P.2d at 538. Voluntariness depends instead on whether that conduct "reasonably appear{s] to a person with the defendant's characteristics to be overbearing, coercive, or deceptive." Id.
142 So, is there any record evidence of police overreaching here sufficient to justify remand? The majority thinks not. I disagree. Munoz-Gutierres had only four months of formal education in Mexico when he was seven years old. Although unemployed, he had last worked milking cows at a dairy. The trial court found that the troopers communicating with Munoz-Gutierrez knew "rudimentary Spanish"; conversely, it was unclear how much English Munoz-Gu-tierres understood. Munoz-Guttieres was tired, having driven for twenty hours before the troopers pulled him over. While two troopers interacted with him, three others were nearby with a drug-sniffing dog. At times, he seemed confused.2 Munoz-Gutierrez testified that the troopers did not ask him orally if they could search his car, and that he did not give them oral consent to do so. He also testified that one of the troopers yelled "sign it" when he put the written consent form before him and that he signed the form because the trooper was "pressuring" him to do so. The trial court made no credibility findings about this part of the encounter, even though it chose to underscore this aspect of Munoz-Gutierrez's testimony.
43 Yes, the record provides evidence to support the majority's determination that Munoz-Gutierrez's consent to search was voluntary: the traffic stop was relatively uneventful; Munoz-Gutierres clearly understood enough English to follow commands and answer basic questions; he testified that he understood Trooper Biesemeier's Spanish; he may even have understood that he could refuse the search, but still agreed to the search, at least with "body language." But, as explained above, the record contains contrary evidence as well-evidence that supports the trial court's determination that Munoz-Gutierrez's consent to search was involuntary because there was some overreaching by the police, based on the totality of the cireumstances. Given this conflicting evidence-and the limited findings by the trial court regarding Munoz-Gutierrez's oral consent due to its misapplication of section 16-3-310-we should invite the trial court's re-evaluation using the correct legal standard. We should let the judicial officer who had the front-row seat for the suppression hearing give us the benefit of his direct observations of the critical witnesses. *451Therefore, I concur in part and join in all aspects of the majority's opinion aside from Part IIL.C, from which I respectfully dissent.
I am authorized to state that JUSTICE HOBBS and JUSTICE MARQUEZ join in the concurrence in part and dissent in part.
. Significantly, this is not a case where we can review a video or audio tape and glean nearly as much critical information as was available to the trial judge. See, eg., People v. Madrid, 179 P.3d 1010, 1014 (Colo.2008) ("[Where the statements sought to be suppressed are audio- and video-recorded, and there are no disputed facts outside the recording controlling the issue of suppression, we are in a similar position as the trial court to determine whether the statements should be suppressed."); People v. Al-Yousif, 49 P.3d 1165, 1171 (Colo.2002) (noting that the existence of a videotaped interview allowed this court to review the adequacy of the Miranda warning in light of a language barrier "not just from the 'cold record," but-at least in part-in precisely the same manner as the trial court"). *450In substituting its judgment for the trial court's, the majority is relying on a cold transcript. Such reliance is not always misplaced. Sometimes the written record is unequivocal. This is not one of those times.
. For instance, Munoz-Gutierrez ultimately signed the wrong line on the consent form (the date line and not the signature line), suggesting that he did not understand the significance of his signature on the form and may even have had difficulties reading the form. Although the majority correctly points out that it does not need to assess the validity of the written consent because of its finding of oral consent, Munoz-Gutierrez's mistake is tangible evidence of his confusion, which sheds light on the totality of the circumstances.