OPINION
MARK A. GOLDSMITH, District Judge.
Appellants Dawn Bishop and Brandi Henry brought this lawsuit under Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., alleging that they were terminated from their positions as corrections officers for Appellee Ohio Department of Rehabilitation and Corrections (ODRC) in retaliation for complaining of gender discrimination perpetrated by their immediate superior, Lt. Yvonne Richardson. Bishop and Henry do not argue that the individual who made the decision to terminate them, Warden Deborah Tim-[687]*687merman-Cooper, held any retaliatory animus toward them. Rather, proceeding on a so-called “cat’s-paw” or “rubber-stamp” theory, they argue that Lt. Richardson, motivated by retaliatory animus, influenced the Warden through the submission of negative performance evaluations on which the Warden relied in reaching her decision. The district court granted summary judgment in favor of ODRC, concluding that Bishop and Henry failed to carry their burden to show that there was a genuine factual dispute whether ODRC’s asserted reason for the terminations was a mere pretext for retaliation. For the reasons that follow, we reverse the district court’s judgment and remand the case for further proceedings.
I. BACKGROUND
Dawn Bishop and Brandi Henry began working for ODRC as corrections officers in January 2005. They were assigned to London Correctional Institution (LCI) and worked the third shift, which ran from 10:00 p.m. to 6:00 a.m.
Deborah Timmerman-Cooper was the warden of LCI at the time and was the ultimate decisionmaker with regard to disciplinary action, as well as hiring and firing. The Warden was assisted by three deputy wardens and a major. The major supervised the captains, who acted as shift commanders and were responsible for the operation of the facility during that shift. Captains were assisted by lieutenants, who directly supervised the corrections officers. Corrections officers were responsible for securing and maintaining control over the inmates of the institution.
The third shift was supervised by two captains and three lieutenants. The lieutenants were responsible for creating schedules, initiating disciplinary action, and evaluating the performance of corrections officers through performance evaluations. The captains reviewed and signed off on schedules and performance evaluations created by the lieutenants.
Newly-hired corrections officers were subject to a one-year probationary period. Bishop and Henry were at all times probationary officers. As probationary officers, Bishop and Henry were at-will employees and subject to termination without just cause. Additionally, their performance was evaluated every 60 days. Each performance review consisted of a rating system in which officers were rated “below,” “meets,” or “above” in various categories. The officers also received an overall rating of “satisfactory” or “unsatisfactory” and specific comments about their performance. In evaluating an officer, the reviewing lieutenant could seek input about the officer’s performance from other lieutenants.
Lt. Yvonne Richardson directly supervised Bishop and Henry during their probationary employment, which began in January 2005 and ended in December 2005. Throughout their tenure at LCI, Bishop and Henry — along with other female officers — claimed to have experienced hostile and discriminatory treatment by Lt. Richardson. For example, Henry testified that Richardson treated her rudely, gave her dirty looks, and berated her in front of inmates. Bishop testified that Lt. Richardson asked her inappropriate personal questions about her ex-boyfriend. Both Henry and Bishop, along with other female corrections officers, testified that Lt. Richardson treated male officers more favorably than female officers.
Bishop and Henry’s main issue with Lt. Richardson was her scheduling system, which they and other female corrections officers felt gave preferential post assignments to male officers. Specifically, Bishop, Henry, and others claimed that fe[688]*688male officers were disproportionately scheduled to work at dormitory posts, which were considered less desirable than other duty stations, and were assigned to outside posts more frequently in inclement weather than were male officers. Bishop, Henry, and five other female corrections officers made their views about Lt. Richardson’s purported discriminatory practices known through a complaint, discussed in detail below, that was jointly submitted directly to the Warden in October 2005.
Prior to submitting their complaint, Bishop and Henry both earned positive performance evaluations in their four-month, six-month, and eight-month reviews, receiving “meets” in all categories and overall ratings of “satisfactory.” Their eight-month evaluations, which cover the time period from January 25, 2005 until September 24, 2005, would be their last unqualifiedly positive reviews.
On August 16, 2005, Bishop was reprimanded for being two minutes late to a training session on July 15, 2005. Although the reprimand occurred within the time period covered by her eight-month performance evaluation, that evaluation mentioned nothing about the reprimand; in fact, in the comments section of Bishop’s eight-month evaluation, Lt. Richardson commended Bishop on her attendance, writing that Bishop had “never missed a day of work.”1
On September 11, 2005, Bishop failed to properly record observations during a suicide-watch post. Lt. O. Barney recommended that no action be taken beyond his counseling of Bishop, but the Warden disagreed and issued a written reprimand to Bishop on September 29, 2005. The Warden felt that a written reprimand was warranted because “suicide watches are extremely important to make sure the documentation is done appropriately ... because that’s the life of an inmate.” Warden Dep. at 135. Bishop received this reprimand during the time period covered by her ten-month evaluation. That evaluation is discussed in more detail below.
In October 2005, Bishop and Henry — as well as five other female corrections officers — submitted a detailed complaint to the Warden, dated October 3, 2005, regarding Lt. Richardson’s alleged discriminatory practices against female “relief’ corrections officers assigned to the third shift.2 In their complaint, which is two and a half single-spaced pages in length, the complaining parties explain their belief that Lt. Richardson favored male officers in creating workplace schedules, using statistical evidence showing that female relief officers on the third shift were less frequently assigned desirable posts and more frequently assigned undesirable posts. In addition, the complainants describe what they believe was a gender-based “hostile work environment” created by Lt. Richardson, writing that they were “forced to walk on egg shells” for fear that they could be “humiliate[d]” and “intimidate[d]” by Lt. Richardson at any time, and that Lt. Richardson “abuse[d] and misuse[d] her authority ... to cause discomfort and fear for our jobs.” The complainants also noted that, on several occasions, Lt. Richardson improperly “question[ed] them about their personal lives.”
The time stamp on the complaint reflects that it was received by the Warden on October 17, 2005, at which time the [689]*689Warden discussed it with Major Bill Kelly. On November 7, 2005, the Warden assigned Kelly Mason, a unit manager, to investigate the matter.3 In addition to the complainants who signed the letter, the Warden instructed Mason to interview two additional corrections officers — Peggy Miller and Tim Caudy — because both had contacted the Warden claiming “they had information that was contrary to what the letter said.”4 The complaint and the ensuing investigation were “the talk at the institution at the time.” Board Dep. at 27.5
Bishop and Henry — along with Cheri Gause and Denise Marsh, two other third-shift female corrections officers who also signed the complaint to the Warden — testified that Lt. Richardson’s hostility toward them intensified after they submitted their complaint to the Warden.6 In particular, Henry testified that, after the complaint was submitted, Lt. Richardson reminded her “a couple of times” that she “was on probation,” and that “[t]he atmosphere got worse” in that the hostile looks from Lt. Richardson were “[m] ore frequent” and “even more hateful.”7 Henry Dep. at 146, 160-161. Gause testified that the “glares” and “dirty looks” from Lt. Richardson “escalate[d]” after the complaint was filed. Gause Dep. at 166, 179. Marsh similarly testified that “everything went to shit” after the complaint was filed. Marsh Dep. at 167. According to Marsh, Lt. Richardson’s post assignments became even more discriminatory, prompting the male officers to openly mock the female officers. Id. at 168.
Meanwhile, on October 22, 2005, Bishop submitted, on behalf of herself only, two incident reports to the Warden complaining of discriminatory post assignments by Lt. Richardson. Bishop submitted another incident report on October 81, 2005, alleging that Lt. Richardson told her on October 28, 2005 — six days after the Warden received the group complaint against Lt. Richardson: “There is a lot of stuff going on around here, don’t get involved in it, remember you are on probation.” On October 24, 2005, the Warden forwarded to Lt. Richardson and others the two incident reports that had been submitted to the Warden on October 22.
On November 1, 2005, Bishop’s probationary period was extended until April 24, 2006, three months beyond her original probationary end date. The memo notifying Bishop of the extension was signed by the Warden on November 2, 2005, and states that Bishop’s probationary status was extended “due to [her] performance.” However, the Warden testified that she [690]*690had no memory of extending Bishop’s probationary period. Warden Dep. at 124.
On November 14, 2005, a male corrections officer reported to Lt. Robert Joy that Henry was not performing an inmate-count correctly. Lt. Joy subsequently observed Henry conducting a count and, according to Henry, Lt. Joy interrupted the count by following Henry and asking her every couple steps what her count was, causing her to miscount. Lt. Joy testified that he was simply showing her how to conduct the count properly. On the advice of a union steward, Henry logged the incident arid completed an incident report on November 15, 2005, the following day.
On November 19, 2005, Lt. Joy learned that Henry had reported the incident and, in the presence of the corrections officer who had initially reported Henry’s purported counting issue, confronted Henry. Henry testified that Lt. Joy yelled at her and threatened to discipline her. On November 20 and 21, respectively, Henry and Lt. Joy each submitted separate incident reports regarding the confrontation. The Warden assigned Major Kelly to investigate. In her December 1, 2005 report to the Warden, Major Kelly expressed concern over Henry’s inability to properly conduct an inmate count, and recommended that “further action” be taken against Henry.
On November 20, 2005, Lt. Richardson signed Bishop and Henry’s ten-month performance reviews, which covered the time period from January 25, 2005 until November 24, 2005. Although Bishop and Henry were both given overall ratings of “satisfactory,” for the first time they received “below” ratings in two of six categories— “cooperation” and “communication” — and negative comments. As to Bishop, Lt. Richardson commented:
Officer Bishop has not effectively communicated with her supervisors this reporting period. Officer Bishop has lacked communicating [sic] with her supervisors on several occasions. Incident reports were written by Officer Bishop concerning her job performance and job assignments before these incidents were discussed or resolved with her supervisors. I expect Officer Bishop to use good judgment on every situation and not blindly follow a suggestion.
Although Bishop’s September 29, 2005 reprimand relating to her failure to properly observe suicidal inmates occurred within the time period covered by this evaluation, the incident was not referenced by Lt. Richardson.
In her ten-month evaluation, Henry received similar criticism from Lt. Richardson:
Officer Henry has not effectively communicated with her supervisors this reporting period. I expect Officer Henry to listen to her supervisors and to use good judgment on every situation and not blindly follow a suggestion of others. Institutional procedures need to be adhere [sic] to and realize that constructive criticism is a tool that is used to make her a better correctional officer.
Although Henry’s inmate counting issues, along with her confrontation with Lt. Joy, occurred within the time frame covered by the ten-month evaluation, the incidents were not referenced by Lt. Richardson.
On December 12, 2005, less than a month after receiving their ten-month evaluations, the Warden terminated Bishop and Henry. Their termination letters, which were signed by the Warden, both contain identical language: “You are being ... removed because you are unable to perform your job duties based on your [691]*691performance evaluations.”8 Later, in her position statements to the Ohio Civil Rights Commission, the Warden amplified on her reasons for terminating Bishop and Henry, writing that they were terminated for below average performance of their duties. Specifically, the Warden noted that Bishop’s termination resulted from “discipline [Bishop] received while serving her probationary period,” and that Henry’s termination was “a result of her ten (10) month evaluation” and, specifically, due to the incident with Lt. Joy.
The Warden testified that, although she did not remember Bishop and Henry on the day of her deposition (September 28, 2009), id. at 100-101, 109, she would, “in general,” ask questions of other command officers before removing an employee based on a negative performance evaluation. Id. at 114, 116. In particular, the Warden testified that she “would have asked questions” of her command officers after reading Bishop and Henry’s performance evaluations to ensure that she was “removing them correctly.” Id. at 114. However, no command officer could remember whether they consulted with the Warden about Bishop and Henry’s performance prior to the Warden’s termination decision. Both Lt. Richardson and Major Kelly, for example, testified that they could not remember whether they were consulted by the Warden about Bishop and Henry’s job performance at the time the termination decision was made. Richardson Dep. at 52-53; Kelly Dep. at 33-35.
In addition, if the Warden did conduct an inquiry beyond reviewing Bishop and Henry’s negative performance evaluations, the individuals with whom the Warden communicated are unknown. The Warden testified that she generally consults with either the major or, in some cases, with the employee’s supervisor, who in this case was Lt. Richardson. Id. at 116-117. The Warden was unable to recall with whom, if anyone, she communicated in the case of Bishop and Henry: “I can’t tell you [with whom I talked in] specific cases. I mean, I have talked to supervisors before. I don’t know if I talked to them or just talked to the major here.” Id. at 117.
Bishop and Henry, along with two other female corrections officers who are not involved in this appeal, filed the present lawsuit in the United States District Court for the Southern District of Ohio. They allege a gender-based hostile work environment and retaliation, both in violation of Title VII.
The district court granted ODRC’s motion for summary judgment as to both claims; however, only the retaliation claim of Bishop and Henry is before us. With regard to that claim, Bishop and Henry do not argue that the Warden personally acted with a retaliatory animus in terminating them. Rather, their theory is that Lt. Richardson harbored a retaliatory animus against them for complaining of Lt. Richardson’s alleged gender discrimination, and the Warden then relied on tainted information provided by Lt. Richardson— namely, the ten-month performance evalu[692]*692ations — in reaching her decision to terminate Bishop and Henry.
The district court concluded that, while Bishop and Henry carried their burden of demonstrating a prima-facie case of retaliation, they failed to satisfy their burden to show that the reasons offered by ODRC for the terminations — namely, performance issues — were a mere pretext for retaliation. On that basis, the district court granted summary judgment in favor of ODRC. This appeal followed.
II. STANDARD OF REVIEW
We review de novo a district court’s grant of summary judgment. Planned Parenthood Southwest Ohio Region v. DeWine, 696 F.3d 490, 503 (6th Cir.2012). In doing so, we review the evidence and draw all inferences in the light most favorable to the non-moving party. Id. “Credibility determinations, the weighing of the evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of a judge.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). Summary judgment is appropriate only if “the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(a).
III. ANALYSIS
A. General Framework for Title VII Retaliation Claims
Title VII prohibits employers from retaliating against employees who oppose an unlawful employment practice. See 42 U.S.C. § 2000e-3 (a). Absent direct evidence of retaliation, retaliation claims are analyzed under the burden-shifting framework set forth in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973) and Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981). Fuhr v. Hazel Park Sch. Dist., 710 F.3d 668, 674 (6th Cir.2013). Under this framework, the employee bears the initial burden of establishing a prima-facie case of retaliation, which consists of four elements: (1) the employee engaged in activity protected under Title VII; (2) the employee’s exercise of his or her protected rights was known to the employer; (3) an adverse employment action was subsequently taken against the employee, and (4) there was a causal connection between the protected activity and the adverse employment action or harassment. Id.
If the employee establishes a prima-facie case, the burden shifts to the employer to articulate a legitimate, nondiscriminatory reason for its action. Id. at 674-75. If the employer does so, the burden then shifts back to the employee to demonstrate, by a preponderance of the evidence, that the proffered reason was a mere pretext, or coverup, for discrimination. Id. at 675.
The district court determined that Bishop and Henry carried their burden of demonstrating a prima-facie case of retaliation, but concluded that they failed to show that ODRC’s proffered legitimate, nondiscriminatory reason for the terminations was pretextual. ODRC agrees with the latter conclusion but challenges the former conclusion, urging us to affirm the district court’s judgment on the alternative ground that Bishop and Henry failed to satisfy their prima-facie burden. See Abercrombie & Fitch Stores, Inc. v. Am. Eagle Outfitters, Inc., 280 F.3d 619, 629 (6th Cir.2002) (“Because this court’s de novo review involves only application of legal propositions to the undisputed facts in the record, we may affirm on any grounds supported by the record even if different from the reasons of the district court.”). Bishop and Henry agree with the district [693]*693court’s conclusion that they satisfied their prima-facie burden, but challenge the district court’s conclusion that they failed to show a triable issue on pretext, urging us to reverse the district’s court grant of summary judgment in favor of ODRC.
B. Prima-Facie Case
ODRC argues that Bishop and Henry have not satisfied their burden as to the second and fourth elements of the prima-facie framework, which require a showing that Lt. Richardson knew that Bishop and Henry had engaged in protected activity and that there is a causal connection between their protected activity and Lt. Richardson’s decision to give them a negative performance evaluation, which lead to their terminations. We agree with the district court that Bishop and Henry have met their burden on those elements.9
1. Knowledge of Protected Activity
Regarding the second element of the prima-facie framework — whether Lt. Richardson knew that Bishop and Henry engaged in protected activity at the time she gave them negative performance reviews — ODRC argues that, while Lt. Richardson may have known generally that a group of female corrections officers complained to the Warden about Lt. Richardson’s scheduling practices, there is no evidence that Lt. Richardson knew specifically that Bishop and Henry were among the officers who submitted the complaint. The district court rejected this argument and so do we.
Although “[i]n most Title VII retaliation cases, [the employee] will be able to produce direct evidence that the decision making officials knew of the [employee’s] protected activity.... [D]irect evidence of such knowledge or awareness is not required, and ... [an employee] may survive summary judgment by producing circumstantial evidence to establish this element of [his or] her claim.” Mulhall v. Ashcroft, 287 F.3d 543, 552 (6th Cir.2002). Here, construing the evidence in the light most favorable to Bishop and Henry, the record before us contains sufficient circumstantial evidence on which a reasonable jury could rely to conclude that Lt. Richardson knew that Bishop and Henry were among those female corrections officers who had submitted the complaint.
First, it is undisputed that the existence of the complaint was widely known by LCI personnel — a fact that ODRC does not dispute.10 Second, the Warden testified that she was contacted by two corrections officers, Peggy Miller and Tim Caudy, who offered to supply information that was inconsistent with the allegations contained in the complaint. This testimony tends to support a reasonable inference that employees other than the complainants knew not only of the existence of the complaint but also of its substance — namely, that female third-shift relief corrections officers were complaining of gender-based discrimination perpetrated by Lt. Richardson through her post-schedules. Given that there were only five female relief corrections officers working the third shift at the [694]*694time, two of whom were Bishop and Henry, a reasonable jury could conclude that Lt. Richardson deduced that Bishop and Henry were among those officers who were behind the complaint.
ODRC argues that Lt. Richardson’s verbal warning to Bishop — that “there is a lot of stuff going on around here, don’t get involved in it, remember you are on probation” — suggests that Lt. Richardson did not know that Bishop was one of the complainants; the theory offered is that, if Lt. Richardson knew of Bishop’s involvement, it would have been unnecessary to warn Bishop not to get involved. We believe a reasonable jury could reach that conclusion, but we also believe a reasonable jury could interpret the statement as a general warning to Bishop not to continue pressing her complaints. Additionally, as the statement was allegedly made only six days after the Warden received the complaint and well before Mason launched her investigation, a reasonable jury might conclude that Lt. Richardson did not yet know of Bishop’s involvement at the time the statement was made. The operative question is whether Lt. Richardson knew of Bishop’s involvement on or before November 20, 2005, the day Lt. Richardson signed Bishop’s negative performance review. We determine that a reasonable jury could conclude that Lt. Richardson knew by November 20, 2005 that Bishop was one of the complainants who submitted the complaint based on the notoriety that the complaint received among LCI personnel, and the fact that Bishop’s October 22 and 31, 2005 incident reports containing allegations similar to those contained in the jointly-filed complaint were forwarded to Lt. Richardson on October 24, 2005.11
For all these reasons, we hold that there is a triable issue as to whether Lt. Richardson knew that Bishop and Henry were among the complainants to submit the complaint.
2. Causation
ODRC argues that Bishop and Henry proffered insufficient evidence demonstrating a causal connection between their protected activity — the October 3, 2005 complaint, which was received by the Warden on October 17, 2005 — and the negative performance evaluations that led to their terminations. The district court concluded that Bishop and Henry satisfied their prima-facie burden on the causation element. We agree.
“A causal connection is established when a plaintiff proffers ‘evidence sufficient to raise the inference that her protected activity was the likely reason for the adverse action.’” Fuhr, 710 F.3d at 675 (quoting Upshaw v. Ford Motor Co., 576 F.3d 576, 588 (6th Cir.2009)). “[W]here some time elapses between when the employer learns of a protected activity and the subsequent adverse employment action, the employee must couple temporal proximity with other evidence of retaliatory conduct to establish causality.” Mickey v. Zeidler Tool & Die Co., 516 F.3d 516, 525 (6th Cir.2008). Here, 34 days elapsed between the date on which the Warden received the complaint (October 17, 2005) and the date on which Lt. Richardson signed the negative evaluations of Bishop and Henry (November 20, 2005), which led to their terminations 22 days thereafter on December 12, 2005. Thus, to satisfy their burden on causation, Bishop and Henry must offer evidence, in addition to temporal proximity, of retaliatory motive.
[695]*695We conclude that Bishop and Henry-have offered sufficient evidence, in addition to temporal proximity, of retaliatory motive. First, six days after the Warden received the complaint, Lt. Richardson allegedly warned Bishop: “There is a lot of stuff going on around here, don’t get involved in it, remember you are on probation.” Second, Henry testified that, after the complaint was submitted, Lt. Richardson reminded her “a couple of times” that she “was on probation.” Third, Henry further testified that “[t]he atmosphere got worse” after the complaint was submitted in that the hostile looks from Lt. Richardson were “[m]ore frequent” and “even more hateful.” Fourth, some two weeks after the Warden received the complaint, Bishop’s probationary period was extended by the Warden under arguably suspicious circumstances in that no reason was given for the extension other than “due to your performance” and, at the time of her deposition, the Warden had no memory of taking such action against Bishop. Finally, Bishop and Henry’s two co-plaintiffs in this action, both of whom also signed the complaint to the Warden, testified that the hostility between them and Lt. Richardson intensified after the complaint was submitted. This testimony supports Bishop and Henry’s argument that Lt. Richardson harbored a grudge against those who submitted the complaint.
For all these reasons, we agree with the district court that, construing the evidence in the light most favorable to Bishop and Henry, they have raised a fact issue as to whether they were given negative performance evaluations and ultimately terminated in retaliation for their protected activity. Our conclusion is consistent with the principle that the burden of establishing a prima-faeie case in retaliation cases is “not onerous.” Taylor v. Geithner, 703 F.3d 328, 336 (6th Cir.2013); see also Dixon v. Gonzales, 481 F.3d 324, 333 (6th Cir.2007) (“The burden of proof at the prima facie stage is minimal; all the plaintiff must do is put forth some credible evidence that enables the court to deduce that there is a causal connection between the retaliatory action and the protected activity”).
C. Pretext
Bishop and Henry challenge the district court’s conclusion that they failed to present a fact issue as to whether ODRC’s stated reason for their terminations — performances issues, as reflected in their performance evaluations — was a mere pretext for unlawful retaliation. For the reasons that follow, we disagree with the district court and conclude that Bishop and Henry have presented sufficient evidence of pretext to survive summary judgment.
An employee may rebut the legitimate, nondiscriminatory reason offered by the employer in support of its decision to take an adverse employment action against an employee by making one or more of three showings:
(1) the employer’s stated reason for terminating the employee has no basis in fact, (2) the reason offered for terminating the employee was not the actual reason for the termination, or (3) the reason offered was insufficient to explain the employer’s action.
Imwalle v. Reliance Med. Prods., Inc., 515 F.3d 531, 545 (6th Cir.2008). Regardless of the method chosen, the employee retains the burden of producing sufficient evidence from which the jury could reasonably reject the employer’s explanation and infer intentional retaliation by the employer. See Johnson v. Kroger Co., 319 F.3d 858, 866 (6th Cir.2003). As the Supreme Court has recently clarified, an employee bringing a Title VII retaliation claim must [696]*696ultimately show “that the desire to retaliate was the but-for cause of the challenged employment action.” Univ. of Texas Sw. Med. Ctr. v. Nassar, — U.S. -, 133 S.Ct. 2517, 2521, 186 L.Ed.2d 503 (2013).
In the present case, Bishop and Henry proceed principally under the second method. Invoking the “cat’s-paw” or “rubber-stamp” theory of liability, Bishop and Henry argue that they were terminated in retaliation for their protected activity— submitting the joint complaint to the Warden complaining of discriminatory treatment by Lt. Richardson — and not due to their job performance. Their theory is that, although the Warden herself did not harbor any discriminatory animus toward them, the Warden was influenced by Lt. Richardson, who did harbor a grudge against Bishop and Henry as a result of their protected activity.
“When an adverse ... decision is made by a supervisor who lacks impermissible bias, but that supervisor was influenced by another individual who was motivated by such bias, this Court has held that the employer may be held liable under a ‘rubber-stamp’ or ‘cat’s paw’ theory of liability.” Arendale v. City of Memphis, 519 F.3d 587, 604 n. 13 (6th Cir.2008). See also Cobbins v. Tenn. Dep’t of Transp., 566 F.3d 582, 586 n. 5 (6th Cir.2009) (“The ‘cat’s paw1 theory refers to a situation in which a biased subordinate, who lacks de-cisionmaking power, influences the unbiased decisionmaker to make an adverse ... decision, thereby hiding the subordinate’s discriminatory intent.”); Staub v. Proctor Hosp., — U.S. -, 131 S.Ct. 1186, 179 L.Ed.2d 144 (2011) (applying cat’s-paw theory in the context of a discrimination claim brought under the Uniformed Services Employment and Reemployment Rights Act, 38 U.S.C. § 4301 et seq.).
To succeed on a cat’s-paw theory, the employee “must offer evidence of a ‘causal nexus’ between the ultimate decisionmaker’s decision to terminate the [employee] and the supervisor’s discriminatory animus.” Madden v. Chattanooga City Wide Serv. Dep’t, 549 F.3d 666, 677 (6th Cir.2008). In other words, the employee must show that, “[b]y relying on this discriminatory information flow, the ultimate deci-sionmakers acted as the conduit of the supervisor’s prejudice — his cat’s paw.” Id. at 678 (internal quotation marks omitted). However, a causal nexus is lacking if the ultimate decision “was based on an independent investigation” and the employee “presented no evidence that the supervisor’s discriminatory animus had influenced the decision.” Id.
Here, we have already determined that Bishop and Henry presented a fact issue as to whether Lt. Richardson acted with a retaliatory animus when she submitted the negative ten-month performance reviews. The analysis, then, focuses on whether Bishop and Henry have established the requisite “causal nexus” between the Warden’s termination decision and Lt. Richardson’s retaliatory animus.
It is undisputed that Lt. Richardson’s evaluation of Bishop and Henry played a role in the Warden’s decision to terminate them. The termination letters themselves state that termination was based on “performance evaluations,” and Lt. Richardson did author performance evaluations critical of both Bishop and Henry. In addition, the Warden admitted in her notarized position statement to the Ohio Civil Rights Commission dated March 3, 2006, that “Richardson provided input [into the termination decision] via her evaluation of the Charging Party’s work performance.” Because Lt. Richardson wrote the performance evaluations on which the Warden relied, and because Bishop and Henry have [697]*697presented a fact issue as to whether the evaluations were motivated by Lt. Richardson’s retaliatory amicus, the required “causal nexus” is established, unless the Warden also conducted an independent investigation severing the causal connection between Lt. Richardson’s evaluations and the Warden’s ultimate decision to terminate Bishop and Henry.
Adopting ODRC’s argument, the district court concluded that “the evidence is un-controverted” that “Warden Timmerman-Cooper engaged in an independent investigation and did not simply rubber stamp Lt. Richardson’s evaluation[s].” However, the district court did not explain the factual basis underlying this conclusion. Although ODRC urges us to accept the district court’s conclusion, it too does not point to any evidence in the record establishing that the Warden conducted an investigation independent of any input from Lt. Richardson. Upon review of the record, we conclude that, construing the evidence in the light most favorable to Bishop and Henry, the existence of genuine issues of material fact relating to the precise nature of the investigation that was conducted by the Warden, if any, render the district court’s grant of summary judgment improper.
We emphasize that there is no evidence in this case that the Warden conducted any independent investigation before terminating Bishop and Henry. Although the Warden testified that, “in general,” she asks questions of her command officers before removing an employee based on a negative performance evaluation, it is unclear whether she actually did so with regard to Bishop and Henry. While ODRC argues that “there is no evidence that [the Warden] did not engage in her ordinary practice of consulting her executive staff when removing [Bishop and Henry],” we note the absence of any evidence that she did, in fact, follow her general practice in this case. In fact, no command officer who was deposed in this matter remembers whether he or she spoke to the Warden about Bishop and Henry’s job performance prior to the Warden’s termination decision, and the Warden likewise does not remember whether she spoke to her command officers about Bishop and Henry prior to terminating them.
More importantly, however, even if the Warden did follow her general practice of consulting her command officers in the case of Bishop and Henry, the record is unclear as to the person or persons with whom she communicated. The Warden testified that she generally inquires with either the major or, in some cases, with the employee’s supervisor, who in this case was Lt. Richardson. In the Warden’s own words: “I can’t tell you [who I talked with in] specific cases. I mean, I have talked to supervisors before. I don’t know if I talked to them or just talked to the major here.” If the Warden did, in fact, consult with Lt. Richardson (or another command officer who had consulted Lt. Richardson) about Bishop and Henry’s performance evaluations — and neither Lt. Richardson nor the Warden could recall whether they consulted regarding the evaluations — that conversation would only serve to exacerbate, not purge, the alleged retaliatory taint.
ODRC argues that untainted input informed the Warden’s decision because the termination letter used the term “performance evaluations,” which ODRC claims encompasses more than just Lt. Richardson’s written performance evaluations. However, Bishop and Henry argue that the term is confined only to her written evaluations. We conclude that there is evidence in the record supporting both views. Bishop and Henry’s view is supported by the language of the termination letters, which states [698]*698that they were terminated based on their “performance evaluations,” without any reference at all to specific incidents or any discipline. Support for ODRC’s view is found in an affidavit submitted by the Warden in connection with this litigation, in which the Warden stated that “[a] poor performance evaluation can be demonstrated through ... the employee’s disciplinary history, or incident reports.” While the Warden’s self-serving explanation in her affidavit might be viewed as suspect because there is no pre-litigation evidence to substantiate it, see Shaw v. Danley, 202 F.3d 270, 2000 WL 64945, at *7 (6th Cir. Jan. 10, 2000) (“self-serving statement disclaiming discriminatory motive alone is not probative, but rather con-clusory”), we leave it to the jury to determine precisely what the Warden meant by the term “performance evaluation.” For the present purposes, construing the evidence in the light most favorable to Bishop and Henry, the non-moving parties, we accept as true their interpretation of the term. Thus, a reasonable jury could conclude that the Warden made her decision solely on Lt. Richardson’s allegedly biased input.
In sum, it is unclear on this record whether the Warden inquired further— beyond merely reading the Richardson evaluations — regarding the negative aspects of Bishop and Henry’s performance evaluations. And even if the Warden did inquire further, she might have consulted Lt. Richardson, the very person alleged to have poisoned the Warden’s decisionmak-ing process, or someone else who was influenced by Lt. Richardson. Construing the evidence in the light most favorable to Bishop and Henry, we conclude that there is a fact issue as to whether the Warden conducted an independent investigation and, if so, whether that investigation was sanitized of Lt. Richardson’s retaliatory bias.
Our conclusion is not altered by the fact that the Warden, at one time, had personal knowledge of the incidents purportedly underlying the negative performance evaluations, as illustrated by the fact that she ordered investigations into some of those incidents and approved disciplinary actions following those investigations. Any personal knowledge that the Warden may have had does not break the casual nexus between the Warden’s termination decisions and Lt. Richardson’s retaliatory animus because there is no evidence that the Warden remembered the incidents and connected them to Bishop and Henry at the time she made her ultimate decision to terminate Bishop and Henry. The Warden testified during her deposition that, as she sat there on that day, she did not remember Bishop and Henry.12 At the very least, the Warden’s complete unfamiliarity with Bishop and Henry would justify a jury in doubting whether she was independently familiar with the incidents purportedly underlying the negative performance evaluations at the time the termination decision was made and, ultimately, whether the Warden conducted any meaningful review, beyond reviewing the performance evaluations, before authorizing the terminations. Taking the facts in the light most favorable to Bishop and Henry, a reasonable jury could conclude that, had the Warden conducted a mean[699]*699ingful inquiry beyond merely reviewing the performance evaluations before making her termination decision, she would have at least remembered Bishop and Henry and the circumstances surrounding their terminations.13
For all these reasons, we conclude that Bishop and Henry have presented a genuine issue of material fact as to whether ODRC’s proffered legitimate reason for the terminations was a mere pretext for unlawful retaliation.
IV. CONCLUSION
For the reasons stated above, we reverse the judgment of the district court granting summary judgment in favor of ODRC and remand the case for further proceedings.