delivered the opinion of the court:
Plaintiffs, Sandra Childred Mtengule (Sandra) and Basil G. Mten-gule (Basil), brought a personal injury action against defendant, City of Chicago (City), for injuries arising from an automobile accident. Plaintiffs alleged that the City failed to maintain the traffic control *324signal at the site of the accident in a reasonably safe manner, and that the failure to do so was a proximate cause of their injuries.1 The case was tried before a jury, which returned a general verdict in favor of the City. The trial court entered judgment on the verdict, and plaintiffs now appeal. We reverse the judgment of the trial court and remand for a new trial.
Plaintiffs allege on their appeal that (1) the jury’s answer to the special interrogatory concerning notice was against the manifest weight of the evidence; (2) the trial court erred in granting defendant’s motion in limine to exclude opinion testimony of Sandra’s treating psychiatrist; (3) the trial court erred in submitting the special interrogatory concerning proximate cause to the jury; (4) the trial court erred in allowing the testimony of a witness disclosed on the last day of trial; and (5) the trial court erred in prohibiting evidence of a subsequent similar accident.
We find that the jury’s answer to the special interrogatory concerning notice was against the manifest weight of the evidence and that the trial court erred in excluding the testimony of Sandra’s treating psychiatrist. Our findings as to these issues provide sufficient cause to remand this case for a new trial. Accordingly, we do not address the remaining issues raised on this appeal.
On March 19, 1982, Sandra was riding in the front passenger seat of a car being driven by Mark Graves. At approximately 8:30 p.m., as they drove westbound on Madison Street in Chicago, they approached the intersection of Madison and Wells Streets. The intersection was controlled by a traffic signal which was green as they entered the intersection. Meanwhile, a car driven by Deborah Dohsi headed northbound on Wells Street. Dohsi thought she had a green light. (The evidence at trial showed there were four traffic signals facing the northbound traffic on Wells Street. Three signals showed red and one showed green.) Both cars proceeded into the intersection. The car driven by Graves struck the car driven by Dohsi. The impact of the collision was described by witnesses as "hard,” and caused Dohsi’s car to spin 180 degrees. Upon impact, Sandra struck her head on the dashboard of Graves’ car. Shortly thereafter a Chicago police officer arrived at the scene. The officer discovered that one of the traffic signals, which should have been facing westbound traffic on Madison Street, was twisted so that it faced the northbound traffic on Wells Street. The signal was stuck in the green position.
After the accident, Sandra began to develop various physical *325disabilities, including memory loss, difficulty in holding objects and a speech disorder. Eventually, Sandra was diagnosed as suffering from a conversion reaction disorder. A conversion reaction disorder is manifested by physical symptoms, but lacks an organic cause. Rather, psychological causation is likely and the onset of the disorder can usually be traced to the occurrence of a traumatic event.
Immediately after the accident Sandra did not seek or accept any medical attention. The next day, however, Sandra went to the emergency room of Michael Reese Hospital complaining of pain in her neck, head and leg. The pain in her neck was located in the area of a previously diagnosed lipoma. A lipoma is a benign, fatty mass of tissue. Sandra was examined, released and told to follow up with her family doctor. On March 25, 1982, she returned to Michael Reese Hospital for treatment.
On March 26, Dr. Michael Goldman examined her, and told her that the lump in her neck (the previously diagnosed lipoma) should be removed in order to determine if the lump was cancerous. On April 8, Dr. Goldman removed the lump and determined it to be benign. In August of 1982, Sandra began treatment with Dr. Hawkins, a psychiatrist. Dr. Hawkins concluded that Sandra’s physical disabilities were the result of a conversion reaction disorder brought on by the automobile accident on March 19. Although the City does not dispute that Sandra suffers from a conversion reaction disorder, it contends that the disorder was triggered by the surgical removal of the lipoma and that the surgery was unrelated to the car accident.
In order for a plaintiff to prove negligence, he or she must first establish that a duty of reasonable care was owed by the defendant to the plaintiff. (Marshall v. City of Centralia (1991), 143 Ill. 2d 1, 6, 570 N.E.2d 315, 318.) Section 3 — 102 of the Local Governmental and Governmental Employees Tort Immunity Act speaks to when a public entity owes a duty and provides:
"(a) Except as otherwise provided in this Article, a local public entity has the duty to exercise ordinary care to maintain its property in a reasonably safe condition for the use in the exercise of ordinary care of people whom the entity intended and permitted to use the property in a manner in which and at such times as it was reasonably foreseeable that it would be used, and shall not be liable for injury unless it is proven that it has actual or constructive notice of the existence of such a condition that is not reasonably safe in a reasonably adequate time prior to an injury to have taken *326 measures to remedy or protect against such condition.” (Emphasis added.) (745 ILCS 10/3 — 102(a) (West 1992).2)
Thus, if the City did not have actual or constructive notice of the twisted traffic signal in adequate time to take precautions against it, the City owed no duty to the plaintiffs.
The jury was given special interrogatory No. 1, which addressed the issue of notice and read:
"Did the City of Chicago have actual or constructive notice of the twisted traffic signal in sufficient time prior to the March 19, 1982 car accident to have taken measures to remedy or protect against it?”
The jury answered "no” to this special interrogatory, thereby concluding that the City did not have sufficient notice and owed no duty to the plaintiffs. Plaintiffs contend that this answer was error. We note that a jury’s answer to a special interrogatory may be disturbed only if it is against the manifest weight of the evidence. (Buford v. Chicago Housing Authority (1985), 131 Ill. App. 3d 235, 249, 476 N.E.2d 427, 438.) After reviewing the evidence produced at trial, we find that the jury’s answer to the special interrogatory was against the manifest weight of the evidence.
Initially, we note that special interrogatory No. 1 may require two findings. The first finding concerns notice, actual or constructive. If the jury makes an affirmative finding of notice, it must then consider whether the notice was had in sufficient time to have allowed the City to have remedied or guarded against the twisted signal. After reviewing the record, we find no evidence on which to disturb the *327jury’s finding as to actual notice. We address the issue of constructive notice below.
Thomas Garvy testified for the plaintiffs concerning the issue of notice. Garvy was an ironworker who worked on a construction site located at the intersection of Madison and Wells Streets. He worked at this site for six months preceding and including the day of the accident. He testified that he saw the traffic signal twisted in the wrong direction at 9 a.m. on the day of the accident. He further testified that he saw a Chicago police officer directing traffic at the intersection of Madison and Wells Streets also at 9 a.m. on the day of the accident. The officer, according to Garvy, directed traffic throughout the day. Garvy stated that he had never seen an officer directing traffic at the intersection before that day.
According to Garvy’s testimony, he left work at 4:30 p.m. and went to Marshall Field’s to shop and meet a friend. He testified that he left Marshall Field’s at about 6:30 p.m. or 7:30 p.m. From there, Garvy said, he stopped at a restaurant near Madison and Wells for a "half-hour or more.” He then testified that he was heading to the train station when he saw the accident at issue at 8:30 p.m. Lastly, Garvy stated that the traffic signal remained in the same twisted position as he had seen it earlier at 9 a.m. Garvy’s testimony taken alone establishes that the traffic signal remained facing the wrong direction for at least 11½ hours and that a Chicago police officer was present at the intersection directing traffic throughout the day.
The City did not present any evidence to controvert Garvy’s testimony. Rather, the City sought at trial to impeach Garvy’s credibility on several fronts. In its brief, the City maintains that the jury was justified in not accepting Garvy’s testimony and that the jury’s answer to special interrogatory No. 1 must stand.
At trial, the City tried to impeach Garvy’s credibility with the testimony of Elwood Stolle, the officer in charge of the Loop traffic division. Stolle testified that it was the division’s practice to have a police officer or traffic aide direct traffic at intersections located at construction sites every weekday from 10 a.m. to 6 p.m. The City contends that the jury was justified in disbelieving Garvy because he testified that he had never seen an officer at the intersection before the day of the accident, despite the City’s practice of assigning an officer or aide to intersections located near construction sites. There was no testimony, however, that whoever was assigned was actually present at the intersection prior to the date of the accident. More importantly, the officer’s testimony does not contradict Garvy’s testimony that he saw the traffic signal at 9 a.m. on the day of the accident and that he saw a Chicago police officer at the intersection throughout the day.
*328The City then cites Garvy’s account of the accident as the main reason for the jury’s refusal to believe his testimony. Garvy testified that Dohsi’s car hit Graves’ car and that it was Graves’ car that was spun around by the collision. Both drivers, however, testified that it was the other way around: that Graves’ car hit Dohsi’s car, and Dohsi’s car was spun around. Garvy also testified as to the number of people in each car. He stated that two people rode in Graves’ car and three to four middle-aged people rode in Dohsi’s car. The evidence, however, showed that Graves’ car had three people and Dohsi’s car had two college students. Here, the inference the City wishes to draw is that since Garvy had confused the two cars, his testimony as to seeing the twisted traffic signal is not to be believed. Garvy’s somewhat confused testimony as to the accident, however, casts no doubt whatsoever on his testimony as to the twisted traffic signal or on his testimony as to the presence of the police officer.
The City also sought to impeach Garvy’s testimony with the testimony of Holly Tattson, an employee of Marshall Field’s. Tattson testified that Marshall Field’s closed at 5:45 p.m. on the day of the accident. Garvy testified that he left the store at 6:30 p.m. or 7:30 p.m. on that day. Here, defendant wishes to make the inference that since Garvy was wrong as to the time he left Marshall Field’s, he was wrong about seeing the twisted traffic signal and the police officer. This argument suffers from the same fatal flaw as the two previous ones. Tattson’s testimony does not contradict Garvy’s testimony that at 9 a.m. on the day of the accident he saw the twisted traffic signal and a Chicago police officer directing traffic throughout the day.
Finally, the City tried to impeach Garvy with his deposition testimony. At trial, Garvy stated that the twisted signal was hanging from the El platform. At his deposition, however, Garvy testified that the twisted signal was on a pole. Further, at trial he testified that he "couldn’t tell what colors were what.” But at his deposition, he stated that the twisted signal showed the same color sequence as the other signals facing the northbound traffic on Wells. Despite these inconsistencies, Garvy’s testimony that he saw the twisted traffic signal at 9 a.m. and a Chicago police officer directing traffic throughout the day on the day of the accident remains uncontro-verted.
A lapse of time may be sufficient cause to impute constructive notice upon a public entity. (Baker v. City of Granite City (1979), 75 Ill. App. 3d 157, 160, 394 N.E.2d 33, 35; Palermo v. City of Chicago Heights (1971), 2 Ill. App. 3d 1004, 1008, 276 N.E.2d 470, 473.) Where a condition has existed for such a length of time that the public entity might have known of the condition by the exercise of reason*329able care and diligence, constructive notice exists. (Baker, 75 Ill. App. 3d at 160, 394 N.E.2d at 35; Palermo, 2 Ill. App. 3d at 1008, 276 N.E.2d at 473.) Among the factors to be considered are the length of time the condition existed and the conspicuity of the condition. (Palermo, 2 Ill. App. 3d at 1008, 276 N.E.2d at 473.) Additionally, constructive notice will be imputed on a public entity where one of its employees had actual knowledge of the defect. Duewel v. Lahman (1981), 103 Ill. App. 3d 220, 227, 430 N.E.2d 662, 667.
On the issue of constructive notice Livings v. City of Chicago proves instructive. The court there found:
"The undisputed and uncontradicted evidence is that the deteriorated portion of the sidewalk was approximately 2 to 3 feet in width, 3 feet in length, and 21h inches deep. The testimony of a disinterested third party, based on her own observations, indicated that the defective condition, located a short distance from her residence, had existed for 1 or 2 years prior to the accident. There was no contrary evidence for the jury to consider.” ( Livings v. City of Chicago (1975), 26 Ill. App. 3d 850, 854, 326 N.E.2d 170, 173-74.)
The court went on to hold that where the facts of a case are undisputed and allow for only one reasonable inference, the issue of constructive notice may be determined as a matter of law. ( Livings, 26 Ill. App. 3d at 854-55, 326 N.E.2d at 174.) Under the facts in Livings, the court upheld the trial court’s ruling that the defendant had constructive notice as a matter of law. Livings, 26 Ill. App. 3d at 854-55, 326 N.E.2d at 174.
Similarly, in the case before us plaintiffs presented the uncontroverted testimony of a disinterested third party based on his own observations. Garvy’s testimony established that the twisted traffic signal, located adjacent to his work site, existed for llx/2 hours prior to the accident. There was no contrary evidence for the jury to consider.
While we are not willing to hold that the City had constructive notice as a matter of law, we hold that the jury’s finding was against the manifest weight of the evidence. Applying the relevant factors of time and conspicuity to the uncontroverted evidence, it is clear the traffic signal remained twisted for a sufficient amount of time and in a sufficiently conspicuous location to impute constructive notice to the City. Garvy’s uncontradicted testimony established that the traffic signal remained twisted for llx/2 hours prior to the accident. Furthermore, one can hardly imagine a more conspicuous location for a twisted traffic signal than the intersection of Madison and Wells in the heart of downtown Chicago.
*330Additionally, we find that the jury’s answer to special interrogatory No. 1 was against the manifest weight of the evidence because the uncontradicted testimony of Garvy established that a City employee had actual knowledge of the twisted signal. In Duewel v. Lahman, the court refused to overturn a jury’s finding of constructive notice where a stop sign was missing from a busy county highway, used frequently by county vehicles, and where the stop sign was missing for 81/2 hours prior to the accident. (Duewel v. Lahman (1981), 103 Ill. App. 3d 220, 430 N.E.2d 662.) Constructive notice was imputed to the defendant public entity because of actual knowledge by one of its employees. (Duewel, 103 Ill. App. 3d at 227, 430 N.E.2d at 667.) Actual knowledge was established by a witness who testified that a county worker mowed the grass along the shoulder of the road at the intersection where and when the stop sign was missing. Duewel, 103 Ill. App. 3d at 227, 430 N.E.2d at 668.
Likewise, in the present case, there was uncontradicted testimony establishing the presence of a city employee directing traffic at the very intersection where and when the traffic signal was twisted. Moreover, the City, itself, buttressed this testimony when it put into evidence its practice of assigning someone to direct traffic at the intersection.
The question remains whether the City had notice of the twisted traffic signal in sufficient time to take proper precautions. The court in Duewel found that the "sufficient time” necessary to take precautions against a missing stop sign was negligible considering the time needed to replace or warn against the missing stop sign. (Duewel, 103 Ill. App. 3d at 228, 430 N.E.2d at 668.) In the instant case, we also believe that the "sufficient time” was negligible. Very little time would have been needed to station an officer at the intersection to direct traffic until the signal could be repaired. In fact, the evidence indicates that there was an officer at the intersection for most of the day. Therefore, we hold that the jury’s answer to special interrogatory No. 1 is against the manifest weight of the evidence.
Plaintiffs next argue that opinion testimony of their main expert was improperly excluded. In support of their theory that the automobile accident triggered Sandra’s conversion reaction disorder, plaintiffs called Sandra’s treating psychiatrist, Dr. David Hawkins, to testify. Dr. Hawkins testified that it was his opinion that Sandra’s conversion reaction disorder was brought on as a result of the car accident.
The City, however, advanced the theory that Sandra’s disorder could be traced to her surgery. Moreover, the City contended that *331her surgery was unrelated to the accident, and moved in limine to preclude Dr. Hawkins from testifying as to the reason for the surgery. At his deposition, Dr. Hawkins stated that it was his opinion that the lipoma was surgically removed because of irritation resulting when Sandra’s cervical collar, worn as result of the accident, rubbed against her neck. The trial court granted the City’s motion, ruling that Dr. Hawkins’ testimony concerning the reason for the surgery was not relevant. We hold that it was reversible error to exclude this testimony.
The proper standard to apply when reviewing a trial court’s ruling on a motion in limine to exclude testimony is whether the trial court abused its discretion, and if so, whether the ruling unfairly prejudices a party or materially affects the outcome of the trial. (Congregation of the Passion, Holy Cross Province v. Touche Ross & Co. (1991), 224 Ill. App. 3d 559, 578, 586 N.E.2d 600, 613.) Here, the trial court clearly abused its discretion by excluding relevant testimony. Additionally, the exclusion of Dr. Hawkins’ testimony materially affected the outcome of the case.
Testimony is relevant if it tends to make a material fact at issue at trial more or less probable than it would be without the testimony. (Lundquist v. Nickels (1992), 238 Ill. App. 3d 410, 427-28, 605 N.E.2d 1373, 1385.) Dr. Hawkins’ proposed testimony meets this test. His proposed testimony tends to establish that the car accident made the surgery necessary. Thus, even if the surgery triggered the disorder as the City argued, it would not operate as an intervening cause. Instead, the surgery would merely be a link in the causal connection between the car accident and the conversion reaction disorder. Plainly, Dr. Hawkins’ proposed testimony was relevant on the issue of proximate cause, and the trial court’s ruling was an abuse of discretion.
That the exclusion of Dr. Hawkins’ testimony materially affected the outcome of the case is demonstrated by the close balance of evidence put on at trial concerning the issue of proximate cause. Moreover, proximate cause is an element necessary to plaintiffs’ case.
The issue of proximate cause was put squarely before the jury in the form of special interrogatory No. 3. Special interrogatory No. 3 read as follows:
"Was the car accident of March 19, 1982 a proximate cause of Sandra Mtengule’s conversion reaction disorder?”
The jury answered, "No.” The jury, however, did not hear relevant testimony from Dr. Hawkins concerning the causal relationship between the automobile accident and the surgery. We are, therefore, faced with a situation where relevant testimony regarding an issue *332both closely balanced at trial and necessary to the plaintiffs’ case was excluded. Had the jury heard Dr. Hawkins’ proposed testimony, it may very well have decided differently on the issue of proximate cause. Clearly, then, the exclusion of Dr. Hawkins’ testimony materially affected the outcome of this case.
For the reasons stated herein, we reverse the judgment of the trial court and remand for a new trial.
Reversed and remanded.
TULLY, P.J., and CERDA, J., concur.