While you are tracking the softball as it travels toward home plate you decide that it is time to swing the bat…
While you are driving along a road the car ahead of you slows down and, suddenly, you decide that it is critical to step on your brakes…
While your doctor views your internal organs on a video monitor during your gall bladder surgery he or she decides just how far to move the fiberoptic scope so that your artery does not get punctured…
Performance in all of these situations relies upon visual perception. By studying performance in these real-world situations, we can better understand the mechanisms that underlie visual perception. By better understanding visual perception, we can improve performance in these real-world situations. This is the essence of human factors psychology: the integration of basic science with applications to the real world. This has been the essence of my research at Texas Tech University. The following sections summarize various research projects that have been conducted in my laboratory in collaboration with outstanding graduate students including Eston Betts, Jennifer Blume, Gregory Liddell, Les Meyer, Kerstan Mork, Tammy Ott, and Anand Tharanathan.
Visual Perception of Collision.
Your decision of when to swing the bat at the ball is a perceptual judgment about when the ball will hit the bat, known in the literature as a judgment of time to contact (TTC). In the three-dimensional world, TTC is computed by dividing distance by velocity. However, it has been shown that TTC is available in the two-dimensional pattern of light that reaches the eye or optic flow. This quantity is known as tau defined as the ratio of an object’s optical size to its rate of optical expansion (see figure below). Tau-based models of performance have been highly influential and prior results suggest that people can use tau in various tasks. Such models raise doubts about traditional (cognitive) theories of depth perception because they do not require mental processes or pictorial depth cues. However, my research demonstrates that depth cues and cognitive processes can influence TTC judgments despite the presence of tau. For example, observers reported that a large far (computer-generated) approaching object would reach them before a small near object that was specified to arrive sooner by tau. Judgments were consistent with the depth cue of relative size rather than with tau. My findings have been recognized as a “challenge to the autonomy of tau.” My work has implications for the design of cockpit displays and has been funded by NASA-Ames Research Center.
Perceptual Factors in Driving and Implications for Models of Space Perception.
Your decision of when to step on the brakes of your car is a judgment of when you would collide with the car ahead—a judgment of time-to-contact. Rear-end collisions account for 25% of all accidents. To avoid such collisions, drivers must detect that the car ahead of them is decelerating. The driver may rely on brake lights. Alternatively, before the brake lights illuminate, the driver may notice the car slowing down due to its growth in the visual field (i.e., optical expansion). Our research using driving simulations (see figure below) indicates that the type of information that drivers use depends on how far ahead the lead car is located. This suggests that the processes that underlie space perception depend on distance. Recently, I incorporated this idea into a conceptual model of space perception which is based on recent neuropsychological and neuroimaging studies. According to this model the perception of very far distances is served by cognitive processes and the perception of vary near distances is served by non-cognitive (direct) processes. The model also incorporates task parameters and the presence and nature of motion, resulting in a multidimensional model of space perceptions.
Perceptual Factors in Minimally-Invasive Surgery.
Your doctor’s ability to perform surgery safely depends on his or her ability to perceive the relative depth between the gall bladder and the surrounding arteries and tissues. The performance of surgery with the use of fiberoptic cameras has brought great benefits to patients, including less injury and faster recovery. However, it makes the job more difficult for surgeons who have to rely on fewer depth cues and a narrower field of view. In short, it is more difficult for the surgeon to move the surgical instruments through the anatomical spaces. Our research aims to apply what we know about depth perception to develop surgical displays. For example, using a surgical simulator, we examined whether three separate views of space—a top view, frontal view, and side view (thus providing more depth information than a single display), would improve perceptual-motor performance (see figure below). Our results show that observers relied upon the top view and did not utilize the other views. In addition, the side view was detrimental to performance. Our most recent work in this areas suggests that the multiple displays that surgeons use in the already cluttered operating room can be integrated into one split-screen display without a detrimental effect on performance.
Visual Memory for Moving Scenes.
As you drive, you take your eyes off the car ahead to view other parts of the traffic scene. While you do so you presumably keep track of the car ahead in memory. If you remember the lead car as being farther away than it really is, you may follow it too closely and cause a rear end collision. My research shows that observers’ memories for moving three-dimensional scenes are distorted in systematic ways. These results extended prior studies of visual memory which were limited to static displays. The implication is that there is a common mechanism that underlies memory for static and moving scenes. Furthermore, my results show that optic flow information due to (simulated) self motion does not eliminate memory distortions obtained with static scenes. This research has important implications for models of scene memory and for transportation safety.
Visual Illusions.
Some theories have argued that visual illusions occur only with static, impoverished laboratory displays. My research shows that illusions can occur with three-dimensional objects and moving observers and can even affect judgments about collisions. Therefore, illusions can have adverse consequences for human safety and it becomes important to predict the occurrence of illusions. This requires theory. I developed a method to evaluate how well several different theories predict the Mueller-Lyer illusion (Shown in figure below; line “a” is equal in length to line “b”, but “a” appears longer). Unlike prior studies which evaluated directional predictions, I evaluated quantitative predictions, which are more critical to evaluations of theory (Meehl, 1967). My results demonstrated difficulties for perspective and centroid theories of the illusion.
Patient Wait Time and Patient Satisfaction.
Satisfaction with healthcare has a direct impact on a patient’s compliance with medical treatment and willingness to attend follow-up visits with healthcare providers. At the request of a local cancer treatment center, we conducted a study on how long it took patients to complete their visit at the center, and how satisfied they were with their wait time. Unlike other studies of patient wait times which asked patients how long they waited, we measured each individual patient’s wait time with a stop watch. In addition, we measured wait times for each component of the visit (reception desk, triage, clinic exam room, etc). Our results replicated prior studies which showed that longer wait times were associated with lower ratings of satisfaction with wait times. New in our study was that longer wait times were also associated with satisfaction with other aspects of the visit. Most important, the longer that patients waited for the doctor in the clinic examination room, the less satisfied they were with the time they spent with the doctor. The implication is that as doctors are pressured to spend less and less time with their patients, they should minimize the time that patients have to wait to see them.
Performance in Nursing.
In collaboration with the Transdisciplinary Research in Patient Safety Team (TRIPS), a joint effort of TTU and TTUHSC, I have developed a line of research on performance in nursing. The Institute of Medicine has identified the work environment of nursing as critical in the delivery of safe patient care. Nurses spend the most time with patients compared with all other health care providers. Their performance has a direct impact on the quality of health care that patients receive and thus patient outcomes. The nurse’s critical task is patient surveillance—monitoring changes in the patient’s status to prevent declines in health. However, due to the design of the work environment, nurses spend only about 25% of their time in direct contact with patients. In addition, they engage in multiple tasks at once, shift their attention from one patient to another frequently, and rely heavily on prospective and procedural memory. In short, they have heavy cognitive loads. Yet, they are frequently interrupted which degrades cognitive performance. There is a pressing need to redesign the work environment of nurses to increase efficiency so that they can spend more time with patients. Our research team hopes to contribute to improve the work environment of nurses so that patient safety can be enhanced.