Reference Information for Human Performance Capabilities

The following information is provided as a reference for human performance capabilities for different sensory and motor modes. It is not intended to be all inclusive, but instead it is intended to provide information that may be useful when considering how to design medical devices to be accessible for users with disabilities.

Capabilities Related to Visual Display Modes: Gaze and Visuomotor Integration

While vision involves spatiotemporal sensing via the retina, high clarity is only possible in a small region of about one degree of total arc from central fixation of the retinal field that maps to the fovea (where there is a higher sensor density of especially cone receptors); the drop-off is such that by 2 degrees of eccentricity, acuity has dropped in half and by 5 degrees, to one-third; the rest of the field of view, the so-called “peripheral vision,” is best viewed as sensing “shades of gray” with lower clarity (e.g., under 10% clarity when over 20 degrees). The illusion of being able to see the entire field of view is due to eye tracking control subsystems that rotate the eyes and then shift the retinal field, plus the use of some specialized optical subsystems to adjust light intensity (pupil neurocontrol system) and to focus light (accommodation neurocontrol system, where small muscles can mildly adjust the shape of the lens for near distance). “Gaze” is also a function of head orientation, and eye and head neurocontrol systems are tightly coordinated (Zangemeister and Stark, 1981); gaze angles (relative to the body axis system) of under 20 degrees tend to be made exclusively by eye movements, with larger gaze orientations involving a progressively greater proportion of gaze being attributed to head rotation. There are four classic types of eye tracking movements, each driven by different locations within the brain that use the same eye muscles:

• Saccadic eye movements : fast, near time-optimal movements of only 20-60 ms duration, which are normally voluntary, involve a reaction time latency of 200 ms during target tracking,
• Vestibulo-ocular reflex (VOR): where as the head turns the eyes automatically move in the opposite direction, driven by a 3-neuron, roughly 12-ms "arc" that initiates in the semicircular canals and normally can stabilize self-rotations of up to at least 100 deg/sec,
• Smooth pursuit eye movements: a velocity feedback system that enables predictive tracking of moderate-speed objects of up to about 30 deg/sec of retina slip, with about 95% of individuals unable to make smooth pursuit movements without a visual stimulus, and
• Vergence eye: convergence/divergence “depth” eye movements that work in conjunction with the accommodation (lens) system to help foveate an image as it goes between near and far.

Any of these six distinct neuromotor systems (four tracking, pupil, and accommodation) can be limited or dysfunctional, and some also diminish in effectiveness with age. Eye tracking movements are often viewed as a “window to the brain” for disease assessments (Jones and Stark, 1983). For example, persons with schizophrenia often presents abnormal saccadic trajectories (Stark, 1983), while persons with aphasia tend to use suboptimal strategies to sample images by focusing on contrast (e.g., edges of objects) rather than context (e.g., features such as eyes and mouth during conversations). The common approach of superimposing spatial eye locations on the image being viewed is often called a scan path.

Some age- and medication-related visual changes that occur (Bailey, 1996):

• Visual acuity is relatively poor for young children, improves as a person reaches young adulthood, slightly decreases from about the middle twenties to fifties, and has a more accelerated decline after that.
• For older adults, pupil size tends to shrink and it opens and closes slower, which causes a chronic reduction in the amount of light that reaches the retina so that vision becomes worse with lower illumination.
• Humor (jellylike fluid in inner eye) can become cloudy, which increases problems associated with glare because light is scattered in the humor so that visual contrast and sensitivity are reduced.
• Some commonly prescribed drugs affect vision, and the source of such effects may be sensory (e.g., retinal, visual cortical), and/or of a visuomotor origin (tracking movements, pupillary reflex, lens accommodation). Often sensory-based dysfunction is more sustained, while motor-based effects are shorter-term while systems adjust. Green and Spencer (1969) list about 200 commonly prescribed drugs that affect vision.

In terms of medical device use, the saccadic system tends to be the most used of these six visuomotor subsystems. Normally an individual can make up to three voluntary saccadic “jumps” per second (30-60 milliseconds for the movement, then at least 200 milliseconds to sample the image). In environments where self-movement is common, the VOR-saccadic-vergence-accommodation systems must work together, while in environments with movement of the device (or within a display, such as a graph) the smooth pursuit and saccadic systems work together. In environments where illumination varies, of note is that the pupil can change its diameter by a factor of about 5 (e.g., from 1.5 mm in bright light to about 8 mm in dim light), suggesting a pupil area change (and thus intensity adjustment) of about a factor of 25, which is impressive. While mild size adjustments can occur within a second, significant dark-to-light or light-to-dark environmental transitions can require seconds or even tens of seconds. Many diseases affect pupillary response, as do many drugs, including some over-the-counter drugs and alcohol. General guidance (in the form of rules of thumb) for device designers is included below.

Vision During Self-Movement

While humans often excel at keeping “an eye on a monitor” while engaged in self-movement activities, largely because of the VOR capability, it becomes very likely that the first saccadic eye movement made toward a monitor by a multi-tasking individual with a primary focus of attention elsewhere will not foveate at the desired monitor location for sampling information, but rather it will take 2-3 saccades to reach on the desired spatial location on a display. This implies that it will likely take over one-half second before visual processing can start to capture the desired information content.

1. Peripheral Vision Contrast Cues. When visual contrast provides position and orientation cues for peripheral vision, such as monitor contrast edges, it is more likely that 1-2 saccadic eye movements will be sufficient to attain the targeted location on a display. Such contrast also benefits individuals with certain types of visual impairments. An example is the recent trend towards black monitor frames, with illuminated monitor screens.
2. Accommodating for Distance Changes. Self-movement implies frequent changes in distance between the user and the display, which requires visuomotor responses from the accommodation and vergence systems. Thus it may take several seconds before the individual can be fully adjusted and view a visual display with optimum clarity. For time-critical information with self-movement, font sizes should be larger and contrast significant so as to not depend on this fine-tuning of visual acuity.
3. Recognizing VOR Limitations Due to Impact or Vibration. For visual displays on hand-held mobile devices, a normally functioning VOR system accommodates for most self-movements, but there are exceptions, such as the moment of heel strike during walking or during impact while using a hand tool; if this is an expected component of a task, the designer should assume temporarily poor visual acuity and use larger fonts, higher contrast, etc.
4. Placing Labels to Minimize Obstruction During Hand Operation of Control. Normally this implies providing a label above a control.

Illumination, Contrast, Colors and Characters

While illumination and contrast are often similar issues for visual acuity in a stationary setting, for dynamically changing environments, they are very different. Humans can function in dramatically different illumination because of the pupillary reflex system changing the size of the pupil; but this does take time, and for some persons the range of pupil size variation is small. Because the pupillary system enables humans to function in both low and high illumination, it is easy for a designer to forget the distinction between light intensity and contrast. Similarly, a significant portion of the population has congenital color blindness, especially for green-red. Designers should be aware of the need to provide reasonable accommodations for these realities.

1. Illumination. It is commonly noted that there should be adequate illumination, but this guidance goes beyond that point. Medical displays that are expected to be used with varying environmental illumination may consider use of monitors that enable the user to adjust the level of base illumination on the display so as to better match the environment, thus lessening possible time-critical delays related to pupil size adjustment.
2. Contrast. Contrast provides important visual cuing, especially for many persons with visual acuity limitations. While it is recognized that this may not be as aesthetically desirable, designs with a color-coordinated set of similar shades of gray or gray-blue is to be avoided if intended users of a device may have a degree of visual limitation. Rather, gray-level contrast that orients the user to the device should be built into the design. When a product permits a user to adjust color and contrast settings, a range of color selections capable of producing a variety of contrast levels should be provided (based on Section 508, §1194.25g).
3. Surface Texture as a Visual Cue. While the main purpose of changes in texture is tactile, the visuomotor system can reinforce tactile sensing.

Capabilities Related to Auditory Communication Modes: Hearing and Speech

There is no question that using the mode of hearing for both consumer products with embedded computing and medical devices in likely to grow as more “intelligent” conversational interfaces emerge. Sound can be characterized by its frequency and magnitude, both of which change over time. The human ear responds to frequencies from about 20-20,000 Hz, with the high end of this range decreasing to about 10,000 Hz by age 65. A person with normal hearing is most sensitive to frequencies in the region from about 1000-4000 Hz, but readily hears radio music that ranges from about 50 Hz to 8,000 Hz, with the full pitch range that is possible by humans going from about 100 Hz (lowest pitch, males) to perhaps 8,000 Hz (shrill sound). An adult male speaks with an average pitch of about 120 Hz, with an adult female at about 250 Hz. Noise-canceling microphones are widely available that can be designed to reject acoustic signals outside of a certain range of interest, such as simple speech. Audio compression technologies such as used for cell phones typically pass digital signals at about 8,000 bits/sec which implies a top frequency of 4,000 Hz that in reality is much more sophisticated in that it takes into account speech patterns.

Sound magnitude is normally given in decibels. Normal conversation is around 60-70 dB, live rock music is about 100 dB, and painful sound is about 140 dB (suggesting rock music is not painful); the useful range for medical devices is about 50-80 dB. Without use of visual cues such as lips, the “signal” of interest normally needs to be at least 6 dB greater than environmental noise, and preferably more than this so as to increase access (Note that hearing sensitivity varies greatly among individuals, by as much as 20 dB, and tends to decrease with age.

About 19% of working-age Americans have hearing limitations, most of which are mild. Importantly, the reason for hearing loss can be due to conduction through stages of the transduction process within the ear or due to nerve deafness, and both are common [Bailey, 1996 - Figure 3-12]. It is important that during usability testing phases, the intended users conducting evaluations be exposed to environments with typical background environmental noise.