CPD Modules Available

Print this page

Defining Near Vision Behaviour: A New Tool for Practice

2M CPD in Australia | 1G in New Zealand | 30 April 2018


By Tim Thurn

A new test gives optometrists the potential to personalise the positioning of the near vision zone based on how the wearer makes use of the lens. This article examines the innovative test; the experimental validation of the test structure as reported in a poster at the AAO in 2016; and describes the measurements a practitioner can perform, highlighting the benefits for wearers.

LEARNING OBJECTIVES

1. Appreciate the specific relationships between posture and reading,

2. Understand how the use of motion capture technology can be used to record precise data on posture related to targets and tasks,

3. Understand how it is possible to use a pseudo reading task to assess direction of gaze; working distance; lateral offset; and visual behaviour without the need for the patient to use a near correction,

4. Understand how this data can be applied to personalise the near vision of a progressive lens.

 

Properly positioning the near vision zone on a progressive lens is critical to ensuring that the presbyopic wearer enjoys optimal comfort for near vision. If you know the wearer’s posture and behaviour while reading, it is possible to personalise the design of the progressive lens in accordance with their needs.

Physiology of Reading and Near Vision Zone Positioning

Reading is a recent activity in comparison to the duration of human evolution. While it is one of the most important near vision activities, it is also very demanding in terms of vision. As well as the ocular path, reading involves a whole body postural response, which also varies across readers.1,2,3 Every person will adopt a distinct body, head and eye posture and will explore the field of near vision with dynamics of gaze that are unique to that person.

The postural response for near vision is more complex than for distance vision as it involves not only the head, neck and trunk (Figure 1) but also the shoulders, wrists and elbows. Visually, all of this is governed by the intricate coordination of eye movements, including lowering of gaze, fixations and saccades, designed to bring the image of the words sequentially onto the fovea so the text can be read.

Figure 1. Postural response for near vision is involves the head, neck, trunk, shoulders, wrists and elbows.

To ensure comfortable vision, the near vision zone on a progressive lens must be optimally located at the very spot where the wearer directs his or her gaze and explores the lens during reading. That zone demands maximum visual acuity, and its dimensions must correspond to the eye’s exploration of the zone.

In the past, the near vision zone was positioned on progressive lenses based on average wearer behaviour or in accordance with optical parameters such as add power, ametropia and/or reading distance. Today, a new tool makes it possible to personalise the position of the near vision zone on each lens, based on the wearer’s exact behaviour in terms of reading posture and overall near vision behaviour.

Measuring the Wearer’s NEAR Vision Behaviour (NVB)

Knowing a wearer’s reading posture and gaze dynamics is certainly very helpful, but obtaining that information is no simple matter. The difficulty comes from measuring the wearer’s natural posture, that is, the posture that he or she would adopt without optical correction. However, it’s precisely that optical correction that presbyopes need in order to read and this can influence behaviour.

To solve that problem, the Essilor research team developed a special vision task that can be performed without correction, for near-vision ametropia ranging from -10.00D to +7.50D. The task, called pseudo reading, accurately reflects the wearer’s reading posture and consists of observing and tracking a large-size object – that is, one that does not require visual acuity. The object is blue against a white background and shown on a tablet. The movement of this target across the screen is similar to the average reading behaviour of an adult, with fixations of 233 milliseconds and saccades of 6.3 characters.5

In practice, wearers grasp the tablet, position it naturally in front of them as they would with a document they are reading, and follow the target’s horizontal movement line by line with their gaze. The measurement lasts about 18 seconds in all, depending on the length of the wearer’s fixations, and the target’s movement guides the wearer: he or she can predict the required saccade movements with the help of grey dots shown on the screen background, which anticipate the target’s movement (Figure 2).

Figure 2. Measuring near-vision behaviour

It should be noted that this measurement is not the wearer’s absolute behaviour, but rather his or her relative behaviour in response to the pseudo reading task. This measured behaviour correlates exactly with the wearer’s actual behaviour, as has been shown in validation studies for the measurement protocol conducted on a sample of presbyopes.6 The calculation is then used to establish the actual position of the wearer’s head and eyes.

In the course of this measurement, horizontal and vertical head movements are measured in real time in order to determine the wearer’s head posture and position of gaze at all times. From this, four critical data points are obtained:

The first three of these data points are:

  • The angle of down gaze,
  • The lateral offset, and
  • The reading distance.

These points describe the wearer’s posture, which is ultimately measured by the average posture during the pseudo reading task, known as the near vision behaviour (NVB) Point (Figure 3).

Figure 3. The wearer’s posture datapoints: lowered angle of gaze, lateral offset, reading distance. The near-vision behaviour point

The fourth data point is referred to as:

  • The wearer’s NVB Ratio

This point describes how the wearer adjusts his or her gaze vertically during the entire measurement process:

  • For wearers who have a strong tendency to lower their eyes whenever they move to the next line, and therefore change their head and body posture or the tablet position only slightly; then the ratio will be close to zero,
  • When wearers maintain a static position of gaze, i.e., they have a strong tendency to change their posture and/or the position of the tablet vertically during the pseudo reading task, then the ratio is close to one (Figure 4).

Figure 4. The wearer’s behaviour datapoint. The near-vision behaviour ratio. 

Establishing and Validating the Pseudo Reading Task

While it is interesting to observe the concepts behind this new measurement, it is also important to assess its validity for use in clinical practice and how it correlates with actual near/reading tasks.

Establishing the concept for the pseudo reading task started from experiments conducted at MOVIS – Motion Capture and Vision Science Laboratory at Essilor’s research centre (Figure 5). Motion capture is commonly used as a cinematic technique to allow 3D modelling of movement for either animatronics or 3D computer rendering in films, advertising etc. Using the same technology, Essilor researchers attached motion capture sensors to their research subjects and with the aid of infrared cameras, recorded their movements in real time and in 3D. In addition, as the data is generated in real time, the research modelling was also possible in real time, allowing for dynamic interaction within the experiments.

Figure 5. Images generated during experiments at MOVIS

Real life scenarios were used for the experiments that covered:

Twenty-two subjects, all presbyopes but corrected with contact lenses,

  • Observation of three different postures: standing, seated and lying down,
  • Seven target objects: smartphone, e-reader, tablet computer, A4 sheet of paper, laptop computer, large screen for a desktop computer and interaction with another person,
  • Seven different tasks including texting, watching video on a tablet, reading an e-reader and A4 sheet etc.

Data gathered from these experiments was very precise and showed that:

  • Each task had an average position,
  • The variations of distance, offset and down gaze equated to a specific ‘volume’ of visual range (Figure 6),
  • Each wearer handled each object differently giving rise to their own specific volume,
  • Reading on an A4 sheet can be used as a reference point, as the changes made when using hand-held objects are proportional.

 

Figure 6. Variations of distance, offset and down gaze equate to a specific volume of visual range.

This final point is important because it also helps to define the difference between wearers. According to Proudlock’s3 and Hartig’s8 research, during a reading task on an A4 sheet of paper, eye movements accounted for most shifts of gaze (close to 70 per cent) in the horizontal direction. However, vertical eye movements accounted for close to two thirds of the shifts of gaze and head movement for one third of the shifts of gaze.7,8 The point being: while each wearer has his or her own behaviour or ‘volume of vision’ for a series of tasks, most of the difference can be seen in the vertical direction (Figures 7 and 8).

Figure 7. Graphic demonstrates different volume for two wearers.

Figure 8. Graphic demonstrates different visual behaviour for two wearers. 

All of this data and knowledge was applied to engineering the pseudo reading task based on the following criteria:

  1. No correction would be required as both the patient’s desire for clarity and the prismatic effect of the spectacle lenses would affect their natural posture.
  2. The stimulus had to be large enough and clear enough, even with low acuity, but in the presence of binocular vision.
  3. Reading necessitates eye movements to place the text sequentially on the fovea. These are not smooth movements, but rather a series of saccades and fixations. The pseudo task had to replicate the eye’s saccades and fixations without the use of text.5
  4. The next position of the stimulus had to be highly predictable, allowing for voluntary saccades as in a real reading task.9

Additionally, the pattern of movement had to be as close as possible to normal reading. The mean duration of fixations and the saccades size were defined based on the data obtained by several researchers and compiled by Rayner (Table 1, below).

The definition of the pseudo reading task was now set as:

  • To follow a dot which shifts across the screen of a tablet,
  • The wearer is sitting on a chair without armrests,
  • They wear a frame without lenses,
  • A positioning clip with markers is set on the frame. 

The following reference measurements are taken:

  • The eyes’ centre of rotation relative to the clip,
  • Distance vision direction of gaze.

The following near vision measurements are taken:

  • The 3D position of the clip is continuously recorded using the tablet’s in-built camera,
  • Direction of down gaze,
  • Near vision task distance.

Stimulus characteristics noted as:

  • A blue dot large enough to be seen without optical correction moves on the screen similarly to normal reading eye movements.5
  • Display duration between 215 and 250ms
  • Distance between successive positions, between 5.8 and 6.8 characters computed on a mean size of 2mm per character.
  • A pattern of grey dots indicates the successive positions of the dot, ensuring the next position of the dot is easily predictable.9 

Validation of the Pseudo Reading Task

With the stimulus and the task protocol defined, it was time for validation. An experimental method was set up to test the following hypothesis:

“Postural data obtained by the pseudo reading method can predict the postural parameters adopted by a wearer during a real reading.”

Method

Twenty-eight subjects participated in randomised crossover study. Their preliminary data – including acuities, binocular vision, Rx etc. – were measured before they were randomised into two groups; A and B. Both groups had their distance reference position measured using the IVS Visoffice2.

Group A then performed a series of three actual reading tasks on a tablet using their Rx, while Group B performed three pseudo reading tasks without Rx.

Their distance reference position was taken again and the tasks were swapped.

Figure 9. Results from a randomised crossover study were highly correlated.

The results, (Figure 9) were highly correlated, leading the research team to conclude:

  • Both for reading distances and direction of down gaze, the data is strongly correlated between real reading and pseudo reading,
  • Even if the data is not strictly equivalent, one can predict the posture the wearer would have adopted during real reading based on pseudo reading values,
  • Even through the wearer’s vision is not corrected during the measurement, the pseudo reading task allows the optometrist to infer the real near vision posture data,
  • This tool may be useful in daily practice.

Pseudo Reading Testing and Application to Progressive Lens Design

In practice, to create a personalised lens, the wearer’s near vision behaviour needs to be measured. This can be done using a tablet, which can be connected to an electronic measuring column or used on its own. The measurement is taken as follows:

  1. The wearer’s frame is positioned in a clip that is used to define the frame’s position in space and, by extension, the wearer’s posture and head movements (shown in Figure 2),
  2. A baseline far vision position is measured in primary position of gaze, using either a measuring column (such as a Visioffice) or a tablet, by taking two photos – a front picture and a three-quarter profile. An algorithm in the app then calculates the primary position of gaze,
  3. The measuring process is demonstrated and explained to the wearer, so that he or she understands the task to be accomplished,
  4. The wearer holds the tablet and fixates on the blue dot in the centre until the clip is detected by the camera; then the movement of the stimulus is activated and the position of gaze and movements are continuously recorded (Figure 10).
  5. The measurement is validated and the near vision posture and behaviour data points are saved.

Figure 10. Measuring the wearer’s near-vision behaviour during a pseudo-reading task.

Once the measurements have been taken, the data used to personalise the lens must be forwarded. The data is sent using a seven-digit alphanumeric code that combines two pieces of information:

  1. The NVB Point, which is the wearer’s average position of gaze during the measurement, representing the wearer’s reading posture;
  2. The NVB Ratio, which is the distribution of measurements around the NVB Point, representing the wearer’s dynamic near vision behaviour.

The lens can then be personalised in a three-step process:

  1. Use the wearer’s data, (prescriptions, pupillary distance, position of the eye’s centre of rotation), and the conditions in which the lenses are worn, (shape and size of the frame, back vertex distance, pantoscopic tilt and wrap angle), combined with the characteristics of the lenses to be produced (front surface, geometry and refractive index),
  2. Identify the optimal position of the near vision zone on the progressive lens, based on the wearer’s posture (as indicated by the NVB Point). Information on ametropia, prismatic effects and binocular vision is taken into account during this stage of the process,
  3. Enhance the progression profile based on the wearer’s gaze dynamics, in light of the NVB Ratio. The objective is to adjust the size and shape of the progressive’s near vision zone in accordance with the wearer’s vertical exploration of that zone.

Figure 11 is an example of a lens with the optimal near-vision zone position and shape based on a wearer’s near vision behaviour. We see how, as a result of this test, the position of the near vision zone on the lens (the point where the add power is at 100 per cent, shown by a blue cross) has been modified, and the size of the so-called arm’s-length vision zone (from 85 per cent to 60 per cent of the add power) has been adjusted.

Figure 11. The optimal near-vision zone position and shape, based on a wearer’s near-vision behaviour.

The exact values for progression length and inset for the near vision zone can be defined and forwarded to the optometrist once the full lens calculation has been performed.

Mapping Near Vision Behaviour

To demonstrate these findings in simple graphic form, Essilor’s research and development specialists designed a way to map the wearer’s results on a graph of possible behaviours (Figure 12).

Figure 12. The optimal near-vision zone position and shape, based on a wearer’s near-vision behaviour.

On this graph:

  • The horizontal axis shows the wearer’s average posture while reading, expressed as the angle of down gaze (from 12 to 30 degrees),
  • The vertical axis shows near-vision behaviour, i.e., the dispersion of the direction of gaze (between zero and one).

Thus, a wearer who adopts a sharp downward gaze while reading and primarily uses his eyes to explore his near vision vertically, will fall at the bottom-right portion of the graph. By contrast, a wearer who lowers his or her eyes only slightly to read, and primarily changes posture or moves the tablet while reading, will fall at the upper left of the graph. Every kind of behaviour between these two extremes can be located on the graph.

The mapping process includes a colour code; there is a significant effect on the optical design of the lens only if the colour codes for two measurements can be differentiated by the naked eye. This offers an immediate way to verify that the measurements are reproducible.

Having performed multiple measurements on numerous presbyopes, the Essilor researchers were able to show that each wearer’s behaviour is reproducible and represents an appropriate data point to use for lens customisation, since it is both specific to each individual and differentiating.

Once the measurement has been taken, the practice provides Essilor with the alphanumeric code for the wearer so that a corresponding custom designed lens can be manufactured. Two codes that are very different may represent two very similar vision behaviours; conversely, two codes that are similar may correspond to very different behaviour patterns. Whatever the case, each code contains all the information needed to manufacture a lens that corresponds very accurately to the wearer’s needs.  

    Tim Thurn B Optom (UNSW); GDip (Photo) (SA School of Art, Adelaide University); GCertB (American University, Paris) is Director of Professional Services for Essilor Australia and New Zealand. A UNSW graduate, he joined Essilor in 1988 having worked in private practice for eight years. From 1991–1998, he worked in Essilor’s International Strategic Marketing Department in Paris, where he obtained his Graduate Certificate in Business from American University. Tim Thurn is a qualified optometrist and the Director of Professional Services for Essilor Australia and New Zealand.


This article was edited from three sources:

1. Escalier G., Perrin JL., Heslouis M., Poulain I., Jolivet V., Rousseau B., Lebrun C., Varilux X Series Lenses - Near Vision behaviour personalization, Points de Vue, International Review of Ophthalmic Optics, online publication, May 2017
2. Rousseau B., Meslin D., Varilux X series: the progressive lens with an expanded field of near vision, Points de Vue, International Review of Ophthalmic Optics, www.pointsdevue.com, July2017
3. Essilor R & D Presentation to US University and College Symposium 2017
 
References
1. Kowler, E., Pizlo, Z., Zhu, G., Erkelens, C. J., Steinman, R. M., & Collewijn, H. (1992). Coordination of head and eyes during the performance of natural (and unnatural) visual tasks. The head-neck sensory motor system, 419-426.
2. Lee, C. (1999). Eye and head coordination in reading: roles of head movement and cognitive control. Vision Research, 39, 3761-3768.
3. Proudlock, F. A., Shekhar, H., & Gottlob, I. (2003). Coordination of eye and head movements during
reading. Investigative Ophthalmology & Visual Science, 44, 2991-2998.
4. Proudlock, F. A. & Gottlob, I. (2007). Physiology and pathology of eye-head coordination. Progress in retinal and eye research, 26, 486-515.
5. Rayner K. Eye movements in reading and information processing: 20 years of research, Psychological Bulletin, 124, 372-422 (1998).
6. Poulain I, Perrin JL, Escalier G, A tool for measuring reading posture with no need for visual correction, Optometry & Visual Science, 93: e-abstract
165190 (2016).
7. Proudlock, F. A., Shekhar, H., & Gottlob, I. (2003). Coordination of eye and head movements during
reading. Investigative Ophthalmology & Visual Science, 44, 2991-2998
8. Hartig, A., Gowen, E., Charman, W. N., Radhakrishnan, H. (2011), Working distance and eye and head movements during near work in myopes and non-myopes, Clinical and Experimental Optometry ,94.6 , November 2011, 536-544
9. Walker, R., Walker, D. G., Husain, M., & Kennard, C. (2000). Control of voluntary and reflexive saccades. Experimental brain research, 130, 540-544.

' Every person will adopt a distinct body, head and eye posture and will explore the field of near vision with dynamics of gaze that are unique to that person '